1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.406 2006/10/10 15:36:50 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
85 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
86 address A is aliased to B and C, where B exists and C does not. Without
87 -v the output is "A verified" because verification stops after a
88 successful redirection if more than one address is generated. However,
89 with -v the child addresses are also verified. Exim was outputting "A
90 failed to verify" and then showing the successful verification for C,
91 with its parentage. It now outputs "B failed to verify", showing B's
92 parentage before showing the successful verification of C.
94 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
95 look up a TXT record in a specific list after matching in a combined
98 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
99 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
100 they consult the DNS. I had assumed they would set it the way they
101 wanted; and indeed my experiments on Linux seem to show that in some
102 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
103 To be on the safe side, however, I have now made the interface to
104 host_find_byname() similar to host_find_bydns(), with an argument
105 containing the DNS resolver options. The host_find_byname() function now
106 sets these options at its start, just as host_find_bydns() does. The smtp
107 transport options dns_qualify_single and dns_search_parents are passed to
108 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
109 of host_find_byname() use the default settings of RES_DEFNAMES
110 (qualify_single) but not RES_DNSRCH (search_parents).
112 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
113 spool_read_header() do less string testing, by means of a preliminary
114 switch on the second character of optional "-foo" lines. (This is
115 overdue, caused by the large number of possibilities that now exist.
116 Originally there were few.) While I was there, I also converted the
117 str(n)cmp tests so they don't re-test the leading "-" and the first
118 character, in the hope this might squeeze out yet more improvement.
120 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
121 flag allowing group syntax was set by the header_syntax check but not
122 turned off, possible causing trouble later; (2) The flag was not being
123 set at all for the header_verify test, causing "group"-style headers to
124 be rejected. I have now set it in this case, and also caused header_
125 verify to ignore an empty address taken from a group. While doing this, I
126 came across some other cases where the code for allowing group syntax
127 while scanning a header line wasn't quite right (mostly, not resetting
128 the flag correctly in the right place). These bugs could have caused
129 trouble for malformed header lines. I hope it is now all correct.
135 SC/01 Use a glob alias rather than an array ref in eximstats generated
136 parser. This improves both readability and performance.
138 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
139 Don't display local sender or destination tables in eximstats unless
140 there is data to show.
141 Added average volumes into the eximstats top table text output.
143 SC/03 Collect data on the number of addresses (recipients) as well
144 as the number of messages in eximstats.
146 TF/01 Correct an error in the documentation for the redirect router. Exim
147 does (usually) call initgroups() when daemonizing.
149 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
150 with consistent privilege compared to when running as a daemon.
152 TF/03 Note in the spec that $authenticated_id is not set for local
153 submissions from trusted users.
155 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
156 Thanks to Dean Brooks <dean@iglou.com> for the patch.
158 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
159 by adding some example configuration directives to the default
160 configuration file. A little bit of work is required to uncomment the
161 directives and define how usernames and passwords are checked, but
162 there is now a framework to start from.
164 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
165 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
166 without this. I don't know how relevant this is to other LDAP libraries.
168 PH/02 Add the verb name to the "unknown ACL verb" error.
170 PH/03 Magnus Holmgren's patch for filter_prepend_home.
172 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
174 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
175 directory not expanded when it should be if an expanded home directory
176 was set for the address (which is overridden by the transport).
178 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
181 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
182 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
183 because it is too late at that time, and has no effect.
185 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
186 security issue with \' (bugzilla #107). I could not use the
187 PQescapeStringConn() function, because it needs a PGconn value as one of
190 PH/08 When testing addresses using -bt, indicate those final addresses that
191 are duplicates that would not cause an additional delivery. At least one
192 person was confused, thinking that -bt output corresponded to deliveries.
193 (Suppressing duplicates isn't a good idea as you lose the information
194 about possibly different redirections that led to the duplicates.)
196 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
197 systems where poll() doesn't work, in particular OS X.
199 PH/10 Added more information to debugging output for retry time not reached.
201 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
202 operations in malware.c.
204 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
207 PH/13 If write_rejectlog was set false when logging was sent to syslog with
208 syslog_duplication set false, log lines that would normally be written
209 both the the main log and to the reject log were not written to syslog at
212 PH/14 In the default configuration, change the use of "message" in ACL warn
213 statements to "add_header".
215 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
216 not followed by a command (e.g. "seen endif").
218 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
219 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
222 PH/17 Added extra conditions to the default value of delay_warning_condition
226 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
227 { match{$h_precedence:}{(?i)bulk|list|junk} } \
228 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
231 The Auto-Submitted: and various List- headers are standardised, whereas I
232 don't think Precedence: ever was.
234 PH/18 Refactored debugging code in route_finduser() to show more information,
235 in particular, the error code if getpwnam() issues one.
237 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
238 This is apparently needed in addition to the PH/07 change above to avoid
239 any possible encoding problems.
241 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
242 but not after initializing Perl.
244 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
245 output them only if debugging. By default they are written stderr,
246 apparently, which is not desirable.
248 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
251 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
254 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
256 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
257 authenticated or an ident call has been made. Suppress the default
258 values for $authenticated_id and $authenticated_sender (but permit -oMai
259 and -oMas) when testing with -bh.
261 PH/24 Re-jigged the order of the tests in the default configuration so that the
262 tests for valid domains and recipients precede the DNS black list and CSA
263 tests, on the grounds that those ones are more expensive.
265 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
266 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
267 This bug exists in every version of Exim that I still have, right back to
270 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
271 However, an attempt to turn on case-sensitivity in a regex key by
272 including (?-i) didn't work because the subject string was already
273 lowercased, and the effects were non-intuitive. It turns out that a
274 one-line patch can be used to allow (?-i) to work as expected.
280 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
281 other effects) broke the use of negated acl sub-conditions.
283 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
286 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
287 "Deny" causes Exim to reject the incoming connection with a 554 error.
288 Unfortunately, if there is a major crisis, such as a disk failure,
289 tcp-wrappers gives "deny", whereas what one would like would be some
290 kind of temporary error. A kludge has been added to help with this.
291 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
292 554 error is used if errno is still zero or contains ENOENT (which occurs
293 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
296 PH/03 Add -lutil to the default FreeBSD LIBS setting.
298 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
299 errors. Otherwise a message that provokes a temporary error (when other
300 messages do not) can cause a whole host to time out.
302 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
303 addresses were routed directly to files or pipes from a redirect router.
304 File deliveries just didn't batch; pipe deliveries might have suffered
307 PH/06 A failure to get a lock for a hints database would erroneously always say
308 "Failed to get write lock", even when it was really a read lock.
310 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
311 of 0600. This has been changed to use the value of the lockfile_mode
312 option (which defaults to 0600).
314 PH/08 Applied small patch from the Sieve maintainer.
316 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
317 folder from quota calculations, a direct delivery into this folder messed
318 up the contents of the maildirsize file. This was because the regex was
319 used only to exclude .Trash (or whatever) when the size of the mailbox
320 was calculated. There was no check that a delivery was happening into an
321 excluded directory. This bug has been fixed by ignoring all quota
322 processing for deliveries into excluded directories.
324 PH/10 Added the maildirfolder_create_regex option to appendfile.
330 PH/01 The code for finding all the local interface addresses on a FreeBSD
331 system running IPv6 was broken. This may well have applied to all BSD
332 systems, as well as to others that have similar system calls. The broken
333 code found IPv4 interfaces correctly, but gave incorrect values for the
334 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
335 that it would not match correctly against @[] and not recognize the IPv6
338 PH/02 The ipliteral router was not recognizing addresses of the form user@
339 [ipv6:....] because it didn't know about the "ipv6:" prefix.
341 PH/03 Added disable_ipv6.
343 PH/04 Changed $reply_address to use the raw form of the headers instead of the
344 decoded form, because it is most often used to construct To: headers
345 lines in autoreplies, and the decoded form may well be syntactically
346 invalid. However, $reply_address has leading white space removed, and all
347 newlines turned into spaces so that the autoreply transport does not
350 PH/05 If group was specified without a user on a router, and no group or user
351 was specified on a transport, the group from the router was ignored.
353 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
354 for visible compile-time settings that can be used to change these
355 numbers, for those that want even more. Backwards compatibility with old
356 spool files has been maintained. However, going back to a previous Exim
357 release will lost any variables that are in spool files.
359 PH/07 Two small changes when running in the test harness: increase delay when
360 passing a TCP/IP connection to a new process, in case the original
361 process has to generate a bounce, and remove special handling of
362 127.0.0.2 (sic), which is no longer necessary.
364 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
365 be the same on different OS.
367 PH/09 Moved a debug statement in filter processing to avoid a race problem when
370 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
371 whether --show-vars was specified or not
373 JJ/02 exipick: Added support for new ACL variable spool format introduced
376 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
377 syntactically invalid From: or Reply-to: line, and a filter used this to
378 generate an autoreply, and therefore failed to obtain an address for the
379 autoreply, Exim could try to deliver to a non-existent relative file
380 name, causing unrelated and misleading errors. What now happens is that
381 it logs this as a hard delivery error, but does not attempt to create a
384 PH/11 The exinext utility has a -C option for testing purposes, but although
385 the given file was scanned by exinext itself; it wasn't being passed on
386 when Exim was called.
388 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
389 an end-of-file indication when reading a command response.
391 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
392 compiled. In many other places in Exim, IPv6 addresses are always
393 recognized, so I have changed this. It also means that IPv4 domain
394 literals of the form [IPV4:n.n.n.n] are now always recognized.
396 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
397 used if the router is not running as root, for example, when verifying at
398 ACL time, or when using -bh. The debugging output from this situation was
399 non-existent - all you got was a failure to exec. I have made two
402 (a) Failures to set uid/gid, the current directory, or a process leader
403 in a subprocess such as that created by queryprogram now generate
404 suitable debugging ouput when -d is set.
406 (b) The queryprogram router detects when it is not running as root,
407 outputs suitable debugging information if -d is set, and then runs
408 the subprocess without attempting to change uid/gid.
410 PH/15 Minor change to Makefile for building test_host (undocumented testing
413 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
414 additional section of a DNS packet that returns MX or SRV records.
415 Instead, it always explicitly searches for A/AAAA records. This avoids
416 major problems that occur when a DNS server includes only records of one
417 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
418 fixed another bug: if SRV records were looked up and the corresponding
419 address records were *not* found in the additional section, the port
420 values from the SRV records were lost.
422 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
423 using the correct key (the original address) when searching the retry
424 rules in order to find which one to use for generating the retry hint.
426 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
427 adding the default one. Similarly, if it contains a Reply-To: header, the
428 errors_reply_to option, if set, is not used.
430 PH/19 When calculating a retry time, Exim used to measure the "time since
431 failure" by looking at the "first failed" field in the retry record. Now
432 it does not use this if it is later than than the arrival time of the
433 message. Instead it uses the arrival time. This makes for better
434 behaviour in cases where some deliveries succeed, thus re-setting the
435 "first failed" field. An example is a quota failure for a huge message
436 when small messages continue to be delivered. Without this change, the
437 "time since failure" will always be short, possible causing more frequent
438 delivery attempts for the huge message than are intended.
439 [Note: This change was subsequently modified - see PH/04 for 4.62.]
441 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
442 $1, $2, $3) because the numerical variables can be reset during some
443 expansion items (e.g. "match"), thereby losing the authentication data.
445 PH/21 Make -bV show the size of off_t variables so that the test suite can
446 decide whether to run tests for quotas > 2G.
448 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
449 mailbox_size, and mailbox_filecount in the appendfile transport. If a
450 filecount value is greater than 2G or if a quota value is greater than 2G
451 on a system where the size of off_t is not greater than 4, a panic error
454 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
455 never match. The debug and -bh output now contains an explicit error
456 message indicating a malformed IPv4 address or mask.
458 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
459 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
462 PH/25 Do not write to syslog when running in the test harness. The only
463 occasion when this arises is a failure to open the main or panic logs
464 (for which there is an explicit test).
466 PH/26 Added the /no_tell option to "control=freeze".
468 PH/27 If a host name lookup failed very early in a connection, for example, if
469 the IP address matched host_lookup and the reverse lookup yielded a name
470 that did not have a forward lookup, an error message of the form "no IP
471 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
472 could be logged. Now it outputs the IP address instead of "NULL".
474 PH/28 An enabling patch from MH: add new function child_open_exim2() which
475 allows the sender and the authenticated sender to be set when
476 submitting a message from within Exim. Since child_open_exim() is
477 documented for local_scan(), the new function should be too.
479 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
480 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
481 results in an empty string is now treated as unset.
483 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
485 PH/31 Added sender_verify_fail logging option.
487 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
488 needed by RFC 822 but not by RFC 2822 was commented out. I have now
489 tidied the source and removed it altogether.
491 PH/33 When a queue run was abandoned because the load average was too high, a
492 log line was always written; now it is written only if the queue_run log
493 selector is set. In addition, the log line for abandonment now contains
494 information about the queue run such as the pid. This is always present
495 in "start" and "stop" lines but was omitted from the "abandon" line.
497 PH/34 Omit spaces between a header name and the colon in the error message that
498 is given when verify = headers_syntax fails (if there are lots of them,
499 the message gets confusing).
501 PH/35 Change the default for dns_check_names_pattern to allow slashes within
502 names, as there are now some PTR records that contain slashes. This check
503 is only to protect against broken name servers that fall over on strange
504 characters, so the fact that it applies to all lookups doesn't matter.
506 PH/36 Now that the new test suite is complete, we can remove some of the
507 special code in Exim that was needed for the old test suite. For example,
508 sorting DNS records because real resolvers return them in an arbitrary
509 order. The new test suite's fake resolver always returns records in the
512 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
513 bounces) except when queue_only is set, to avoid logging races between
514 the different processes.
516 PH/38 Panic-die if .include specifies a non-absolute path.
518 PH/39 A tweak to the "H" retry rule from its user.
520 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
521 a label. They prevented compilation on older perls.
523 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
524 a warning to be raised on newish perls.
526 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
527 on queue. Changes to match documented behaviour of showing count of
528 messages matching specified criteria.
530 PH/40 Changed the default ident timeout from 30s to 5s.
532 PH/41 Added support for the use of login_cap features, on those BSD systems
533 that have them, for controlling the resources used by pipe deliveries.
535 PH/42 The content-scanning code uses fopen() to create files in which to put
536 message data. Previously it was not paying any attention to the mode of
537 the files. Exim runs with umask(0) because the rest of the code creates
538 files with open(), and sets the required mode explicitly. Thus, these
539 files were ending up world-writeable. This was not a big issue, because,
540 being within the spool directory, they were not world-accessible. I have
541 created a function called modefopen, which takes an additional mode
542 argument. It sets umask(777), creates the file, chmods it to the required
543 mode, then resets the umask. All the relevant calls to fopen() in the
544 content scanning code have been changed to use this function.
546 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
547 to 24 hours. This avoids potential overflow problems when processing G
548 and H retry rules. I suspect nobody ever tinkers with this value.
550 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
552 PH/45 When the plaintext authenticator is running as a client, the server's
553 challenges are checked to ensure they are valid base64 strings. By
554 default, the authentication attempt is cancelled if an invalid string is
555 received. Setting client_ignore_invalid_base64 true ignores these errors.
556 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
557 they are received. Thus, the responses can be made to depend on the
558 challenges. If an invalid string is ignored, an empty string is placed in
561 PH/46 Messages that are created by the autoreply transport now contains a
562 References: header, in accordance with RFCs 2822 and 3834.
564 PH/47 Added authenticated_sender_force to the smtp transport.
566 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
568 PH/49 Installed latest patch from the Sieve maintainer.
570 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
571 was also set, the appendfile transport was unnecessarily scanning a
572 directory of message files (e.g. for maildir delivery) to find the count
573 of files (along with the size), even though it did not need this
574 information. It now does the scan only if it needs to find either the
575 size of the count of files.
577 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
579 PH/52 Two bugs concerned with error handling when the smtp transport is
582 (i) Exim was not creating retry information for temporary errors given
583 for individual recipients after the DATA command when the smtp transport
584 was used in LMTP mode. This meant that they could be retried too
585 frequently, and not timed out correctly.
587 (ii) Exim was setting the flag that allows error details to be returned
588 for LMTP errors on RCPT commands, but not for LMTP errors for individual
589 recipients that were returned after the DATA command.
591 PH/53 This is related to PH/52, but is more general: for any failing address,
592 when detailed error information was permitted to be returned to the
593 sender, but the error was temporary, then after the final timeout, only
594 "retry timeout exceeded" was returned. Now it returns the full error as
595 well as "retry timeout exceeded".
597 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
598 do this, and (what is worse) MTAs that accept it.
600 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
601 will now be deprecated.
603 PH/56 New os.c-cygwin from the Cygwin maintainer.
605 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
606 formats (previously only available in exim formats via -bpr, -bpru,
607 and -bpra. Now also available in native and exiqgrep formats)
609 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
610 with very large, slow to parse queues
612 JJ/08 exipick: added ! as generic prefix to negate any criteria format
614 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
616 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
617 responses to authentication challenges, though it was showing the
618 challenges; (ii) I've removed the CR characters from the debug output for
621 PH/58 Allow for the insertion of a newline as well as a space when a string
622 is turned into more than one encoded-word during RFC 2047 encoding. The
623 Sieve code now uses this.
625 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
626 data_4xx, lost_connection, tls_required.
628 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
629 message was being sent as an SMTP response.
631 PH/61 Add -l and -k options to exicyclog.
633 PH/62 When verifying, if an address was redirected to one new address, so that
634 verification continued, and the new address failed or deferred after
635 having set something in $address_data, the value of $address_data was not
636 passed back to the ACL. This was different to the case when no
637 redirection occurred. The value is now passed back in both cases.
639 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
640 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
641 use login_cap.h, so on its own it isn't the distinguishing feature. The
642 new name refers directly to the setclassresources() function.
644 PH/65 Added configuration files for NetBSD3.
646 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
648 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
649 is preferred over IPv4.
651 PH/68 The bounce_return_message and bounce_return_body options were not being
652 honoured for bounces generated during the reception of non-SMTP messages.
653 In particular, this applied to messages rejected by the ACL. This bug has
654 been fixed. However, if bounce_return_message is true and bounce_return_
655 body is false, the headers that are returned for a non-SMTP message
656 include only those that have been read before the error was detected.
657 (In the case of an ACL rejection, they have all been read.)
659 PH/69 The HTML version of the specification is now built in a directory called
660 spec_html instead of spec.html, because the latter looks like a path with
661 a MIME-type, and this confuses some software.
663 PH/70 Catch two compiler warnings in sieve.c.
665 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
666 function verify_get_ident() calls ip_connect() to connect a socket, but
667 if the "connect()" function timed out, ip_connect() used to close the
668 socket. However, verify_get_ident() also closes the socket later, and in
669 between Exim writes to the log, which may get opened at this point. When
670 the socket was closed in ip_connect(), the log could get the same file
671 descriptor number as the socket. This naturally causes chaos. The fix is
672 not to close the socket in ip_connect(); the socket should be closed by
673 the function that creates it. There was only one place in the code where
674 this was missing, in the iplookup router, which I don't think anybody now
675 uses, but I've fixed it anyway.
677 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
678 well as to direct DNS lookups. Otherwise the handling of names in host
679 lists is inconsistent and therefore confusing.
685 PH/01 Two changes to the default runtime configuration:
687 (1) Move the checks for relay_from_hosts and authenticated clients from
688 after to before the (commented out) DNS black list checks.
690 (2) Add control=submission to the relay_from_hosts and authenticated
691 clients checks, on the grounds that messages accepted by these
692 statements are most likely to be submissions.
694 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
696 (1) Generate an error if the third argument for the ${prvs expansion is
699 (2) Treat a missing third argument of ${prvscheck as if it were an empty
702 (3) Reset the variables that are obtained from the first argument of
703 ${prvscheck and used in the second argument before leaving the code,
704 because their memory is reclaimed, so using them afterwards may do
707 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
708 one (it's much easier than Tom thought :-).
710 (5) Because of (4), we can now allow for the use of $prvscheck_result
711 inside the third argument.
713 PH/03 For some reason, the default setting of PATH when running a command from
714 a pipe transport was just "/usr/bin". I have changed it to
717 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
718 anything to be listed in the output from -bV.
720 PH/05 When a filter generated an autoreply, the entire To: header line was
721 quoted in the delivery log line, like this:
723 => >A.N.Other <ano@some.domain> <original@ddress> ...
725 This has been changed so that it extracts the operative address. There
726 may be more than one such address. If so, they are comma-separated, like
729 => >ano@some.domain,ona@other.domain <original@ddress> ...
731 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
732 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
733 not being looked up in the rDNS to get a host name, Exim was showing the
734 IP address twice in Received: lines, even though the IP addresses were
735 identical. For example:
737 Received: from [1.2.3.4] (helo=[1.2.3.4])
739 However, if the real host name was known, it was omitting the HELO data
740 if it matched the actual IP address. This has been tidied up so that it
741 doesn't show the same IP address twice.
743 PH/07 When both +timestamp and +memory debugging was on, the value given by
744 $tod_xxx expansions could be wrong, because the tod_stamp() function was
745 called by the debug printing, thereby overwriting the timestamp buffer.
746 Debugging no longer uses the tod_stamp() function when +timestamp is set.
748 PH/08 When the original message was included in an autoreply transport, it
749 always said "this is a copy of the message, including all the headers",
750 even if body_only or headers_only was set. It now gives an appropriate
753 PH/09 Applied a patch from the Sieve maintainer which:
755 o fixes some comments
756 o adds the (disabled) notify extension core
757 o adds some debug output for the result of if/elsif tests
758 o points to the current vacation draft in the documentation
759 and documents the missing references header update
763 o fixes a bug in processing the envelope test (when testing
764 multiple envelope elements, the last element determinted the
767 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
768 Electronic Mail") by including:
770 Auto-submitted: auto-generated
772 in the messages that it generates (bounce messages and others, such as
773 warnings). In the case of bounce messages for non-SMTP mesages, there was
774 also a typo: it was using "Auto_submitted" (underscore instead of
775 hyphen). Since every message generated by Exim is necessarily in response
776 to another message, thes have all been changed to:
778 Auto-Submitted: auto-replied
780 in accordance with these statements in the RFC:
782 The auto-replied keyword:
784 - SHOULD be used on messages sent in direct response to another
785 message by an automatic process,
787 - MUST NOT be used on manually-generated messages,
789 - MAY be used on Delivery Status Notifications (DSNs) and Message
790 Disposition Notifications (MDNs),
792 - MUST NOT be used on messages generated by automatic or periodic
793 processes, except for messages which are automatic responses to
796 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
797 to the default Received: header definition.
799 PH/12 Added log selector acl_warn_skipped (default on).
801 PH/13 After a successful wildlsearch lookup, discard the values of numeric
802 variables because (a) they are in the wrong storage pool and (b) even if
803 they were copied, it wouldn't work properly because of the caching.
805 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
806 checking when decoding. Apparently there are clients that generate
807 overlong encoded strings. Why am I not surprised?
809 PH/15 If the first argument of "${if match_address" was not empty, but did not
810 contain an "@" character, Exim crashed. Now it writes a panic log message
811 and treats the condition as false.
813 PH/16 In autoreply, treat an empty string for "once" the same as unset.
815 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
816 extension "envelope-auth". The code is finished and in agreement with
817 other implementations, but there is no documentation so far and in fact,
818 nobody wrote the draft yet. This extension is currently #undef'ed, thus
819 not changing the active code.
821 Print executed "if" and "elsif" statements when debugging is used. This
822 helps a great deal to understand what a filter does.
824 Document more things not specified clearly in RFC3028. I had all this
825 sorted out, when out of a sudden new issues came to my mind. Oops."
827 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
830 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
831 canonical form (as documented). However, after a host name lookup from
832 the IP address, check_host() was doing a simple string comparison with
833 addresses acquired from the DNS when checking that the found name did
834 have the original IP as one of its addresses. Since any found IPv6
835 addresses are likely to be in abbreviated form, the comparison could
836 fail. Luckily, there already exists a function for doing the comparison
837 by converting both addresses to binary, so now that is used instead of
840 PH/20 There was another similar case to PH/19, when a complete host name was
841 given in a host list; looking up its IP address could give an abbreviated
842 form, whereas the current host's name might or might not be abbreviated.
843 The same fix has been applied.
849 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
850 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
853 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
854 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
856 PH/03 Typo: missing ".o" in src/pcre/Makefile.
858 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
859 header line, restrict the check to what is listed in RFCs 2369 and 2929.
860 Also, for "Auto-Submitted", treat anything other than "no" as
861 non-personal, in accordance with RFC 3834. (Previously it treated
862 anything starting "auto-" as non-personal.)
864 TF/01 The control=submission/name=... option had a problem with syntax
865 errors if the name included a slash character. The /name= option
866 now slurps the rest of the string, so it can include any characters
867 but it must come last in the list of options (after /sender_retain
870 PH/05 Some modifications to the interface to the fake nameserver for the new
878 TK/01 Added the "success_on_redirect" address verification option. See
879 NewStuff for rationale and an example.
881 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
883 PH/02 Patch to exigrep to allow it to work on syslog lines.
885 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
886 fread() to skip over the body file's header line, because in Cygwin the
887 header line is locked and is inaccessible.
889 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
890 co-exist for some time) to make it clear that it is the Exim ID that is
891 referenced, not the Message-ID: header line.
893 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
894 string_format() function, because snprintf() does not exist on all
897 PH/06 The use of forbid_filter_existstest now also locks out the use of the
898 ${stat: expansion item.
900 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
901 protocol synchronization error", to keep the pedants happy.
903 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
904 well as for IRIX systems, when gcc is being used. See the host.c source
907 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
909 PH/10 Named domain lists were not working if used in a queue_smtp_domains
912 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
913 transport and to the smtp transport in LMTP mode.
915 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
917 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
918 run a filter in a subprocess. This could lead to confusion in subsequent
919 lookups in the parent process. There should also be a search_tidyup() at
920 the end of the subprocess.
922 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
923 only if the host matched helo_try_verify_hosts, which caused the
924 verification to occur when the EHLO/HELO command was issued. The ACL just
925 tested the remembered result. Now, if a previous verification attempt has
926 not happened, "verify = helo" does it there and then.
928 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
930 TK/03 Fix log output including CR from clamd.
932 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
933 exist provoked a memory error which could cause a segfault.
935 PH/15 Installed PCRE 6.2
937 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
939 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
940 of the problem. Specifically, suggested +O2 rather than +O1 for the
943 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
945 PH/20 If a delivery was routed to a non-standard port by means of an SRV
946 record, the port was not correctly logged when the outgoing_port log
947 selector was set (it logged the transort's default port).
949 PH/21 Added support for host-specific ports to manualroute, queryprogram,
950 fallback_hosts, and "hosts" in the smtp transport.
952 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
953 host errors such as "Connection refused".
955 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
956 authentication with radiusclient 0.4.9:
958 - Error returned from rc_read_config was caught wrongly
959 - Username/password not passed on to radius server due to wrong length.
961 The presumption is that some radiusclient API changes for 4.51/PH/17
962 were not taken care of correctly. The code is still untested by me (my
963 Linux distribution still has 0.3.2 of radiusclient), but it was
964 contributed by a Radius user.
966 PH/24 When doing a callout, the value of $domain wasn't set correctly when
967 expanding the "port" option of the smtp transport.
969 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
970 while reading a MIME header. Thanks to Tom Hughes for a patch.
972 PH/24 Include config.h inside local_scan.h so that configuration settings are
975 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
976 that in an ACL for RCPT (for example), you can examine exactly what was
979 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
980 commands, but it was not correctly comparing the address with the actual
981 client host address. Thus, it would show the EHLO address in Received:
982 header lines when this was not necessary.
984 PH/27 Added the % operator to ${eval:}.
986 PH/28 Exim tries to create and chdir to its spool directory when it starts;
987 it should be ignoring failures (because with -C, for example, it has lost
988 privilege). It wasn't ignoring creation failures other than "already
991 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
994 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
995 because an input line was too long, either on its own, or by virtue of
996 too many continuations, the temporary file was not being removed, and the
997 return code was incorrect.
999 PH/31 Missing "BOOL" in function definition in filtertest.c.
1001 PH/32 Applied Sieve patches from the maintainer.
1003 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1005 PH/33 Added "verify = not_blind".
1007 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1008 Local/Makefile (with some defaults set). These are used in built scripts
1009 such as exicyclog, but they have never been used in the exim_install
1010 script (though there are many overriding facilities there). I have
1011 arranged that the exim_install script now takes note of these two
1014 PH/35 Installed configuration files for Dragonfly.
1016 PH/36 When a locally submitted message by a trusted user did not contain a
1017 From: header, and the sender address was obtained from -f or from an SMTP
1018 MAIL command, and the trusted user did not use -F to supply a sender
1019 name, $originator_name was incorrectly used when constructing a From:
1020 header. Furthermore, $originator_name was used for submission mode
1021 messages from external hosts without From: headers in a similar way,
1022 which is clearly wrong.
1024 PH/37 Added control=suppress_local_fixups.
1026 PH/38 When log_selector = +received_sender was set, and the addition of the
1027 sender made the log line's construction buffer exactly full, or one byte
1028 less than full, an overflow happened when the terminating "\n" was
1031 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1032 when the result of a list match is failure because a DNS lookup failed.
1034 PH/40 RM_COMMAND is now used in the building process.
1036 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1037 the "build-* directories that it finds.
1039 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1040 address in a domain literal was a prefix of an interface address.
1042 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1043 when verifying a sender address, unless rewrite_headers is false.
1045 PH/44 Wrote a long comment about why errors_to addresses are verified as
1046 recipients, not senders.
1048 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1049 the ratelimit ACL was added.
1051 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1053 PH/46 Added extra information about PostgreSQL errors to the error string.
1055 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1056 avoiding the need to install special zones in a real server. This is
1057 backwards compatible; if it can't find the fake resolver, it drops back.
1058 Thus, both old and new test suites can be run.
1060 TF/02 Added util/ratelimit.pl
1062 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1063 clock is set back in time.
1065 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1066 Candler <B.Candler@pobox.com>.
1068 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1069 OK for addresses that are the result of redirections.
1071 PH/48 A number of further additions for the benefit of the new test suite,
1072 including a fake gethostbyname() that interfaces to the fake DNS resolver
1075 TF/06 The fix for widen_domains has also been applied to qualify_single and
1076 search_parents which are the other dnslookup options that can cause
1079 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1082 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1084 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1085 service" if it encountered an MX record with an empty target hostname.
1086 The message is now "an MX or SRV record indicated no SMTP service".
1088 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1089 if the DNS of the sending site is misconfigured. This is quite a
1090 common situation. This change restores the behaviour of treating a
1091 helo verification defer as a failure.
1093 PH/51 If self=fail was set on a router, the bounce message did not include the
1094 actual error message.
1100 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1102 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1103 command itself ended in error, the underlying message about the transport
1104 filter timeout was being overwritten with the pipe command error. Now the
1105 underlying error message should be appended to the second error message.
1107 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1109 PH/02 Reduce the amount of output that "make" produces by default. Full output
1110 can still be requested.
1112 PH/03 The warning log line about a condition test deferring for a "warn" verb
1113 was being output only once per connection, rather than after each
1114 occurrence (because it was using the same function as for successful
1115 "warn" verbs). This seems wrong, so I have changed it.
1117 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1118 it should not have, which might have caused a crash in the right
1119 circumstances, but probably never did.
1121 PH/04 Installed a modified version of Tony Finch's patch to make submission
1122 mode fix the return path as well as the Sender: header line, and to
1123 add a /name= option so that you can make the user's friendly name appear
1126 TF/03 Added the control = fakedefer ACL modifier.
1128 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1129 Mark Lowes for thorough testing.
1131 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1133 TK/03 Merged latest SRS patch from Miles Wilton.
1135 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1136 with the definition in sysexits.h (which is #included earlier).
1137 Fortunately, Exim does not actually use EX_OK. The code used to try to
1138 preserve the sysexits.h value, by assumimg that macro definitions were
1139 scanned for macro replacements. I have been disabused of this notion,
1140 so now the code just undefines EX_OK before #including unistd.h.
1142 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1143 in the smtp transport. When a block could not be written in a single
1144 write() function, the timeout was being re-applied to each part-write.
1145 This seems wrong - if the receiver was accepting one byte at a time it
1146 would take for ever. The timeout is now adjusted when this happens. It
1147 doesn't have to be particularly precise.
1149 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1150 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1152 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1153 without a domain if the check to <postmaster@domain> fails.
1155 SC/01 Eximstats: added -xls and the ability to specify output files
1156 (patch written by Frank Heydlauf).
1158 SC/02 Eximstats: use FileHandles for outputing results.
1160 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1162 SC/04 Eximstats: fixed display of large numbers with -nvr option
1164 SC/05 Eximstats: fixed merging of reports with empty tables.
1166 SC/06 Eximstats: added the -include_original_destination flag
1168 SC/07 Eximstats: removed tabs and trailing whitespace.
1170 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1172 TK/06 MBOX spool code: Add real "From " MBOX separator line
1173 so the .eml file is really in mbox format (even though
1174 most programs do not really care). Patch from Alex Miller.
1176 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1177 The latter is generated from $received_to and is only set if the
1178 message has one envelope recipient. SA can use these headers,
1179 obviously out-of-the-box. Patch from Alex Miller.
1181 PH/08 The ${def test on a variable was returning false if the variable's
1182 value was "0", contrary to what the specification has always said!
1183 The result should be true unless the variable is empty.
1185 PH/09 The syntax error of a character other than { following "${if
1186 def:variable_name" (after optional whitespace) was not being diagnosed.
1187 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1188 accidental colon was present, for example, could give incorrect results.
1190 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1191 result is used (not including appendfile, where other changes are about
1194 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1195 This involved changing a lot of size variables from int to off_t. It
1196 should work with maildirs and everything.
1198 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1199 spamd dying while we are connected to it.
1201 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1204 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1205 function for BDB 4.3.
1207 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1209 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1210 the sources for PCRE have been re-arranged and more files are now
1213 PH/15 The code I had for printing potentially long long variables in PH/11
1214 above was not the best (it lost precision). The length of off_t variables
1215 is now inspected at build time, and an appropriate printing format (%ld
1216 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1217 to be "long long int" or "long int". This is needed for the internal
1218 formatting function string_vformat().
1220 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1221 the configuration file to be ":syslog", then the script "guesses" where
1222 the logs files are, rather than using the compiled in default. In our
1223 case the guess is not the same as the compiled default, so the script
1224 suddenly stopped working when I started to use syslog. The patch checks
1225 to see if log_file_path is "". If so, it attempts to read it from exim
1226 with no configuration file to get the compiled in version, before it
1227 falls back to the previous guessing code."
1229 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1230 implementing BATV in an Exim configuration. See NewStuff for the gory
1233 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1234 Makefile that are specific to HP-UX.
1236 PH/18 If the "use_postmaster" option was set for a recipient callout together
1237 with the "random" option, the postmaster address was used as the MAIL
1238 FROM address for the random test, but not for the subsequent recipient
1239 test. It is now used for both.
1241 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1242 patch removes a few documentation additions to RFC 3028, because the
1243 latest draft now contains them. It adds the new en;ascii-case comparator
1244 and a new error check for 8bit text in MIME parts. Comparator and
1245 require names are now matched exactly. I enabled the subaddress
1246 extension, but it is not well tested yet (read: it works for me)."
1248 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1249 rework some of the code of TK/09 above to avoid the hardwired use of
1250 "%lld" and "long long". Replaced the call to snprintf() with a call to
1253 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1254 records point to non-existent hosts", "retry timeout exceeded", and
1255 "retry time not reached for any host after a long failure period".
1257 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1258 experimental DomainKeys support:
1260 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1261 (2) On an error such as an illegally used "control", the wrong name for
1262 the control was given.
1264 These problems did NOT occur unless DomainKeys support was compiled.
1266 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1268 PH/24 Added ${if match_ip condition.
1270 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1271 databases so that it will be absolutely obvious if a crash occurs in the
1272 DB library. This is a regular occurrence (often caused by mis-matched
1275 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1276 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1277 picked up on a user's system that detects such things. There doesn't seem
1278 to be a gcc warning option for this - only an attribute that has to be
1279 put on the function's prototype. It seems that in Fedora Core 4 they have
1280 set this on a number of new functions. No doubt there will be more in due
1283 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1284 specify a transport. However, if an address that was verified by such a
1285 router was the subject of a callout, Exim crashed because it tried to
1286 read the rcpt_include_affixes from the non-existent transport. Now it
1287 just assumes that the setting of that option is false. This bug was
1288 introduced by 4.51/PH/31.
1290 PH/28 Changed -d+all to exclude +memory, because that information is very
1291 rarely of interest, but it makes the output a lot bigger. People tend to
1292 do -d+all out of habit.
1294 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1295 code in os-type was giving problems when libc.so lives in lib64, like on
1298 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1299 aren't the modern standard, and it seems that some systems' include files
1300 don't always have them. Exim was already checking for some of the newer
1301 ones like T_AAAA, and defining it itself. I've added checks for all the
1302 record types that Exim uses.
1304 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1305 not automatically generating a new one, as it is supposed to. This
1306 prevented TLS from working. If the file did exist, but contained invalid
1307 data, a new version was generated, as expected. It was only the case of a
1308 non-existent file that was broken.
1310 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1311 with a change in libdomainkeys > 0.64.
1313 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1314 from DNS. If the selector record carries the flag, it now has
1315 precedence over the domain-wide flag.
1317 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1319 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1320 the use of a port name that isn't defined in /etc/services) occurred, the
1321 message was deferred as in a normal delivery, and thus remained on the
1322 spool, instead of being failed because of the mua_wrapper setting. This
1323 is now fixed, and I tidied up some of the mua_wrapper messages at the
1326 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1327 the messages in a hash of arrays rather than using individual hashes.
1328 This is a bit cleaner and results in dramatic memory savings, albeit
1329 at a slight CPU cost.
1331 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1332 as requested by Marc Sherman.
1334 SC/10 Eximstats: added histograms for user specified patterns as requested
1337 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1339 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1340 fopen() in the content-scanning modules that did not already have it.
1346 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1347 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1349 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1351 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1353 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1354 to test Sieve filters that use "vacation".
1356 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1357 that changes the way the GnuTLS parameters are stored in the cache file.
1358 The new format can be generated externally. For backward compatibility,
1359 if the data in the cache doesn't make sense, Exim assumes it has read an
1360 old-format file, and it generates new data and writes a new file. This
1361 means that you can't go back to an older release without removing the
1364 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1365 work if there are any delivery delays because "one_time" forces the
1366 parent to be marked "delivered", so its unseen clone is never tried
1367 again. For this reason, Exim now forbids the simultaneous setting of
1370 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1371 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1372 entry if you want to know the details. The fix, however, neglected to
1373 consider the case when local delivery batching is involved. The test for
1374 "previously delivered" was not happening when checking to see if an
1375 address could be batched with a previous (undelivered) one; under
1376 certain circumstances this could lead to multiple deliveries to the same
1379 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1380 in its include files, and this causes problems building Exim.
1382 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1383 header_syntax) but Exim was just ignoring anything given after a slash.
1384 In particular, this caused confusion with an attempt to use "verify =
1385 reverse_host_lookup/defer_ok". An error is now given when options are
1386 supplied for verify items that do not have them. (Maybe reverse_host_
1387 lookup should have a defer_ok option, but that's a different point.)
1389 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1390 defined by RFC 821) to 2048, because there were problems with some AUTH
1391 commands, and RFC 1869 says the size should be increased for extended
1392 SMTP commands that take arguments.
1394 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1397 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1398 "unknown" error; now it says that the functionality isn't in the binary.
1400 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1401 an address' error message when a string expansion fails (syntax or
1402 whatever). Otherwise the password may appear in the log. Following change
1403 PH/42 below, there is no longer a chance of it appearing in a bounce
1406 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1408 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1409 of its addresses was ever tried. (Bugzilla bug #2.)
1411 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1412 the result incorrectly in the debug output. (It correctly added a newline
1413 to what was transported.)
1415 TF/01 Added $received_time.
1417 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1418 commented out examples of how to interface to a virus scanner and to
1419 SpamAssassin. Also added commented examples of av_scanner and
1420 spamd_address settings.
1422 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1423 and controls are allowed in which ACLs. There were a couple of minor
1424 errors. Some of the entries in the conditions table (which is a table of
1425 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1426 negation of where the condition IS allowed.
1428 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1430 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1431 header file does not have a version number, so I've had to invent a new
1432 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1433 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1434 radiusclient), but it was contributed by a Radius user.
1436 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1437 files or directories, for OpenSSL.
1439 PH/19 When an Exim process that is running as root has to create an Exim log
1440 file, it does so in a subprocess that runs as exim:exim so as to get the
1441 ownership right at creation (otherwise, other Exim processes might see
1442 the file with the wrong ownership). There was no test for failure of this
1443 fork() call, which would lead to the process getting stuck as it waited
1444 for a non-existent subprocess. Forks do occasionally fail when resources
1445 run out. I reviewed all the other calls to fork(); they all seem to check
1448 PH/20 When checking for unexpected SMTP input at connect time (before writing
1449 the banner), Exim was not dealing correctly with a non-positive return
1450 from the read() function. If the client had disconnected by this time,
1451 the result was a log entry for a synchronization error with an empty
1452 string after "input=" when read() returned zero. If read() returned -1
1453 (an event I could not check), uninitialized data bytes were printed.
1454 There were reports of junk text (parts of files, etc) appearing after
1457 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1459 PH/22 Added support for macro redefinition, and (re)definition in between
1460 driver and ACL definitions.
1462 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1463 forgetting to use the resulting value; it was using the unexpanded value.
1465 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1466 hadn't been configured. The fix is from Juergen Kreileder, who
1467 understands it better than I do:
1469 "Here's what I see happening with three configured cyrus_sasl
1470 authenticators configured (plain, login, cram-md5):
1472 On startup auth_cyrus_sasl_init() gets called for each of these.
1473 This means three calls to sasl_listmech() without a specified mech_list.
1474 => SASL tests which mechs of all available mechs actually work
1475 => three warnings about OTP not working
1476 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1478 With the patch, sasl_listmech() also gets called three times. But now
1479 SASL's mech_list option is set to the server_mech specified in the the
1480 authenticator. Or in other words, the answer from sasl_listmech()
1481 gets limited to just the mech you're testing for (which is different
1483 => the return list contains just 'plain' or 'login', 'cram-md5' or
1484 nothing depending on the value of ob->server_mech.
1486 I've just tested the patch: Authentication still works fine,
1487 unavailable mechs specified in the exim configuration are still
1488 caught, and the auth.log warnings about OTP are gone."
1490 PH/25 When debugging is enabled, the contents of the command line are added
1491 to the debugging output, even when log_selector=+arguments is not
1494 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1495 answer is "GNU", and only if the return is "GNU/something" is the answer
1498 PH/27 $acl_verify_message is now set immediately after the failure of a
1499 verification in an ACL, and so is available in subsequent modifiers. In
1500 particular, the message can be preserved by coding like this:
1502 warn !verify = sender
1503 set acl_m0 = $acl_verify_message
1505 Previously, $acl_verify_message was set only while expanding "message"
1506 and "log_message" when a very denied access.
1508 PH/28 Modified OS/os.c-Linux with
1510 -#ifndef OS_LOAD_AVERAGE
1511 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1513 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1514 nomenclature these days.)
1516 PH/29 Installed patch from the Sieve maintainer that adds the options
1517 sieve_useraddress and sieve_subaddress to the redirect router.
1519 PH/30 In these circumstances:
1520 . Two addresses routed to the same list of hosts;
1521 . First host does not offer TLS;
1522 . First host accepts first address;
1523 . First host gives temporary error to second address;
1524 . Second host offers TLS and a TLS session is established;
1525 . Second host accepts second address.
1526 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1527 and peerdn, if requested) that were in fact used only for the second
1530 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1531 attention to any local part prefix or suffix that was matched by the
1532 router that accepted the address. It now behaves in the same way as it
1533 does for delivery: the affixes are removed from the local part unless
1534 rcpt_include_affixes is set on the transport.
1536 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1537 timeout during the DATA phase of an incoming message.
1539 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1540 applied a patch sanctioned by the Sieve maintainer.
1542 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1543 the uid or gid is negative. A case of a negative gid caused this to be
1544 noticed. The fix allows for either to be negative.
1546 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1547 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1550 PH/36 The change PH/12 above was broken. Fixed it.
1552 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1553 the grounds that routing the same address twice would always produce the
1554 same answer. This might have been true once, but it is certainly no
1555 longer true now. Routing a child address may depend on the previous
1556 routing that produced that child. Some complicated redirection strategies
1557 went wrong when messages had multiple recipients, and made Exim's
1558 behaviour dependent on the order in which the addresses were given.
1560 I have moved the duplicate checking until after the routing is complete.
1561 Exim scans the addresses that are assigned to local and remote
1562 transports, and removes any duplicates. This means that more work will be
1563 done, as duplicates will always all be routed, but duplicates are
1564 presumably rare, so I don't expect this is of any significance.
1566 For deliveries to pipes, files, and autoreplies, the duplicate checking
1567 still happens during the routing process, since they are not going to be
1570 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1571 It corrects a timeout issue with spamd. This is Ian's comment: "The
1572 background is that sometimes spamd either never reads data from a
1573 connection it has accepted, or it never writes response data. The exiscan
1574 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1575 blindly assumes that writes won't block so it may never time out."
1577 PH/39 Allow G after quota size as well as K and M.
1579 PH/40 The value set for $authenticated_id in an authenticator may not contain
1580 binary zeroes or newlines because the value is written to log lines and
1581 to spool files. There was no check on this. Now the value is run through
1582 the string_printing() function so that such characters are converted to
1583 printable escape sequences.
1585 PH/41 $message_linecount is a new variable that contains the total number of
1586 lines in the message. Compare $body_linecount, which is the count for the
1589 PH/42 Exim no longer gives details of delivery errors for specific addresses in
1590 bounce and delay warning messages, except in certain special cases, which
1593 (a) An SMTP error message from a remote host;
1594 (b) A message specified in a :fail: redirection;
1595 (c) A message specified in a "fail" command in a system filter;
1596 (d) A message specified in a FAIL return from the queryprogram router;
1597 (e) A message specified by the cannot_route_message router option.
1599 In these cases only, Exim does include the error details in bounce and
1600 warning messages. There are also a few cases where bland messages such
1601 as "unrouteable address" or "local delivery error" are given.
1603 PH/43 $value is now also set for the "else" part of a ${run expansion.
1605 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1606 being worked on, but at least Exim now implements the latest version to
1609 PH/45 In a pipe transport, although a timeout while waiting for the pipe
1610 process to complete was treated as a delivery failure, a timeout while
1611 writing the message to the pipe was logged, but erroneously treated as a
1612 successful delivery. Such timeouts include transport filter timeouts. For
1613 consistency with the overall process timeout, these timeouts are now
1614 treated as errors, giving rise to delivery failures by default. However,
1615 there is now a new Boolean option for the pipe transport called
1616 timeout_defer, which, if set TRUE, converts the failures into defers for
1617 both kinds of timeout. A transport filter timeout is now identified in
1620 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1621 systems where "make" and "gmake" are different, calling "gmake" at top
1622 level broke things. I've arranged for the value of $(MAKE) to be passed
1623 from the Makefile to this script so that it can call the same version of
1627 A note about Exim versions 4.44 and 4.50
1628 ----------------------------------------
1630 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1631 changes of various kinds. As a consequence, a big documentation update was
1632 needed. This delayed the release for rather longer than seemed good, especially
1633 in the light of a couple of (minor) security issues. Therefore, the changes
1634 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1635 release. So 4.44 and 4.50 are in effect two different branches that both start
1638 I have left the 4.50 change log unchanged; it contains all the changes since
1639 4.43. The change log for 4.44 is below; many of its items are identical to
1640 those for 4.50. This seems to be the most sensible way to preserve the
1641 historical information.
1647 1. Minor wording change to the doc/README.SIEVE file.
1649 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
1650 computation of the current number of files was incorrect.
1652 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1653 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1656 4. Give more explanation in the error message when the command for a transport
1657 filter fails to execute.
1659 5. There are several places where Exim runs a non-Exim command in a
1660 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1661 was being done only for the command run by the queryprogram router. It is
1662 now done for all such subprocesses. The other cases are: ${run, transport
1663 filters, and the commands run by the lmtp and pipe transports.
1665 6. Added CONFIGURE_GROUP build-time option.
1667 7. Some older OS have a limit of 256 on the maximum number of file
1668 descriptors. Exim was using setrlimit() to set 1000 as a large value
1669 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1670 systems. I've change it so that if it can't get 1000, it tries for 256.
1672 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1673 was an oversight, and furthermore, ever since the addition of extra
1674 controls (e.g. 4.43/32), the checks on when to allow different forms of
1675 "control" were broken. There should now be diagnostics for all cases when a
1676 control that does not make sense is encountered.
1678 9. Added the /retain_sender option to "control=submission".
1680 10. $recipients is now available in the predata ACL (oversight).
1682 11. Tidy the search cache before the fork to do a delivery from a message
1683 received from the command line. Otherwise the child will trigger a lookup
1684 failure and thereby defer the delivery if it tries to use (for example) a
1685 cached ldap connection that the parent has called unbind on.
1687 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1688 of $address_data from the recipient verification was clobbered by the
1689 sender verification.
1691 13. The value of address_data from a sender verification is now available in
1692 $sender_address_data in subsequent conditions in the ACL statement.
1694 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1696 15. Added a new option "connect=<time>" to callout options, to set a different
1699 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1700 was its contents. (It was OK if the option was not defined at all.)
1702 17. A "Completed" log line is now written for messages that are removed from
1703 the spool by the -Mrm option.
1705 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1706 information about exactly what failed.
1708 19. Added -dd to debug only the daemon process.
1710 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1711 handles timeouts, both on the server side and network timeouts. Renamed the
1712 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1714 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1715 It is now set to "smtps".
1717 22. $host_address is now set to the target address during the checking of
1718 ignore_target_hosts.
1720 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1721 being passed; this would have caused $sender_host_name to have been used if
1722 matching the list had actually called for a host name (not very likely,
1723 since this list is usually IP addresses). A host name is now passed as
1726 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1727 code that specifies a non-restarting handler (typically sigaction() in
1728 modern systems) in an attempt to fix a rare and obscure crash bug.
1730 25. Narrowed the window for a race in the daemon that could cause it to ignore
1731 SIGCHLD signals. This is not a major problem, because they are used only to
1732 wake it up if nothing else does.
1734 26. A malformed maildirsize file could cause Exim to calculate negative values
1735 for the mailbox size or file count. Odd effects could occur as a result.
1736 The maildirsize information is now recalculated if the size or filecount
1739 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1740 support for a long time. Removed HAVE_SYS_VFS_H.
1742 28. Installed the latest version of exipick from John Jetmore.
1744 29. In an address list, if the pattern was not a regular expression, an empty
1745 subject address (from a bounce message) matched only if the pattern was an
1746 empty string. Non-empty patterns were not even tested. This was the wrong
1747 because it is perfectly reasonable to use an empty address as part of a
1748 database query. An empty address is now tested by patterns that are
1749 lookups. However, all the other forms of pattern expect the subject to
1750 contain a local part and a domain, and therefore, for them, an empty
1751 address still always fails if the pattern is not itself empty.
1753 30. Exim went into a mad DNS loop when attempting to do a callout where the
1754 host was specified on an smtp transport, and looking it up yielded more
1755 than one IP address.
1757 31. Re-factored the code for checking spool and log partition space into a
1758 function that finds that data and another that does the check. The former
1759 is then used to implement four new variables: $spool_space, $log_space,
1760 $spool_inodes, and $log_inodes.
1762 32. The RFC2047 encoding function was originally intended for short strings
1763 such as real names; it was not keeping to the 75-character limit for
1764 encoded words that the RFC imposes. It now respects the limit, and
1765 generates multiple encoded words if necessary. To be on the safe side, I
1766 have increased the buffer size for the ${rfc2047: expansion operator from
1769 33. It is now permitted to omit both strings after an "if" condition; if the
1770 condition is true, the result is "true". As before, when the second string
1771 is omitted, a false condition yields an empty string. This makes it less
1772 cumbersome to write custom ACL and router conditions.
1774 34. Failure to deliver a bounce message always caused it to be frozen, even if
1775 there was an errors_to setting on the router. The errors_to setting is now
1778 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1779 canonical form (fully expanded) before being placed in
1780 $sender_host_address.
1782 36. The table in the code that translates DNS record types into text (T_A to
1783 "A" for instance) was missing entries for NS and CNAME. It is just possible
1784 that this could have caused confusion if both these types were looked up
1785 for the same domain, because the text type is used as part of Exim's
1786 per-process caching. But the chance of anyone hitting this buglet seems
1789 37. The dnsdb lookup has been extended in a number of ways.
1791 (1) There is a new type, "zns", which walks up the domain tree until it
1792 finds some nameserver records. It should be used with care.
1794 (2) There is a new type, "mxh", which is like "mx" except that it returns
1795 just the host names, not the priorities.
1797 (3) It is now possible to give a list of domains (or IP addresses) to be
1798 looked up. The behaviour when one of the lookups defers can be
1799 controlled by a keyword.
1801 (4) It is now possible to specify the separator character for use when
1802 multiple records are returned.
1804 38. The dnslists ACL condition has been extended: it is now possible to supply
1805 a list of IP addresses and/or domains to be looked up in a particular DNS
1808 39. Added log_selector=+queue_time_overall.
1810 40. When running the queue in the test harness, wait just a tad after forking a
1811 delivery process, to get repeatability of debugging output.
1813 41. Include certificate and key file names in error message when GnuTLS fails
1814 to set them up, because the GnuTLS error message doesn't include the name
1815 of the failing file when there is a problem reading it.
1817 42. Allow both -bf and -bF in the same test run.
1819 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1821 44. The "Exiscan patch" is now merged into the mainline Exim source.
1823 45. Sometimes the final signoff response after QUIT could fail to get
1824 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1825 < 0 before doing a fflush(). This bug looks as though it goes back to the
1826 introduction of TLS in release 3.20, but "sometimes" must have been rare
1827 because the tests only now provoked it.
1829 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1830 (this can affect the format of dates).
1832 47. exim_tidydb, when checking for the continued existence of a message for
1833 which it has found a message-specific retry record, was not finding
1834 messages that were in split spool directories. Consequently, it was
1835 deleting retry records that should have stayed in existence.
1837 48. Steve fixed some bugs in eximstats.
1839 49. The SPA authentication driver was not abandoning authentication and moving
1840 on to the next authenticator when an expansion was forced to fail,
1841 contradicting the general specification for all authenticators. Instead it
1842 was generating a temporary error. It now behaves as specified.
1844 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1845 (the order specifies the preference for clients). The order is now AES256,
1846 AES128, 3DES, ARCFOUR128.
1848 51. Small patch to Sieve code - explicitly set From: when generating an
1851 52. Exim crashed if a remote delivery caused a very long error message to be
1852 recorded - for instance if somebody sent an entire SpamAssassin report back
1853 as a large number of 550 error lines. This bug was coincidentally fixed by
1854 increasing the size of one of Exim's internal buffers (big_buffer) that
1855 happened as part of the Exiscan merge. However, to be on the safe side, I
1856 have made the code more robust (and fixed the comments that describe what
1859 53. Now that there can be additional text after "Completed" in log lines (if
1860 the queue_time_overall log selector is set), a one-byte patch to exigrep
1861 was needed to allow it to recognize "Completed" as not the last thing in
1864 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1865 patch that reportedly fixes this has been added. I am not expert enough to
1866 create a test for it. This is what the patch creator wrote:
1868 "I found a little strange behaviour of ldap code when working with
1869 Windows 2003 AD Domain, where users was placed in more than one
1870 Organization Units. When I tried to give exim partial DN, the exit code
1871 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1872 But simultaneously result of request was absolutely normal ldap result,
1873 so I produce this patch..."
1875 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1876 REFERENCE, so I have modified the code to exclude the patch when that macro
1879 55. Some experimental protocols are using DNS PTR records for new purposes. The
1880 keys for these records are domain names, not reversed IP addresses. The
1881 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1882 leaves it alone. Component reversal etc. now happens only for IP addresses.
1885 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1887 57. Double the size of the debug message buffer (to 2048) so that more of very
1888 long debug lines gets shown.
1890 58. The exicyclog utility now does better if the number of log files to keep
1891 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1893 59. Two changes related to the smtp_active_hostname option:
1895 (1) $smtp_active_hostname is now available as a variable.
1896 (2) The default for smtp_banner uses $smtp_active_hostname instead
1897 of $primary_hostname.
1899 60. The host_aton() function is supposed to be passed a string that is known
1900 to be a valid IP address. However, in the case of IPv6 addresses, it was
1901 not checking this. This is a hostage to fortune. Exim now panics and dies
1902 if the condition is not met. A case was found where this could be provoked
1903 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1904 components; fortuitously, this particular loophole had already been fixed
1905 by change 4.50/55 above.
1907 If there are any other similar loopholes, the new check in host_aton()
1908 itself should stop them being exploited. The report I received stated that
1909 data on the command line could provoke the exploit when Exim was running as
1910 exim, but did not say which command line option was involved. All I could
1911 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1912 running as the user.
1915 61. There was a buffer overflow vulnerability in the SPA authentication code
1916 (which came originally from the Samba project). I have added a test to the
1917 spa_base64_to_bits() function which I hope fixes it.
1920 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1921 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1923 63. The daemon start-up calls getloadavg() while still root for those OS that
1924 need the first call to be done as root, but it missed one case: when
1925 deliver_queue_load_max is set with deliver_drop_privilege. This is
1926 necessary for the benefit of the queue runner, because there is no re-exec
1927 when deliver_drop_privilege is set.
1929 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1930 This has been fixed.
1932 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1933 was in use, was not putting the data itself into the right store pool;
1934 consequently, it could be overwritten for a subsequent message in the same
1935 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1938 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1940 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1941 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1942 the function were treating the return as a boolean value, which happened to
1943 work because 0=false and not-0=true, but is not correct code.
1945 68. The host_aton() function was not handling scoped IPv6 addresses (those
1946 with, for example, "%eth0" on the end) correctly.
1948 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1949 negated items (that is, ~something) in unsigned ints. Some compilers
1950 apparently mutter when there is no cast.
1952 70. If an address verification called from an ACL failed, and did not produce a
1953 user-specific message (i.e. there was only a "system" message), nothing was
1954 put in $acl_verify_message. In this situation, it now puts the system
1957 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1958 session; change 4.31/43 added the unwanted input to the log line - except
1959 that it did not do this in the start of session case. It now does.
1961 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1962 This is wrong and can cause the other end to generate a synchronization
1963 error if it is another Exim or anything else that does the synchronization
1964 check. A QUIT command is no longer sent after a timeout.
1966 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1967 during host lookups.
1969 74. The defer_ok option of callout verification was not working if it was used
1970 when verifying addresses in header lines, that is, for this case:
1972 verify = header_sender/callout=defer_ok
1974 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1975 those file descriptors could be used for SMTP connections. If anything
1976 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1977 could be sent to the SMTP client, causing chaos. The daemon now opens
1978 stdin, stdout, and stderr to /dev/null when it puts itself into the
1981 76. Arrange for output from Perl's "warn" command to be written to Exim's main
1982 log by default. The user can override this with suitable Perl magic.
1984 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
1985 the log message when discard triggers, was not working for the DATA ACL or
1986 for the non-SMTP ACL.
1988 78. Error message wording change in sieve.c.
1990 79. If smtp_accept_max_per_host was set, the number of connections could be
1991 restricted to fewer than expected, because the daemon was trying to set up
1992 a new connection before checking whether the processes handling previous
1993 connections had finished. The check for completed processes is now done
1994 earlier. On busy systems, this bug wouldn't be noticed because something
1995 else would have woken the daemon, and it would have reaped the completed
1998 80. If a message was submitted locally by a user whose login name contained one
1999 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2000 It caused a spool format error. I have fixed the spool reading code. A
2001 related problem was that the "from" clause in the Received: line became
2002 illegal because of the space(s). It is now covered by ${quote_local_part.
2004 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2007 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2009 83. Patch from Sieve maintainer to fix unterminated string problem in
2010 "vacation" handling.
2012 84. Some minor changes to the Linux configuration files to help with other
2013 OS variants using glibc.
2015 85. One more patch for Sieve to update vacation handling to latest spec.
2018 ----------------------------------------------------
2019 See the note above about the 4.44 and 4.50 releases.
2020 ----------------------------------------------------
2026 1. Change 4.43/35 introduced a bug that caused file counts to be
2027 incorrectly computed when quota_filecount was set in an appendfile
2030 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2031 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2034 3. Give more explanation in the error message when the command for a transport
2035 filter fails to execute.
2037 4. There are several places where Exim runs a non-Exim command in a
2038 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2039 was being done only for the command run by the queryprogram router. It is
2040 now done for all such subprocesses. The other cases are: ${run, transport
2041 filters, and the commands run by the lmtp and pipe transports.
2043 5. Some older OS have a limit of 256 on the maximum number of file
2044 descriptors. Exim was using setrlimit() to set 1000 as a large value
2045 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2046 systems. I've change it so that if it can't get 1000, it tries for 256.
2048 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2049 was an oversight, and furthermore, ever since the addition of extra
2050 controls (e.g. 4.43/32), the checks on when to allow different forms of
2051 "control" were broken. There should now be diagnostics for all cases when a
2052 control that does not make sense is encountered.
2054 7. $recipients is now available in the predata ACL (oversight).
2056 8. Tidy the search cache before the fork to do a delivery from a message
2057 received from the command line. Otherwise the child will trigger a lookup
2058 failure and thereby defer the delivery if it tries to use (for example) a
2059 cached ldap connection that the parent has called unbind on.
2061 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2062 of $address_data from the recipient verification was clobbered by the
2063 sender verification.
2065 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2066 was its contents. (It was OK if the option was not defined at all.)
2068 11. A "Completed" log line is now written for messages that are removed from
2069 the spool by the -Mrm option.
2071 12. $host_address is now set to the target address during the checking of
2072 ignore_target_hosts.
2074 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2075 being passed; this would have caused $sender_host_name to have been used if
2076 matching the list had actually called for a host name (not very likely,
2077 since this list is usually IP addresses). A host name is now passed as
2080 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2081 code that specifies a non-restarting handler (typically sigaction() in
2082 modern systems) in an attempt to fix a rare and obscure crash bug.
2084 15. Narrowed the window for a race in the daemon that could cause it to ignore
2085 SIGCHLD signals. This is not a major problem, because they are used only to
2086 wake it up if nothing else does.
2088 16. A malformed maildirsize file could cause Exim to calculate negative values
2089 for the mailbox size or file count. Odd effects could occur as a result.
2090 The maildirsize information is now recalculated if the size or filecount
2093 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2094 support for a long time. Removed HAVE_SYS_VFS_H.
2096 18. Updated exipick to current release from John Jetmore.
2098 19. Allow an empty sender to be matched against a lookup in an address list.
2099 Previously the only cases considered were a regular expression, or an
2102 20. Exim went into a mad DNS lookup loop when doing a callout where the
2103 host was specified on the transport, if the DNS lookup yielded more than
2106 21. The RFC2047 encoding function was originally intended for short strings
2107 such as real names; it was not keeping to the 75-character limit for
2108 encoded words that the RFC imposes. It now respects the limit, and
2109 generates multiple encoded words if necessary. To be on the safe side, I
2110 have increased the buffer size for the ${rfc2047: expansion operator from
2113 22. Failure to deliver a bounce message always caused it to be frozen, even if
2114 there was an errors_to setting on the router. The errors_to setting is now
2117 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2118 canonical form (fully expanded) before being placed in
2119 $sender_host_address.
2121 24. Updated eximstats to version 1.33
2123 25. Include certificate and key file names in error message when GnuTLS fails
2124 to set them up, because the GnuTLS error message doesn't include the name
2125 of the failing file when there is a problem reading it.
2127 26. Expand error message when OpenSSL has problems setting up cert/key files.
2130 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2131 (this can affect the format of dates).
2133 28. exim_tidydb, when checking for the continued existence of a message for
2134 which it has found a message-specific retry record, was not finding
2135 messages that were in split spool directories. Consequently, it was
2136 deleting retry records that should have stayed in existence.
2138 29. eximstats updated to version 1.35
2139 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2140 1.35 - bugfix such that pie charts by volume are generated correctly
2142 30. The SPA authentication driver was not abandoning authentication and moving
2143 on to the next authenticator when an expansion was forced to fail,
2144 contradicting the general specification for all authenticators. Instead it
2145 was generating a temporary error. It now behaves as specified.
2147 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2148 (the order specifies the preference for clients). The order is now AES256,
2149 AES128, 3DES, ARCFOUR128.
2151 31. Small patch to Sieve code - explicitly set From: when generating an
2154 32. Exim crashed if a remote delivery caused a very long error message to be
2155 recorded - for instance if somebody sent an entire SpamAssassin report back
2156 as a large number of 550 error lines. This bug was coincidentally fixed by
2157 increasing the size of one of Exim's internal buffers (big_buffer) that
2158 happened as part of the Exiscan merge. However, to be on the safe side, I
2159 have made the code more robust (and fixed the comments that describe what
2162 33. Some experimental protocols are using DNS PTR records for new purposes. The
2163 keys for these records are domain names, not reversed IP addresses. The
2164 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2165 leaves it alone. Component reversal etc. now happens only for IP addresses.
2168 34. The host_aton() function is supposed to be passed a string that is known
2169 to be a valid IP address. However, in the case of IPv6 addresses, it was
2170 not checking this. This is a hostage to fortune. Exim now panics and dies
2171 if the condition is not met. A case was found where this could be provoked
2172 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2173 components; fortuitously, this particular loophole had already been fixed
2174 by change 4.50/55 or 4.44/33 above.
2176 If there are any other similar loopholes, the new check in host_aton()
2177 itself should stop them being exploited. The report I received stated that
2178 data on the command line could provoke the exploit when Exim was running as
2179 exim, but did not say which command line option was involved. All I could
2180 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2181 running as the user.
2184 35. There was a buffer overflow vulnerability in the SPA authentication code
2185 (which came originally from the Samba project). I have added a test to the
2186 spa_base64_to_bits() function which I hope fixes it.
2189 36. The daemon start-up calls getloadavg() while still root for those OS that
2190 need the first call to be done as root, but it missed one case: when
2191 deliver_queue_load_max is set with deliver_drop_privilege. This is
2192 necessary for the benefit of the queue runner, because there is no re-exec
2193 when deliver_drop_privilege is set.
2195 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2196 was in use, was not putting the data itself into the right store pool;
2197 consequently, it could be overwritten for a subsequent message in the same
2198 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2201 38. Sometimes the final signoff response after QUIT could fail to get
2202 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2203 < 0 before doing a fflush(). This bug looks as though it goes back to the
2204 introduction of TLS in release 3.20, but "sometimes" must have been rare
2205 because the tests only now provoked it.
2211 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2212 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2213 Now it returns an integer. A number of calls were still expecting a T/F
2214 return. Fortuitously, in all cases, the tests worked in OK situations,
2215 which is the norm. However, things would have gone wrong on any write
2216 failures on the smtp file descriptor. This function is used when sending
2217 messages over SMTP and also when doing verify callouts.
2219 2. When Exim is called to do synchronous delivery of a locally submitted
2220 message (the -odf or -odi options), it no longer closes stderr before doing
2223 3. Implemented the mua_wrapper option.
2225 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2227 5. Implemented the functions header_remove(), header_testname(),
2228 header_add_at_position(), and receive_remove_recipient(), and exported them
2231 6. If an ACL "warn" statement specified the addition of headers, Exim already
2232 inserted X-ACL-Warn: at the start if there was no header name. However, it
2233 was not making this test for the second and subsequent header lines if
2234 there were newlines in the string. This meant that an invalid header could
2235 be inserted if Exim was badly configured.
2237 7. Allow an ACL "warn" statement to add header lines at the start or after all
2238 the Received: headers, as well as at the end.
2240 8. Added the rcpt_4xx retry error code.
2242 9. Added postmaster_mailfrom=xxx to callout verification option.
2244 10. Added mailfrom=xxxx to the callout verification option, for verify=
2247 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2248 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2250 12. Inserted some casts to stop certain compilers warning when using pointer
2251 differences as field lengths or precisions in printf-type calls (mostly
2252 affecting debugging statements).
2254 13. Added optional readline() support for -be (dynamically loaded).
2256 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2257 same clock tick as a message's arrival, so that its received time was the
2258 same as the "first fail" time on the retry record, and that message
2259 remained on the queue past the ultimate address timeout, every queue runner
2260 would try a delivery (because it was past the ultimate address timeout) but
2261 after another failure, the ultimate address timeout, which should have then
2262 bounced the address, did not kick in. This was a "< instead of <=" error;
2263 in most cases the first failure would have been in the next clock tick
2264 after the received time, and all would be well.
2266 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2267 being recognized when the domain list was tested by the match_domain
2268 condition in an expansion string.
2270 16. Added the ${str2b64: operator.
2272 17. Exim was always calling setrlimit() to set a large limit for the number of
2273 processes, without checking whether the existing limit was already
2274 adequate. (It did check for the limit on file descriptors.) Furthermore,
2275 errors from getrlimit() and setrlimit() were being ignored. Now they are
2276 logged to the main and panic logs, but Exim does carry on, to try to do its
2277 job under whatever limits there are.
2279 18. Imported PCRE 5.0.
2281 19. Trivial typo in log message " temporarily refused connection" (the leading
2284 20. If the log selector return_path_on_delivery was set and an address was
2285 redirected to /dev/null, the delivery process crashed because it assumed
2286 that a return path would always be set for a "successful" delivery. In this
2287 case, the whole delivery is bypassed as an optimization, and therefore no
2290 21. Internal re-arrangement: the function for sending a challenge and reading
2291 a response while authentication was assuming a zero-terminated challenge
2292 string. It's now changed to take a pointer and a length, to allow for
2293 binary data in such strings.
2295 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2297 23. Exim was not respecting finduser_retries when seeking the login of the
2298 uid under which it was called; it was always trying 10 times. (The default
2299 setting of finduser_retries is zero.) Also, it was sleeping after the final
2300 failure, which is pointless.
2302 24. Implemented tls_on_connect_ports.
2304 25. Implemented acl_smtp_predata.
2306 26. If the domain in control=submission is set empty, Exim assumes that the
2307 authenticated id is a complete email address when it generates From: or
2308 Sender: header lines.
2310 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2311 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2312 chown and chgrp in /bin and hostname in /usr/bin.
2314 28. Exim was keeping the "process log" file open after each use, just as it
2315 does for the main log. This opens the possibility of it remaining open for
2316 long periods when the USR1 signal hits a daemon. Occasional processlog
2317 errors were reported, that could have been caused by this. Anyway, it seems
2318 much more sensible not to leave this file open at all, so that is what now
2321 29. The long-running daemon process does not normally write to the log once it
2322 has entered its main loop, and it closes the log before doing so. This is
2323 so that log files can straightforwardly be renamed and moved. However,
2324 there are a couple of unusual error situations where the daemon does write
2325 log entries, and I had neglected to close the log afterwards.
2327 30. The text of an SMTP error response that was received during a remote
2328 delivery was being truncated at 512 bytes. This is too short for some of
2329 the long messages that one sometimes sees. I've increased the limit to
2332 31. It is now possible to make retry rules that apply only when a message has a
2333 specific sender, in particular, an empty sender.
2335 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2336 it possible to be selective about when SMTP synchronization is enforced.
2338 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2340 32. Implemented hosts_connection_nolog.
2342 33. Added an ACL for QUIT.
2344 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2347 35. Added mailbox_size and mailbox_filecount to appendfile.
2349 36. Added control = no_multiline_responses to ACLs.
2351 37. There was a bug in the logic of the code that waits for the clock to tick
2352 in the case where the clock went backwards by a substantial amount such
2353 that the microsecond fraction of "now" was more than the microsecond
2354 fraction of "then" (but the whole seconds number was less).
2356 38. Added support for the libradius Radius client library this is found on
2357 FreeBSD (previously only the radiusclient library was supported).
2363 1. When certain lookups returned multiple values in the form name=value, the
2364 quoting of the values was not always being done properly. Specifically:
2365 (a) If the value started with a double quote, but contained no whitespace,
2367 (b) If the value contained whitespace other than a space character (i.e.
2368 tabs or newlines or carriage returns) it was not quoted.
2369 This fix has been applied to the mysql and pgsql lookups by writing a
2370 separate quoting function and calling it from the lookup code. The fix
2371 should probably also be applied to nisplus, ibase and oracle lookups, but
2372 since I cannot test any of those, I have not disturbed their existing code.
2374 2. A hit in the callout cache for a specific address caused a log line with no
2375 reason for rejecting RCPT. Now it says "Previous (cached) callout
2376 verification failure".
2378 3. There was an off-by-one bug in the queryprogram router. An over-long
2379 return line was truncated at 256 instead of 255 characters, thereby
2380 overflowing its buffer with the terminating zero. As well as fixing this, I
2381 have increased the buffer size to 1024 (and made a note to document this).
2383 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2384 when Exim is waiting for an SMTP response from a remote server, Exim
2385 restarts its select() call on the socket, thereby resetting its timeout.
2386 This is not a problem when such interrupts are rare. Somebody set up a cron
2387 job to run exiwhat every 2 minutes, which is less than the normal select()
2388 timeout (5 or 10 minutes). This meant that the select() timeout never
2389 kicked in because it was always reset. I have fixed this by comparing the
2390 time when an interrupt arrives with the time at the start of the first call
2391 to select(). If more time than the timeout has elapsed, the interrupt is
2392 treated as a timeout.
2394 5. Some internal re-factoring in preparation for the addition of Sieve
2395 extensions (by MH). In particular, the "personal" test is moved to a
2396 separate function, and given an option for scanning Cc: and Bcc: (which is
2397 not set for Exim filters).
2399 6. When Exim created an email address using the login of the caller as the
2400 local part (e.g. when creating a From: or Sender: header line), it was not
2401 quoting the local part when it contained special characters such as @.
2403 7. Installed new OpenBSD configuration files.
2405 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2406 try to make them clearer.
2408 9. Callout options, other than the timeout value, were being ignored when
2409 verifying sender addresses in header lines. For example, when using
2411 verify = header_sender/callout=no_cache
2413 the cache was (incorrectly) being used.
2415 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2416 only the Cygwin environment.
2418 11. When return_path_on_delivery was set as a log selector, if different remote
2419 addresses in the same message used different return paths and parallel
2420 remote delivery occurred, the wrong values would sometimes be logged.
2421 (Whenever a remote delivery process finished, the return path value from
2422 the most recently started remote delivery process was logged.)
2424 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2425 lines when AUTH and/or TLS are in use. This is the "received protocol"
2426 field. Exim used to use "asmtp" for authenticated SMTP, without any
2427 indication (in the protocol name) for TLS use. Now it follows the RFC and
2428 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2429 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2430 names appear in log lines as well as in Received: header lines.
2432 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2433 extensions, and comparison tests, and to fix some bugs.
2435 14. Changes to the "personal" filter test:
2437 (1) The test was buggy in that it was just doing the equivalent of
2438 "contains" tests on header lines. For example, if a user's address was
2439 anne@some.where, the "personal" test would incorrectly be true for
2441 To: susanne@some.where
2443 This test is now done by extracting each address from the header in turn,
2444 and checking the entire address. Other tests that are part of "personal"
2445 are now done using regular expressions (for example, to check local parts
2446 of addresses in From: header lines).
2448 (2) The list of non-personal local parts in From: addresses has been
2449 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2450 taken from the Sieve specification recommendations.
2452 (3) If the message contains any header line starting with "List-" it is
2453 treated as non-personal.
2455 (4) The test for "circular" in the Subject: header line has been removed
2456 because it now seems ill-conceived.
2458 15. Minor typos in src/EDITME comments corrected.
2460 16. Installed latest exipick from John Jetmore.
2462 17. If headers_add on a router specified a text string that was too long for
2463 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2464 of string_sprintf() is now avoided.
2466 18. $message_body_size was not set (it was always zero) when running the DATA
2467 ACL and the local_scan() function.
2469 19. For the "mail" command in an Exim filter, no default was being set for
2470 the once_repeat time, causing a random time value to be used if "once" was
2471 specified. (If the value happened to be <= 0, no repeat happened.) The
2472 default is now 0s, meaning "never repeat". The "vacation" command was OK
2473 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2474 (I found it when inspecting the code).
2476 20. There is now an overall timeout for performing a callout verification. It
2477 defaults to 4 times the callout timeout, which applies to individual SMTP
2478 commands during the callout. The overall timeout applies when there is more
2479 than one host that can be tried. The timeout is checked before trying the
2480 next host. This prevents very long delays if there are a large number of
2481 hosts and all are timing out (e.g. when the network connections are timing
2482 out). The value of the overall timeout can be changed by specifying an
2483 additional sub-option for "callout", called "maxwait". For example:
2485 verify = sender/callout=5s,maxwait=20s
2487 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2488 to the end before writing, but this should make it even safer).
2490 22. Exim was forgetting that it had advertised PIPELINING for the second and
2491 subsequent messages on an SMTP connection. It was also not resetting its
2492 memory on STARTTLS and an internal HELO.
2494 23. When Exim logs an SMTP synchronization error within a session, it now
2495 records whether PIPELINING has been advertised or not.
2497 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2498 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2499 rather than long int.
2501 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2503 26. Added the never_mail option to autoreply.
2509 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2510 crash if the getsockname() call failed; this can happen if a connection is
2511 closed very soon after it is established. The problem was simply in the
2512 order in which certain operations were done, causing Exim to try to write
2513 to the SMTP stream before it had set up the file descriptor. The bug has
2514 been fixed by making things happen in the correct order.
2520 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2521 before the connection was closed, thus losing the rejection response.
2523 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2524 some early Solaris releases, but causes trouble in current releases where
2525 socklen_t is defined.
2527 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2530 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2533 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2534 that releasing the top of it at the end releases what was used for sub-
2535 expansions (unless the block got too big). However, discard this block if
2536 the first thing is a variable or header, so that we can use its block when
2537 it is dynamic (useful for very large $message_headers, for example).
2539 6. Lookups now cache *every* query, not just the most recent. A new, separate
2540 store pool is used for this. It can be recovered when all lookup caches are
2541 flushed. Lookups now release memory at the end of their result strings.
2542 This has involved some general refactoring of the lookup sources.
2544 7. Some code has been added to the store_xxx() functions to reduce the amount
2545 of flapping under certain conditions.
2547 8. log_incoming_interface used to affect only the <= reception log lines. Now
2548 it causes the local interface and port to be added to several more SMTP log
2549 lines, for example "SMTP connection from", and rejection lines.
2551 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2553 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2555 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2556 could be overwritten at the end of the current message (or the start of a
2557 new message if it was set in a HELO ACL). The value is now preserved for
2558 the duration of the SMTP connection.
2560 12. If a transport had a headers_rewrite setting, and a matching header line
2561 contained an unqualified address, that address was qualified, even if it
2562 did not match any rewriting rules. The underlying bug was that the values
2563 of the flags that permit the existence of unqualified sender and recipient
2564 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2565 non-local messages, and by -bnq for local messages) were not being
2566 preserved with the message after it was received.
2568 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2569 "next input=" as part of the text comprising the host identity instead of
2570 the correct text. The code was using the same buffer for two different
2571 strings. However, depending on which order the printing function evaluated
2572 its arguments, the bug did not always show up. Under Linux, for example, my
2573 test suite worked just fine.
2575 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2576 doesn't work with some older versions of Perl. It has been changed to "my",
2577 which in any case is probably the better facility to use.
2579 15. A really picky compiler found some instances of statements for creating
2580 error messages that either had too many or two few arguments for the format
2583 16. The size of the buffer for calls to the DNS resolver has been increased
2584 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2585 for addresses that have a lot of PTR records. This alleviates a problem; it
2586 does not fully solve it.
2588 17. A dnsdb lookup for PTR records that receives more data than will fit in the
2589 buffer now truncates the list and logs the incident, which is the same
2590 action as happens when Exim is looking up a host name and its aliases.
2591 Previously in this situation something unpredictable would happen;
2592 sometimes it was "internal error: store_reset failed".
2594 18. If a server dropped the connection unexpectedly when an Exim client was
2595 using GnuTLS and trying to read a response, the client delivery process
2596 crashed while trying to generate an error log message.
2598 19. If a "warn" verb in an ACL added multiple headers to a message in a single
2599 string, for example:
2601 warn message = H1: something\nH2: something
2603 the text was added as a single header line from Exim's point of view
2604 though it ended up OK in the delivered message. However, searching for the
2605 second and subsequent header lines using $h_h2: did not work. This has been
2606 fixed. Similarly, if a system filter added multiple headers in this way,
2607 the routers could not see them.
2609 20. Expanded the error message when iplsearch is called with an invalid key to
2610 suggest using net-iplsearch in a host list.
2612 21. When running tests using -bh, any delays imposed by "delay" modifiers in
2613 ACLs are no longer actually imposed (and a message to that effect is
2616 22. If a "gecos" field in a passwd entry contained escaped characters, in
2617 particular, if it contained a \" sequence, Exim got it wrong when building
2618 a From: or a Sender: header from that name. A second bug also caused
2619 incorrect handling when an unquoted " was present following a character
2620 that needed quoting.
2622 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2623 was not being matched caselessly.
2625 24. Arranged for all hyphens in the exim.8 source to be escaped with
2628 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2629 properly. Recipient callout cache records were still being keyed to include
2630 the sender, even when use_sender was set false. This led to far more
2631 callouts that were necessary. The sender is no longer included in the key
2632 when use_sender is false.
2634 26. Added "control = submission" modifier to ACLs.
2636 27. Added the ${base62d: operator to decode base 62 numbers.
2638 28. dnsdb lookups can now access SRV records.
2640 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2641 the configuration file.
2643 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2644 (-v) mode. This makes the output for a verbose queue run more intelligible.
2646 31. Added a use_postmaster feature to recipient callouts.
2648 32. Added the $body_zerocount variable, containing the number of binary zero
2649 bytes in the message body.
2651 33. The time of last modification of the "new" subdirectory is now used as the
2652 "mailbox time last read" when there is a quota error for a maildir
2655 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2657 35. Added +ignore_unknown as a special item in host lists.
2659 36. Code for decoding IPv6 addresses in host lists is now included, even if
2660 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2661 address was recognized as an IP address, but was then not correctly decoded
2662 into binary, causing unexpected and incorrect effects when compared with
2669 1. Very minor rewording of debugging text in manualroute to say "list of
2670 hosts" instead of "hostlist".
2672 2. If verify=header_syntax was set, and a header line with an unqualified
2673 address (no domain) and a large number of spaces between the end of the
2674 name and the colon was received, the reception process suffered a buffer
2675 overflow, and (when I tested it) crashed. This was caused by some obsolete
2676 code that should have been removed. The fix is to remove it!
2678 3. When running in the test harness, delay a bit after writing a bounce
2679 message to get a bit more predictability in the log output.
2681 4. Added a call to search_tidyup() just before forking a reception process. In
2682 theory, someone could use a lookup in the expansion of smtp_accept_max_
2683 per_host which, without the tidyup, could leave open a database connection.
2685 5. Added the variables $recipient_data and $sender_data which get set from a
2686 lookup success in an ACL "recipients" or "senders" condition, or a router
2687 "senders" option, similar to $domain_data and $local_part_data.
2689 6. Moved the writing of debug_print from before to after the "senders" test
2692 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2693 problems for message scanning, either using a data ACL, or using
2694 local_scan() because the Received: header was not generated till after they
2695 were called (in order to set the time as the time of reception completion).
2696 I have revised the way this works. The header is now generated after the
2697 body is received, but before the ACL or local_scan() are called. After they
2698 are run, the timestamp in the header is updated.
2704 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2705 before starting a queue runner without re-exec. This happened only when
2706 deliver_drop_privilege was set or when the Exim user was set to root. The
2707 effect of the bug was that timeouts during subsequent deliveries caused
2708 crashes instead of being properly handled. The handler is now left at its
2709 default (and expected) setting.
2711 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2712 message, again when deliver_drop_privilege is set or Exim is run as root.
2713 The bug described in (1) was not present in this case, but the tidying up
2714 of the other signals was missing. I have made the two cases consistent.
2716 3. The ignore_target_hosts setting on a manualroute router was being ignored
2717 for hosts that were looked up using the /MX notation.
2719 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2722 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2723 operated on the sender address. After changing the $sender_address to <>
2724 for the sender address verify, Exim was re-instated it as the original
2725 (before rewriting) address, but remembering that it had rewritten it, so it
2726 wasn't rewriting it again. This bug also had the effect of breaking the
2727 sender address verification caching when the sender address was rewritten.
2729 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2730 This has been changed so that if the ip literal address matches
2731 ignore_target_hosts, the router declines.
2733 7. Added expansion conditions match_domain, match_address, and match_local_
2734 part (NOT match_host).
2736 8. The placeholder for the Received: header didn't have a length field set.
2738 9. Added code to Exim itself and to exim_lock to test for a specific race
2739 condition that could lead to file corruption when using MBX delivery. The
2740 issue is with the lockfile that is created in /tmp. If this file is removed
2741 after a process has opened it but before that process has acquired a lock,
2742 there is the potential for a second process to recreate the file and also
2743 acquire a lock. This could lead to two Exim processes writing to the file
2744 at the same time. The added code performs the same test as UW imapd; it
2745 checks after acquiring the lock that its file descriptor still refers to
2746 the same named file.
2748 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2749 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2752 11. Added the smtp_active_hostname option. If used, this will typically be made
2753 to depend on the incoming interface address. Because $interface_address is
2754 not set up until the daemon has forked a reception process, error responses
2755 that can happen earlier (such as "too many connections") no longer contain
2758 12. If an expansion in a condition on a "warn" statement fails because a lookup
2759 defers, the "warn" statement is abandoned, and the next ACL statement is
2760 processed. Previously this caused the whole ACL to be aborted.
2762 13. Added the iplsearch lookup type.
2764 14. Added ident_timeout as a log selector.
2766 15. Added tls_certificate_verified as a log selector.
2768 16. Added a global option tls_require_ciphers (compare the smtp transport
2769 option of the same name). This controls incoming TLS connections.
2771 17. I finally figured out how to make tls_require_ciphers do a similar thing
2772 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2773 before starting the TLS session.
2775 18. Tabs are now shown as \t in -bP output.
2777 19. If the log selector return_path_on_delivery was set, Exim crashed when
2778 bouncing a message because it had too many Received: header lines.
2780 20. If two routers both had headers_remove settings, and the first one included
2781 a superfluous trailing colon, the final name in the first list and the
2782 first name in the second list were incorrectly joined into one item (with a
2783 colon in the middle).
2789 1. Added -C and -D options to the exinext utility, mainly to make it easier
2790 to include in the automated testing, but these could be helpful when
2791 multiple configurations are in use.
2793 2. The exinext utility was not formatting the output nicely when there was
2794 an alternate port involved in the retry record key, nor when there was a
2795 message id as well (for retries that were specific to a specific message
2796 and a specific host). It was also confused by IPv6 addresses, because of
2797 the additional colons they contain. I have fixed the IPv4 problem, and
2798 patched it up to do a reasonable job for IPv6.
2800 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2801 delivery, the log line now contains "pipelined" if PIPELINING was used.
2803 4. An SMTP transport process used to panic and die if the bind() call to set
2804 an explicit outgoing interface failed. This has been changed; it is now
2805 treated in the same way as a connect() failure.
2807 5. A reference to $sender_host_name in the part of a conditional expansion
2808 that was being skipped was still causing a DNS lookup. This no longer
2811 6. The def: expansion condition was not recognizing references to header lines
2812 that used bh_ and bheader_.
2814 7. Added the _cache feature to named lists.
2816 8. The code for checking quota_filecount in the appendfile transport was
2817 allowing one more file than it should have been.
2819 9. For compatibility with Sendmail, the command line option
2827 and sets the incoming protocol and host name (for trusted callers). The
2828 host name and its colon can be omitted when only the protocol is to be set.
2829 Note the Exim already has two private options, -pd and -ps, that refer to
2830 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2831 "s", but I don't think that's a major issue.
2833 10. A number of refactoring changes to the code, none of which should affect
2836 (a) The number of logging options was getting close to filling up the
2837 32-bit word that was used as a bit map. I have split them into two classes:
2838 those that are passed in the argument to log_write(), and those that are
2839 only ever tested independently outside of that function. These are now in
2840 separate 32-bit words, so there is plenty of room for expansion again.
2841 There is no change in the user interface or the logging behaviour.
2843 (b) When building, for example, log lines, the code previously used a
2844 macro that called string_cat() twice, in order to add two strings. This is
2845 not really sufficiently general. Furthermore, there was one instance where
2846 it was actually wrong because one of the argument was used twice, and in
2847 one call a function was used. (As it happened, calling the function twice
2848 did not affect the overall behaviour.) The macro has been replaced by a
2849 function that can join an arbitrary number of extra strings onto a growing
2852 (c) The code for expansion conditions now uses a table and a binary chop
2853 instead of a serial search (which was left over from when there were very
2854 few conditions). Also, it now recognizes conditions like "pam" even when
2855 the relevant support is not compiled in: a suitably worded error message is
2856 given if an attempt is made to use such a condition.
2858 11. Added ${time_interval:xxxxx}.
2860 12. A bug was causing one of the ddress fields not to be passed back correctly
2861 from remote delivery subprocesses. The field in question was not being
2862 subsequently used, so this caused to problems in practice.
2864 13. Added new log selectors queue_time and deliver_time.
2866 14. Might have fixed a bug in maildirsizefile handling that threw up
2867 "unexpected character" debug warnings, and recalculated the data
2868 unnecessarily. In any case, I expanded the warning message to give more
2871 15. Added the message "Restricted characters in address" to the statements in
2872 the default ACL that block characters like @ and % in local parts.
2874 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2875 Three changes have been made:
2877 (a) There was a serious bug; a negative response to MAIL caused the whole
2878 recipient domain to be cached as invalid, thereby blocking all messages
2879 to all local parts at the same domain, from all senders. This bug has
2880 been fixed. The domain is no longer cached after a negative response to
2881 MAIL if the sender used is not empty.
2883 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2886 (c) A new callout option, "use_sender" has been added for people who want
2887 the modified behaviour.
2893 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2896 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2897 indeed breaks things for older releases.
2899 3. Added additional logging to the case where there is a problem reading data
2900 from a filter that is running in a subprocess using a pipe, in order to
2901 try to track down a specific problem.
2903 4. Testing facility fudge: when running in the test harness and attempting
2904 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2905 getting "No route to host". Convert this to a timeout.
2907 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2910 6. Some OS don't have socklen_t but use size_t instead. This affects the
2911 fifth argument of getsockopt() amongst other things. This is now
2912 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2913 can be set for individual OS. I have set it for SunOS5, OSF1, and
2914 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2915 some earlier ones do not.
2917 7. Change 4.30/15 was not doing the test caselessly.
2919 8. The standard form for an IPv6 address literal was being rejected by address
2920 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2921 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2922 this, as well as the form without the "IPv6" on the front (but only when
2923 address literals are enabled, of course).
2925 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2927 10. Exim crashed if a message with an empty sender address specified by -f
2928 encountered a router with an errors_to setting. This could be provoked only
2929 by a command such as
2933 where an empty string was supplied; "<>" did not hit this bug.
2935 11. Installed PCRE release 4.5.
2937 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2938 remained set. It is now erased.
2940 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2941 times from message ids (which are base 36 rather than the normal 62).
2943 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2944 were being counted as actual protocol errors, and logged if the log
2945 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2946 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2947 and DATA following a set of rejected RCPTs do not count as protocol errors.
2948 In other words, Exim assumes they were pipelined, though this may not
2949 actually be the case. Of course, in all cases the client gets an
2950 appropriate error code.
2952 15. If a lookup fails in an ACL condition, a message about the failure may
2953 be available; it is used if testing the ACL cannot continue, because most
2954 such messages specify what the cause of the deferral is. However, some
2955 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2956 that caused an old message to be retained and used if a later statement
2957 caused a defer, replacing the real cause of the deferral.
2959 16. If an IP address had so many PTR records that the DNS lookup buffer
2960 was not large enough to hold them, Exim could crash while trying to process
2961 the truncated data. It now detects and logs this case.
2963 17. Further to 4.21/58, another change has been made: if (and only if) the
2964 first line of a message (the first header line) ends with CRLF, a bare LF
2965 in a subsequent header line has a space inserted after it, so as not to
2966 terminate the header.
2968 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
2969 unnecessarily, used atoi() instead of strtol(), and didn't check the
2970 termination when getting file sizes from file names by regex.
2972 19. Completely re-implemented the support for maildirsize files, in the light
2973 of a number of problems with the previous contributed implementation
2974 (4.30/29). In particular:
2976 . If the quota is zero, the maildirsize file is maintained, but no quota is
2979 . If the maildir directory does not exist, it is created before any attempt
2980 to write a maildirsize file.
2982 . The quota value in the file is just a cache; if the quota is changed in
2983 the transport, the new value overrides.
2985 . A regular expression is available for excluding directories from the
2988 20. The autoreply transport checks the characters in options that define the
2989 message's headers; it allows continued headers, but it was checking with
2990 isspace() after an embedded newline instead of explicitly looking for a
2993 21. If all the "regular" hosts to which an address was routed had passed their
2994 expiry times, and had not reached their retry times, the address was
2995 bounced, even if fallback hosts were defined. Now Exim should go on to try
2998 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2999 equivalent code in the SMTP transport. Some hosts send humungous responses
3000 to HELO/EHLO, more than 1024 it seems.
3002 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3003 gives compiler warnings in some environments. I've now done it "properly",
3006 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3007 (because of problems with the built-in one) was declared to return uschar *
3008 instead of char *, causing compiler failure.
3010 25. Fixed a file descriptor leak when processing alias/forward files.
3012 26. Fixed a minor format string issue in dbfn.c.
3014 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3016 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3017 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3020 29. When a sender address is verified, it is cached, to save repeating the test
3021 when there is more than one recipient in a message. However, when the
3022 verification involves a callout, it is possible for different callout
3023 options to be set for different recipients. It is too complicated to keep
3024 track of this in the cache, so now Exim always runs a verification when a
3025 callout is required, relying on the callout cache for the optimization.
3026 The overhead is duplication of the address routing, but this should not be
3029 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3030 to be verified with callout=postmaster, and the main callout worked but the
3031 postmaster check failed, the verification correctly failed. However, if a
3032 subsequent RCPT command asked for sender verification *without* the
3033 postmaster check, incorrect caching caused this verification also to fail,
3036 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3037 it was not caching the DNS options (qualify_single, search_parents) that
3038 were used when the lookup failed. A subsequent lookup with different
3039 options therefore always gave the same answer, though there were cases
3040 where it should not have. (Example: a "domains = !$mx_any" option on a
3041 dnslookup router: the "domains" option is always processed without any
3042 widening, but the router might have qualify_single set.) Now Exim uses the
3043 cached value only when the same options are set.
3045 32. Added John Jetmore's "exipick" utility to the distribution.
3047 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3048 than a timeout (e.g. a certificate is required, and is not provided), an
3049 Exim server now closes the connection immediately. Previously it waited for
3050 the client to close - but if the client is SSL, it seems that they each
3051 wait for each other, leading to a delay before one of them times out.
3053 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3054 maintained 0.8.x compatibility because I don't think many are using it, and
3055 it is clearly obsolete.
3057 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3060 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3061 client certificate was expired. A simple patch fixes this, though I don't
3062 understand the full logic of why the verify callback is called multiple
3065 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3066 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3067 which causes problems with some clients (such as the Certicom SSL Plus
3068 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3069 disables the coutermeasure allowing Eudora to connect."
3071 38. Exim was not checking that a write() to a log file succeeded. This could
3072 lead to Bad Things if a log got too big, in particular if it hit a file
3073 size limit. Exim now panics and dies if it cannot write to a log file, just
3074 as it does if it cannot open a log file.
3076 39. Modified OS/Makefile-Linux so that it now contains
3078 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3080 The two -D definitions ensure that Exim is compiled with large file
3081 support, which makes it possible to handle log files that are bigger than
3084 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3085 instance) a domain was checked against a named list that involved a lookup,
3086 causing $domain_data to be set, then another domain was checked against the
3087 same list, then the first domain was re-checked, the value of $domain_data
3088 after the final check could be wrong. In particular, if the second check
3089 failed, it could be set empty. This bug probably also applied to
3092 41. The strip_trailing_dot option was not being applied to the address given
3093 with the -f command-line option.
3095 42. The code for reading a message's header from the spool was incrementing
3096 $received_count, but never initializing it. This meant that the value was
3097 incorrect (doubled) while delivering a message in the same process in which
3098 it was received. In the most common configuration of Exim, this never
3099 happens - a fresh exec is done - but it can happen when
3100 deliver_drop_privilege is set.
3102 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3103 it now includes up to 150 characters of the unexpected data in the log
3106 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3107 and building data strings. The size of both of these buffers was 10 000
3108 bytes - far larger than anybody would *ever* want, thought I. Needless to
3109 say, somebody hit the limit. I have increased the maximum line length to
3110 20 000 and the maximum data length of concatenated lines to 100 000. I have
3111 also fixed two bugs, because there was no checking on these buffers. Tsk,
3112 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3113 buffer is too small.
3115 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3116 lsearch lookups. Now it does.
3118 46. When parsing a route_list item in a manualroute router, a fixed-length
3119 buffer was used for the list of hosts. I made this 1024 bytes long,
3120 thinking that nobody would ever have a list of hosts that long. Wrong.
3121 Somebody had a whole pile of complicated expansion conditions, and the
3122 string was silently truncated, leading to an expansion error. It turns out
3123 that it is easier to change to an unlimited length (owing to other changes
3124 that have happened since this code was originally written) than to build
3125 structure for giving a limitation error. The length of the item that
3126 expands into the list of hosts is now unlimited.
3128 47. The lsearch lookup could not handle data where the length of text line was
3129 more than 4095 characters. Such lines were truncated, leading to shortened
3130 data being returned. It should now handle lines of any length.
3132 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3133 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3136 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3137 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3138 no longer happens in comment lines. A list of replacements is now placed
3139 at the head of all of the source files, except those whose only change is
3140 to replace PERL_COMMAND in the very first #! line.
3142 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3143 messages on the queue, with a bottom-up merge sort, using code contributed
3144 by Michael Haardt. This should make operations like -bp somewhat faster on
3145 large queues. It won't affect queue runners, except when queue_run_in_order
3148 51. Installed eximstats 1.31 in the distribution.
3150 52. Added support for SRV lookups to the dnslookup router.
3152 53. If an ACL referred to $message_body or $message_body_end, the value was not
3153 reset for any messages that followed in the same SMTP session.
3155 54. The store-handling optimization for building very long strings was not
3156 differentiating between the different store pools. I don't think this
3157 actually made any difference in practice, but I've tidied it.
3159 55. While running the routers to verify a sender address, $sender_address
3160 was still set to the sender address. This is wrong, because when routing to
3161 send a bounce to the sender, it would be empty. Therefore, I have changed
3162 it so that, while verifying a sender address, $sender_address is set to <>.
3163 (There is no change to what happens when verifying a recipient address.)
3165 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3166 target A or AAAA records (if not already returned) without resetting the
3167 qualify_single or search_parents options of the DNS resolver. These are
3168 inappropriate in this case because the targets of MX and SRV records must
3169 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3170 target that, when qualified, matched something in the local domain. These
3171 two options are now turned off when doing these lookups.
3173 57. It seems that at least some releases of Reiserfs (which does not have the
3174 concept of a fixed number of inodes) returns zero and not -1 for the
3175 number of available inodes. This interacted badly with check_spool_inodes,
3176 which assumed that -1 was the "no such thing" setting. What I have done is
3177 to check that the total number of inodes is greater than zero before doing
3178 the test of how many are available.
3180 58. When a "warn" ACL statement has a log_message modifier, the message is
3181 remembered, and not repeated. This is to avoid a lot of repetition when a
3182 message has many recipients that cause the same warning to be written.
3183 Howewer, Exim was preserving the list of already written lines for an
3184 entire SMTP session, which doesn't seem right. The memory is now reset if a
3185 new message is started.
3187 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3188 debugging output unless log_rewrite was also set.
3190 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3191 of (int)(handle) when we know that handle contains (void *)(-1).
3193 61. The Exim daemon panic-logs an error return when it closes the incoming
3194 connection. However "connection reset by peer" seems to be common, and
3195 isn't really an error worthy of noting specially, so that particular error
3198 62. When Exim is trying to find all the local interfaces, it used to panic and
3199 die if the ioctl to get the interface flags failed. However, it seems that
3200 on at least one OS (Solaris 9) it is possible to have an interface that is
3201 included in the list of interfaces, but for which you get a failure error
3202 for this call. This happens when the interface is not "plumbed" into a
3203 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3204 failure of the "get flags" call assumes that the interface is down.
3206 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3207 makes life easier for people who are doing arithmetic on fields extracted
3208 from dates, where you often get leading zeros that should not be
3209 interpreted as octal.
3211 64. Added qualify_domain to the redirect router, to override the global
3214 65. If a pathologically long header line contained very many addresses (the
3215 report of this problem mentioned 10 000) and each of them was rewritten,
3216 Exim could use up a very large amount of memory. (It kept on making new
3217 copies of the header line as it rewrote, and never released the old ones.)
3218 At the expense of a bit more processing, the header rewriting function has
3219 been changed so that it no longer eats memory in this way.
3221 66. The generation of the Received: header has been moved from the time that a
3222 message starts to be received, to the time that it finishes. The timestamp
3223 in the Received: header should now be very close to that of the <= log
3224 line. There are two side-effects of this change:
3226 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3227 logged header lines no longer include the local Received: line, because
3228 it has not yet been created. The same applies to a copy of the message
3229 that is returned to a non-SMTP sender when a message is rejected.
3231 (b) When a filter file is tested using -bf, no additional Received: header
3232 is added to the test message. After some thought, I decided that this
3235 This change does not affect the value of $received_for. It is still set
3236 after address rewriting, but before local_scan() is called.
3238 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3240 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3241 gave an unhelpful panic error message, and a defer error. I have managed to
3242 change this behaviour so that it now rejects any supplied certificate,
3243 which seems right, as the list of acceptable certificates is empty.
3245 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3246 gave an unhelpful defer error. I have not managed to make this reject any
3247 supplied certificates, but the error message it gives is "no certificate
3248 supplied", which is not helpful.
3250 70. exigrep's output now also includes lines that are not associated with any
3251 message, but which match the given pattern. Implemented by a patch from
3252 Martin Sluka, which also tidied up the Perl a bit.
3254 71. Recipient callout verification, like sender verification, was using <> in
3255 the MAIL FROM command. This isn't really the right thing, since the actual
3256 sender may affect whether the remote host accepts the recipient or not. I
3257 have changed it to use the actual sender in the callout; this means that
3258 the cache record is now keyed on a recipient/sender pair, not just the
3259 recipient address. There doesn't seem to be a real danger of callout loops,
3260 since a callout by the remote host to check the sender would use <>.
3261 [SEE ABOVE: changed after hitting problems.]
3263 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3264 temporary errors. However, in the case of such a code being given after
3265 the end of a data transmission (i.e. after ".") Exim was failing to write
3266 a retry record for the message. (Yes, there was some broken host that was
3267 actually sending 8xx at this point.)
3269 73. An unknown lookup type in a host list could cause Exim to panic-die when
3270 the list was checked. (An example that provoked this was putting <; in the
3271 middle of a list instead of at the start.) If this happened during a DATA
3272 ACL check, a -D file could be left lying around. This kind of configuration
3273 error no longer causes Exim to die; instead it causes a defer errror. The
3274 incident is still logged to the main and panic logs.
3276 74. Buglet left over from Exim 3 conversion. The message "too many messages
3277 in one connection" was written to the rejectlog but not the mainlog, except
3278 when address rewriting (yes!) was being logged.
3280 75. Added write_rejectlog option.
3282 76. When a system filter was run not as root (that is, when system_filter_user
3283 was set), the values of the $n variables were not being returned to the
3284 main process; thus, they were not subsequently available in the $sn
3287 77. Added +return_path_on_delivery log selector.
3289 78. A connection timeout was being treated differently from recipients deferred
3290 when testing hosts_max_try with a message that was older than the host's
3291 retry timeout. (The host should not be counted, thus allowing all hosts to
3292 be tried at least once before bouncing.) This may have been the cause of an
3293 occasionally reported bug whereby a message would remain on the queue
3294 longer than the retry timeout, but would be bounced if a delivery was
3295 forced. I say "may" because I never totally pinned down the problem;
3296 setting up timeout/retry tests is difficult. See also the next item.
3298 79. The ultimate address timeout was not being applied to errors that involved
3299 a combination of host plus message (for example, a timeout on a MAIL
3300 command). When an address resolved to a number of possible hosts, and they
3301 were not all tried for each delivery (e.g. because of hosts_max_try), a
3302 message could remain on the queue longer than the retry timeout.
3304 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3307 81. Fixed an obscure SMTP outgoing bug which required at least the following
3308 conditions: (a) there was another message waiting for the same server;
3309 (b) the server returned 5xx to all RCPT commands in the first message so
3310 that the message was not completed; (c) the server dropped the connection
3311 or gave a negative response to the RSET that Exim sends to abort the
3312 transaction. The observed case was a dropped connection after DATA that had
3313 been sent in pipelining mode. That is, the server had advertised PIPELINING
3314 but was not implementing it correctly. The effect of the bug was incorrect
3315 behaviour, such as trying another host, and this could lead to a crash.
3321 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3322 and daemon.c were passed as pointers to ints; they should have been
3323 pointers to socklen_t variables (which are typically unsigned ints).
3325 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3328 3. Fixed a really odd bug that affected only the testing scheme; patching a
3329 certain fixed string in the binary changed the value of another string that
3330 happened to be identical to the end of the original first string.
3332 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3333 name", it returns that address as the IP address. On some operating
3334 systems (e.g. Solaris), it also passes back the IP address string as the
3335 "host name". However, on others (e.g. Linux), it passes back an empty
3336 string. Exim wasn't checking for this, and was changing the host name to an
3337 empty string, assuming it had been canonicized.
3339 5. Although rare, it is permitted to have more than one PTR record for a given
3340 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3341 all the names associated with an address, because they do in Solaris.
3342 However, it seems that they do not in Linux for data that comes from the
3343 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3344 I found this out when I moved to a new Linux workstation and tried to run
3345 the Exim test suite.
3347 To get round this problem I have changed the code so that it now does its
3348 own call to the DNS to look up PTR records when searching for a host name.
3349 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3350 addresses that are only in /etc/hosts are still found.
3352 This behaviour is, however, controlled by an option called host_lookup_
3353 order, which defaults to "bydns:byaddr". If people want to use the other
3354 order, or indeed, just use one or the other means of lookup, they can
3355 specify it in this variable.
3357 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3358 some operating systems, this comes back from gethostbyaddr() as an empty
3359 string, and this is what Exim used to test for. However, it seems that in
3360 other systems, "." is yielded. Exim now tests for this case too.
3362 7. The values of check_spool_space and check_log_space are now held internally
3363 as a number of kilobytes instead of an absolute number of bytes. If a
3364 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3365 kilobyte. This means that much larger values can be stored.
3367 8. Exim monitor: an attempt to get the action menu when not actually pointing
3368 at a message produces an empty menu entitled "No message selected". This
3369 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3370 no entries in it ("Shell widget menu has zero width and/or height"). So I
3371 have added a single, blank menu entry in this case.
3373 9. Added ${quote_local_part.
3375 10. MIME decoding is now applied to the contents of Subject: header lines when
3378 11. Now that a reference to $sender_host_address automatically causes a reverse
3379 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3380 host lookup before query-style lookups in lists that might use this
3381 variable. This has therefore been abolished, and the "net-" prefix is no
3382 longer necessary for query-style lookups.
3384 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3385 have been a typo for LFLAGS, so it has been changed.
3387 13. The install script calls Exim with "-C /dev/null" in order to find the
3388 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3389 to be output. Howeve, since Exim outputs its version number before the
3390 error, it didn't break the script. It just looked ugly. I fixed this by
3391 always allowing "-C /dev/null" if the caller is root.
3393 14. Ignore overlarge ACL variable number when reading spool file - insurance
3394 against a later release with more variables having written the file.
3396 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3397 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3398 this, as well as the form without the "IPv6" on the front.
3400 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3401 OS/Makefile-Darwin file.
3403 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3404 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3406 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3407 variable. However, some OS use ints for inodes. I've added cast to long int
3408 to get rid of the compiler warning.
3410 19. I had forgotten to lock out "/../" in configuration file names when
3411 ALT_CONFIG_PREFIX was set.
3413 20. Routers used for verification do not need to specify transports. However,
3414 if such a router generated a host list, and callout was configured, Exim
3415 crashed, because it could not find a port number from the (non-existent)
3416 transport. It now assumes port 25 in this circumstance.
3418 21. Added the -t option to exigrep.
3420 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3421 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3422 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3423 them is set, LOOKUP_LSEARCH is forced.
3425 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3427 24. Added sender and host information to the "rejected by local_scan()" log
3428 line; previously there was no indication of these.
3430 25. Added .include_if_exists.
3432 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3433 Linux. It turns out that not all file systems support this. Apparently some
3434 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3435 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3436 not supported on the file descriptor, is now ignored when Exim is trying to
3437 sync a directory. This applies only to Linux.
3439 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3441 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3442 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3445 29. Added support for maildirsize files from supplied patch (modified a bit).
3447 30. The use of :fail: followed by an empty string could lead Exim to respond to
3448 sender verification failures with (e.g.):
3450 550 Verification failed for <xxx>
3451 550 Sender verify failed
3453 where the first response line was missing the '-' that indicates it is not
3454 the final line of the response.
3456 31. The loop for finding the name of the user that called Exim had a hardwired
3457 limit of 10; it now uses the value of finduser_retries, which is used for
3458 all other user lookups.
3460 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3463 33. Exim was neglecting to zero errno before one call of strtol() when
3464 expanding a string and expecting an integer value. On some systems this
3465 resulted in spurious "integer overflow" errors. Also, it was casting the
3466 result into an int without checking.
3468 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3469 did not work. The code looks as if it has *never* worked, though it appears
3470 to have been documented since at least releast 1.62. I have made it work.
3472 35. The "timeout_DNS" error in retry rules, also documented since at least
3473 1.62, also never worked. As it isn't clear exactly what this means, and
3474 clearly it isn't a major issue, I have abolished the feature by treating it
3475 as "timeout", and writing a warning to the main and panic logs.
3477 36. The display of retry rules for -brt wasn't always showing the error code
3480 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3481 timeout_connect_A, timeout_connect_MX.
3483 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3484 to the empty sender.
3486 39. The daemon was not analysing the content of -oX till after it had closed
3487 stderr and disconnected from the controlling terminal. This meant that any
3488 syntax errors were only noted on the panic log, and the return code from
3489 the command was 0. By re-arranging the code a little, I've made the
3490 decoding happen first, so such errors now appear on stderr, and the return
3491 code is 1. However, the actual setting up of the sockets still happens in
3492 the disconnected process, so errors there are still only recorded on the
3495 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3496 connections (as happens on some IP stacks) was logged at start up time as
3497 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3498 it from "IPv6 and IPv4", which means that two separate sockets are being
3501 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3502 says whether AF_INET or AF_INET6 was passed as an argument.
3504 42. Exiwhat output was messed up when time zones were included in log
3507 43. Exiwhat now gives more information about the daemon's listening ports,
3508 and whether -tls-on-connect was used.
3510 44. The "port" option of the smtp transport is now expanded.
3512 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3513 silently ignored. Now an error message is written to the main and panic
3516 46. There's a new ACL modifier called "logwrite" which writes to a log file
3517 as soon as it is encountered.
3519 47. Added $local_user_uid and $local_user_gid at routing time.
3521 48. Exim crashed when trying to verify a sender address that was being
3524 49. Exim was recognizing only a space character after ".include". It now also
3525 recognizes a tab character.
3527 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3528 extracting the relevant information from the specification. The man page no
3529 longer contains scrambled data for the -d option, and I've added a section
3530 at the front about calling Exim under different names.
3532 51. Added "extra_headers" argument to the "mail" command in filter files.
3534 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3537 53. Installed eximstats 1.29.
3539 54. Added transport_filter_timeout as a generic transport option.
3541 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3542 Cc: header lines. This was required by RFC 822, but it not required by RFC
3545 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3546 incoming messages that did not have them. Now it does so only if the
3547 message originates locally, that is, if there is no associated remote host
3548 address. When Resent- header lines are present, this applies to the Resent-
3549 lines rather than the non-Resent- lines.
3551 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3552 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3554 58. Messages for configuration errors now include the name of the main
3555 configuration file - useful now that there may be more than one file in a
3556 list (.included file names were always shown).
3558 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3559 for those rare installations that do not start the daemon as root or run it
3560 setuid root. I've cut out the call to initgroups() if the daemon is not
3563 60. The Exim user and group can now be bound into the binary as text strings
3564 that are looked up at the start of Exim's processing.
3566 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3568 62. Added $mailstore_basename variable.
3570 63. Installed patch to sieve.c from Michael Haardt.
3572 64. When Exim failed to open the panic log after failing to open the main log,
3573 the original message it was trying to log was written to stderr and debug
3574 output, but if they were not available (the usual case in production), it
3575 was lost. Now it is written to syslog before the two lines that record the
3576 failures to open the logs.
3578 65. Users' Exim filters run in subprocesses under the user's uid. It is
3579 possible for a "deliver" command or an alias in a "personal" command to
3580 provoke an address rewrite. If logging of address rewriting is configured,
3581 this fails because the process is not running as root or exim. There may be
3582 a better way of dealing with this, but for the moment (because 4.30 needs
3583 to be released), I have disabled address rewrite logging when running a
3584 filter in a non-root, non-exim process.
3590 1. The buildconfig auxiliary program wasn't quoting the value set for
3591 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3592 not defined. This bug was masked in 4.22 by the effect that was fixed in
3595 2. Some messages that were rejected after a message id was allocated were
3596 shown as "incomplete" by exigrep. It no longer does this for messages that
3597 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3599 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3600 have allow_domain_literals set, the ID did not get logged in the <= line.
3601 Domain literals are now always recognized in Message-ID: header lines.
3603 4. The first argument for a ${extract expansion item is the key name or field
3604 number. Leading and trailing spaces in this item were not being ignored,
3605 causing some misleading effects.
3607 5. When deliver_drop_privilege was set, single queue runner processes started
3608 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3609 same command in the process it spins off) were not dropping privilege.
3611 6. When the daemon running as "exim" started a queue runner, it always
3612 re-executed Exim in the spun-off process. This is a waste of effort when
3613 deliver_drop_privilege is set. The new process now just calls the
3614 queue-runner function directly.
3620 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3623 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3624 ignored. Though the use of -oP was forcing the writing of a pid file, it
3625 was always written to the default place.
3627 3. If the message "no IP address found for host xxxx" is generated during
3628 incoming verification, it is now followed by identification of the incoming
3629 connection (so you can more easily find what provoked it).
3631 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3633 5. Added some features to "harden" Exim a bit more against certain attacks:
3635 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3636 be put in Local/Makefile. This is like the never_users runtime option,
3637 but it cannot be overridden. The default setting is "root".
3639 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3640 prefix string with which any file named in a -C command line option
3643 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3644 is retained for -C and -D only if the caller of Exim is root. Without
3645 it, the exim user may also use -C and -D and retain privilege.
3647 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3648 command line option is disabled.
3650 6. Macro names set by the -D option must start with an upper case letter, just
3651 like macro names defined in the configuration file.
3653 7. Added "dereference=" facility to LDAP.
3655 8. Two instances of the typo "uknown" in the source files are fixed.
3657 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3658 the Configure-Makefile script screwed up while processing it.
3660 10. Incorporated PCRE 4.4.
3662 11. The SMTP synchronization check was not operating right at the start of an
3663 SMTP session. For example, it could not catch a HELO sent before the client
3664 waited for the greeting. There is now a check for outstanding input at the
3665 point when the greeting is written. Because of the duplex, asynchronous
3666 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3667 way, but not yet received, when the check is performed.
3669 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3670 on TCP/IP sockets, because this apparently causes some broken clients to
3673 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3674 unchanged) from the Cygwin maintainer.
3676 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
3677 was supported instead of testing for mbx. Effectively a typo.
3679 15. The spa authenticator server code was not checking that the input it
3680 received was valid base64.
3682 16. The debug output line for the "set" modifier in ACLs was not showing the
3683 name of the variable that was being set.
3685 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
3687 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
3688 lookup on its own. Something else was needed to trigger the lookup. For
3689 example, a match in host_lookup or the need for a host name in a host list.
3690 Now, if $sender_host_name is referenced and the host name has not yet been
3691 looked up, a lookup is performed. If the lookup fails, the variable remains
3692 empty, and $host_lookup_failed is set to "1".
3694 19. Added "eqi" as a case-independent comparison operator.
3696 20. The saslauthd authentication condition could segfault if neither service
3697 nor realm was specified.
3699 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3700 error message that was logged was misleading, and incoming SMTP
3701 connections were dropped. The message is now more accurate, and temporary
3702 errors are given to SMTP connections.
3704 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3705 (except RSET) with a 421 error, until QUIT is received. However, it was
3706 failing to send a response to QUIT.
3708 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3709 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3710 ACL is now run after the helo_try_verify_hosts code.
3712 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3713 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3714 case-independent, so other case variants are also recognized). Apparently
3715 some systems use these upper case variants.
3717 25. If more than two messages were waiting for the same host, and a transport
3718 filter was specified for the transport, Exim sent two messages over the
3719 same TCP/IP connection, and then failed with "socket operation on non-
3720 socket" when it tried to send the third.
3722 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3724 27. The extern definition of crypt16() in expand.c was not being excluded when
3725 the OS had its own crypt16() function.
3727 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3728 as a preferred synonym for return_size_limit, both as an option and as an
3731 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3733 30. Changed the default configuration ACL to relax the local part checking rule
3734 for addresses that are not in any local domains. For these addresses,
3735 slashes and pipe symbols are allowed within local parts, but the sequence
3736 /../ is explicitly forbidden.
3738 31. SPA server authentication was not clearing the challenge buffer before
3741 32. log_message in a "warn" ACL statement was writing to the reject log as
3742 well as to the main log, which contradicts the documentation and doesn't
3743 seem right (because no rejection is happening). So I have stopped it.
3745 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3746 However, I am unable to do any testing of this.
3748 34. Fixed an infelicity in the appendfile transport. When checking directories
3749 for a mailbox, to see if any needed to be created, it was accidentally
3750 using path names with one or more superfluous leading slashes; tracing
3751 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3753 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3754 contents are added to the log line that is written for every discarded
3755 recipient. (Previously a log_message setting was ignored.)
3757 36. The ${quote: operator now quotes the string if it is empty.
3759 37. The install script runs exim in order to find its version number. If for
3760 some reason other than non-existence or emptiness, which it checks, it
3761 could not run './exim', it was installing it with an empty version number,
3762 i.e. as "exim-". This error state is now caught, and the installation is
3765 38. An argument was missing from the function that creates an error message
3766 when Exim fails to connect to the socket for saslauthd authentication.
3767 This could cause Exim to crash, or give a corrupted message.
3769 39. Added isip, isip4, and isip6 to ${if conditions.
3771 40. The ACL variables $acl_xx are now saved with the message, and can be
3772 accessed later in routers, transports, and filters.
3774 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3775 strings in the file are not string-expanded.
3777 42. If a MAIL command specified a SIZE value that was too large to fit into an
3778 int variable, the check against message_size_limit failed. Such values are
3779 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3780 day this will have to be increased, but I don't think I want to be around
3781 when emails are that large.
3788 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3789 iconv() is not standard in FreeBSD.
3791 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3792 IPv6 enabled. The observed symptom was a segmentation fault on return from
3793 the function os_common_find_running_interfaces() in src/os.c.
3795 3. In the check_special_case() function in daemon.c I had used "errno" as an
3796 argument name, which causes warnings on some systems. This was basically a
3797 typo, since it was named "eno" in the comments!
3799 4. The code that waits for the clock to tick (at a resolution of some fraction
3800 of a second) so as to ensure message-id uniqueness was always waiting for
3801 at least one whole tick, when it could have waited for less. [This is
3802 almost certainly not relevant at current processor speeds, where it is
3803 unlikely to ever wait at all. But we try to future-proof.]
3805 5. The function that sleeps for a time interval that includes fractions of a
3806 second contained a race. It did not block SIGALRM between setting the
3807 timer, and suspending (a couple of lines later). If the interval was short
3808 and the sigsuspend() was delayed until after it had expired, the suspension
3809 never ended. On busy systems this could lead to processes getting stuck for
3812 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3813 process, before it forks any delivery processes. The open lookup caching
3814 mechanism meant that the open file or database connection was passed into
3815 the delivery process. The problem was that delivery processes always tidy
3816 up cached lookup data. This could cause a problem for the next delivery
3817 process started by the queue runner, because the external queue runner
3818 process does not know about the closure. So the next delivery process
3819 still has data in the lookup cache. In the case of a file lookup, there was
3820 no problem because closing a file descriptor in a subprocess doesn't affect
3821 the parent. However, if the lookup was caching a connection to a database,
3822 the connection was closed, and the second delivery process was likely to
3823 see errors such as "PGSQL: query failed: server closed the connection
3824 unexpectedly". The problem has been fixed by closing all cached lookups
3825 in a queue runner before running a delivery process.
3827 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3828 seem to have the "const" qualifier which it has on other OS. I've
3831 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3832 *of the same type* (client or server) with the same public name that an
3833 error should be diagnosed.
3835 9. When Exim looked up a host name for an IP address, but failed to find the
3836 original IP address when looking up the host name (a safety check), it
3837 output the message "<ip address> does not match any IP for NULL", which was
3838 confusing, to say the least. The bug was that the host name should have
3839 appeared instead of "NULL".
3841 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3842 user that is built into the binary, and the -C or -D options is used, root
3843 privilege is dropped before the configuration file is read. In addition,
3844 logging is switched to stderr instead of the normal log files. If the
3845 configuration then re-defines the Exim user, the unprivileged environment
3846 is probably not what is expected, so Exim logs a panic warning message (but
3849 However, if deliver_drop_privilege is set, the unprivileged state may well
3850 be exactly what is intended, so the warning has been cut out in that case,
3851 and Exim is allowed to try to write to its normal log files.
3857 1. smtp_return_error_details was not giving details for temporary sender
3858 or receiver verification errors.
3860 2. Diagnose a configuration error if two authenticators have the same public
3863 3. Exim used not to create the message log file for a message until the first
3864 delivery attempt. This could be confusing when incoming messages were held
3865 for policy or load reasons. The message log file is now created at the time
3866 the message is received, and an initial "Received" line is written to it.
3868 4. The automatically generated man page for command line options had a minor
3869 bug that caused no ill effects; however, a more serious problem was that
3870 the procedure for building the man page automatically didn't always
3871 operate. Consequently, release 4.20 contains an out-of-date version. This
3872 shouldn't happen again.
3874 5. When building Exim with embedded Perl support, the script that builds the
3875 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3876 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3878 6. The freeze_tell option was not being used for messages that were frozen on
3879 arrival, either by an ACL or by local_scan().
3881 7. Added the smtp_incomplete_transaction log selector.
3883 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3884 was accepting AUTH without a new EHLO.
3886 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3887 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3890 10. Logging of TCP/IP connections (when configured) now happens in the main
3891 daemon process instead of the child process, so that the TCP/IP connection
3892 count is more accurate (but it can never be perfect).
3894 11. The use of "drop" in a nested ACL was not being handled correctly in the
3895 outer ACL. Now, if condition failure induced by the nested "drop" causes
3896 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3897 or "require"), the connection is dropped.
3899 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3900 that yield "discard" can now be used with an "accept" or a "discard" verb,
3901 but an error is generated for any others (because I can't see a useful way
3902 to define what should happen).
3904 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3905 are now processed in the same way as lines in the Exim configuration file.
3906 In particular, continuation lines are supported.
3908 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3910 15. Added -ti meaning -t -i.
3912 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3913 domains, and warn by logging if others are found.
3915 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3916 returned from GIFCONF when Exim is trying to find the list of interfaces on
3917 a host. The code in os.c has been modified to copy each ifreq to an aligned
3918 structure in all cases.
3920 Also, in some cases, the returned ifreq's were being copied to a 'struct
3921 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3922 means the last couple of bytes of an IPv6 address could be chopped if the
3923 ifreq contained only a normal sockaddr (14 bytes storage).
3925 18. Named domain lists were not supported in the hosts_treat_as_local option.
3926 An entry such as +xxxx was not recognized, and was treated as a literal
3929 19. Ensure that header lines added by a DATA ACL are included in the reject log
3930 if the ACL subsequently rejects the message.
3932 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3933 MD5 (which is deprecated).
3935 21. When testing a filter file using -bf, Exim was writing a message when it
3936 took the sender from a "From " line in the message, but it was not doing so
3937 when it took $return_path from a Return-Path: header line. It now does.
3939 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3940 with a valid header line field name (a series of printing characters
3941 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3943 23. Changed "disc" in the source to "disk" to conform to the documentation and
3944 the book and for uniformity.
3946 24. Ignore Sendmail's -Ooption=value command line item.
3948 25. When execve() failed while trying to run a command in a pipe transport,
3949 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3950 could be confused with a return value of 69 from the command itself. This
3951 has been changed to 127, the value the shell returns if it is asked to run
3952 a non-existent command. The wording for the related log line suggests a
3953 non-existent command as the problem.
3955 26. If received_header_text expands to an empty string, do not add a Received:
3956 header line to the message. (Well, it adds a token one on the spool, but
3957 marks it "old" so that it doesn't get used or transmitted.)
3959 27. Installed eximstats 1.28 (addition of -nt option).
3961 28. There was no check for failure on the call to getsockname() in the daemon
3962 code. This can fail if there is a shortage of resources on the system, with
3963 ENOMEM, for example. A temporary error is now given on failure.
3965 29. Contrary to the C standard, it seems that in some environments, the
3966 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3967 program. Exim now does this explicitly; it affects the formatting of
3968 timestamps using strftime().
3970 30. If exiqsumm was given junk data, it threw up some uninitialized variable
3971 complaints. I've now initialized all the variables, to avoid this.
3973 32. Header lines added by a system filter were not being "seen" during
3974 transport-time rewrites.
3976 33. The info_callback() function passed to OpenSSL is set up with type void
3977 (*)(SSL *, int, int), as described somewhere. However, when calling the
3978 function (actually a macro) that sets it up, the type void(*)() is
3979 expected. I've put in a cast to prevent warnings from picky compilers.
3981 34. If a DNS black list lookup found a CNAME record, but there were no A
3982 records associated with the domain it pointed at, Exim crashed.
3984 35. If a DNS black list lookup returned more than one A record, Exim ignored
3985 all but the first. It now scans all returned addresses if a particular IP
3986 value is being sought. In this situation, the contents of the
3987 $dnslist_value variable are a list of all the addresses, separated by a
3990 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
3991 to accept a host name and all its aliases if the forward lookup for any of
3992 them yielded the IP address of the incoming connection. Now it accepts only
3993 those names whose forward lookup yields the correct IP address. Any other
3994 names are discarded. This closes a loophole whereby a rogue DNS
3995 administrator could create reverse DNS records to break through a
3996 wildcarded host restriction in an ACL.
3998 37. If a user filter or a system filter that ran in a subprocess used any of
3999 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4000 the wrong values were passed to the pipe command ($thisaddress had the
4001 value of $0, $0 had the value of $1, etc). This bug was introduced by
4002 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4004 38. Improved the line breaking for long SMTP error messages from ACLs.
4005 Previously, if there was no break point between 40 and 75 characters, Exim
4006 left the rest of the message alone. Two changes have been made: (a) I've
4007 reduced the minimum length to 35 characters; (b) if it can't find a break
4008 point between 35 and 75 characters, it looks ahead and uses the first one
4009 that it finds. This may give the occasional overlong line, but at least the
4010 remaining text gets split now.
4012 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4013 file descriptors might be low, and that setting 1000 would always raise it.
4014 It turns out that in some environments, the limit is already over 1000 and
4015 that lowering it causes trouble. So now Exim takes care not to decrease it.
4017 40. When delivering a message, the value of $return_path is set to $sender_
4018 address at the start of routing (routers may change the value). By an
4019 oversight, this default was not being set up when an address was tested by
4020 -bt or -bv, which affected the outcome if any router or filter referred to
4023 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4024 log only when "message" or "log_message" are set. However, if one of the
4025 conditions was an address verification, or a call to a nested ACL, the
4026 messages generated by the underlying test were being passed through. This
4027 no longer happens. The underlying message is available in $acl_verify_
4028 message for both "message" and "log_message" expansions, so it can be
4029 passed through if needed.
4031 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4032 new expansion $bh_ to give the encoded byte string without charset
4033 translation. Translation happens only if iconv() is available; HAVE_ICONV
4034 indicates this at build time. HEADERS_CHARSET gives the charset to
4035 translate to; headers_charset can change it in the configuration, and
4036 "headers charset" can change it in an individual filter file.
4038 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4039 that creates RFC 2047 encoded "words" labels them as that charset instead
4040 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4041 expansion operator; (ii) when Exim creates a From: line for a local
4042 message; (iii) when a header line is rewritten to include a "phrase" part.
4044 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4045 buggy, causing it to skip the first lines of messages whose message ID
4046 ended in 'D'. This would not have bitten before Exim release 4.14, because
4047 message IDs were unlikely to end in 'D' before then. The effect was to have
4048 incorrect size information for certain domains.
4050 45. #include "config.h" was missing at the start of the crypt16.c module. This
4051 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4054 46. If there was a timeout during a "random" callout check, Exim treated it as
4055 a failure of the random address, and carried on sending RSET and the real
4056 address. If the delay was just some slowness somewhere, the response to the
4057 original RCPT would be taken as a response to RSET and so on, causing
4058 mayhem of various kinds.
4060 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4061 when I implemented it. It didn't allow for the fact that some option values
4062 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4063 the right test for positive values.
4065 48. Domain names in DNS records are case-independent. Exim always looks them up
4066 in lower case. Some resolvers return domain names in exactly the case they
4067 appear in the zone file, that is, they may contain uppercase letters. Not
4068 all resolvers do this - some return always lower case. Exim was treating a
4069 change of case by a resolver as a change of domain, similar to a widening
4070 of a domain abbreviation. This triggered its re-routing code and so it was
4071 trying to route what was effectively the same domain again. This normally
4072 caused routing to fail (because the router wouldn't handle the domain
4073 twice). Now Exim checks for this case specially, and just changes the
4074 casing of the domain that it ultimately uses when it transmits the message
4077 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4080 50. If a filter generated a file delivery with a non-absolute name (possible if
4081 no home directory exists for the router), the forbid_file option was not
4084 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4085 the existing equality matching.
4087 52. Exim was using ints instead of ino_t variables in some places where it was
4088 dealing with inode numbers.
4090 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4091 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4092 called TMPDIR, and if it finds it is different, it changes its value.
4094 54. The smtp_printf() function is now made available to local_scan() so
4095 additional output lines can be written before returning. There is also an
4096 smtp_fflush() function to enable the detection of a dropped connection.
4097 The variables smtp_input and smtp_batched_input are exported to
4100 55. Changed the default runtime configuration: the message "Unknown user"
4101 has been removed from the ACL, and instead placed on the localuser router,
4102 using the cannot_route_message feature. This means that any verification
4103 failures that generate their own messages won't get overridden. Similarly,
4104 the "Unrouteable address" message that was in the ACL for unverifiable
4105 relay addresses has also been removed.
4107 56. Added hosts_avoid_esmtp to the smtp transport.
4109 57. The exicyclog script was not checking for the esoteric option
4110 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4111 will work only if exicyclog is run under the appropriate euid.
4113 58. Following a discussion on the list, the rules by which Exim recognises line
4114 endings on incoming messages have been changed. The -dropcr and drop_cr
4115 options are now no-ops, retained only for backwards compatibility. The
4116 following line terminators are recognized: LF CRLF CR. However, special
4117 processing applies to CR:
4119 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4120 nor a local message in the state where . is a terminator.
4122 (ii) If a bare CR is encountered in a header line, an extra space is added
4123 after the line terminator so as not to end the header. The reasoning
4124 behind this is that bare CRs in header lines are most likely either
4125 to be mistakes, or people trying to play silly games.
4127 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4128 was being incorrectly given as 18 bytes larger than it should have been.
4129 This is a VOB (very old bug).
4131 60. This may never have affected anything current, but just in case it has:
4132 When the local host is found other than at the start of a list of hosts,
4133 the local host, those with the same MX, and any that follow, are discarded.
4134 When the list in question was part of a longer list of hosts, the following
4135 hosts (not currently being processed) were also being discarded. This no
4136 longer happens. I'm not sure if this situation could ever has previously
4139 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4142 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4143 header. This is something that is recommended in a new Internet Draft, and
4144 is something that is documented as being done by Sendmail. There are two
4145 possible values. For messages generated by the autoreply transport, Exim
4148 Auto-Submitted: auto-replied
4150 whereas for all other generated messages (e.g. bounces) it adds
4152 Auto-Submitted: auto-generated
4154 63. The "personal" condition in filters now includes a test for the
4155 Auto-Submitted: header. If it contains the string "auto-" the message it
4156 not considered personal.
4158 64. Added rcpt_include_affixes as a generic transport option.
4160 65. Added queue_only_override (default true).
4162 66. Added the syslog_duplication option.
4164 67. If what should have been the first header line of a message consisted of
4165 a space followed by a colon, Exim was mis-interpreting it as a header line.
4166 It isn't of course - it is syntactically invalid and should therefore be
4167 treated as the start of the message body. The misbehaviour could have
4168 caused a number of strange effects, including loss of data in subsequent
4169 header lines, and spool format errors.
4171 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4172 client host had authenticated. This control can now be exercised by an ACL
4173 for more flexibility.
4175 69. By default, callouts do not happen when testing with -bh. There is now a
4176 variant, -bhc, which does actually run the callout code, including
4177 consulting and updating the callout cache.
4179 70. Added support for saslauthd authentication, courtesy of Alexander
4182 71. If statvfs() failed on the spool or log directories while checking their
4183 size for availability, Exim confusingly gave the error "space shortage".
4184 Furthermore, in debugging mode it crashed with a floating point exception.
4185 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4186 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4187 serious problem, Exim now writes to the main and panic logs when this
4188 happens, with details of the failure. It then refuses to accept the
4189 incoming message, giving the message "spool directory problem" or "log
4190 directory problem" with a 421 code for SMTP messages.
4192 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4193 0, 1, and 2 exist, because some OS complain for execs without them (see
4194 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4195 descriptors. However, the code omitted to check that the open succeeded,
4196 causing mysterious errors if for some reason the permissions on /dev/null
4197 got screwed. Now Exim writes a message to the main and panic logs, and
4198 bombs out if it can't open /dev/null.
4200 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4201 interact so that it is all more flexible. It is supposed to remain
4202 backwards compatible. Also added extra_local_interfaces.
4204 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4205 to bomb out with an assertion failure - to the client this appears as a
4206 connection drop. This problem occurs in the part of the code that was taken
4207 from the Samba project. Fortunately, the assertion is in a very simple
4208 function, so I have fixed this by reproducing the function inline in the
4209 one place where it is called, and arranging for authentication to fail
4210 instead of killing the process with assert().
4212 75. The SPA client code was not working when the server requested OEM rather
4213 than Unicode encoding.
4215 76. Added code to make require_files with a specific uid setting more usable in
4216 the case where statting the file as root fails - usually a non-root-mounted
4217 NFS file system. When this happens and the failure is EACCES, Exim now
4218 forks a subprocess and does the per-uid checking as the relevant uid.
4220 77. Added process_log_path.
4222 78. If log_file_path was not explicitly set, a setting of check_log_space or
4223 check_log_inodes was ignored.
4225 79. If a space check for the spool or log partitions fails, the incident is now
4226 logged. Of course, in the latter case the data may get lost...
4228 80. Added the %p formatting code to string_format() so that it can be used to
4229 print addresses in debug_print(). Adjusted all the address printing in the
4230 debugging in store.c to use %p rather than %d.
4232 81. There was a concern that a line of code in smtp_in.c could overflow a
4233 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4234 initially expressed, the concern was not well-founded, because trailing
4235 spaces are removed early. However, if the trailing spaces were followed by
4236 a NULL, they did not get removed, so the overflow was possible. Two fixes
4239 (a) I re-wrote the offending code in a cleaner fashion.
4240 (b) If an incoming SMTP command contains a NULL character, it is rejected
4243 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4244 runs initgroups(), so that if the Exim user is in any additional groups,
4245 they will be used during message reception.
4251 The change log for 4.20 and earlier releases has been archived.