1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.405 2006/10/10 11:15:12 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.
124 SC/01 Use a glob alias rather than an array ref in eximstats generated
125 parser. This improves both readability and performance.
127 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
128 Don't display local sender or destination tables in eximstats unless
129 there is data to show.
130 Added average volumes into the eximstats top table text output.
132 SC/03 Collect data on the number of addresses (recipients) as well
133 as the number of messages in eximstats.
135 TF/01 Correct an error in the documentation for the redirect router. Exim
136 does (usually) call initgroups() when daemonizing.
138 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
139 with consistent privilege compared to when running as a daemon.
141 TF/03 Note in the spec that $authenticated_id is not set for local
142 submissions from trusted users.
144 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
145 Thanks to Dean Brooks <dean@iglou.com> for the patch.
147 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
148 by adding some example configuration directives to the default
149 configuration file. A little bit of work is required to uncomment the
150 directives and define how usernames and passwords are checked, but
151 there is now a framework to start from.
153 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
154 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
155 without this. I don't know how relevant this is to other LDAP libraries.
157 PH/02 Add the verb name to the "unknown ACL verb" error.
159 PH/03 Magnus Holmgren's patch for filter_prepend_home.
161 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
163 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
164 directory not expanded when it should be if an expanded home directory
165 was set for the address (which is overridden by the transport).
167 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
170 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
171 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
172 because it is too late at that time, and has no effect.
174 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
175 security issue with \' (bugzilla #107). I could not use the
176 PQescapeStringConn() function, because it needs a PGconn value as one of
179 PH/08 When testing addresses using -bt, indicate those final addresses that
180 are duplicates that would not cause an additional delivery. At least one
181 person was confused, thinking that -bt output corresponded to deliveries.
182 (Suppressing duplicates isn't a good idea as you lose the information
183 about possibly different redirections that led to the duplicates.)
185 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
186 systems where poll() doesn't work, in particular OS X.
188 PH/10 Added more information to debugging output for retry time not reached.
190 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
191 operations in malware.c.
193 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
196 PH/13 If write_rejectlog was set false when logging was sent to syslog with
197 syslog_duplication set false, log lines that would normally be written
198 both the the main log and to the reject log were not written to syslog at
201 PH/14 In the default configuration, change the use of "message" in ACL warn
202 statements to "add_header".
204 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
205 not followed by a command (e.g. "seen endif").
207 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
208 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
211 PH/17 Added extra conditions to the default value of delay_warning_condition
215 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
216 { match{$h_precedence:}{(?i)bulk|list|junk} } \
217 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
220 The Auto-Submitted: and various List- headers are standardised, whereas I
221 don't think Precedence: ever was.
223 PH/18 Refactored debugging code in route_finduser() to show more information,
224 in particular, the error code if getpwnam() issues one.
226 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
227 This is apparently needed in addition to the PH/07 change above to avoid
228 any possible encoding problems.
230 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
231 but not after initializing Perl.
233 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
234 output them only if debugging. By default they are written stderr,
235 apparently, which is not desirable.
237 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
240 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
243 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
245 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
246 authenticated or an ident call has been made. Suppress the default
247 values for $authenticated_id and $authenticated_sender (but permit -oMai
248 and -oMas) when testing with -bh.
250 PH/24 Re-jigged the order of the tests in the default configuration so that the
251 tests for valid domains and recipients precede the DNS black list and CSA
252 tests, on the grounds that those ones are more expensive.
254 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
255 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
256 This bug exists in every version of Exim that I still have, right back to
259 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
260 However, an attempt to turn on case-sensitivity in a regex key by
261 including (?-i) didn't work because the subject string was already
262 lowercased, and the effects were non-intuitive. It turns out that a
263 one-line patch can be used to allow (?-i) to work as expected.
269 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
270 other effects) broke the use of negated acl sub-conditions.
272 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
275 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
276 "Deny" causes Exim to reject the incoming connection with a 554 error.
277 Unfortunately, if there is a major crisis, such as a disk failure,
278 tcp-wrappers gives "deny", whereas what one would like would be some
279 kind of temporary error. A kludge has been added to help with this.
280 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
281 554 error is used if errno is still zero or contains ENOENT (which occurs
282 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
285 PH/03 Add -lutil to the default FreeBSD LIBS setting.
287 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
288 errors. Otherwise a message that provokes a temporary error (when other
289 messages do not) can cause a whole host to time out.
291 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
292 addresses were routed directly to files or pipes from a redirect router.
293 File deliveries just didn't batch; pipe deliveries might have suffered
296 PH/06 A failure to get a lock for a hints database would erroneously always say
297 "Failed to get write lock", even when it was really a read lock.
299 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
300 of 0600. This has been changed to use the value of the lockfile_mode
301 option (which defaults to 0600).
303 PH/08 Applied small patch from the Sieve maintainer.
305 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
306 folder from quota calculations, a direct delivery into this folder messed
307 up the contents of the maildirsize file. This was because the regex was
308 used only to exclude .Trash (or whatever) when the size of the mailbox
309 was calculated. There was no check that a delivery was happening into an
310 excluded directory. This bug has been fixed by ignoring all quota
311 processing for deliveries into excluded directories.
313 PH/10 Added the maildirfolder_create_regex option to appendfile.
319 PH/01 The code for finding all the local interface addresses on a FreeBSD
320 system running IPv6 was broken. This may well have applied to all BSD
321 systems, as well as to others that have similar system calls. The broken
322 code found IPv4 interfaces correctly, but gave incorrect values for the
323 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
324 that it would not match correctly against @[] and not recognize the IPv6
327 PH/02 The ipliteral router was not recognizing addresses of the form user@
328 [ipv6:....] because it didn't know about the "ipv6:" prefix.
330 PH/03 Added disable_ipv6.
332 PH/04 Changed $reply_address to use the raw form of the headers instead of the
333 decoded form, because it is most often used to construct To: headers
334 lines in autoreplies, and the decoded form may well be syntactically
335 invalid. However, $reply_address has leading white space removed, and all
336 newlines turned into spaces so that the autoreply transport does not
339 PH/05 If group was specified without a user on a router, and no group or user
340 was specified on a transport, the group from the router was ignored.
342 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
343 for visible compile-time settings that can be used to change these
344 numbers, for those that want even more. Backwards compatibility with old
345 spool files has been maintained. However, going back to a previous Exim
346 release will lost any variables that are in spool files.
348 PH/07 Two small changes when running in the test harness: increase delay when
349 passing a TCP/IP connection to a new process, in case the original
350 process has to generate a bounce, and remove special handling of
351 127.0.0.2 (sic), which is no longer necessary.
353 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
354 be the same on different OS.
356 PH/09 Moved a debug statement in filter processing to avoid a race problem when
359 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
360 whether --show-vars was specified or not
362 JJ/02 exipick: Added support for new ACL variable spool format introduced
365 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
366 syntactically invalid From: or Reply-to: line, and a filter used this to
367 generate an autoreply, and therefore failed to obtain an address for the
368 autoreply, Exim could try to deliver to a non-existent relative file
369 name, causing unrelated and misleading errors. What now happens is that
370 it logs this as a hard delivery error, but does not attempt to create a
373 PH/11 The exinext utility has a -C option for testing purposes, but although
374 the given file was scanned by exinext itself; it wasn't being passed on
375 when Exim was called.
377 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
378 an end-of-file indication when reading a command response.
380 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
381 compiled. In many other places in Exim, IPv6 addresses are always
382 recognized, so I have changed this. It also means that IPv4 domain
383 literals of the form [IPV4:n.n.n.n] are now always recognized.
385 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
386 used if the router is not running as root, for example, when verifying at
387 ACL time, or when using -bh. The debugging output from this situation was
388 non-existent - all you got was a failure to exec. I have made two
391 (a) Failures to set uid/gid, the current directory, or a process leader
392 in a subprocess such as that created by queryprogram now generate
393 suitable debugging ouput when -d is set.
395 (b) The queryprogram router detects when it is not running as root,
396 outputs suitable debugging information if -d is set, and then runs
397 the subprocess without attempting to change uid/gid.
399 PH/15 Minor change to Makefile for building test_host (undocumented testing
402 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
403 additional section of a DNS packet that returns MX or SRV records.
404 Instead, it always explicitly searches for A/AAAA records. This avoids
405 major problems that occur when a DNS server includes only records of one
406 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
407 fixed another bug: if SRV records were looked up and the corresponding
408 address records were *not* found in the additional section, the port
409 values from the SRV records were lost.
411 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
412 using the correct key (the original address) when searching the retry
413 rules in order to find which one to use for generating the retry hint.
415 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
416 adding the default one. Similarly, if it contains a Reply-To: header, the
417 errors_reply_to option, if set, is not used.
419 PH/19 When calculating a retry time, Exim used to measure the "time since
420 failure" by looking at the "first failed" field in the retry record. Now
421 it does not use this if it is later than than the arrival time of the
422 message. Instead it uses the arrival time. This makes for better
423 behaviour in cases where some deliveries succeed, thus re-setting the
424 "first failed" field. An example is a quota failure for a huge message
425 when small messages continue to be delivered. Without this change, the
426 "time since failure" will always be short, possible causing more frequent
427 delivery attempts for the huge message than are intended.
428 [Note: This change was subsequently modified - see PH/04 for 4.62.]
430 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
431 $1, $2, $3) because the numerical variables can be reset during some
432 expansion items (e.g. "match"), thereby losing the authentication data.
434 PH/21 Make -bV show the size of off_t variables so that the test suite can
435 decide whether to run tests for quotas > 2G.
437 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
438 mailbox_size, and mailbox_filecount in the appendfile transport. If a
439 filecount value is greater than 2G or if a quota value is greater than 2G
440 on a system where the size of off_t is not greater than 4, a panic error
443 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
444 never match. The debug and -bh output now contains an explicit error
445 message indicating a malformed IPv4 address or mask.
447 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
448 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
451 PH/25 Do not write to syslog when running in the test harness. The only
452 occasion when this arises is a failure to open the main or panic logs
453 (for which there is an explicit test).
455 PH/26 Added the /no_tell option to "control=freeze".
457 PH/27 If a host name lookup failed very early in a connection, for example, if
458 the IP address matched host_lookup and the reverse lookup yielded a name
459 that did not have a forward lookup, an error message of the form "no IP
460 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
461 could be logged. Now it outputs the IP address instead of "NULL".
463 PH/28 An enabling patch from MH: add new function child_open_exim2() which
464 allows the sender and the authenticated sender to be set when
465 submitting a message from within Exim. Since child_open_exim() is
466 documented for local_scan(), the new function should be too.
468 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
469 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
470 results in an empty string is now treated as unset.
472 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
474 PH/31 Added sender_verify_fail logging option.
476 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
477 needed by RFC 822 but not by RFC 2822 was commented out. I have now
478 tidied the source and removed it altogether.
480 PH/33 When a queue run was abandoned because the load average was too high, a
481 log line was always written; now it is written only if the queue_run log
482 selector is set. In addition, the log line for abandonment now contains
483 information about the queue run such as the pid. This is always present
484 in "start" and "stop" lines but was omitted from the "abandon" line.
486 PH/34 Omit spaces between a header name and the colon in the error message that
487 is given when verify = headers_syntax fails (if there are lots of them,
488 the message gets confusing).
490 PH/35 Change the default for dns_check_names_pattern to allow slashes within
491 names, as there are now some PTR records that contain slashes. This check
492 is only to protect against broken name servers that fall over on strange
493 characters, so the fact that it applies to all lookups doesn't matter.
495 PH/36 Now that the new test suite is complete, we can remove some of the
496 special code in Exim that was needed for the old test suite. For example,
497 sorting DNS records because real resolvers return them in an arbitrary
498 order. The new test suite's fake resolver always returns records in the
501 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
502 bounces) except when queue_only is set, to avoid logging races between
503 the different processes.
505 PH/38 Panic-die if .include specifies a non-absolute path.
507 PH/39 A tweak to the "H" retry rule from its user.
509 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
510 a label. They prevented compilation on older perls.
512 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
513 a warning to be raised on newish perls.
515 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
516 on queue. Changes to match documented behaviour of showing count of
517 messages matching specified criteria.
519 PH/40 Changed the default ident timeout from 30s to 5s.
521 PH/41 Added support for the use of login_cap features, on those BSD systems
522 that have them, for controlling the resources used by pipe deliveries.
524 PH/42 The content-scanning code uses fopen() to create files in which to put
525 message data. Previously it was not paying any attention to the mode of
526 the files. Exim runs with umask(0) because the rest of the code creates
527 files with open(), and sets the required mode explicitly. Thus, these
528 files were ending up world-writeable. This was not a big issue, because,
529 being within the spool directory, they were not world-accessible. I have
530 created a function called modefopen, which takes an additional mode
531 argument. It sets umask(777), creates the file, chmods it to the required
532 mode, then resets the umask. All the relevant calls to fopen() in the
533 content scanning code have been changed to use this function.
535 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
536 to 24 hours. This avoids potential overflow problems when processing G
537 and H retry rules. I suspect nobody ever tinkers with this value.
539 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
541 PH/45 When the plaintext authenticator is running as a client, the server's
542 challenges are checked to ensure they are valid base64 strings. By
543 default, the authentication attempt is cancelled if an invalid string is
544 received. Setting client_ignore_invalid_base64 true ignores these errors.
545 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
546 they are received. Thus, the responses can be made to depend on the
547 challenges. If an invalid string is ignored, an empty string is placed in
550 PH/46 Messages that are created by the autoreply transport now contains a
551 References: header, in accordance with RFCs 2822 and 3834.
553 PH/47 Added authenticated_sender_force to the smtp transport.
555 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
557 PH/49 Installed latest patch from the Sieve maintainer.
559 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
560 was also set, the appendfile transport was unnecessarily scanning a
561 directory of message files (e.g. for maildir delivery) to find the count
562 of files (along with the size), even though it did not need this
563 information. It now does the scan only if it needs to find either the
564 size of the count of files.
566 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
568 PH/52 Two bugs concerned with error handling when the smtp transport is
571 (i) Exim was not creating retry information for temporary errors given
572 for individual recipients after the DATA command when the smtp transport
573 was used in LMTP mode. This meant that they could be retried too
574 frequently, and not timed out correctly.
576 (ii) Exim was setting the flag that allows error details to be returned
577 for LMTP errors on RCPT commands, but not for LMTP errors for individual
578 recipients that were returned after the DATA command.
580 PH/53 This is related to PH/52, but is more general: for any failing address,
581 when detailed error information was permitted to be returned to the
582 sender, but the error was temporary, then after the final timeout, only
583 "retry timeout exceeded" was returned. Now it returns the full error as
584 well as "retry timeout exceeded".
586 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
587 do this, and (what is worse) MTAs that accept it.
589 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
590 will now be deprecated.
592 PH/56 New os.c-cygwin from the Cygwin maintainer.
594 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
595 formats (previously only available in exim formats via -bpr, -bpru,
596 and -bpra. Now also available in native and exiqgrep formats)
598 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
599 with very large, slow to parse queues
601 JJ/08 exipick: added ! as generic prefix to negate any criteria format
603 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
605 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
606 responses to authentication challenges, though it was showing the
607 challenges; (ii) I've removed the CR characters from the debug output for
610 PH/58 Allow for the insertion of a newline as well as a space when a string
611 is turned into more than one encoded-word during RFC 2047 encoding. The
612 Sieve code now uses this.
614 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
615 data_4xx, lost_connection, tls_required.
617 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
618 message was being sent as an SMTP response.
620 PH/61 Add -l and -k options to exicyclog.
622 PH/62 When verifying, if an address was redirected to one new address, so that
623 verification continued, and the new address failed or deferred after
624 having set something in $address_data, the value of $address_data was not
625 passed back to the ACL. This was different to the case when no
626 redirection occurred. The value is now passed back in both cases.
628 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
629 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
630 use login_cap.h, so on its own it isn't the distinguishing feature. The
631 new name refers directly to the setclassresources() function.
633 PH/65 Added configuration files for NetBSD3.
635 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
637 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
638 is preferred over IPv4.
640 PH/68 The bounce_return_message and bounce_return_body options were not being
641 honoured for bounces generated during the reception of non-SMTP messages.
642 In particular, this applied to messages rejected by the ACL. This bug has
643 been fixed. However, if bounce_return_message is true and bounce_return_
644 body is false, the headers that are returned for a non-SMTP message
645 include only those that have been read before the error was detected.
646 (In the case of an ACL rejection, they have all been read.)
648 PH/69 The HTML version of the specification is now built in a directory called
649 spec_html instead of spec.html, because the latter looks like a path with
650 a MIME-type, and this confuses some software.
652 PH/70 Catch two compiler warnings in sieve.c.
654 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
655 function verify_get_ident() calls ip_connect() to connect a socket, but
656 if the "connect()" function timed out, ip_connect() used to close the
657 socket. However, verify_get_ident() also closes the socket later, and in
658 between Exim writes to the log, which may get opened at this point. When
659 the socket was closed in ip_connect(), the log could get the same file
660 descriptor number as the socket. This naturally causes chaos. The fix is
661 not to close the socket in ip_connect(); the socket should be closed by
662 the function that creates it. There was only one place in the code where
663 this was missing, in the iplookup router, which I don't think anybody now
664 uses, but I've fixed it anyway.
666 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
667 well as to direct DNS lookups. Otherwise the handling of names in host
668 lists is inconsistent and therefore confusing.
674 PH/01 Two changes to the default runtime configuration:
676 (1) Move the checks for relay_from_hosts and authenticated clients from
677 after to before the (commented out) DNS black list checks.
679 (2) Add control=submission to the relay_from_hosts and authenticated
680 clients checks, on the grounds that messages accepted by these
681 statements are most likely to be submissions.
683 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
685 (1) Generate an error if the third argument for the ${prvs expansion is
688 (2) Treat a missing third argument of ${prvscheck as if it were an empty
691 (3) Reset the variables that are obtained from the first argument of
692 ${prvscheck and used in the second argument before leaving the code,
693 because their memory is reclaimed, so using them afterwards may do
696 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
697 one (it's much easier than Tom thought :-).
699 (5) Because of (4), we can now allow for the use of $prvscheck_result
700 inside the third argument.
702 PH/03 For some reason, the default setting of PATH when running a command from
703 a pipe transport was just "/usr/bin". I have changed it to
706 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
707 anything to be listed in the output from -bV.
709 PH/05 When a filter generated an autoreply, the entire To: header line was
710 quoted in the delivery log line, like this:
712 => >A.N.Other <ano@some.domain> <original@ddress> ...
714 This has been changed so that it extracts the operative address. There
715 may be more than one such address. If so, they are comma-separated, like
718 => >ano@some.domain,ona@other.domain <original@ddress> ...
720 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
721 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
722 not being looked up in the rDNS to get a host name, Exim was showing the
723 IP address twice in Received: lines, even though the IP addresses were
724 identical. For example:
726 Received: from [1.2.3.4] (helo=[1.2.3.4])
728 However, if the real host name was known, it was omitting the HELO data
729 if it matched the actual IP address. This has been tidied up so that it
730 doesn't show the same IP address twice.
732 PH/07 When both +timestamp and +memory debugging was on, the value given by
733 $tod_xxx expansions could be wrong, because the tod_stamp() function was
734 called by the debug printing, thereby overwriting the timestamp buffer.
735 Debugging no longer uses the tod_stamp() function when +timestamp is set.
737 PH/08 When the original message was included in an autoreply transport, it
738 always said "this is a copy of the message, including all the headers",
739 even if body_only or headers_only was set. It now gives an appropriate
742 PH/09 Applied a patch from the Sieve maintainer which:
744 o fixes some comments
745 o adds the (disabled) notify extension core
746 o adds some debug output for the result of if/elsif tests
747 o points to the current vacation draft in the documentation
748 and documents the missing references header update
752 o fixes a bug in processing the envelope test (when testing
753 multiple envelope elements, the last element determinted the
756 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
757 Electronic Mail") by including:
759 Auto-submitted: auto-generated
761 in the messages that it generates (bounce messages and others, such as
762 warnings). In the case of bounce messages for non-SMTP mesages, there was
763 also a typo: it was using "Auto_submitted" (underscore instead of
764 hyphen). Since every message generated by Exim is necessarily in response
765 to another message, thes have all been changed to:
767 Auto-Submitted: auto-replied
769 in accordance with these statements in the RFC:
771 The auto-replied keyword:
773 - SHOULD be used on messages sent in direct response to another
774 message by an automatic process,
776 - MUST NOT be used on manually-generated messages,
778 - MAY be used on Delivery Status Notifications (DSNs) and Message
779 Disposition Notifications (MDNs),
781 - MUST NOT be used on messages generated by automatic or periodic
782 processes, except for messages which are automatic responses to
785 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
786 to the default Received: header definition.
788 PH/12 Added log selector acl_warn_skipped (default on).
790 PH/13 After a successful wildlsearch lookup, discard the values of numeric
791 variables because (a) they are in the wrong storage pool and (b) even if
792 they were copied, it wouldn't work properly because of the caching.
794 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
795 checking when decoding. Apparently there are clients that generate
796 overlong encoded strings. Why am I not surprised?
798 PH/15 If the first argument of "${if match_address" was not empty, but did not
799 contain an "@" character, Exim crashed. Now it writes a panic log message
800 and treats the condition as false.
802 PH/16 In autoreply, treat an empty string for "once" the same as unset.
804 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
805 extension "envelope-auth". The code is finished and in agreement with
806 other implementations, but there is no documentation so far and in fact,
807 nobody wrote the draft yet. This extension is currently #undef'ed, thus
808 not changing the active code.
810 Print executed "if" and "elsif" statements when debugging is used. This
811 helps a great deal to understand what a filter does.
813 Document more things not specified clearly in RFC3028. I had all this
814 sorted out, when out of a sudden new issues came to my mind. Oops."
816 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
819 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
820 canonical form (as documented). However, after a host name lookup from
821 the IP address, check_host() was doing a simple string comparison with
822 addresses acquired from the DNS when checking that the found name did
823 have the original IP as one of its addresses. Since any found IPv6
824 addresses are likely to be in abbreviated form, the comparison could
825 fail. Luckily, there already exists a function for doing the comparison
826 by converting both addresses to binary, so now that is used instead of
829 PH/20 There was another similar case to PH/19, when a complete host name was
830 given in a host list; looking up its IP address could give an abbreviated
831 form, whereas the current host's name might or might not be abbreviated.
832 The same fix has been applied.
838 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
839 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
842 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
843 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
845 PH/03 Typo: missing ".o" in src/pcre/Makefile.
847 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
848 header line, restrict the check to what is listed in RFCs 2369 and 2929.
849 Also, for "Auto-Submitted", treat anything other than "no" as
850 non-personal, in accordance with RFC 3834. (Previously it treated
851 anything starting "auto-" as non-personal.)
853 TF/01 The control=submission/name=... option had a problem with syntax
854 errors if the name included a slash character. The /name= option
855 now slurps the rest of the string, so it can include any characters
856 but it must come last in the list of options (after /sender_retain
859 PH/05 Some modifications to the interface to the fake nameserver for the new
867 TK/01 Added the "success_on_redirect" address verification option. See
868 NewStuff for rationale and an example.
870 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
872 PH/02 Patch to exigrep to allow it to work on syslog lines.
874 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
875 fread() to skip over the body file's header line, because in Cygwin the
876 header line is locked and is inaccessible.
878 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
879 co-exist for some time) to make it clear that it is the Exim ID that is
880 referenced, not the Message-ID: header line.
882 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
883 string_format() function, because snprintf() does not exist on all
886 PH/06 The use of forbid_filter_existstest now also locks out the use of the
887 ${stat: expansion item.
889 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
890 protocol synchronization error", to keep the pedants happy.
892 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
893 well as for IRIX systems, when gcc is being used. See the host.c source
896 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
898 PH/10 Named domain lists were not working if used in a queue_smtp_domains
901 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
902 transport and to the smtp transport in LMTP mode.
904 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
906 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
907 run a filter in a subprocess. This could lead to confusion in subsequent
908 lookups in the parent process. There should also be a search_tidyup() at
909 the end of the subprocess.
911 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
912 only if the host matched helo_try_verify_hosts, which caused the
913 verification to occur when the EHLO/HELO command was issued. The ACL just
914 tested the remembered result. Now, if a previous verification attempt has
915 not happened, "verify = helo" does it there and then.
917 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
919 TK/03 Fix log output including CR from clamd.
921 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
922 exist provoked a memory error which could cause a segfault.
924 PH/15 Installed PCRE 6.2
926 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
928 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
929 of the problem. Specifically, suggested +O2 rather than +O1 for the
932 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
934 PH/20 If a delivery was routed to a non-standard port by means of an SRV
935 record, the port was not correctly logged when the outgoing_port log
936 selector was set (it logged the transort's default port).
938 PH/21 Added support for host-specific ports to manualroute, queryprogram,
939 fallback_hosts, and "hosts" in the smtp transport.
941 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
942 host errors such as "Connection refused".
944 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
945 authentication with radiusclient 0.4.9:
947 - Error returned from rc_read_config was caught wrongly
948 - Username/password not passed on to radius server due to wrong length.
950 The presumption is that some radiusclient API changes for 4.51/PH/17
951 were not taken care of correctly. The code is still untested by me (my
952 Linux distribution still has 0.3.2 of radiusclient), but it was
953 contributed by a Radius user.
955 PH/24 When doing a callout, the value of $domain wasn't set correctly when
956 expanding the "port" option of the smtp transport.
958 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
959 while reading a MIME header. Thanks to Tom Hughes for a patch.
961 PH/24 Include config.h inside local_scan.h so that configuration settings are
964 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
965 that in an ACL for RCPT (for example), you can examine exactly what was
968 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
969 commands, but it was not correctly comparing the address with the actual
970 client host address. Thus, it would show the EHLO address in Received:
971 header lines when this was not necessary.
973 PH/27 Added the % operator to ${eval:}.
975 PH/28 Exim tries to create and chdir to its spool directory when it starts;
976 it should be ignoring failures (because with -C, for example, it has lost
977 privilege). It wasn't ignoring creation failures other than "already
980 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
983 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
984 because an input line was too long, either on its own, or by virtue of
985 too many continuations, the temporary file was not being removed, and the
986 return code was incorrect.
988 PH/31 Missing "BOOL" in function definition in filtertest.c.
990 PH/32 Applied Sieve patches from the maintainer.
992 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
994 PH/33 Added "verify = not_blind".
996 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
997 Local/Makefile (with some defaults set). These are used in built scripts
998 such as exicyclog, but they have never been used in the exim_install
999 script (though there are many overriding facilities there). I have
1000 arranged that the exim_install script now takes note of these two
1003 PH/35 Installed configuration files for Dragonfly.
1005 PH/36 When a locally submitted message by a trusted user did not contain a
1006 From: header, and the sender address was obtained from -f or from an SMTP
1007 MAIL command, and the trusted user did not use -F to supply a sender
1008 name, $originator_name was incorrectly used when constructing a From:
1009 header. Furthermore, $originator_name was used for submission mode
1010 messages from external hosts without From: headers in a similar way,
1011 which is clearly wrong.
1013 PH/37 Added control=suppress_local_fixups.
1015 PH/38 When log_selector = +received_sender was set, and the addition of the
1016 sender made the log line's construction buffer exactly full, or one byte
1017 less than full, an overflow happened when the terminating "\n" was
1020 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1021 when the result of a list match is failure because a DNS lookup failed.
1023 PH/40 RM_COMMAND is now used in the building process.
1025 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1026 the "build-* directories that it finds.
1028 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1029 address in a domain literal was a prefix of an interface address.
1031 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1032 when verifying a sender address, unless rewrite_headers is false.
1034 PH/44 Wrote a long comment about why errors_to addresses are verified as
1035 recipients, not senders.
1037 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1038 the ratelimit ACL was added.
1040 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1042 PH/46 Added extra information about PostgreSQL errors to the error string.
1044 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1045 avoiding the need to install special zones in a real server. This is
1046 backwards compatible; if it can't find the fake resolver, it drops back.
1047 Thus, both old and new test suites can be run.
1049 TF/02 Added util/ratelimit.pl
1051 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1052 clock is set back in time.
1054 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1055 Candler <B.Candler@pobox.com>.
1057 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1058 OK for addresses that are the result of redirections.
1060 PH/48 A number of further additions for the benefit of the new test suite,
1061 including a fake gethostbyname() that interfaces to the fake DNS resolver
1064 TF/06 The fix for widen_domains has also been applied to qualify_single and
1065 search_parents which are the other dnslookup options that can cause
1068 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1071 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1073 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1074 service" if it encountered an MX record with an empty target hostname.
1075 The message is now "an MX or SRV record indicated no SMTP service".
1077 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1078 if the DNS of the sending site is misconfigured. This is quite a
1079 common situation. This change restores the behaviour of treating a
1080 helo verification defer as a failure.
1082 PH/51 If self=fail was set on a router, the bounce message did not include the
1083 actual error message.
1089 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1091 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1092 command itself ended in error, the underlying message about the transport
1093 filter timeout was being overwritten with the pipe command error. Now the
1094 underlying error message should be appended to the second error message.
1096 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1098 PH/02 Reduce the amount of output that "make" produces by default. Full output
1099 can still be requested.
1101 PH/03 The warning log line about a condition test deferring for a "warn" verb
1102 was being output only once per connection, rather than after each
1103 occurrence (because it was using the same function as for successful
1104 "warn" verbs). This seems wrong, so I have changed it.
1106 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1107 it should not have, which might have caused a crash in the right
1108 circumstances, but probably never did.
1110 PH/04 Installed a modified version of Tony Finch's patch to make submission
1111 mode fix the return path as well as the Sender: header line, and to
1112 add a /name= option so that you can make the user's friendly name appear
1115 TF/03 Added the control = fakedefer ACL modifier.
1117 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1118 Mark Lowes for thorough testing.
1120 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1122 TK/03 Merged latest SRS patch from Miles Wilton.
1124 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1125 with the definition in sysexits.h (which is #included earlier).
1126 Fortunately, Exim does not actually use EX_OK. The code used to try to
1127 preserve the sysexits.h value, by assumimg that macro definitions were
1128 scanned for macro replacements. I have been disabused of this notion,
1129 so now the code just undefines EX_OK before #including unistd.h.
1131 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1132 in the smtp transport. When a block could not be written in a single
1133 write() function, the timeout was being re-applied to each part-write.
1134 This seems wrong - if the receiver was accepting one byte at a time it
1135 would take for ever. The timeout is now adjusted when this happens. It
1136 doesn't have to be particularly precise.
1138 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1139 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1141 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1142 without a domain if the check to <postmaster@domain> fails.
1144 SC/01 Eximstats: added -xls and the ability to specify output files
1145 (patch written by Frank Heydlauf).
1147 SC/02 Eximstats: use FileHandles for outputing results.
1149 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1151 SC/04 Eximstats: fixed display of large numbers with -nvr option
1153 SC/05 Eximstats: fixed merging of reports with empty tables.
1155 SC/06 Eximstats: added the -include_original_destination flag
1157 SC/07 Eximstats: removed tabs and trailing whitespace.
1159 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1161 TK/06 MBOX spool code: Add real "From " MBOX separator line
1162 so the .eml file is really in mbox format (even though
1163 most programs do not really care). Patch from Alex Miller.
1165 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1166 The latter is generated from $received_to and is only set if the
1167 message has one envelope recipient. SA can use these headers,
1168 obviously out-of-the-box. Patch from Alex Miller.
1170 PH/08 The ${def test on a variable was returning false if the variable's
1171 value was "0", contrary to what the specification has always said!
1172 The result should be true unless the variable is empty.
1174 PH/09 The syntax error of a character other than { following "${if
1175 def:variable_name" (after optional whitespace) was not being diagnosed.
1176 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1177 accidental colon was present, for example, could give incorrect results.
1179 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1180 result is used (not including appendfile, where other changes are about
1183 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1184 This involved changing a lot of size variables from int to off_t. It
1185 should work with maildirs and everything.
1187 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1188 spamd dying while we are connected to it.
1190 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1193 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1194 function for BDB 4.3.
1196 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1198 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1199 the sources for PCRE have been re-arranged and more files are now
1202 PH/15 The code I had for printing potentially long long variables in PH/11
1203 above was not the best (it lost precision). The length of off_t variables
1204 is now inspected at build time, and an appropriate printing format (%ld
1205 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1206 to be "long long int" or "long int". This is needed for the internal
1207 formatting function string_vformat().
1209 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1210 the configuration file to be ":syslog", then the script "guesses" where
1211 the logs files are, rather than using the compiled in default. In our
1212 case the guess is not the same as the compiled default, so the script
1213 suddenly stopped working when I started to use syslog. The patch checks
1214 to see if log_file_path is "". If so, it attempts to read it from exim
1215 with no configuration file to get the compiled in version, before it
1216 falls back to the previous guessing code."
1218 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1219 implementing BATV in an Exim configuration. See NewStuff for the gory
1222 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1223 Makefile that are specific to HP-UX.
1225 PH/18 If the "use_postmaster" option was set for a recipient callout together
1226 with the "random" option, the postmaster address was used as the MAIL
1227 FROM address for the random test, but not for the subsequent recipient
1228 test. It is now used for both.
1230 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1231 patch removes a few documentation additions to RFC 3028, because the
1232 latest draft now contains them. It adds the new en;ascii-case comparator
1233 and a new error check for 8bit text in MIME parts. Comparator and
1234 require names are now matched exactly. I enabled the subaddress
1235 extension, but it is not well tested yet (read: it works for me)."
1237 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1238 rework some of the code of TK/09 above to avoid the hardwired use of
1239 "%lld" and "long long". Replaced the call to snprintf() with a call to
1242 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1243 records point to non-existent hosts", "retry timeout exceeded", and
1244 "retry time not reached for any host after a long failure period".
1246 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1247 experimental DomainKeys support:
1249 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1250 (2) On an error such as an illegally used "control", the wrong name for
1251 the control was given.
1253 These problems did NOT occur unless DomainKeys support was compiled.
1255 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1257 PH/24 Added ${if match_ip condition.
1259 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1260 databases so that it will be absolutely obvious if a crash occurs in the
1261 DB library. This is a regular occurrence (often caused by mis-matched
1264 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1265 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1266 picked up on a user's system that detects such things. There doesn't seem
1267 to be a gcc warning option for this - only an attribute that has to be
1268 put on the function's prototype. It seems that in Fedora Core 4 they have
1269 set this on a number of new functions. No doubt there will be more in due
1272 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1273 specify a transport. However, if an address that was verified by such a
1274 router was the subject of a callout, Exim crashed because it tried to
1275 read the rcpt_include_affixes from the non-existent transport. Now it
1276 just assumes that the setting of that option is false. This bug was
1277 introduced by 4.51/PH/31.
1279 PH/28 Changed -d+all to exclude +memory, because that information is very
1280 rarely of interest, but it makes the output a lot bigger. People tend to
1281 do -d+all out of habit.
1283 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1284 code in os-type was giving problems when libc.so lives in lib64, like on
1287 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1288 aren't the modern standard, and it seems that some systems' include files
1289 don't always have them. Exim was already checking for some of the newer
1290 ones like T_AAAA, and defining it itself. I've added checks for all the
1291 record types that Exim uses.
1293 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1294 not automatically generating a new one, as it is supposed to. This
1295 prevented TLS from working. If the file did exist, but contained invalid
1296 data, a new version was generated, as expected. It was only the case of a
1297 non-existent file that was broken.
1299 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1300 with a change in libdomainkeys > 0.64.
1302 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1303 from DNS. If the selector record carries the flag, it now has
1304 precedence over the domain-wide flag.
1306 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1308 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1309 the use of a port name that isn't defined in /etc/services) occurred, the
1310 message was deferred as in a normal delivery, and thus remained on the
1311 spool, instead of being failed because of the mua_wrapper setting. This
1312 is now fixed, and I tidied up some of the mua_wrapper messages at the
1315 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1316 the messages in a hash of arrays rather than using individual hashes.
1317 This is a bit cleaner and results in dramatic memory savings, albeit
1318 at a slight CPU cost.
1320 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1321 as requested by Marc Sherman.
1323 SC/10 Eximstats: added histograms for user specified patterns as requested
1326 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1328 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1329 fopen() in the content-scanning modules that did not already have it.
1335 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1336 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1338 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1340 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1342 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1343 to test Sieve filters that use "vacation".
1345 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1346 that changes the way the GnuTLS parameters are stored in the cache file.
1347 The new format can be generated externally. For backward compatibility,
1348 if the data in the cache doesn't make sense, Exim assumes it has read an
1349 old-format file, and it generates new data and writes a new file. This
1350 means that you can't go back to an older release without removing the
1353 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1354 work if there are any delivery delays because "one_time" forces the
1355 parent to be marked "delivered", so its unseen clone is never tried
1356 again. For this reason, Exim now forbids the simultaneous setting of
1359 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1360 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1361 entry if you want to know the details. The fix, however, neglected to
1362 consider the case when local delivery batching is involved. The test for
1363 "previously delivered" was not happening when checking to see if an
1364 address could be batched with a previous (undelivered) one; under
1365 certain circumstances this could lead to multiple deliveries to the same
1368 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1369 in its include files, and this causes problems building Exim.
1371 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1372 header_syntax) but Exim was just ignoring anything given after a slash.
1373 In particular, this caused confusion with an attempt to use "verify =
1374 reverse_host_lookup/defer_ok". An error is now given when options are
1375 supplied for verify items that do not have them. (Maybe reverse_host_
1376 lookup should have a defer_ok option, but that's a different point.)
1378 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1379 defined by RFC 821) to 2048, because there were problems with some AUTH
1380 commands, and RFC 1869 says the size should be increased for extended
1381 SMTP commands that take arguments.
1383 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1386 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1387 "unknown" error; now it says that the functionality isn't in the binary.
1389 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1390 an address' error message when a string expansion fails (syntax or
1391 whatever). Otherwise the password may appear in the log. Following change
1392 PH/42 below, there is no longer a chance of it appearing in a bounce
1395 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1397 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1398 of its addresses was ever tried. (Bugzilla bug #2.)
1400 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1401 the result incorrectly in the debug output. (It correctly added a newline
1402 to what was transported.)
1404 TF/01 Added $received_time.
1406 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1407 commented out examples of how to interface to a virus scanner and to
1408 SpamAssassin. Also added commented examples of av_scanner and
1409 spamd_address settings.
1411 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1412 and controls are allowed in which ACLs. There were a couple of minor
1413 errors. Some of the entries in the conditions table (which is a table of
1414 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1415 negation of where the condition IS allowed.
1417 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1419 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1420 header file does not have a version number, so I've had to invent a new
1421 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1422 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1423 radiusclient), but it was contributed by a Radius user.
1425 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1426 files or directories, for OpenSSL.
1428 PH/19 When an Exim process that is running as root has to create an Exim log
1429 file, it does so in a subprocess that runs as exim:exim so as to get the
1430 ownership right at creation (otherwise, other Exim processes might see
1431 the file with the wrong ownership). There was no test for failure of this
1432 fork() call, which would lead to the process getting stuck as it waited
1433 for a non-existent subprocess. Forks do occasionally fail when resources
1434 run out. I reviewed all the other calls to fork(); they all seem to check
1437 PH/20 When checking for unexpected SMTP input at connect time (before writing
1438 the banner), Exim was not dealing correctly with a non-positive return
1439 from the read() function. If the client had disconnected by this time,
1440 the result was a log entry for a synchronization error with an empty
1441 string after "input=" when read() returned zero. If read() returned -1
1442 (an event I could not check), uninitialized data bytes were printed.
1443 There were reports of junk text (parts of files, etc) appearing after
1446 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1448 PH/22 Added support for macro redefinition, and (re)definition in between
1449 driver and ACL definitions.
1451 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1452 forgetting to use the resulting value; it was using the unexpanded value.
1454 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1455 hadn't been configured. The fix is from Juergen Kreileder, who
1456 understands it better than I do:
1458 "Here's what I see happening with three configured cyrus_sasl
1459 authenticators configured (plain, login, cram-md5):
1461 On startup auth_cyrus_sasl_init() gets called for each of these.
1462 This means three calls to sasl_listmech() without a specified mech_list.
1463 => SASL tests which mechs of all available mechs actually work
1464 => three warnings about OTP not working
1465 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1467 With the patch, sasl_listmech() also gets called three times. But now
1468 SASL's mech_list option is set to the server_mech specified in the the
1469 authenticator. Or in other words, the answer from sasl_listmech()
1470 gets limited to just the mech you're testing for (which is different
1472 => the return list contains just 'plain' or 'login', 'cram-md5' or
1473 nothing depending on the value of ob->server_mech.
1475 I've just tested the patch: Authentication still works fine,
1476 unavailable mechs specified in the exim configuration are still
1477 caught, and the auth.log warnings about OTP are gone."
1479 PH/25 When debugging is enabled, the contents of the command line are added
1480 to the debugging output, even when log_selector=+arguments is not
1483 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1484 answer is "GNU", and only if the return is "GNU/something" is the answer
1487 PH/27 $acl_verify_message is now set immediately after the failure of a
1488 verification in an ACL, and so is available in subsequent modifiers. In
1489 particular, the message can be preserved by coding like this:
1491 warn !verify = sender
1492 set acl_m0 = $acl_verify_message
1494 Previously, $acl_verify_message was set only while expanding "message"
1495 and "log_message" when a very denied access.
1497 PH/28 Modified OS/os.c-Linux with
1499 -#ifndef OS_LOAD_AVERAGE
1500 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1502 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1503 nomenclature these days.)
1505 PH/29 Installed patch from the Sieve maintainer that adds the options
1506 sieve_useraddress and sieve_subaddress to the redirect router.
1508 PH/30 In these circumstances:
1509 . Two addresses routed to the same list of hosts;
1510 . First host does not offer TLS;
1511 . First host accepts first address;
1512 . First host gives temporary error to second address;
1513 . Second host offers TLS and a TLS session is established;
1514 . Second host accepts second address.
1515 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1516 and peerdn, if requested) that were in fact used only for the second
1519 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1520 attention to any local part prefix or suffix that was matched by the
1521 router that accepted the address. It now behaves in the same way as it
1522 does for delivery: the affixes are removed from the local part unless
1523 rcpt_include_affixes is set on the transport.
1525 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1526 timeout during the DATA phase of an incoming message.
1528 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1529 applied a patch sanctioned by the Sieve maintainer.
1531 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1532 the uid or gid is negative. A case of a negative gid caused this to be
1533 noticed. The fix allows for either to be negative.
1535 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1536 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1539 PH/36 The change PH/12 above was broken. Fixed it.
1541 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1542 the grounds that routing the same address twice would always produce the
1543 same answer. This might have been true once, but it is certainly no
1544 longer true now. Routing a child address may depend on the previous
1545 routing that produced that child. Some complicated redirection strategies
1546 went wrong when messages had multiple recipients, and made Exim's
1547 behaviour dependent on the order in which the addresses were given.
1549 I have moved the duplicate checking until after the routing is complete.
1550 Exim scans the addresses that are assigned to local and remote
1551 transports, and removes any duplicates. This means that more work will be
1552 done, as duplicates will always all be routed, but duplicates are
1553 presumably rare, so I don't expect this is of any significance.
1555 For deliveries to pipes, files, and autoreplies, the duplicate checking
1556 still happens during the routing process, since they are not going to be
1559 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1560 It corrects a timeout issue with spamd. This is Ian's comment: "The
1561 background is that sometimes spamd either never reads data from a
1562 connection it has accepted, or it never writes response data. The exiscan
1563 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1564 blindly assumes that writes won't block so it may never time out."
1566 PH/39 Allow G after quota size as well as K and M.
1568 PH/40 The value set for $authenticated_id in an authenticator may not contain
1569 binary zeroes or newlines because the value is written to log lines and
1570 to spool files. There was no check on this. Now the value is run through
1571 the string_printing() function so that such characters are converted to
1572 printable escape sequences.
1574 PH/41 $message_linecount is a new variable that contains the total number of
1575 lines in the message. Compare $body_linecount, which is the count for the
1578 PH/42 Exim no longer gives details of delivery errors for specific addresses in
1579 bounce and delay warning messages, except in certain special cases, which
1582 (a) An SMTP error message from a remote host;
1583 (b) A message specified in a :fail: redirection;
1584 (c) A message specified in a "fail" command in a system filter;
1585 (d) A message specified in a FAIL return from the queryprogram router;
1586 (e) A message specified by the cannot_route_message router option.
1588 In these cases only, Exim does include the error details in bounce and
1589 warning messages. There are also a few cases where bland messages such
1590 as "unrouteable address" or "local delivery error" are given.
1592 PH/43 $value is now also set for the "else" part of a ${run expansion.
1594 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1595 being worked on, but at least Exim now implements the latest version to
1598 PH/45 In a pipe transport, although a timeout while waiting for the pipe
1599 process to complete was treated as a delivery failure, a timeout while
1600 writing the message to the pipe was logged, but erroneously treated as a
1601 successful delivery. Such timeouts include transport filter timeouts. For
1602 consistency with the overall process timeout, these timeouts are now
1603 treated as errors, giving rise to delivery failures by default. However,
1604 there is now a new Boolean option for the pipe transport called
1605 timeout_defer, which, if set TRUE, converts the failures into defers for
1606 both kinds of timeout. A transport filter timeout is now identified in
1609 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1610 systems where "make" and "gmake" are different, calling "gmake" at top
1611 level broke things. I've arranged for the value of $(MAKE) to be passed
1612 from the Makefile to this script so that it can call the same version of
1616 A note about Exim versions 4.44 and 4.50
1617 ----------------------------------------
1619 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1620 changes of various kinds. As a consequence, a big documentation update was
1621 needed. This delayed the release for rather longer than seemed good, especially
1622 in the light of a couple of (minor) security issues. Therefore, the changes
1623 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1624 release. So 4.44 and 4.50 are in effect two different branches that both start
1627 I have left the 4.50 change log unchanged; it contains all the changes since
1628 4.43. The change log for 4.44 is below; many of its items are identical to
1629 those for 4.50. This seems to be the most sensible way to preserve the
1630 historical information.
1636 1. Minor wording change to the doc/README.SIEVE file.
1638 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
1639 computation of the current number of files was incorrect.
1641 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1642 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1645 4. Give more explanation in the error message when the command for a transport
1646 filter fails to execute.
1648 5. There are several places where Exim runs a non-Exim command in a
1649 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1650 was being done only for the command run by the queryprogram router. It is
1651 now done for all such subprocesses. The other cases are: ${run, transport
1652 filters, and the commands run by the lmtp and pipe transports.
1654 6. Added CONFIGURE_GROUP build-time option.
1656 7. Some older OS have a limit of 256 on the maximum number of file
1657 descriptors. Exim was using setrlimit() to set 1000 as a large value
1658 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1659 systems. I've change it so that if it can't get 1000, it tries for 256.
1661 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1662 was an oversight, and furthermore, ever since the addition of extra
1663 controls (e.g. 4.43/32), the checks on when to allow different forms of
1664 "control" were broken. There should now be diagnostics for all cases when a
1665 control that does not make sense is encountered.
1667 9. Added the /retain_sender option to "control=submission".
1669 10. $recipients is now available in the predata ACL (oversight).
1671 11. Tidy the search cache before the fork to do a delivery from a message
1672 received from the command line. Otherwise the child will trigger a lookup
1673 failure and thereby defer the delivery if it tries to use (for example) a
1674 cached ldap connection that the parent has called unbind on.
1676 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1677 of $address_data from the recipient verification was clobbered by the
1678 sender verification.
1680 13. The value of address_data from a sender verification is now available in
1681 $sender_address_data in subsequent conditions in the ACL statement.
1683 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1685 15. Added a new option "connect=<time>" to callout options, to set a different
1688 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1689 was its contents. (It was OK if the option was not defined at all.)
1691 17. A "Completed" log line is now written for messages that are removed from
1692 the spool by the -Mrm option.
1694 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1695 information about exactly what failed.
1697 19. Added -dd to debug only the daemon process.
1699 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1700 handles timeouts, both on the server side and network timeouts. Renamed the
1701 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1703 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1704 It is now set to "smtps".
1706 22. $host_address is now set to the target address during the checking of
1707 ignore_target_hosts.
1709 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1710 being passed; this would have caused $sender_host_name to have been used if
1711 matching the list had actually called for a host name (not very likely,
1712 since this list is usually IP addresses). A host name is now passed as
1715 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1716 code that specifies a non-restarting handler (typically sigaction() in
1717 modern systems) in an attempt to fix a rare and obscure crash bug.
1719 25. Narrowed the window for a race in the daemon that could cause it to ignore
1720 SIGCHLD signals. This is not a major problem, because they are used only to
1721 wake it up if nothing else does.
1723 26. A malformed maildirsize file could cause Exim to calculate negative values
1724 for the mailbox size or file count. Odd effects could occur as a result.
1725 The maildirsize information is now recalculated if the size or filecount
1728 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1729 support for a long time. Removed HAVE_SYS_VFS_H.
1731 28. Installed the latest version of exipick from John Jetmore.
1733 29. In an address list, if the pattern was not a regular expression, an empty
1734 subject address (from a bounce message) matched only if the pattern was an
1735 empty string. Non-empty patterns were not even tested. This was the wrong
1736 because it is perfectly reasonable to use an empty address as part of a
1737 database query. An empty address is now tested by patterns that are
1738 lookups. However, all the other forms of pattern expect the subject to
1739 contain a local part and a domain, and therefore, for them, an empty
1740 address still always fails if the pattern is not itself empty.
1742 30. Exim went into a mad DNS loop when attempting to do a callout where the
1743 host was specified on an smtp transport, and looking it up yielded more
1744 than one IP address.
1746 31. Re-factored the code for checking spool and log partition space into a
1747 function that finds that data and another that does the check. The former
1748 is then used to implement four new variables: $spool_space, $log_space,
1749 $spool_inodes, and $log_inodes.
1751 32. The RFC2047 encoding function was originally intended for short strings
1752 such as real names; it was not keeping to the 75-character limit for
1753 encoded words that the RFC imposes. It now respects the limit, and
1754 generates multiple encoded words if necessary. To be on the safe side, I
1755 have increased the buffer size for the ${rfc2047: expansion operator from
1758 33. It is now permitted to omit both strings after an "if" condition; if the
1759 condition is true, the result is "true". As before, when the second string
1760 is omitted, a false condition yields an empty string. This makes it less
1761 cumbersome to write custom ACL and router conditions.
1763 34. Failure to deliver a bounce message always caused it to be frozen, even if
1764 there was an errors_to setting on the router. The errors_to setting is now
1767 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1768 canonical form (fully expanded) before being placed in
1769 $sender_host_address.
1771 36. The table in the code that translates DNS record types into text (T_A to
1772 "A" for instance) was missing entries for NS and CNAME. It is just possible
1773 that this could have caused confusion if both these types were looked up
1774 for the same domain, because the text type is used as part of Exim's
1775 per-process caching. But the chance of anyone hitting this buglet seems
1778 37. The dnsdb lookup has been extended in a number of ways.
1780 (1) There is a new type, "zns", which walks up the domain tree until it
1781 finds some nameserver records. It should be used with care.
1783 (2) There is a new type, "mxh", which is like "mx" except that it returns
1784 just the host names, not the priorities.
1786 (3) It is now possible to give a list of domains (or IP addresses) to be
1787 looked up. The behaviour when one of the lookups defers can be
1788 controlled by a keyword.
1790 (4) It is now possible to specify the separator character for use when
1791 multiple records are returned.
1793 38. The dnslists ACL condition has been extended: it is now possible to supply
1794 a list of IP addresses and/or domains to be looked up in a particular DNS
1797 39. Added log_selector=+queue_time_overall.
1799 40. When running the queue in the test harness, wait just a tad after forking a
1800 delivery process, to get repeatability of debugging output.
1802 41. Include certificate and key file names in error message when GnuTLS fails
1803 to set them up, because the GnuTLS error message doesn't include the name
1804 of the failing file when there is a problem reading it.
1806 42. Allow both -bf and -bF in the same test run.
1808 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1810 44. The "Exiscan patch" is now merged into the mainline Exim source.
1812 45. Sometimes the final signoff response after QUIT could fail to get
1813 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1814 < 0 before doing a fflush(). This bug looks as though it goes back to the
1815 introduction of TLS in release 3.20, but "sometimes" must have been rare
1816 because the tests only now provoked it.
1818 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1819 (this can affect the format of dates).
1821 47. exim_tidydb, when checking for the continued existence of a message for
1822 which it has found a message-specific retry record, was not finding
1823 messages that were in split spool directories. Consequently, it was
1824 deleting retry records that should have stayed in existence.
1826 48. Steve fixed some bugs in eximstats.
1828 49. The SPA authentication driver was not abandoning authentication and moving
1829 on to the next authenticator when an expansion was forced to fail,
1830 contradicting the general specification for all authenticators. Instead it
1831 was generating a temporary error. It now behaves as specified.
1833 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1834 (the order specifies the preference for clients). The order is now AES256,
1835 AES128, 3DES, ARCFOUR128.
1837 51. Small patch to Sieve code - explicitly set From: when generating an
1840 52. Exim crashed if a remote delivery caused a very long error message to be
1841 recorded - for instance if somebody sent an entire SpamAssassin report back
1842 as a large number of 550 error lines. This bug was coincidentally fixed by
1843 increasing the size of one of Exim's internal buffers (big_buffer) that
1844 happened as part of the Exiscan merge. However, to be on the safe side, I
1845 have made the code more robust (and fixed the comments that describe what
1848 53. Now that there can be additional text after "Completed" in log lines (if
1849 the queue_time_overall log selector is set), a one-byte patch to exigrep
1850 was needed to allow it to recognize "Completed" as not the last thing in
1853 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1854 patch that reportedly fixes this has been added. I am not expert enough to
1855 create a test for it. This is what the patch creator wrote:
1857 "I found a little strange behaviour of ldap code when working with
1858 Windows 2003 AD Domain, where users was placed in more than one
1859 Organization Units. When I tried to give exim partial DN, the exit code
1860 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1861 But simultaneously result of request was absolutely normal ldap result,
1862 so I produce this patch..."
1864 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1865 REFERENCE, so I have modified the code to exclude the patch when that macro
1868 55. Some experimental protocols are using DNS PTR records for new purposes. The
1869 keys for these records are domain names, not reversed IP addresses. The
1870 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1871 leaves it alone. Component reversal etc. now happens only for IP addresses.
1874 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1876 57. Double the size of the debug message buffer (to 2048) so that more of very
1877 long debug lines gets shown.
1879 58. The exicyclog utility now does better if the number of log files to keep
1880 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1882 59. Two changes related to the smtp_active_hostname option:
1884 (1) $smtp_active_hostname is now available as a variable.
1885 (2) The default for smtp_banner uses $smtp_active_hostname instead
1886 of $primary_hostname.
1888 60. The host_aton() function is supposed to be passed a string that is known
1889 to be a valid IP address. However, in the case of IPv6 addresses, it was
1890 not checking this. This is a hostage to fortune. Exim now panics and dies
1891 if the condition is not met. A case was found where this could be provoked
1892 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1893 components; fortuitously, this particular loophole had already been fixed
1894 by change 4.50/55 above.
1896 If there are any other similar loopholes, the new check in host_aton()
1897 itself should stop them being exploited. The report I received stated that
1898 data on the command line could provoke the exploit when Exim was running as
1899 exim, but did not say which command line option was involved. All I could
1900 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1901 running as the user.
1904 61. There was a buffer overflow vulnerability in the SPA authentication code
1905 (which came originally from the Samba project). I have added a test to the
1906 spa_base64_to_bits() function which I hope fixes it.
1909 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1910 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1912 63. The daemon start-up calls getloadavg() while still root for those OS that
1913 need the first call to be done as root, but it missed one case: when
1914 deliver_queue_load_max is set with deliver_drop_privilege. This is
1915 necessary for the benefit of the queue runner, because there is no re-exec
1916 when deliver_drop_privilege is set.
1918 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1919 This has been fixed.
1921 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1922 was in use, was not putting the data itself into the right store pool;
1923 consequently, it could be overwritten for a subsequent message in the same
1924 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1927 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1929 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1930 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1931 the function were treating the return as a boolean value, which happened to
1932 work because 0=false and not-0=true, but is not correct code.
1934 68. The host_aton() function was not handling scoped IPv6 addresses (those
1935 with, for example, "%eth0" on the end) correctly.
1937 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1938 negated items (that is, ~something) in unsigned ints. Some compilers
1939 apparently mutter when there is no cast.
1941 70. If an address verification called from an ACL failed, and did not produce a
1942 user-specific message (i.e. there was only a "system" message), nothing was
1943 put in $acl_verify_message. In this situation, it now puts the system
1946 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1947 session; change 4.31/43 added the unwanted input to the log line - except
1948 that it did not do this in the start of session case. It now does.
1950 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1951 This is wrong and can cause the other end to generate a synchronization
1952 error if it is another Exim or anything else that does the synchronization
1953 check. A QUIT command is no longer sent after a timeout.
1955 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1956 during host lookups.
1958 74. The defer_ok option of callout verification was not working if it was used
1959 when verifying addresses in header lines, that is, for this case:
1961 verify = header_sender/callout=defer_ok
1963 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1964 those file descriptors could be used for SMTP connections. If anything
1965 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1966 could be sent to the SMTP client, causing chaos. The daemon now opens
1967 stdin, stdout, and stderr to /dev/null when it puts itself into the
1970 76. Arrange for output from Perl's "warn" command to be written to Exim's main
1971 log by default. The user can override this with suitable Perl magic.
1973 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
1974 the log message when discard triggers, was not working for the DATA ACL or
1975 for the non-SMTP ACL.
1977 78. Error message wording change in sieve.c.
1979 79. If smtp_accept_max_per_host was set, the number of connections could be
1980 restricted to fewer than expected, because the daemon was trying to set up
1981 a new connection before checking whether the processes handling previous
1982 connections had finished. The check for completed processes is now done
1983 earlier. On busy systems, this bug wouldn't be noticed because something
1984 else would have woken the daemon, and it would have reaped the completed
1987 80. If a message was submitted locally by a user whose login name contained one
1988 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1989 It caused a spool format error. I have fixed the spool reading code. A
1990 related problem was that the "from" clause in the Received: line became
1991 illegal because of the space(s). It is now covered by ${quote_local_part.
1993 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
1996 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
1998 83. Patch from Sieve maintainer to fix unterminated string problem in
1999 "vacation" handling.
2001 84. Some minor changes to the Linux configuration files to help with other
2002 OS variants using glibc.
2004 85. One more patch for Sieve to update vacation handling to latest spec.
2007 ----------------------------------------------------
2008 See the note above about the 4.44 and 4.50 releases.
2009 ----------------------------------------------------
2015 1. Change 4.43/35 introduced a bug that caused file counts to be
2016 incorrectly computed when quota_filecount was set in an appendfile
2019 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2020 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2023 3. Give more explanation in the error message when the command for a transport
2024 filter fails to execute.
2026 4. There are several places where Exim runs a non-Exim command in a
2027 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2028 was being done only for the command run by the queryprogram router. It is
2029 now done for all such subprocesses. The other cases are: ${run, transport
2030 filters, and the commands run by the lmtp and pipe transports.
2032 5. Some older OS have a limit of 256 on the maximum number of file
2033 descriptors. Exim was using setrlimit() to set 1000 as a large value
2034 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2035 systems. I've change it so that if it can't get 1000, it tries for 256.
2037 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2038 was an oversight, and furthermore, ever since the addition of extra
2039 controls (e.g. 4.43/32), the checks on when to allow different forms of
2040 "control" were broken. There should now be diagnostics for all cases when a
2041 control that does not make sense is encountered.
2043 7. $recipients is now available in the predata ACL (oversight).
2045 8. Tidy the search cache before the fork to do a delivery from a message
2046 received from the command line. Otherwise the child will trigger a lookup
2047 failure and thereby defer the delivery if it tries to use (for example) a
2048 cached ldap connection that the parent has called unbind on.
2050 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2051 of $address_data from the recipient verification was clobbered by the
2052 sender verification.
2054 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2055 was its contents. (It was OK if the option was not defined at all.)
2057 11. A "Completed" log line is now written for messages that are removed from
2058 the spool by the -Mrm option.
2060 12. $host_address is now set to the target address during the checking of
2061 ignore_target_hosts.
2063 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2064 being passed; this would have caused $sender_host_name to have been used if
2065 matching the list had actually called for a host name (not very likely,
2066 since this list is usually IP addresses). A host name is now passed as
2069 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2070 code that specifies a non-restarting handler (typically sigaction() in
2071 modern systems) in an attempt to fix a rare and obscure crash bug.
2073 15. Narrowed the window for a race in the daemon that could cause it to ignore
2074 SIGCHLD signals. This is not a major problem, because they are used only to
2075 wake it up if nothing else does.
2077 16. A malformed maildirsize file could cause Exim to calculate negative values
2078 for the mailbox size or file count. Odd effects could occur as a result.
2079 The maildirsize information is now recalculated if the size or filecount
2082 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2083 support for a long time. Removed HAVE_SYS_VFS_H.
2085 18. Updated exipick to current release from John Jetmore.
2087 19. Allow an empty sender to be matched against a lookup in an address list.
2088 Previously the only cases considered were a regular expression, or an
2091 20. Exim went into a mad DNS lookup loop when doing a callout where the
2092 host was specified on the transport, if the DNS lookup yielded more than
2095 21. The RFC2047 encoding function was originally intended for short strings
2096 such as real names; it was not keeping to the 75-character limit for
2097 encoded words that the RFC imposes. It now respects the limit, and
2098 generates multiple encoded words if necessary. To be on the safe side, I
2099 have increased the buffer size for the ${rfc2047: expansion operator from
2102 22. Failure to deliver a bounce message always caused it to be frozen, even if
2103 there was an errors_to setting on the router. The errors_to setting is now
2106 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2107 canonical form (fully expanded) before being placed in
2108 $sender_host_address.
2110 24. Updated eximstats to version 1.33
2112 25. Include certificate and key file names in error message when GnuTLS fails
2113 to set them up, because the GnuTLS error message doesn't include the name
2114 of the failing file when there is a problem reading it.
2116 26. Expand error message when OpenSSL has problems setting up cert/key files.
2119 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2120 (this can affect the format of dates).
2122 28. exim_tidydb, when checking for the continued existence of a message for
2123 which it has found a message-specific retry record, was not finding
2124 messages that were in split spool directories. Consequently, it was
2125 deleting retry records that should have stayed in existence.
2127 29. eximstats updated to version 1.35
2128 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2129 1.35 - bugfix such that pie charts by volume are generated correctly
2131 30. The SPA authentication driver was not abandoning authentication and moving
2132 on to the next authenticator when an expansion was forced to fail,
2133 contradicting the general specification for all authenticators. Instead it
2134 was generating a temporary error. It now behaves as specified.
2136 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2137 (the order specifies the preference for clients). The order is now AES256,
2138 AES128, 3DES, ARCFOUR128.
2140 31. Small patch to Sieve code - explicitly set From: when generating an
2143 32. Exim crashed if a remote delivery caused a very long error message to be
2144 recorded - for instance if somebody sent an entire SpamAssassin report back
2145 as a large number of 550 error lines. This bug was coincidentally fixed by
2146 increasing the size of one of Exim's internal buffers (big_buffer) that
2147 happened as part of the Exiscan merge. However, to be on the safe side, I
2148 have made the code more robust (and fixed the comments that describe what
2151 33. Some experimental protocols are using DNS PTR records for new purposes. The
2152 keys for these records are domain names, not reversed IP addresses. The
2153 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2154 leaves it alone. Component reversal etc. now happens only for IP addresses.
2157 34. The host_aton() function is supposed to be passed a string that is known
2158 to be a valid IP address. However, in the case of IPv6 addresses, it was
2159 not checking this. This is a hostage to fortune. Exim now panics and dies
2160 if the condition is not met. A case was found where this could be provoked
2161 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2162 components; fortuitously, this particular loophole had already been fixed
2163 by change 4.50/55 or 4.44/33 above.
2165 If there are any other similar loopholes, the new check in host_aton()
2166 itself should stop them being exploited. The report I received stated that
2167 data on the command line could provoke the exploit when Exim was running as
2168 exim, but did not say which command line option was involved. All I could
2169 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2170 running as the user.
2173 35. There was a buffer overflow vulnerability in the SPA authentication code
2174 (which came originally from the Samba project). I have added a test to the
2175 spa_base64_to_bits() function which I hope fixes it.
2178 36. The daemon start-up calls getloadavg() while still root for those OS that
2179 need the first call to be done as root, but it missed one case: when
2180 deliver_queue_load_max is set with deliver_drop_privilege. This is
2181 necessary for the benefit of the queue runner, because there is no re-exec
2182 when deliver_drop_privilege is set.
2184 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2185 was in use, was not putting the data itself into the right store pool;
2186 consequently, it could be overwritten for a subsequent message in the same
2187 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2190 38. Sometimes the final signoff response after QUIT could fail to get
2191 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2192 < 0 before doing a fflush(). This bug looks as though it goes back to the
2193 introduction of TLS in release 3.20, but "sometimes" must have been rare
2194 because the tests only now provoked it.
2200 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2201 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2202 Now it returns an integer. A number of calls were still expecting a T/F
2203 return. Fortuitously, in all cases, the tests worked in OK situations,
2204 which is the norm. However, things would have gone wrong on any write
2205 failures on the smtp file descriptor. This function is used when sending
2206 messages over SMTP and also when doing verify callouts.
2208 2. When Exim is called to do synchronous delivery of a locally submitted
2209 message (the -odf or -odi options), it no longer closes stderr before doing
2212 3. Implemented the mua_wrapper option.
2214 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2216 5. Implemented the functions header_remove(), header_testname(),
2217 header_add_at_position(), and receive_remove_recipient(), and exported them
2220 6. If an ACL "warn" statement specified the addition of headers, Exim already
2221 inserted X-ACL-Warn: at the start if there was no header name. However, it
2222 was not making this test for the second and subsequent header lines if
2223 there were newlines in the string. This meant that an invalid header could
2224 be inserted if Exim was badly configured.
2226 7. Allow an ACL "warn" statement to add header lines at the start or after all
2227 the Received: headers, as well as at the end.
2229 8. Added the rcpt_4xx retry error code.
2231 9. Added postmaster_mailfrom=xxx to callout verification option.
2233 10. Added mailfrom=xxxx to the callout verification option, for verify=
2236 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2237 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2239 12. Inserted some casts to stop certain compilers warning when using pointer
2240 differences as field lengths or precisions in printf-type calls (mostly
2241 affecting debugging statements).
2243 13. Added optional readline() support for -be (dynamically loaded).
2245 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2246 same clock tick as a message's arrival, so that its received time was the
2247 same as the "first fail" time on the retry record, and that message
2248 remained on the queue past the ultimate address timeout, every queue runner
2249 would try a delivery (because it was past the ultimate address timeout) but
2250 after another failure, the ultimate address timeout, which should have then
2251 bounced the address, did not kick in. This was a "< instead of <=" error;
2252 in most cases the first failure would have been in the next clock tick
2253 after the received time, and all would be well.
2255 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2256 being recognized when the domain list was tested by the match_domain
2257 condition in an expansion string.
2259 16. Added the ${str2b64: operator.
2261 17. Exim was always calling setrlimit() to set a large limit for the number of
2262 processes, without checking whether the existing limit was already
2263 adequate. (It did check for the limit on file descriptors.) Furthermore,
2264 errors from getrlimit() and setrlimit() were being ignored. Now they are
2265 logged to the main and panic logs, but Exim does carry on, to try to do its
2266 job under whatever limits there are.
2268 18. Imported PCRE 5.0.
2270 19. Trivial typo in log message " temporarily refused connection" (the leading
2273 20. If the log selector return_path_on_delivery was set and an address was
2274 redirected to /dev/null, the delivery process crashed because it assumed
2275 that a return path would always be set for a "successful" delivery. In this
2276 case, the whole delivery is bypassed as an optimization, and therefore no
2279 21. Internal re-arrangement: the function for sending a challenge and reading
2280 a response while authentication was assuming a zero-terminated challenge
2281 string. It's now changed to take a pointer and a length, to allow for
2282 binary data in such strings.
2284 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2286 23. Exim was not respecting finduser_retries when seeking the login of the
2287 uid under which it was called; it was always trying 10 times. (The default
2288 setting of finduser_retries is zero.) Also, it was sleeping after the final
2289 failure, which is pointless.
2291 24. Implemented tls_on_connect_ports.
2293 25. Implemented acl_smtp_predata.
2295 26. If the domain in control=submission is set empty, Exim assumes that the
2296 authenticated id is a complete email address when it generates From: or
2297 Sender: header lines.
2299 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2300 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2301 chown and chgrp in /bin and hostname in /usr/bin.
2303 28. Exim was keeping the "process log" file open after each use, just as it
2304 does for the main log. This opens the possibility of it remaining open for
2305 long periods when the USR1 signal hits a daemon. Occasional processlog
2306 errors were reported, that could have been caused by this. Anyway, it seems
2307 much more sensible not to leave this file open at all, so that is what now
2310 29. The long-running daemon process does not normally write to the log once it
2311 has entered its main loop, and it closes the log before doing so. This is
2312 so that log files can straightforwardly be renamed and moved. However,
2313 there are a couple of unusual error situations where the daemon does write
2314 log entries, and I had neglected to close the log afterwards.
2316 30. The text of an SMTP error response that was received during a remote
2317 delivery was being truncated at 512 bytes. This is too short for some of
2318 the long messages that one sometimes sees. I've increased the limit to
2321 31. It is now possible to make retry rules that apply only when a message has a
2322 specific sender, in particular, an empty sender.
2324 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2325 it possible to be selective about when SMTP synchronization is enforced.
2327 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2329 32. Implemented hosts_connection_nolog.
2331 33. Added an ACL for QUIT.
2333 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2336 35. Added mailbox_size and mailbox_filecount to appendfile.
2338 36. Added control = no_multiline_responses to ACLs.
2340 37. There was a bug in the logic of the code that waits for the clock to tick
2341 in the case where the clock went backwards by a substantial amount such
2342 that the microsecond fraction of "now" was more than the microsecond
2343 fraction of "then" (but the whole seconds number was less).
2345 38. Added support for the libradius Radius client library this is found on
2346 FreeBSD (previously only the radiusclient library was supported).
2352 1. When certain lookups returned multiple values in the form name=value, the
2353 quoting of the values was not always being done properly. Specifically:
2354 (a) If the value started with a double quote, but contained no whitespace,
2356 (b) If the value contained whitespace other than a space character (i.e.
2357 tabs or newlines or carriage returns) it was not quoted.
2358 This fix has been applied to the mysql and pgsql lookups by writing a
2359 separate quoting function and calling it from the lookup code. The fix
2360 should probably also be applied to nisplus, ibase and oracle lookups, but
2361 since I cannot test any of those, I have not disturbed their existing code.
2363 2. A hit in the callout cache for a specific address caused a log line with no
2364 reason for rejecting RCPT. Now it says "Previous (cached) callout
2365 verification failure".
2367 3. There was an off-by-one bug in the queryprogram router. An over-long
2368 return line was truncated at 256 instead of 255 characters, thereby
2369 overflowing its buffer with the terminating zero. As well as fixing this, I
2370 have increased the buffer size to 1024 (and made a note to document this).
2372 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2373 when Exim is waiting for an SMTP response from a remote server, Exim
2374 restarts its select() call on the socket, thereby resetting its timeout.
2375 This is not a problem when such interrupts are rare. Somebody set up a cron
2376 job to run exiwhat every 2 minutes, which is less than the normal select()
2377 timeout (5 or 10 minutes). This meant that the select() timeout never
2378 kicked in because it was always reset. I have fixed this by comparing the
2379 time when an interrupt arrives with the time at the start of the first call
2380 to select(). If more time than the timeout has elapsed, the interrupt is
2381 treated as a timeout.
2383 5. Some internal re-factoring in preparation for the addition of Sieve
2384 extensions (by MH). In particular, the "personal" test is moved to a
2385 separate function, and given an option for scanning Cc: and Bcc: (which is
2386 not set for Exim filters).
2388 6. When Exim created an email address using the login of the caller as the
2389 local part (e.g. when creating a From: or Sender: header line), it was not
2390 quoting the local part when it contained special characters such as @.
2392 7. Installed new OpenBSD configuration files.
2394 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2395 try to make them clearer.
2397 9. Callout options, other than the timeout value, were being ignored when
2398 verifying sender addresses in header lines. For example, when using
2400 verify = header_sender/callout=no_cache
2402 the cache was (incorrectly) being used.
2404 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2405 only the Cygwin environment.
2407 11. When return_path_on_delivery was set as a log selector, if different remote
2408 addresses in the same message used different return paths and parallel
2409 remote delivery occurred, the wrong values would sometimes be logged.
2410 (Whenever a remote delivery process finished, the return path value from
2411 the most recently started remote delivery process was logged.)
2413 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2414 lines when AUTH and/or TLS are in use. This is the "received protocol"
2415 field. Exim used to use "asmtp" for authenticated SMTP, without any
2416 indication (in the protocol name) for TLS use. Now it follows the RFC and
2417 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2418 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2419 names appear in log lines as well as in Received: header lines.
2421 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2422 extensions, and comparison tests, and to fix some bugs.
2424 14. Changes to the "personal" filter test:
2426 (1) The test was buggy in that it was just doing the equivalent of
2427 "contains" tests on header lines. For example, if a user's address was
2428 anne@some.where, the "personal" test would incorrectly be true for
2430 To: susanne@some.where
2432 This test is now done by extracting each address from the header in turn,
2433 and checking the entire address. Other tests that are part of "personal"
2434 are now done using regular expressions (for example, to check local parts
2435 of addresses in From: header lines).
2437 (2) The list of non-personal local parts in From: addresses has been
2438 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2439 taken from the Sieve specification recommendations.
2441 (3) If the message contains any header line starting with "List-" it is
2442 treated as non-personal.
2444 (4) The test for "circular" in the Subject: header line has been removed
2445 because it now seems ill-conceived.
2447 15. Minor typos in src/EDITME comments corrected.
2449 16. Installed latest exipick from John Jetmore.
2451 17. If headers_add on a router specified a text string that was too long for
2452 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2453 of string_sprintf() is now avoided.
2455 18. $message_body_size was not set (it was always zero) when running the DATA
2456 ACL and the local_scan() function.
2458 19. For the "mail" command in an Exim filter, no default was being set for
2459 the once_repeat time, causing a random time value to be used if "once" was
2460 specified. (If the value happened to be <= 0, no repeat happened.) The
2461 default is now 0s, meaning "never repeat". The "vacation" command was OK
2462 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2463 (I found it when inspecting the code).
2465 20. There is now an overall timeout for performing a callout verification. It
2466 defaults to 4 times the callout timeout, which applies to individual SMTP
2467 commands during the callout. The overall timeout applies when there is more
2468 than one host that can be tried. The timeout is checked before trying the
2469 next host. This prevents very long delays if there are a large number of
2470 hosts and all are timing out (e.g. when the network connections are timing
2471 out). The value of the overall timeout can be changed by specifying an
2472 additional sub-option for "callout", called "maxwait". For example:
2474 verify = sender/callout=5s,maxwait=20s
2476 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2477 to the end before writing, but this should make it even safer).
2479 22. Exim was forgetting that it had advertised PIPELINING for the second and
2480 subsequent messages on an SMTP connection. It was also not resetting its
2481 memory on STARTTLS and an internal HELO.
2483 23. When Exim logs an SMTP synchronization error within a session, it now
2484 records whether PIPELINING has been advertised or not.
2486 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2487 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2488 rather than long int.
2490 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2492 26. Added the never_mail option to autoreply.
2498 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2499 crash if the getsockname() call failed; this can happen if a connection is
2500 closed very soon after it is established. The problem was simply in the
2501 order in which certain operations were done, causing Exim to try to write
2502 to the SMTP stream before it had set up the file descriptor. The bug has
2503 been fixed by making things happen in the correct order.
2509 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2510 before the connection was closed, thus losing the rejection response.
2512 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2513 some early Solaris releases, but causes trouble in current releases where
2514 socklen_t is defined.
2516 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2519 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2522 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2523 that releasing the top of it at the end releases what was used for sub-
2524 expansions (unless the block got too big). However, discard this block if
2525 the first thing is a variable or header, so that we can use its block when
2526 it is dynamic (useful for very large $message_headers, for example).
2528 6. Lookups now cache *every* query, not just the most recent. A new, separate
2529 store pool is used for this. It can be recovered when all lookup caches are
2530 flushed. Lookups now release memory at the end of their result strings.
2531 This has involved some general refactoring of the lookup sources.
2533 7. Some code has been added to the store_xxx() functions to reduce the amount
2534 of flapping under certain conditions.
2536 8. log_incoming_interface used to affect only the <= reception log lines. Now
2537 it causes the local interface and port to be added to several more SMTP log
2538 lines, for example "SMTP connection from", and rejection lines.
2540 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2542 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2544 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2545 could be overwritten at the end of the current message (or the start of a
2546 new message if it was set in a HELO ACL). The value is now preserved for
2547 the duration of the SMTP connection.
2549 12. If a transport had a headers_rewrite setting, and a matching header line
2550 contained an unqualified address, that address was qualified, even if it
2551 did not match any rewriting rules. The underlying bug was that the values
2552 of the flags that permit the existence of unqualified sender and recipient
2553 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2554 non-local messages, and by -bnq for local messages) were not being
2555 preserved with the message after it was received.
2557 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2558 "next input=" as part of the text comprising the host identity instead of
2559 the correct text. The code was using the same buffer for two different
2560 strings. However, depending on which order the printing function evaluated
2561 its arguments, the bug did not always show up. Under Linux, for example, my
2562 test suite worked just fine.
2564 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2565 doesn't work with some older versions of Perl. It has been changed to "my",
2566 which in any case is probably the better facility to use.
2568 15. A really picky compiler found some instances of statements for creating
2569 error messages that either had too many or two few arguments for the format
2572 16. The size of the buffer for calls to the DNS resolver has been increased
2573 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2574 for addresses that have a lot of PTR records. This alleviates a problem; it
2575 does not fully solve it.
2577 17. A dnsdb lookup for PTR records that receives more data than will fit in the
2578 buffer now truncates the list and logs the incident, which is the same
2579 action as happens when Exim is looking up a host name and its aliases.
2580 Previously in this situation something unpredictable would happen;
2581 sometimes it was "internal error: store_reset failed".
2583 18. If a server dropped the connection unexpectedly when an Exim client was
2584 using GnuTLS and trying to read a response, the client delivery process
2585 crashed while trying to generate an error log message.
2587 19. If a "warn" verb in an ACL added multiple headers to a message in a single
2588 string, for example:
2590 warn message = H1: something\nH2: something
2592 the text was added as a single header line from Exim's point of view
2593 though it ended up OK in the delivered message. However, searching for the
2594 second and subsequent header lines using $h_h2: did not work. This has been
2595 fixed. Similarly, if a system filter added multiple headers in this way,
2596 the routers could not see them.
2598 20. Expanded the error message when iplsearch is called with an invalid key to
2599 suggest using net-iplsearch in a host list.
2601 21. When running tests using -bh, any delays imposed by "delay" modifiers in
2602 ACLs are no longer actually imposed (and a message to that effect is
2605 22. If a "gecos" field in a passwd entry contained escaped characters, in
2606 particular, if it contained a \" sequence, Exim got it wrong when building
2607 a From: or a Sender: header from that name. A second bug also caused
2608 incorrect handling when an unquoted " was present following a character
2609 that needed quoting.
2611 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2612 was not being matched caselessly.
2614 24. Arranged for all hyphens in the exim.8 source to be escaped with
2617 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2618 properly. Recipient callout cache records were still being keyed to include
2619 the sender, even when use_sender was set false. This led to far more
2620 callouts that were necessary. The sender is no longer included in the key
2621 when use_sender is false.
2623 26. Added "control = submission" modifier to ACLs.
2625 27. Added the ${base62d: operator to decode base 62 numbers.
2627 28. dnsdb lookups can now access SRV records.
2629 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2630 the configuration file.
2632 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2633 (-v) mode. This makes the output for a verbose queue run more intelligible.
2635 31. Added a use_postmaster feature to recipient callouts.
2637 32. Added the $body_zerocount variable, containing the number of binary zero
2638 bytes in the message body.
2640 33. The time of last modification of the "new" subdirectory is now used as the
2641 "mailbox time last read" when there is a quota error for a maildir
2644 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2646 35. Added +ignore_unknown as a special item in host lists.
2648 36. Code for decoding IPv6 addresses in host lists is now included, even if
2649 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2650 address was recognized as an IP address, but was then not correctly decoded
2651 into binary, causing unexpected and incorrect effects when compared with
2658 1. Very minor rewording of debugging text in manualroute to say "list of
2659 hosts" instead of "hostlist".
2661 2. If verify=header_syntax was set, and a header line with an unqualified
2662 address (no domain) and a large number of spaces between the end of the
2663 name and the colon was received, the reception process suffered a buffer
2664 overflow, and (when I tested it) crashed. This was caused by some obsolete
2665 code that should have been removed. The fix is to remove it!
2667 3. When running in the test harness, delay a bit after writing a bounce
2668 message to get a bit more predictability in the log output.
2670 4. Added a call to search_tidyup() just before forking a reception process. In
2671 theory, someone could use a lookup in the expansion of smtp_accept_max_
2672 per_host which, without the tidyup, could leave open a database connection.
2674 5. Added the variables $recipient_data and $sender_data which get set from a
2675 lookup success in an ACL "recipients" or "senders" condition, or a router
2676 "senders" option, similar to $domain_data and $local_part_data.
2678 6. Moved the writing of debug_print from before to after the "senders" test
2681 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2682 problems for message scanning, either using a data ACL, or using
2683 local_scan() because the Received: header was not generated till after they
2684 were called (in order to set the time as the time of reception completion).
2685 I have revised the way this works. The header is now generated after the
2686 body is received, but before the ACL or local_scan() are called. After they
2687 are run, the timestamp in the header is updated.
2693 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2694 before starting a queue runner without re-exec. This happened only when
2695 deliver_drop_privilege was set or when the Exim user was set to root. The
2696 effect of the bug was that timeouts during subsequent deliveries caused
2697 crashes instead of being properly handled. The handler is now left at its
2698 default (and expected) setting.
2700 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2701 message, again when deliver_drop_privilege is set or Exim is run as root.
2702 The bug described in (1) was not present in this case, but the tidying up
2703 of the other signals was missing. I have made the two cases consistent.
2705 3. The ignore_target_hosts setting on a manualroute router was being ignored
2706 for hosts that were looked up using the /MX notation.
2708 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2711 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2712 operated on the sender address. After changing the $sender_address to <>
2713 for the sender address verify, Exim was re-instated it as the original
2714 (before rewriting) address, but remembering that it had rewritten it, so it
2715 wasn't rewriting it again. This bug also had the effect of breaking the
2716 sender address verification caching when the sender address was rewritten.
2718 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2719 This has been changed so that if the ip literal address matches
2720 ignore_target_hosts, the router declines.
2722 7. Added expansion conditions match_domain, match_address, and match_local_
2723 part (NOT match_host).
2725 8. The placeholder for the Received: header didn't have a length field set.
2727 9. Added code to Exim itself and to exim_lock to test for a specific race
2728 condition that could lead to file corruption when using MBX delivery. The
2729 issue is with the lockfile that is created in /tmp. If this file is removed
2730 after a process has opened it but before that process has acquired a lock,
2731 there is the potential for a second process to recreate the file and also
2732 acquire a lock. This could lead to two Exim processes writing to the file
2733 at the same time. The added code performs the same test as UW imapd; it
2734 checks after acquiring the lock that its file descriptor still refers to
2735 the same named file.
2737 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2738 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2741 11. Added the smtp_active_hostname option. If used, this will typically be made
2742 to depend on the incoming interface address. Because $interface_address is
2743 not set up until the daemon has forked a reception process, error responses
2744 that can happen earlier (such as "too many connections") no longer contain
2747 12. If an expansion in a condition on a "warn" statement fails because a lookup
2748 defers, the "warn" statement is abandoned, and the next ACL statement is
2749 processed. Previously this caused the whole ACL to be aborted.
2751 13. Added the iplsearch lookup type.
2753 14. Added ident_timeout as a log selector.
2755 15. Added tls_certificate_verified as a log selector.
2757 16. Added a global option tls_require_ciphers (compare the smtp transport
2758 option of the same name). This controls incoming TLS connections.
2760 17. I finally figured out how to make tls_require_ciphers do a similar thing
2761 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2762 before starting the TLS session.
2764 18. Tabs are now shown as \t in -bP output.
2766 19. If the log selector return_path_on_delivery was set, Exim crashed when
2767 bouncing a message because it had too many Received: header lines.
2769 20. If two routers both had headers_remove settings, and the first one included
2770 a superfluous trailing colon, the final name in the first list and the
2771 first name in the second list were incorrectly joined into one item (with a
2772 colon in the middle).
2778 1. Added -C and -D options to the exinext utility, mainly to make it easier
2779 to include in the automated testing, but these could be helpful when
2780 multiple configurations are in use.
2782 2. The exinext utility was not formatting the output nicely when there was
2783 an alternate port involved in the retry record key, nor when there was a
2784 message id as well (for retries that were specific to a specific message
2785 and a specific host). It was also confused by IPv6 addresses, because of
2786 the additional colons they contain. I have fixed the IPv4 problem, and
2787 patched it up to do a reasonable job for IPv6.
2789 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2790 delivery, the log line now contains "pipelined" if PIPELINING was used.
2792 4. An SMTP transport process used to panic and die if the bind() call to set
2793 an explicit outgoing interface failed. This has been changed; it is now
2794 treated in the same way as a connect() failure.
2796 5. A reference to $sender_host_name in the part of a conditional expansion
2797 that was being skipped was still causing a DNS lookup. This no longer
2800 6. The def: expansion condition was not recognizing references to header lines
2801 that used bh_ and bheader_.
2803 7. Added the _cache feature to named lists.
2805 8. The code for checking quota_filecount in the appendfile transport was
2806 allowing one more file than it should have been.
2808 9. For compatibility with Sendmail, the command line option
2816 and sets the incoming protocol and host name (for trusted callers). The
2817 host name and its colon can be omitted when only the protocol is to be set.
2818 Note the Exim already has two private options, -pd and -ps, that refer to
2819 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2820 "s", but I don't think that's a major issue.
2822 10. A number of refactoring changes to the code, none of which should affect
2825 (a) The number of logging options was getting close to filling up the
2826 32-bit word that was used as a bit map. I have split them into two classes:
2827 those that are passed in the argument to log_write(), and those that are
2828 only ever tested independently outside of that function. These are now in
2829 separate 32-bit words, so there is plenty of room for expansion again.
2830 There is no change in the user interface or the logging behaviour.
2832 (b) When building, for example, log lines, the code previously used a
2833 macro that called string_cat() twice, in order to add two strings. This is
2834 not really sufficiently general. Furthermore, there was one instance where
2835 it was actually wrong because one of the argument was used twice, and in
2836 one call a function was used. (As it happened, calling the function twice
2837 did not affect the overall behaviour.) The macro has been replaced by a
2838 function that can join an arbitrary number of extra strings onto a growing
2841 (c) The code for expansion conditions now uses a table and a binary chop
2842 instead of a serial search (which was left over from when there were very
2843 few conditions). Also, it now recognizes conditions like "pam" even when
2844 the relevant support is not compiled in: a suitably worded error message is
2845 given if an attempt is made to use such a condition.
2847 11. Added ${time_interval:xxxxx}.
2849 12. A bug was causing one of the ddress fields not to be passed back correctly
2850 from remote delivery subprocesses. The field in question was not being
2851 subsequently used, so this caused to problems in practice.
2853 13. Added new log selectors queue_time and deliver_time.
2855 14. Might have fixed a bug in maildirsizefile handling that threw up
2856 "unexpected character" debug warnings, and recalculated the data
2857 unnecessarily. In any case, I expanded the warning message to give more
2860 15. Added the message "Restricted characters in address" to the statements in
2861 the default ACL that block characters like @ and % in local parts.
2863 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2864 Three changes have been made:
2866 (a) There was a serious bug; a negative response to MAIL caused the whole
2867 recipient domain to be cached as invalid, thereby blocking all messages
2868 to all local parts at the same domain, from all senders. This bug has
2869 been fixed. The domain is no longer cached after a negative response to
2870 MAIL if the sender used is not empty.
2872 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2875 (c) A new callout option, "use_sender" has been added for people who want
2876 the modified behaviour.
2882 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2885 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2886 indeed breaks things for older releases.
2888 3. Added additional logging to the case where there is a problem reading data
2889 from a filter that is running in a subprocess using a pipe, in order to
2890 try to track down a specific problem.
2892 4. Testing facility fudge: when running in the test harness and attempting
2893 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2894 getting "No route to host". Convert this to a timeout.
2896 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2899 6. Some OS don't have socklen_t but use size_t instead. This affects the
2900 fifth argument of getsockopt() amongst other things. This is now
2901 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2902 can be set for individual OS. I have set it for SunOS5, OSF1, and
2903 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2904 some earlier ones do not.
2906 7. Change 4.30/15 was not doing the test caselessly.
2908 8. The standard form for an IPv6 address literal was being rejected by address
2909 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2910 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2911 this, as well as the form without the "IPv6" on the front (but only when
2912 address literals are enabled, of course).
2914 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2916 10. Exim crashed if a message with an empty sender address specified by -f
2917 encountered a router with an errors_to setting. This could be provoked only
2918 by a command such as
2922 where an empty string was supplied; "<>" did not hit this bug.
2924 11. Installed PCRE release 4.5.
2926 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2927 remained set. It is now erased.
2929 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2930 times from message ids (which are base 36 rather than the normal 62).
2932 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2933 were being counted as actual protocol errors, and logged if the log
2934 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2935 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2936 and DATA following a set of rejected RCPTs do not count as protocol errors.
2937 In other words, Exim assumes they were pipelined, though this may not
2938 actually be the case. Of course, in all cases the client gets an
2939 appropriate error code.
2941 15. If a lookup fails in an ACL condition, a message about the failure may
2942 be available; it is used if testing the ACL cannot continue, because most
2943 such messages specify what the cause of the deferral is. However, some
2944 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2945 that caused an old message to be retained and used if a later statement
2946 caused a defer, replacing the real cause of the deferral.
2948 16. If an IP address had so many PTR records that the DNS lookup buffer
2949 was not large enough to hold them, Exim could crash while trying to process
2950 the truncated data. It now detects and logs this case.
2952 17. Further to 4.21/58, another change has been made: if (and only if) the
2953 first line of a message (the first header line) ends with CRLF, a bare LF
2954 in a subsequent header line has a space inserted after it, so as not to
2955 terminate the header.
2957 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
2958 unnecessarily, used atoi() instead of strtol(), and didn't check the
2959 termination when getting file sizes from file names by regex.
2961 19. Completely re-implemented the support for maildirsize files, in the light
2962 of a number of problems with the previous contributed implementation
2963 (4.30/29). In particular:
2965 . If the quota is zero, the maildirsize file is maintained, but no quota is
2968 . If the maildir directory does not exist, it is created before any attempt
2969 to write a maildirsize file.
2971 . The quota value in the file is just a cache; if the quota is changed in
2972 the transport, the new value overrides.
2974 . A regular expression is available for excluding directories from the
2977 20. The autoreply transport checks the characters in options that define the
2978 message's headers; it allows continued headers, but it was checking with
2979 isspace() after an embedded newline instead of explicitly looking for a
2982 21. If all the "regular" hosts to which an address was routed had passed their
2983 expiry times, and had not reached their retry times, the address was
2984 bounced, even if fallback hosts were defined. Now Exim should go on to try
2987 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2988 equivalent code in the SMTP transport. Some hosts send humungous responses
2989 to HELO/EHLO, more than 1024 it seems.
2991 23. Refactoring: code in filter.c used (void *) for "any old type" but this
2992 gives compiler warnings in some environments. I've now done it "properly",
2995 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2996 (because of problems with the built-in one) was declared to return uschar *
2997 instead of char *, causing compiler failure.
2999 25. Fixed a file descriptor leak when processing alias/forward files.
3001 26. Fixed a minor format string issue in dbfn.c.
3003 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3005 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3006 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3009 29. When a sender address is verified, it is cached, to save repeating the test
3010 when there is more than one recipient in a message. However, when the
3011 verification involves a callout, it is possible for different callout
3012 options to be set for different recipients. It is too complicated to keep
3013 track of this in the cache, so now Exim always runs a verification when a
3014 callout is required, relying on the callout cache for the optimization.
3015 The overhead is duplication of the address routing, but this should not be
3018 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3019 to be verified with callout=postmaster, and the main callout worked but the
3020 postmaster check failed, the verification correctly failed. However, if a
3021 subsequent RCPT command asked for sender verification *without* the
3022 postmaster check, incorrect caching caused this verification also to fail,
3025 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3026 it was not caching the DNS options (qualify_single, search_parents) that
3027 were used when the lookup failed. A subsequent lookup with different
3028 options therefore always gave the same answer, though there were cases
3029 where it should not have. (Example: a "domains = !$mx_any" option on a
3030 dnslookup router: the "domains" option is always processed without any
3031 widening, but the router might have qualify_single set.) Now Exim uses the
3032 cached value only when the same options are set.
3034 32. Added John Jetmore's "exipick" utility to the distribution.
3036 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3037 than a timeout (e.g. a certificate is required, and is not provided), an
3038 Exim server now closes the connection immediately. Previously it waited for
3039 the client to close - but if the client is SSL, it seems that they each
3040 wait for each other, leading to a delay before one of them times out.
3042 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3043 maintained 0.8.x compatibility because I don't think many are using it, and
3044 it is clearly obsolete.
3046 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3049 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3050 client certificate was expired. A simple patch fixes this, though I don't
3051 understand the full logic of why the verify callback is called multiple
3054 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3055 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3056 which causes problems with some clients (such as the Certicom SSL Plus
3057 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3058 disables the coutermeasure allowing Eudora to connect."
3060 38. Exim was not checking that a write() to a log file succeeded. This could
3061 lead to Bad Things if a log got too big, in particular if it hit a file
3062 size limit. Exim now panics and dies if it cannot write to a log file, just
3063 as it does if it cannot open a log file.
3065 39. Modified OS/Makefile-Linux so that it now contains
3067 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3069 The two -D definitions ensure that Exim is compiled with large file
3070 support, which makes it possible to handle log files that are bigger than
3073 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3074 instance) a domain was checked against a named list that involved a lookup,
3075 causing $domain_data to be set, then another domain was checked against the
3076 same list, then the first domain was re-checked, the value of $domain_data
3077 after the final check could be wrong. In particular, if the second check
3078 failed, it could be set empty. This bug probably also applied to
3081 41. The strip_trailing_dot option was not being applied to the address given
3082 with the -f command-line option.
3084 42. The code for reading a message's header from the spool was incrementing
3085 $received_count, but never initializing it. This meant that the value was
3086 incorrect (doubled) while delivering a message in the same process in which
3087 it was received. In the most common configuration of Exim, this never
3088 happens - a fresh exec is done - but it can happen when
3089 deliver_drop_privilege is set.
3091 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3092 it now includes up to 150 characters of the unexpected data in the log
3095 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3096 and building data strings. The size of both of these buffers was 10 000
3097 bytes - far larger than anybody would *ever* want, thought I. Needless to
3098 say, somebody hit the limit. I have increased the maximum line length to
3099 20 000 and the maximum data length of concatenated lines to 100 000. I have
3100 also fixed two bugs, because there was no checking on these buffers. Tsk,
3101 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3102 buffer is too small.
3104 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3105 lsearch lookups. Now it does.
3107 46. When parsing a route_list item in a manualroute router, a fixed-length
3108 buffer was used for the list of hosts. I made this 1024 bytes long,
3109 thinking that nobody would ever have a list of hosts that long. Wrong.
3110 Somebody had a whole pile of complicated expansion conditions, and the
3111 string was silently truncated, leading to an expansion error. It turns out
3112 that it is easier to change to an unlimited length (owing to other changes
3113 that have happened since this code was originally written) than to build
3114 structure for giving a limitation error. The length of the item that
3115 expands into the list of hosts is now unlimited.
3117 47. The lsearch lookup could not handle data where the length of text line was
3118 more than 4095 characters. Such lines were truncated, leading to shortened
3119 data being returned. It should now handle lines of any length.
3121 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3122 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3125 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3126 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3127 no longer happens in comment lines. A list of replacements is now placed
3128 at the head of all of the source files, except those whose only change is
3129 to replace PERL_COMMAND in the very first #! line.
3131 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3132 messages on the queue, with a bottom-up merge sort, using code contributed
3133 by Michael Haardt. This should make operations like -bp somewhat faster on
3134 large queues. It won't affect queue runners, except when queue_run_in_order
3137 51. Installed eximstats 1.31 in the distribution.
3139 52. Added support for SRV lookups to the dnslookup router.
3141 53. If an ACL referred to $message_body or $message_body_end, the value was not
3142 reset for any messages that followed in the same SMTP session.
3144 54. The store-handling optimization for building very long strings was not
3145 differentiating between the different store pools. I don't think this
3146 actually made any difference in practice, but I've tidied it.
3148 55. While running the routers to verify a sender address, $sender_address
3149 was still set to the sender address. This is wrong, because when routing to
3150 send a bounce to the sender, it would be empty. Therefore, I have changed
3151 it so that, while verifying a sender address, $sender_address is set to <>.
3152 (There is no change to what happens when verifying a recipient address.)
3154 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3155 target A or AAAA records (if not already returned) without resetting the
3156 qualify_single or search_parents options of the DNS resolver. These are
3157 inappropriate in this case because the targets of MX and SRV records must
3158 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3159 target that, when qualified, matched something in the local domain. These
3160 two options are now turned off when doing these lookups.
3162 57. It seems that at least some releases of Reiserfs (which does not have the
3163 concept of a fixed number of inodes) returns zero and not -1 for the
3164 number of available inodes. This interacted badly with check_spool_inodes,
3165 which assumed that -1 was the "no such thing" setting. What I have done is
3166 to check that the total number of inodes is greater than zero before doing
3167 the test of how many are available.
3169 58. When a "warn" ACL statement has a log_message modifier, the message is
3170 remembered, and not repeated. This is to avoid a lot of repetition when a
3171 message has many recipients that cause the same warning to be written.
3172 Howewer, Exim was preserving the list of already written lines for an
3173 entire SMTP session, which doesn't seem right. The memory is now reset if a
3174 new message is started.
3176 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3177 debugging output unless log_rewrite was also set.
3179 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3180 of (int)(handle) when we know that handle contains (void *)(-1).
3182 61. The Exim daemon panic-logs an error return when it closes the incoming
3183 connection. However "connection reset by peer" seems to be common, and
3184 isn't really an error worthy of noting specially, so that particular error
3187 62. When Exim is trying to find all the local interfaces, it used to panic and
3188 die if the ioctl to get the interface flags failed. However, it seems that
3189 on at least one OS (Solaris 9) it is possible to have an interface that is
3190 included in the list of interfaces, but for which you get a failure error
3191 for this call. This happens when the interface is not "plumbed" into a
3192 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3193 failure of the "get flags" call assumes that the interface is down.
3195 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3196 makes life easier for people who are doing arithmetic on fields extracted
3197 from dates, where you often get leading zeros that should not be
3198 interpreted as octal.
3200 64. Added qualify_domain to the redirect router, to override the global
3203 65. If a pathologically long header line contained very many addresses (the
3204 report of this problem mentioned 10 000) and each of them was rewritten,
3205 Exim could use up a very large amount of memory. (It kept on making new
3206 copies of the header line as it rewrote, and never released the old ones.)
3207 At the expense of a bit more processing, the header rewriting function has
3208 been changed so that it no longer eats memory in this way.
3210 66. The generation of the Received: header has been moved from the time that a
3211 message starts to be received, to the time that it finishes. The timestamp
3212 in the Received: header should now be very close to that of the <= log
3213 line. There are two side-effects of this change:
3215 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3216 logged header lines no longer include the local Received: line, because
3217 it has not yet been created. The same applies to a copy of the message
3218 that is returned to a non-SMTP sender when a message is rejected.
3220 (b) When a filter file is tested using -bf, no additional Received: header
3221 is added to the test message. After some thought, I decided that this
3224 This change does not affect the value of $received_for. It is still set
3225 after address rewriting, but before local_scan() is called.
3227 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3229 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3230 gave an unhelpful panic error message, and a defer error. I have managed to
3231 change this behaviour so that it now rejects any supplied certificate,
3232 which seems right, as the list of acceptable certificates is empty.
3234 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3235 gave an unhelpful defer error. I have not managed to make this reject any
3236 supplied certificates, but the error message it gives is "no certificate
3237 supplied", which is not helpful.
3239 70. exigrep's output now also includes lines that are not associated with any
3240 message, but which match the given pattern. Implemented by a patch from
3241 Martin Sluka, which also tidied up the Perl a bit.
3243 71. Recipient callout verification, like sender verification, was using <> in
3244 the MAIL FROM command. This isn't really the right thing, since the actual
3245 sender may affect whether the remote host accepts the recipient or not. I
3246 have changed it to use the actual sender in the callout; this means that
3247 the cache record is now keyed on a recipient/sender pair, not just the
3248 recipient address. There doesn't seem to be a real danger of callout loops,
3249 since a callout by the remote host to check the sender would use <>.
3250 [SEE ABOVE: changed after hitting problems.]
3252 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3253 temporary errors. However, in the case of such a code being given after
3254 the end of a data transmission (i.e. after ".") Exim was failing to write
3255 a retry record for the message. (Yes, there was some broken host that was
3256 actually sending 8xx at this point.)
3258 73. An unknown lookup type in a host list could cause Exim to panic-die when
3259 the list was checked. (An example that provoked this was putting <; in the
3260 middle of a list instead of at the start.) If this happened during a DATA
3261 ACL check, a -D file could be left lying around. This kind of configuration
3262 error no longer causes Exim to die; instead it causes a defer errror. The
3263 incident is still logged to the main and panic logs.
3265 74. Buglet left over from Exim 3 conversion. The message "too many messages
3266 in one connection" was written to the rejectlog but not the mainlog, except
3267 when address rewriting (yes!) was being logged.
3269 75. Added write_rejectlog option.
3271 76. When a system filter was run not as root (that is, when system_filter_user
3272 was set), the values of the $n variables were not being returned to the
3273 main process; thus, they were not subsequently available in the $sn
3276 77. Added +return_path_on_delivery log selector.
3278 78. A connection timeout was being treated differently from recipients deferred
3279 when testing hosts_max_try with a message that was older than the host's
3280 retry timeout. (The host should not be counted, thus allowing all hosts to
3281 be tried at least once before bouncing.) This may have been the cause of an
3282 occasionally reported bug whereby a message would remain on the queue
3283 longer than the retry timeout, but would be bounced if a delivery was
3284 forced. I say "may" because I never totally pinned down the problem;
3285 setting up timeout/retry tests is difficult. See also the next item.
3287 79. The ultimate address timeout was not being applied to errors that involved
3288 a combination of host plus message (for example, a timeout on a MAIL
3289 command). When an address resolved to a number of possible hosts, and they
3290 were not all tried for each delivery (e.g. because of hosts_max_try), a
3291 message could remain on the queue longer than the retry timeout.
3293 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3296 81. Fixed an obscure SMTP outgoing bug which required at least the following
3297 conditions: (a) there was another message waiting for the same server;
3298 (b) the server returned 5xx to all RCPT commands in the first message so
3299 that the message was not completed; (c) the server dropped the connection
3300 or gave a negative response to the RSET that Exim sends to abort the
3301 transaction. The observed case was a dropped connection after DATA that had
3302 been sent in pipelining mode. That is, the server had advertised PIPELINING
3303 but was not implementing it correctly. The effect of the bug was incorrect
3304 behaviour, such as trying another host, and this could lead to a crash.
3310 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3311 and daemon.c were passed as pointers to ints; they should have been
3312 pointers to socklen_t variables (which are typically unsigned ints).
3314 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3317 3. Fixed a really odd bug that affected only the testing scheme; patching a
3318 certain fixed string in the binary changed the value of another string that
3319 happened to be identical to the end of the original first string.
3321 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3322 name", it returns that address as the IP address. On some operating
3323 systems (e.g. Solaris), it also passes back the IP address string as the
3324 "host name". However, on others (e.g. Linux), it passes back an empty
3325 string. Exim wasn't checking for this, and was changing the host name to an
3326 empty string, assuming it had been canonicized.
3328 5. Although rare, it is permitted to have more than one PTR record for a given
3329 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3330 all the names associated with an address, because they do in Solaris.
3331 However, it seems that they do not in Linux for data that comes from the
3332 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3333 I found this out when I moved to a new Linux workstation and tried to run
3334 the Exim test suite.
3336 To get round this problem I have changed the code so that it now does its
3337 own call to the DNS to look up PTR records when searching for a host name.
3338 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3339 addresses that are only in /etc/hosts are still found.
3341 This behaviour is, however, controlled by an option called host_lookup_
3342 order, which defaults to "bydns:byaddr". If people want to use the other
3343 order, or indeed, just use one or the other means of lookup, they can
3344 specify it in this variable.
3346 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3347 some operating systems, this comes back from gethostbyaddr() as an empty
3348 string, and this is what Exim used to test for. However, it seems that in
3349 other systems, "." is yielded. Exim now tests for this case too.
3351 7. The values of check_spool_space and check_log_space are now held internally
3352 as a number of kilobytes instead of an absolute number of bytes. If a
3353 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3354 kilobyte. This means that much larger values can be stored.
3356 8. Exim monitor: an attempt to get the action menu when not actually pointing
3357 at a message produces an empty menu entitled "No message selected". This
3358 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3359 no entries in it ("Shell widget menu has zero width and/or height"). So I
3360 have added a single, blank menu entry in this case.
3362 9. Added ${quote_local_part.
3364 10. MIME decoding is now applied to the contents of Subject: header lines when
3367 11. Now that a reference to $sender_host_address automatically causes a reverse
3368 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3369 host lookup before query-style lookups in lists that might use this
3370 variable. This has therefore been abolished, and the "net-" prefix is no
3371 longer necessary for query-style lookups.
3373 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3374 have been a typo for LFLAGS, so it has been changed.
3376 13. The install script calls Exim with "-C /dev/null" in order to find the
3377 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3378 to be output. Howeve, since Exim outputs its version number before the
3379 error, it didn't break the script. It just looked ugly. I fixed this by
3380 always allowing "-C /dev/null" if the caller is root.
3382 14. Ignore overlarge ACL variable number when reading spool file - insurance
3383 against a later release with more variables having written the file.
3385 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3386 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3387 this, as well as the form without the "IPv6" on the front.
3389 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3390 OS/Makefile-Darwin file.
3392 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3393 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3395 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3396 variable. However, some OS use ints for inodes. I've added cast to long int
3397 to get rid of the compiler warning.
3399 19. I had forgotten to lock out "/../" in configuration file names when
3400 ALT_CONFIG_PREFIX was set.
3402 20. Routers used for verification do not need to specify transports. However,
3403 if such a router generated a host list, and callout was configured, Exim
3404 crashed, because it could not find a port number from the (non-existent)
3405 transport. It now assumes port 25 in this circumstance.
3407 21. Added the -t option to exigrep.
3409 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3410 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3411 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3412 them is set, LOOKUP_LSEARCH is forced.
3414 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3416 24. Added sender and host information to the "rejected by local_scan()" log
3417 line; previously there was no indication of these.
3419 25. Added .include_if_exists.
3421 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3422 Linux. It turns out that not all file systems support this. Apparently some
3423 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3424 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3425 not supported on the file descriptor, is now ignored when Exim is trying to
3426 sync a directory. This applies only to Linux.
3428 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3430 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3431 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3434 29. Added support for maildirsize files from supplied patch (modified a bit).
3436 30. The use of :fail: followed by an empty string could lead Exim to respond to
3437 sender verification failures with (e.g.):
3439 550 Verification failed for <xxx>
3440 550 Sender verify failed
3442 where the first response line was missing the '-' that indicates it is not
3443 the final line of the response.
3445 31. The loop for finding the name of the user that called Exim had a hardwired
3446 limit of 10; it now uses the value of finduser_retries, which is used for
3447 all other user lookups.
3449 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3452 33. Exim was neglecting to zero errno before one call of strtol() when
3453 expanding a string and expecting an integer value. On some systems this
3454 resulted in spurious "integer overflow" errors. Also, it was casting the
3455 result into an int without checking.
3457 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3458 did not work. The code looks as if it has *never* worked, though it appears
3459 to have been documented since at least releast 1.62. I have made it work.
3461 35. The "timeout_DNS" error in retry rules, also documented since at least
3462 1.62, also never worked. As it isn't clear exactly what this means, and
3463 clearly it isn't a major issue, I have abolished the feature by treating it
3464 as "timeout", and writing a warning to the main and panic logs.
3466 36. The display of retry rules for -brt wasn't always showing the error code
3469 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3470 timeout_connect_A, timeout_connect_MX.
3472 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3473 to the empty sender.
3475 39. The daemon was not analysing the content of -oX till after it had closed
3476 stderr and disconnected from the controlling terminal. This meant that any
3477 syntax errors were only noted on the panic log, and the return code from
3478 the command was 0. By re-arranging the code a little, I've made the
3479 decoding happen first, so such errors now appear on stderr, and the return
3480 code is 1. However, the actual setting up of the sockets still happens in
3481 the disconnected process, so errors there are still only recorded on the
3484 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3485 connections (as happens on some IP stacks) was logged at start up time as
3486 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3487 it from "IPv6 and IPv4", which means that two separate sockets are being
3490 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3491 says whether AF_INET or AF_INET6 was passed as an argument.
3493 42. Exiwhat output was messed up when time zones were included in log
3496 43. Exiwhat now gives more information about the daemon's listening ports,
3497 and whether -tls-on-connect was used.
3499 44. The "port" option of the smtp transport is now expanded.
3501 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3502 silently ignored. Now an error message is written to the main and panic
3505 46. There's a new ACL modifier called "logwrite" which writes to a log file
3506 as soon as it is encountered.
3508 47. Added $local_user_uid and $local_user_gid at routing time.
3510 48. Exim crashed when trying to verify a sender address that was being
3513 49. Exim was recognizing only a space character after ".include". It now also
3514 recognizes a tab character.
3516 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3517 extracting the relevant information from the specification. The man page no
3518 longer contains scrambled data for the -d option, and I've added a section
3519 at the front about calling Exim under different names.
3521 51. Added "extra_headers" argument to the "mail" command in filter files.
3523 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3526 53. Installed eximstats 1.29.
3528 54. Added transport_filter_timeout as a generic transport option.
3530 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3531 Cc: header lines. This was required by RFC 822, but it not required by RFC
3534 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3535 incoming messages that did not have them. Now it does so only if the
3536 message originates locally, that is, if there is no associated remote host
3537 address. When Resent- header lines are present, this applies to the Resent-
3538 lines rather than the non-Resent- lines.
3540 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3541 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3543 58. Messages for configuration errors now include the name of the main
3544 configuration file - useful now that there may be more than one file in a
3545 list (.included file names were always shown).
3547 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3548 for those rare installations that do not start the daemon as root or run it
3549 setuid root. I've cut out the call to initgroups() if the daemon is not
3552 60. The Exim user and group can now be bound into the binary as text strings
3553 that are looked up at the start of Exim's processing.
3555 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3557 62. Added $mailstore_basename variable.
3559 63. Installed patch to sieve.c from Michael Haardt.
3561 64. When Exim failed to open the panic log after failing to open the main log,
3562 the original message it was trying to log was written to stderr and debug
3563 output, but if they were not available (the usual case in production), it
3564 was lost. Now it is written to syslog before the two lines that record the
3565 failures to open the logs.
3567 65. Users' Exim filters run in subprocesses under the user's uid. It is
3568 possible for a "deliver" command or an alias in a "personal" command to
3569 provoke an address rewrite. If logging of address rewriting is configured,
3570 this fails because the process is not running as root or exim. There may be
3571 a better way of dealing with this, but for the moment (because 4.30 needs
3572 to be released), I have disabled address rewrite logging when running a
3573 filter in a non-root, non-exim process.
3579 1. The buildconfig auxiliary program wasn't quoting the value set for
3580 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3581 not defined. This bug was masked in 4.22 by the effect that was fixed in
3584 2. Some messages that were rejected after a message id was allocated were
3585 shown as "incomplete" by exigrep. It no longer does this for messages that
3586 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3588 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3589 have allow_domain_literals set, the ID did not get logged in the <= line.
3590 Domain literals are now always recognized in Message-ID: header lines.
3592 4. The first argument for a ${extract expansion item is the key name or field
3593 number. Leading and trailing spaces in this item were not being ignored,
3594 causing some misleading effects.
3596 5. When deliver_drop_privilege was set, single queue runner processes started
3597 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3598 same command in the process it spins off) were not dropping privilege.
3600 6. When the daemon running as "exim" started a queue runner, it always
3601 re-executed Exim in the spun-off process. This is a waste of effort when
3602 deliver_drop_privilege is set. The new process now just calls the
3603 queue-runner function directly.
3609 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3612 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3613 ignored. Though the use of -oP was forcing the writing of a pid file, it
3614 was always written to the default place.
3616 3. If the message "no IP address found for host xxxx" is generated during
3617 incoming verification, it is now followed by identification of the incoming
3618 connection (so you can more easily find what provoked it).
3620 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3622 5. Added some features to "harden" Exim a bit more against certain attacks:
3624 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3625 be put in Local/Makefile. This is like the never_users runtime option,
3626 but it cannot be overridden. The default setting is "root".
3628 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3629 prefix string with which any file named in a -C command line option
3632 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3633 is retained for -C and -D only if the caller of Exim is root. Without
3634 it, the exim user may also use -C and -D and retain privilege.
3636 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3637 command line option is disabled.
3639 6. Macro names set by the -D option must start with an upper case letter, just
3640 like macro names defined in the configuration file.
3642 7. Added "dereference=" facility to LDAP.
3644 8. Two instances of the typo "uknown" in the source files are fixed.
3646 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3647 the Configure-Makefile script screwed up while processing it.
3649 10. Incorporated PCRE 4.4.
3651 11. The SMTP synchronization check was not operating right at the start of an
3652 SMTP session. For example, it could not catch a HELO sent before the client
3653 waited for the greeting. There is now a check for outstanding input at the
3654 point when the greeting is written. Because of the duplex, asynchronous
3655 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3656 way, but not yet received, when the check is performed.
3658 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3659 on TCP/IP sockets, because this apparently causes some broken clients to
3662 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3663 unchanged) from the Cygwin maintainer.
3665 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
3666 was supported instead of testing for mbx. Effectively a typo.
3668 15. The spa authenticator server code was not checking that the input it
3669 received was valid base64.
3671 16. The debug output line for the "set" modifier in ACLs was not showing the
3672 name of the variable that was being set.
3674 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
3676 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
3677 lookup on its own. Something else was needed to trigger the lookup. For
3678 example, a match in host_lookup or the need for a host name in a host list.
3679 Now, if $sender_host_name is referenced and the host name has not yet been
3680 looked up, a lookup is performed. If the lookup fails, the variable remains
3681 empty, and $host_lookup_failed is set to "1".
3683 19. Added "eqi" as a case-independent comparison operator.
3685 20. The saslauthd authentication condition could segfault if neither service
3686 nor realm was specified.
3688 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3689 error message that was logged was misleading, and incoming SMTP
3690 connections were dropped. The message is now more accurate, and temporary
3691 errors are given to SMTP connections.
3693 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3694 (except RSET) with a 421 error, until QUIT is received. However, it was
3695 failing to send a response to QUIT.
3697 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3698 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3699 ACL is now run after the helo_try_verify_hosts code.
3701 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3702 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3703 case-independent, so other case variants are also recognized). Apparently
3704 some systems use these upper case variants.
3706 25. If more than two messages were waiting for the same host, and a transport
3707 filter was specified for the transport, Exim sent two messages over the
3708 same TCP/IP connection, and then failed with "socket operation on non-
3709 socket" when it tried to send the third.
3711 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3713 27. The extern definition of crypt16() in expand.c was not being excluded when
3714 the OS had its own crypt16() function.
3716 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3717 as a preferred synonym for return_size_limit, both as an option and as an
3720 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3722 30. Changed the default configuration ACL to relax the local part checking rule
3723 for addresses that are not in any local domains. For these addresses,
3724 slashes and pipe symbols are allowed within local parts, but the sequence
3725 /../ is explicitly forbidden.
3727 31. SPA server authentication was not clearing the challenge buffer before
3730 32. log_message in a "warn" ACL statement was writing to the reject log as
3731 well as to the main log, which contradicts the documentation and doesn't
3732 seem right (because no rejection is happening). So I have stopped it.
3734 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3735 However, I am unable to do any testing of this.
3737 34. Fixed an infelicity in the appendfile transport. When checking directories
3738 for a mailbox, to see if any needed to be created, it was accidentally
3739 using path names with one or more superfluous leading slashes; tracing
3740 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3742 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3743 contents are added to the log line that is written for every discarded
3744 recipient. (Previously a log_message setting was ignored.)
3746 36. The ${quote: operator now quotes the string if it is empty.
3748 37. The install script runs exim in order to find its version number. If for
3749 some reason other than non-existence or emptiness, which it checks, it
3750 could not run './exim', it was installing it with an empty version number,
3751 i.e. as "exim-". This error state is now caught, and the installation is
3754 38. An argument was missing from the function that creates an error message
3755 when Exim fails to connect to the socket for saslauthd authentication.
3756 This could cause Exim to crash, or give a corrupted message.
3758 39. Added isip, isip4, and isip6 to ${if conditions.
3760 40. The ACL variables $acl_xx are now saved with the message, and can be
3761 accessed later in routers, transports, and filters.
3763 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3764 strings in the file are not string-expanded.
3766 42. If a MAIL command specified a SIZE value that was too large to fit into an
3767 int variable, the check against message_size_limit failed. Such values are
3768 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3769 day this will have to be increased, but I don't think I want to be around
3770 when emails are that large.
3777 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3778 iconv() is not standard in FreeBSD.
3780 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3781 IPv6 enabled. The observed symptom was a segmentation fault on return from
3782 the function os_common_find_running_interfaces() in src/os.c.
3784 3. In the check_special_case() function in daemon.c I had used "errno" as an
3785 argument name, which causes warnings on some systems. This was basically a
3786 typo, since it was named "eno" in the comments!
3788 4. The code that waits for the clock to tick (at a resolution of some fraction
3789 of a second) so as to ensure message-id uniqueness was always waiting for
3790 at least one whole tick, when it could have waited for less. [This is
3791 almost certainly not relevant at current processor speeds, where it is
3792 unlikely to ever wait at all. But we try to future-proof.]
3794 5. The function that sleeps for a time interval that includes fractions of a
3795 second contained a race. It did not block SIGALRM between setting the
3796 timer, and suspending (a couple of lines later). If the interval was short
3797 and the sigsuspend() was delayed until after it had expired, the suspension
3798 never ended. On busy systems this could lead to processes getting stuck for
3801 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3802 process, before it forks any delivery processes. The open lookup caching
3803 mechanism meant that the open file or database connection was passed into
3804 the delivery process. The problem was that delivery processes always tidy
3805 up cached lookup data. This could cause a problem for the next delivery
3806 process started by the queue runner, because the external queue runner
3807 process does not know about the closure. So the next delivery process
3808 still has data in the lookup cache. In the case of a file lookup, there was
3809 no problem because closing a file descriptor in a subprocess doesn't affect
3810 the parent. However, if the lookup was caching a connection to a database,
3811 the connection was closed, and the second delivery process was likely to
3812 see errors such as "PGSQL: query failed: server closed the connection
3813 unexpectedly". The problem has been fixed by closing all cached lookups
3814 in a queue runner before running a delivery process.
3816 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3817 seem to have the "const" qualifier which it has on other OS. I've
3820 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3821 *of the same type* (client or server) with the same public name that an
3822 error should be diagnosed.
3824 9. When Exim looked up a host name for an IP address, but failed to find the
3825 original IP address when looking up the host name (a safety check), it
3826 output the message "<ip address> does not match any IP for NULL", which was
3827 confusing, to say the least. The bug was that the host name should have
3828 appeared instead of "NULL".
3830 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3831 user that is built into the binary, and the -C or -D options is used, root
3832 privilege is dropped before the configuration file is read. In addition,
3833 logging is switched to stderr instead of the normal log files. If the
3834 configuration then re-defines the Exim user, the unprivileged environment
3835 is probably not what is expected, so Exim logs a panic warning message (but
3838 However, if deliver_drop_privilege is set, the unprivileged state may well
3839 be exactly what is intended, so the warning has been cut out in that case,
3840 and Exim is allowed to try to write to its normal log files.
3846 1. smtp_return_error_details was not giving details for temporary sender
3847 or receiver verification errors.
3849 2. Diagnose a configuration error if two authenticators have the same public
3852 3. Exim used not to create the message log file for a message until the first
3853 delivery attempt. This could be confusing when incoming messages were held
3854 for policy or load reasons. The message log file is now created at the time
3855 the message is received, and an initial "Received" line is written to it.
3857 4. The automatically generated man page for command line options had a minor
3858 bug that caused no ill effects; however, a more serious problem was that
3859 the procedure for building the man page automatically didn't always
3860 operate. Consequently, release 4.20 contains an out-of-date version. This
3861 shouldn't happen again.
3863 5. When building Exim with embedded Perl support, the script that builds the
3864 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3865 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3867 6. The freeze_tell option was not being used for messages that were frozen on
3868 arrival, either by an ACL or by local_scan().
3870 7. Added the smtp_incomplete_transaction log selector.
3872 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3873 was accepting AUTH without a new EHLO.
3875 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3876 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3879 10. Logging of TCP/IP connections (when configured) now happens in the main
3880 daemon process instead of the child process, so that the TCP/IP connection
3881 count is more accurate (but it can never be perfect).
3883 11. The use of "drop" in a nested ACL was not being handled correctly in the
3884 outer ACL. Now, if condition failure induced by the nested "drop" causes
3885 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3886 or "require"), the connection is dropped.
3888 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3889 that yield "discard" can now be used with an "accept" or a "discard" verb,
3890 but an error is generated for any others (because I can't see a useful way
3891 to define what should happen).
3893 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3894 are now processed in the same way as lines in the Exim configuration file.
3895 In particular, continuation lines are supported.
3897 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3899 15. Added -ti meaning -t -i.
3901 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3902 domains, and warn by logging if others are found.
3904 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3905 returned from GIFCONF when Exim is trying to find the list of interfaces on
3906 a host. The code in os.c has been modified to copy each ifreq to an aligned
3907 structure in all cases.
3909 Also, in some cases, the returned ifreq's were being copied to a 'struct
3910 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3911 means the last couple of bytes of an IPv6 address could be chopped if the
3912 ifreq contained only a normal sockaddr (14 bytes storage).
3914 18. Named domain lists were not supported in the hosts_treat_as_local option.
3915 An entry such as +xxxx was not recognized, and was treated as a literal
3918 19. Ensure that header lines added by a DATA ACL are included in the reject log
3919 if the ACL subsequently rejects the message.
3921 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3922 MD5 (which is deprecated).
3924 21. When testing a filter file using -bf, Exim was writing a message when it
3925 took the sender from a "From " line in the message, but it was not doing so
3926 when it took $return_path from a Return-Path: header line. It now does.
3928 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3929 with a valid header line field name (a series of printing characters
3930 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3932 23. Changed "disc" in the source to "disk" to conform to the documentation and
3933 the book and for uniformity.
3935 24. Ignore Sendmail's -Ooption=value command line item.
3937 25. When execve() failed while trying to run a command in a pipe transport,
3938 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3939 could be confused with a return value of 69 from the command itself. This
3940 has been changed to 127, the value the shell returns if it is asked to run
3941 a non-existent command. The wording for the related log line suggests a
3942 non-existent command as the problem.
3944 26. If received_header_text expands to an empty string, do not add a Received:
3945 header line to the message. (Well, it adds a token one on the spool, but
3946 marks it "old" so that it doesn't get used or transmitted.)
3948 27. Installed eximstats 1.28 (addition of -nt option).
3950 28. There was no check for failure on the call to getsockname() in the daemon
3951 code. This can fail if there is a shortage of resources on the system, with
3952 ENOMEM, for example. A temporary error is now given on failure.
3954 29. Contrary to the C standard, it seems that in some environments, the
3955 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3956 program. Exim now does this explicitly; it affects the formatting of
3957 timestamps using strftime().
3959 30. If exiqsumm was given junk data, it threw up some uninitialized variable
3960 complaints. I've now initialized all the variables, to avoid this.
3962 32. Header lines added by a system filter were not being "seen" during
3963 transport-time rewrites.
3965 33. The info_callback() function passed to OpenSSL is set up with type void
3966 (*)(SSL *, int, int), as described somewhere. However, when calling the
3967 function (actually a macro) that sets it up, the type void(*)() is
3968 expected. I've put in a cast to prevent warnings from picky compilers.
3970 34. If a DNS black list lookup found a CNAME record, but there were no A
3971 records associated with the domain it pointed at, Exim crashed.
3973 35. If a DNS black list lookup returned more than one A record, Exim ignored
3974 all but the first. It now scans all returned addresses if a particular IP
3975 value is being sought. In this situation, the contents of the
3976 $dnslist_value variable are a list of all the addresses, separated by a
3979 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
3980 to accept a host name and all its aliases if the forward lookup for any of
3981 them yielded the IP address of the incoming connection. Now it accepts only
3982 those names whose forward lookup yields the correct IP address. Any other
3983 names are discarded. This closes a loophole whereby a rogue DNS
3984 administrator could create reverse DNS records to break through a
3985 wildcarded host restriction in an ACL.
3987 37. If a user filter or a system filter that ran in a subprocess used any of
3988 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3989 the wrong values were passed to the pipe command ($thisaddress had the
3990 value of $0, $0 had the value of $1, etc). This bug was introduced by
3991 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3993 38. Improved the line breaking for long SMTP error messages from ACLs.
3994 Previously, if there was no break point between 40 and 75 characters, Exim
3995 left the rest of the message alone. Two changes have been made: (a) I've
3996 reduced the minimum length to 35 characters; (b) if it can't find a break
3997 point between 35 and 75 characters, it looks ahead and uses the first one
3998 that it finds. This may give the occasional overlong line, but at least the
3999 remaining text gets split now.
4001 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4002 file descriptors might be low, and that setting 1000 would always raise it.
4003 It turns out that in some environments, the limit is already over 1000 and
4004 that lowering it causes trouble. So now Exim takes care not to decrease it.
4006 40. When delivering a message, the value of $return_path is set to $sender_
4007 address at the start of routing (routers may change the value). By an
4008 oversight, this default was not being set up when an address was tested by
4009 -bt or -bv, which affected the outcome if any router or filter referred to
4012 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4013 log only when "message" or "log_message" are set. However, if one of the
4014 conditions was an address verification, or a call to a nested ACL, the
4015 messages generated by the underlying test were being passed through. This
4016 no longer happens. The underlying message is available in $acl_verify_
4017 message for both "message" and "log_message" expansions, so it can be
4018 passed through if needed.
4020 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4021 new expansion $bh_ to give the encoded byte string without charset
4022 translation. Translation happens only if iconv() is available; HAVE_ICONV
4023 indicates this at build time. HEADERS_CHARSET gives the charset to
4024 translate to; headers_charset can change it in the configuration, and
4025 "headers charset" can change it in an individual filter file.
4027 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4028 that creates RFC 2047 encoded "words" labels them as that charset instead
4029 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4030 expansion operator; (ii) when Exim creates a From: line for a local
4031 message; (iii) when a header line is rewritten to include a "phrase" part.
4033 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4034 buggy, causing it to skip the first lines of messages whose message ID
4035 ended in 'D'. This would not have bitten before Exim release 4.14, because
4036 message IDs were unlikely to end in 'D' before then. The effect was to have
4037 incorrect size information for certain domains.
4039 45. #include "config.h" was missing at the start of the crypt16.c module. This
4040 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4043 46. If there was a timeout during a "random" callout check, Exim treated it as
4044 a failure of the random address, and carried on sending RSET and the real
4045 address. If the delay was just some slowness somewhere, the response to the
4046 original RCPT would be taken as a response to RSET and so on, causing
4047 mayhem of various kinds.
4049 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4050 when I implemented it. It didn't allow for the fact that some option values
4051 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4052 the right test for positive values.
4054 48. Domain names in DNS records are case-independent. Exim always looks them up
4055 in lower case. Some resolvers return domain names in exactly the case they
4056 appear in the zone file, that is, they may contain uppercase letters. Not
4057 all resolvers do this - some return always lower case. Exim was treating a
4058 change of case by a resolver as a change of domain, similar to a widening
4059 of a domain abbreviation. This triggered its re-routing code and so it was
4060 trying to route what was effectively the same domain again. This normally
4061 caused routing to fail (because the router wouldn't handle the domain
4062 twice). Now Exim checks for this case specially, and just changes the
4063 casing of the domain that it ultimately uses when it transmits the message
4066 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4069 50. If a filter generated a file delivery with a non-absolute name (possible if
4070 no home directory exists for the router), the forbid_file option was not
4073 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4074 the existing equality matching.
4076 52. Exim was using ints instead of ino_t variables in some places where it was
4077 dealing with inode numbers.
4079 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4080 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4081 called TMPDIR, and if it finds it is different, it changes its value.
4083 54. The smtp_printf() function is now made available to local_scan() so
4084 additional output lines can be written before returning. There is also an
4085 smtp_fflush() function to enable the detection of a dropped connection.
4086 The variables smtp_input and smtp_batched_input are exported to
4089 55. Changed the default runtime configuration: the message "Unknown user"
4090 has been removed from the ACL, and instead placed on the localuser router,
4091 using the cannot_route_message feature. This means that any verification
4092 failures that generate their own messages won't get overridden. Similarly,
4093 the "Unrouteable address" message that was in the ACL for unverifiable
4094 relay addresses has also been removed.
4096 56. Added hosts_avoid_esmtp to the smtp transport.
4098 57. The exicyclog script was not checking for the esoteric option
4099 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4100 will work only if exicyclog is run under the appropriate euid.
4102 58. Following a discussion on the list, the rules by which Exim recognises line
4103 endings on incoming messages have been changed. The -dropcr and drop_cr
4104 options are now no-ops, retained only for backwards compatibility. The
4105 following line terminators are recognized: LF CRLF CR. However, special
4106 processing applies to CR:
4108 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4109 nor a local message in the state where . is a terminator.
4111 (ii) If a bare CR is encountered in a header line, an extra space is added
4112 after the line terminator so as not to end the header. The reasoning
4113 behind this is that bare CRs in header lines are most likely either
4114 to be mistakes, or people trying to play silly games.
4116 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4117 was being incorrectly given as 18 bytes larger than it should have been.
4118 This is a VOB (very old bug).
4120 60. This may never have affected anything current, but just in case it has:
4121 When the local host is found other than at the start of a list of hosts,
4122 the local host, those with the same MX, and any that follow, are discarded.
4123 When the list in question was part of a longer list of hosts, the following
4124 hosts (not currently being processed) were also being discarded. This no
4125 longer happens. I'm not sure if this situation could ever has previously
4128 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4131 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4132 header. This is something that is recommended in a new Internet Draft, and
4133 is something that is documented as being done by Sendmail. There are two
4134 possible values. For messages generated by the autoreply transport, Exim
4137 Auto-Submitted: auto-replied
4139 whereas for all other generated messages (e.g. bounces) it adds
4141 Auto-Submitted: auto-generated
4143 63. The "personal" condition in filters now includes a test for the
4144 Auto-Submitted: header. If it contains the string "auto-" the message it
4145 not considered personal.
4147 64. Added rcpt_include_affixes as a generic transport option.
4149 65. Added queue_only_override (default true).
4151 66. Added the syslog_duplication option.
4153 67. If what should have been the first header line of a message consisted of
4154 a space followed by a colon, Exim was mis-interpreting it as a header line.
4155 It isn't of course - it is syntactically invalid and should therefore be
4156 treated as the start of the message body. The misbehaviour could have
4157 caused a number of strange effects, including loss of data in subsequent
4158 header lines, and spool format errors.
4160 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4161 client host had authenticated. This control can now be exercised by an ACL
4162 for more flexibility.
4164 69. By default, callouts do not happen when testing with -bh. There is now a
4165 variant, -bhc, which does actually run the callout code, including
4166 consulting and updating the callout cache.
4168 70. Added support for saslauthd authentication, courtesy of Alexander
4171 71. If statvfs() failed on the spool or log directories while checking their
4172 size for availability, Exim confusingly gave the error "space shortage".
4173 Furthermore, in debugging mode it crashed with a floating point exception.
4174 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4175 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4176 serious problem, Exim now writes to the main and panic logs when this
4177 happens, with details of the failure. It then refuses to accept the
4178 incoming message, giving the message "spool directory problem" or "log
4179 directory problem" with a 421 code for SMTP messages.
4181 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4182 0, 1, and 2 exist, because some OS complain for execs without them (see
4183 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4184 descriptors. However, the code omitted to check that the open succeeded,
4185 causing mysterious errors if for some reason the permissions on /dev/null
4186 got screwed. Now Exim writes a message to the main and panic logs, and
4187 bombs out if it can't open /dev/null.
4189 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4190 interact so that it is all more flexible. It is supposed to remain
4191 backwards compatible. Also added extra_local_interfaces.
4193 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4194 to bomb out with an assertion failure - to the client this appears as a
4195 connection drop. This problem occurs in the part of the code that was taken
4196 from the Samba project. Fortunately, the assertion is in a very simple
4197 function, so I have fixed this by reproducing the function inline in the
4198 one place where it is called, and arranging for authentication to fail
4199 instead of killing the process with assert().
4201 75. The SPA client code was not working when the server requested OEM rather
4202 than Unicode encoding.
4204 76. Added code to make require_files with a specific uid setting more usable in
4205 the case where statting the file as root fails - usually a non-root-mounted
4206 NFS file system. When this happens and the failure is EACCES, Exim now
4207 forks a subprocess and does the per-uid checking as the relevant uid.
4209 77. Added process_log_path.
4211 78. If log_file_path was not explicitly set, a setting of check_log_space or
4212 check_log_inodes was ignored.
4214 79. If a space check for the spool or log partitions fails, the incident is now
4215 logged. Of course, in the latter case the data may get lost...
4217 80. Added the %p formatting code to string_format() so that it can be used to
4218 print addresses in debug_print(). Adjusted all the address printing in the
4219 debugging in store.c to use %p rather than %d.
4221 81. There was a concern that a line of code in smtp_in.c could overflow a
4222 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4223 initially expressed, the concern was not well-founded, because trailing
4224 spaces are removed early. However, if the trailing spaces were followed by
4225 a NULL, they did not get removed, so the overflow was possible. Two fixes
4228 (a) I re-wrote the offending code in a cleaner fashion.
4229 (b) If an incoming SMTP command contains a NULL character, it is rejected
4232 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4233 runs initgroups(), so that if the Exim user is in any additional groups,
4234 they will be used during message reception.
4240 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob