1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.407 2006/10/16 10:37:19 ph10 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 TF/01 In the add_headers option to the mail command in an Exim filter,
10 there was a bug that Exim would claim a syntax error in any
11 header after the first one which had an odd number of characters
14 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
15 callout verification, Exim cached a "reject" for the entire domain. This
16 is correct for most verifications, but it is not correct for a recipient
17 verification with use_sender or use_postmaster set, because in that case
18 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
19 case of MAIL FROM:<> rejection from other early rejections (e.g.
20 rejection of HELO). When verifying a recipient using a non-null MAIL
21 address, the cache is ignored if it shows MAIL FROM:<> rejection.
22 Whatever the result of the callout, the value of the domain cache is
23 left unchanged (for any other kind of callout, getting as far as trying
24 RCPT means that the domain itself is ok).
26 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
29 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
30 manifest itself as EPIPE rather than ECONNECT. When tidying away a
31 session, the daemon ignores ECONNECT errors and logs others; it now
32 ignores EPIPE as well.
34 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
35 (quoted-printable decoding).
37 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
38 later the small subsequent patch to fix an introduced bug.
40 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
42 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
44 PH/08 An error is now given if message_size_limit is specified negative.
46 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
47 to be given (somewhat) arbitrary names.
49 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
52 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
53 miscellaneous code fixes
55 PH/10 Added the log_reject_target ACL modifier to specify where to log
58 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
59 hostname. This is wrong, because it relates to the incoming message (and
60 probably the interface on which it is arriving) and not to the outgoing
61 callout (which could be using a different interface). This has been
62 changed to use the value of the helo_data option from the smtp transport
63 instead - this is what is used when a message is actually being sent. If
64 there is no remote transport (possible with a router that sets up host
65 addresses), $smtp_active_hostname is used.
67 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
68 tweaks were necessary in order to get it to work:
69 (a) The code assumed that strncpy() returns a negative number on buffer
70 overflow, which isn't the case. Replaced with Exim's string_format()
72 (b) There were several signed/unsigned issues. I just did the minimum
73 hacking in of casts. There is scope for a larger refactoring.
74 (c) The code used strcasecmp() which is not a standard C function.
75 Replaced with Exim's strcmpic() function.
76 (d) The code set only $1; it now sets $auth1 as well.
77 (e) A simple test gave the error "authentication client didn't specify
78 service in request". It would seem that Dovecot has changed its
79 interface. Fortunately there's a specification; I followed it and
80 changed what the client sends and it appears to be working now.
82 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
85 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
86 address A is aliased to B and C, where B exists and C does not. Without
87 -v the output is "A verified" because verification stops after a
88 successful redirection if more than one address is generated. However,
89 with -v the child addresses are also verified. Exim was outputting "A
90 failed to verify" and then showing the successful verification for C,
91 with its parentage. It now outputs "B failed to verify", showing B's
92 parentage before showing the successful verification of C.
94 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
95 look up a TXT record in a specific list after matching in a combined
98 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
99 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
100 they consult the DNS. I had assumed they would set it the way they
101 wanted; and indeed my experiments on Linux seem to show that in some
102 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
103 To be on the safe side, however, I have now made the interface to
104 host_find_byname() similar to host_find_bydns(), with an argument
105 containing the DNS resolver options. The host_find_byname() function now
106 sets these options at its start, just as host_find_bydns() does. The smtp
107 transport options dns_qualify_single and dns_search_parents are passed to
108 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
109 of host_find_byname() use the default settings of RES_DEFNAMES
110 (qualify_single) but not RES_DNSRCH (search_parents).
112 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
113 spool_read_header() do less string testing, by means of a preliminary
114 switch on the second character of optional "-foo" lines. (This is
115 overdue, caused by the large number of possibilities that now exist.
116 Originally there were few.) While I was there, I also converted the
117 str(n)cmp tests so they don't re-test the leading "-" and the first
118 character, in the hope this might squeeze out yet more improvement.
120 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
121 flag allowing group syntax was set by the header_syntax check but not
122 turned off, possible causing trouble later; (2) The flag was not being
123 set at all for the header_verify test, causing "group"-style headers to
124 be rejected. I have now set it in this case, and also caused header_
125 verify to ignore an empty address taken from a group. While doing this, I
126 came across some other cases where the code for allowing group syntax
127 while scanning a header line wasn't quite right (mostly, not resetting
128 the flag correctly in the right place). These bugs could have caused
129 trouble for malformed header lines. I hope it is now all correct.
131 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
132 with the "reply" argument non-NULL. The code, however (which originally
133 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
134 but it didn't always do it. This confused somebody who was copying the
135 code for some other use. I have removed all the tests.
141 SC/01 Use a glob alias rather than an array ref in eximstats generated
142 parser. This improves both readability and performance.
144 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
145 Don't display local sender or destination tables in eximstats unless
146 there is data to show.
147 Added average volumes into the eximstats top table text output.
149 SC/03 Collect data on the number of addresses (recipients) as well
150 as the number of messages in eximstats.
152 TF/01 Correct an error in the documentation for the redirect router. Exim
153 does (usually) call initgroups() when daemonizing.
155 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
156 with consistent privilege compared to when running as a daemon.
158 TF/03 Note in the spec that $authenticated_id is not set for local
159 submissions from trusted users.
161 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
162 Thanks to Dean Brooks <dean@iglou.com> for the patch.
164 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
165 by adding some example configuration directives to the default
166 configuration file. A little bit of work is required to uncomment the
167 directives and define how usernames and passwords are checked, but
168 there is now a framework to start from.
170 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
171 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
172 without this. I don't know how relevant this is to other LDAP libraries.
174 PH/02 Add the verb name to the "unknown ACL verb" error.
176 PH/03 Magnus Holmgren's patch for filter_prepend_home.
178 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
180 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
181 directory not expanded when it should be if an expanded home directory
182 was set for the address (which is overridden by the transport).
184 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
187 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
188 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
189 because it is too late at that time, and has no effect.
191 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
192 security issue with \' (bugzilla #107). I could not use the
193 PQescapeStringConn() function, because it needs a PGconn value as one of
196 PH/08 When testing addresses using -bt, indicate those final addresses that
197 are duplicates that would not cause an additional delivery. At least one
198 person was confused, thinking that -bt output corresponded to deliveries.
199 (Suppressing duplicates isn't a good idea as you lose the information
200 about possibly different redirections that led to the duplicates.)
202 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
203 systems where poll() doesn't work, in particular OS X.
205 PH/10 Added more information to debugging output for retry time not reached.
207 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
208 operations in malware.c.
210 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
213 PH/13 If write_rejectlog was set false when logging was sent to syslog with
214 syslog_duplication set false, log lines that would normally be written
215 both the the main log and to the reject log were not written to syslog at
218 PH/14 In the default configuration, change the use of "message" in ACL warn
219 statements to "add_header".
221 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
222 not followed by a command (e.g. "seen endif").
224 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
225 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
228 PH/17 Added extra conditions to the default value of delay_warning_condition
232 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
233 { match{$h_precedence:}{(?i)bulk|list|junk} } \
234 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
237 The Auto-Submitted: and various List- headers are standardised, whereas I
238 don't think Precedence: ever was.
240 PH/18 Refactored debugging code in route_finduser() to show more information,
241 in particular, the error code if getpwnam() issues one.
243 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
244 This is apparently needed in addition to the PH/07 change above to avoid
245 any possible encoding problems.
247 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
248 but not after initializing Perl.
250 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
251 output them only if debugging. By default they are written stderr,
252 apparently, which is not desirable.
254 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
257 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
260 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
262 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
263 authenticated or an ident call has been made. Suppress the default
264 values for $authenticated_id and $authenticated_sender (but permit -oMai
265 and -oMas) when testing with -bh.
267 PH/24 Re-jigged the order of the tests in the default configuration so that the
268 tests for valid domains and recipients precede the DNS black list and CSA
269 tests, on the grounds that those ones are more expensive.
271 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
272 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
273 This bug exists in every version of Exim that I still have, right back to
276 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
277 However, an attempt to turn on case-sensitivity in a regex key by
278 including (?-i) didn't work because the subject string was already
279 lowercased, and the effects were non-intuitive. It turns out that a
280 one-line patch can be used to allow (?-i) to work as expected.
286 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
287 other effects) broke the use of negated acl sub-conditions.
289 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
292 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
293 "Deny" causes Exim to reject the incoming connection with a 554 error.
294 Unfortunately, if there is a major crisis, such as a disk failure,
295 tcp-wrappers gives "deny", whereas what one would like would be some
296 kind of temporary error. A kludge has been added to help with this.
297 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
298 554 error is used if errno is still zero or contains ENOENT (which occurs
299 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
302 PH/03 Add -lutil to the default FreeBSD LIBS setting.
304 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
305 errors. Otherwise a message that provokes a temporary error (when other
306 messages do not) can cause a whole host to time out.
308 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
309 addresses were routed directly to files or pipes from a redirect router.
310 File deliveries just didn't batch; pipe deliveries might have suffered
313 PH/06 A failure to get a lock for a hints database would erroneously always say
314 "Failed to get write lock", even when it was really a read lock.
316 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
317 of 0600. This has been changed to use the value of the lockfile_mode
318 option (which defaults to 0600).
320 PH/08 Applied small patch from the Sieve maintainer.
322 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
323 folder from quota calculations, a direct delivery into this folder messed
324 up the contents of the maildirsize file. This was because the regex was
325 used only to exclude .Trash (or whatever) when the size of the mailbox
326 was calculated. There was no check that a delivery was happening into an
327 excluded directory. This bug has been fixed by ignoring all quota
328 processing for deliveries into excluded directories.
330 PH/10 Added the maildirfolder_create_regex option to appendfile.
336 PH/01 The code for finding all the local interface addresses on a FreeBSD
337 system running IPv6 was broken. This may well have applied to all BSD
338 systems, as well as to others that have similar system calls. The broken
339 code found IPv4 interfaces correctly, but gave incorrect values for the
340 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
341 that it would not match correctly against @[] and not recognize the IPv6
344 PH/02 The ipliteral router was not recognizing addresses of the form user@
345 [ipv6:....] because it didn't know about the "ipv6:" prefix.
347 PH/03 Added disable_ipv6.
349 PH/04 Changed $reply_address to use the raw form of the headers instead of the
350 decoded form, because it is most often used to construct To: headers
351 lines in autoreplies, and the decoded form may well be syntactically
352 invalid. However, $reply_address has leading white space removed, and all
353 newlines turned into spaces so that the autoreply transport does not
356 PH/05 If group was specified without a user on a router, and no group or user
357 was specified on a transport, the group from the router was ignored.
359 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
360 for visible compile-time settings that can be used to change these
361 numbers, for those that want even more. Backwards compatibility with old
362 spool files has been maintained. However, going back to a previous Exim
363 release will lost any variables that are in spool files.
365 PH/07 Two small changes when running in the test harness: increase delay when
366 passing a TCP/IP connection to a new process, in case the original
367 process has to generate a bounce, and remove special handling of
368 127.0.0.2 (sic), which is no longer necessary.
370 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
371 be the same on different OS.
373 PH/09 Moved a debug statement in filter processing to avoid a race problem when
376 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
377 whether --show-vars was specified or not
379 JJ/02 exipick: Added support for new ACL variable spool format introduced
382 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
383 syntactically invalid From: or Reply-to: line, and a filter used this to
384 generate an autoreply, and therefore failed to obtain an address for the
385 autoreply, Exim could try to deliver to a non-existent relative file
386 name, causing unrelated and misleading errors. What now happens is that
387 it logs this as a hard delivery error, but does not attempt to create a
390 PH/11 The exinext utility has a -C option for testing purposes, but although
391 the given file was scanned by exinext itself; it wasn't being passed on
392 when Exim was called.
394 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
395 an end-of-file indication when reading a command response.
397 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
398 compiled. In many other places in Exim, IPv6 addresses are always
399 recognized, so I have changed this. It also means that IPv4 domain
400 literals of the form [IPV4:n.n.n.n] are now always recognized.
402 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
403 used if the router is not running as root, for example, when verifying at
404 ACL time, or when using -bh. The debugging output from this situation was
405 non-existent - all you got was a failure to exec. I have made two
408 (a) Failures to set uid/gid, the current directory, or a process leader
409 in a subprocess such as that created by queryprogram now generate
410 suitable debugging ouput when -d is set.
412 (b) The queryprogram router detects when it is not running as root,
413 outputs suitable debugging information if -d is set, and then runs
414 the subprocess without attempting to change uid/gid.
416 PH/15 Minor change to Makefile for building test_host (undocumented testing
419 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
420 additional section of a DNS packet that returns MX or SRV records.
421 Instead, it always explicitly searches for A/AAAA records. This avoids
422 major problems that occur when a DNS server includes only records of one
423 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
424 fixed another bug: if SRV records were looked up and the corresponding
425 address records were *not* found in the additional section, the port
426 values from the SRV records were lost.
428 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
429 using the correct key (the original address) when searching the retry
430 rules in order to find which one to use for generating the retry hint.
432 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
433 adding the default one. Similarly, if it contains a Reply-To: header, the
434 errors_reply_to option, if set, is not used.
436 PH/19 When calculating a retry time, Exim used to measure the "time since
437 failure" by looking at the "first failed" field in the retry record. Now
438 it does not use this if it is later than than the arrival time of the
439 message. Instead it uses the arrival time. This makes for better
440 behaviour in cases where some deliveries succeed, thus re-setting the
441 "first failed" field. An example is a quota failure for a huge message
442 when small messages continue to be delivered. Without this change, the
443 "time since failure" will always be short, possible causing more frequent
444 delivery attempts for the huge message than are intended.
445 [Note: This change was subsequently modified - see PH/04 for 4.62.]
447 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
448 $1, $2, $3) because the numerical variables can be reset during some
449 expansion items (e.g. "match"), thereby losing the authentication data.
451 PH/21 Make -bV show the size of off_t variables so that the test suite can
452 decide whether to run tests for quotas > 2G.
454 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
455 mailbox_size, and mailbox_filecount in the appendfile transport. If a
456 filecount value is greater than 2G or if a quota value is greater than 2G
457 on a system where the size of off_t is not greater than 4, a panic error
460 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
461 never match. The debug and -bh output now contains an explicit error
462 message indicating a malformed IPv4 address or mask.
464 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
465 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
468 PH/25 Do not write to syslog when running in the test harness. The only
469 occasion when this arises is a failure to open the main or panic logs
470 (for which there is an explicit test).
472 PH/26 Added the /no_tell option to "control=freeze".
474 PH/27 If a host name lookup failed very early in a connection, for example, if
475 the IP address matched host_lookup and the reverse lookup yielded a name
476 that did not have a forward lookup, an error message of the form "no IP
477 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
478 could be logged. Now it outputs the IP address instead of "NULL".
480 PH/28 An enabling patch from MH: add new function child_open_exim2() which
481 allows the sender and the authenticated sender to be set when
482 submitting a message from within Exim. Since child_open_exim() is
483 documented for local_scan(), the new function should be too.
485 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
486 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
487 results in an empty string is now treated as unset.
489 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
491 PH/31 Added sender_verify_fail logging option.
493 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
494 needed by RFC 822 but not by RFC 2822 was commented out. I have now
495 tidied the source and removed it altogether.
497 PH/33 When a queue run was abandoned because the load average was too high, a
498 log line was always written; now it is written only if the queue_run log
499 selector is set. In addition, the log line for abandonment now contains
500 information about the queue run such as the pid. This is always present
501 in "start" and "stop" lines but was omitted from the "abandon" line.
503 PH/34 Omit spaces between a header name and the colon in the error message that
504 is given when verify = headers_syntax fails (if there are lots of them,
505 the message gets confusing).
507 PH/35 Change the default for dns_check_names_pattern to allow slashes within
508 names, as there are now some PTR records that contain slashes. This check
509 is only to protect against broken name servers that fall over on strange
510 characters, so the fact that it applies to all lookups doesn't matter.
512 PH/36 Now that the new test suite is complete, we can remove some of the
513 special code in Exim that was needed for the old test suite. For example,
514 sorting DNS records because real resolvers return them in an arbitrary
515 order. The new test suite's fake resolver always returns records in the
518 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
519 bounces) except when queue_only is set, to avoid logging races between
520 the different processes.
522 PH/38 Panic-die if .include specifies a non-absolute path.
524 PH/39 A tweak to the "H" retry rule from its user.
526 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
527 a label. They prevented compilation on older perls.
529 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
530 a warning to be raised on newish perls.
532 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
533 on queue. Changes to match documented behaviour of showing count of
534 messages matching specified criteria.
536 PH/40 Changed the default ident timeout from 30s to 5s.
538 PH/41 Added support for the use of login_cap features, on those BSD systems
539 that have them, for controlling the resources used by pipe deliveries.
541 PH/42 The content-scanning code uses fopen() to create files in which to put
542 message data. Previously it was not paying any attention to the mode of
543 the files. Exim runs with umask(0) because the rest of the code creates
544 files with open(), and sets the required mode explicitly. Thus, these
545 files were ending up world-writeable. This was not a big issue, because,
546 being within the spool directory, they were not world-accessible. I have
547 created a function called modefopen, which takes an additional mode
548 argument. It sets umask(777), creates the file, chmods it to the required
549 mode, then resets the umask. All the relevant calls to fopen() in the
550 content scanning code have been changed to use this function.
552 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
553 to 24 hours. This avoids potential overflow problems when processing G
554 and H retry rules. I suspect nobody ever tinkers with this value.
556 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
558 PH/45 When the plaintext authenticator is running as a client, the server's
559 challenges are checked to ensure they are valid base64 strings. By
560 default, the authentication attempt is cancelled if an invalid string is
561 received. Setting client_ignore_invalid_base64 true ignores these errors.
562 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
563 they are received. Thus, the responses can be made to depend on the
564 challenges. If an invalid string is ignored, an empty string is placed in
567 PH/46 Messages that are created by the autoreply transport now contains a
568 References: header, in accordance with RFCs 2822 and 3834.
570 PH/47 Added authenticated_sender_force to the smtp transport.
572 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
574 PH/49 Installed latest patch from the Sieve maintainer.
576 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
577 was also set, the appendfile transport was unnecessarily scanning a
578 directory of message files (e.g. for maildir delivery) to find the count
579 of files (along with the size), even though it did not need this
580 information. It now does the scan only if it needs to find either the
581 size of the count of files.
583 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
585 PH/52 Two bugs concerned with error handling when the smtp transport is
588 (i) Exim was not creating retry information for temporary errors given
589 for individual recipients after the DATA command when the smtp transport
590 was used in LMTP mode. This meant that they could be retried too
591 frequently, and not timed out correctly.
593 (ii) Exim was setting the flag that allows error details to be returned
594 for LMTP errors on RCPT commands, but not for LMTP errors for individual
595 recipients that were returned after the DATA command.
597 PH/53 This is related to PH/52, but is more general: for any failing address,
598 when detailed error information was permitted to be returned to the
599 sender, but the error was temporary, then after the final timeout, only
600 "retry timeout exceeded" was returned. Now it returns the full error as
601 well as "retry timeout exceeded".
603 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
604 do this, and (what is worse) MTAs that accept it.
606 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
607 will now be deprecated.
609 PH/56 New os.c-cygwin from the Cygwin maintainer.
611 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
612 formats (previously only available in exim formats via -bpr, -bpru,
613 and -bpra. Now also available in native and exiqgrep formats)
615 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
616 with very large, slow to parse queues
618 JJ/08 exipick: added ! as generic prefix to negate any criteria format
620 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
622 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
623 responses to authentication challenges, though it was showing the
624 challenges; (ii) I've removed the CR characters from the debug output for
627 PH/58 Allow for the insertion of a newline as well as a space when a string
628 is turned into more than one encoded-word during RFC 2047 encoding. The
629 Sieve code now uses this.
631 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
632 data_4xx, lost_connection, tls_required.
634 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
635 message was being sent as an SMTP response.
637 PH/61 Add -l and -k options to exicyclog.
639 PH/62 When verifying, if an address was redirected to one new address, so that
640 verification continued, and the new address failed or deferred after
641 having set something in $address_data, the value of $address_data was not
642 passed back to the ACL. This was different to the case when no
643 redirection occurred. The value is now passed back in both cases.
645 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
646 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
647 use login_cap.h, so on its own it isn't the distinguishing feature. The
648 new name refers directly to the setclassresources() function.
650 PH/65 Added configuration files for NetBSD3.
652 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
654 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
655 is preferred over IPv4.
657 PH/68 The bounce_return_message and bounce_return_body options were not being
658 honoured for bounces generated during the reception of non-SMTP messages.
659 In particular, this applied to messages rejected by the ACL. This bug has
660 been fixed. However, if bounce_return_message is true and bounce_return_
661 body is false, the headers that are returned for a non-SMTP message
662 include only those that have been read before the error was detected.
663 (In the case of an ACL rejection, they have all been read.)
665 PH/69 The HTML version of the specification is now built in a directory called
666 spec_html instead of spec.html, because the latter looks like a path with
667 a MIME-type, and this confuses some software.
669 PH/70 Catch two compiler warnings in sieve.c.
671 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
672 function verify_get_ident() calls ip_connect() to connect a socket, but
673 if the "connect()" function timed out, ip_connect() used to close the
674 socket. However, verify_get_ident() also closes the socket later, and in
675 between Exim writes to the log, which may get opened at this point. When
676 the socket was closed in ip_connect(), the log could get the same file
677 descriptor number as the socket. This naturally causes chaos. The fix is
678 not to close the socket in ip_connect(); the socket should be closed by
679 the function that creates it. There was only one place in the code where
680 this was missing, in the iplookup router, which I don't think anybody now
681 uses, but I've fixed it anyway.
683 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
684 well as to direct DNS lookups. Otherwise the handling of names in host
685 lists is inconsistent and therefore confusing.
691 PH/01 Two changes to the default runtime configuration:
693 (1) Move the checks for relay_from_hosts and authenticated clients from
694 after to before the (commented out) DNS black list checks.
696 (2) Add control=submission to the relay_from_hosts and authenticated
697 clients checks, on the grounds that messages accepted by these
698 statements are most likely to be submissions.
700 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
702 (1) Generate an error if the third argument for the ${prvs expansion is
705 (2) Treat a missing third argument of ${prvscheck as if it were an empty
708 (3) Reset the variables that are obtained from the first argument of
709 ${prvscheck and used in the second argument before leaving the code,
710 because their memory is reclaimed, so using them afterwards may do
713 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
714 one (it's much easier than Tom thought :-).
716 (5) Because of (4), we can now allow for the use of $prvscheck_result
717 inside the third argument.
719 PH/03 For some reason, the default setting of PATH when running a command from
720 a pipe transport was just "/usr/bin". I have changed it to
723 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
724 anything to be listed in the output from -bV.
726 PH/05 When a filter generated an autoreply, the entire To: header line was
727 quoted in the delivery log line, like this:
729 => >A.N.Other <ano@some.domain> <original@ddress> ...
731 This has been changed so that it extracts the operative address. There
732 may be more than one such address. If so, they are comma-separated, like
735 => >ano@some.domain,ona@other.domain <original@ddress> ...
737 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
738 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
739 not being looked up in the rDNS to get a host name, Exim was showing the
740 IP address twice in Received: lines, even though the IP addresses were
741 identical. For example:
743 Received: from [1.2.3.4] (helo=[1.2.3.4])
745 However, if the real host name was known, it was omitting the HELO data
746 if it matched the actual IP address. This has been tidied up so that it
747 doesn't show the same IP address twice.
749 PH/07 When both +timestamp and +memory debugging was on, the value given by
750 $tod_xxx expansions could be wrong, because the tod_stamp() function was
751 called by the debug printing, thereby overwriting the timestamp buffer.
752 Debugging no longer uses the tod_stamp() function when +timestamp is set.
754 PH/08 When the original message was included in an autoreply transport, it
755 always said "this is a copy of the message, including all the headers",
756 even if body_only or headers_only was set. It now gives an appropriate
759 PH/09 Applied a patch from the Sieve maintainer which:
761 o fixes some comments
762 o adds the (disabled) notify extension core
763 o adds some debug output for the result of if/elsif tests
764 o points to the current vacation draft in the documentation
765 and documents the missing references header update
769 o fixes a bug in processing the envelope test (when testing
770 multiple envelope elements, the last element determinted the
773 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
774 Electronic Mail") by including:
776 Auto-submitted: auto-generated
778 in the messages that it generates (bounce messages and others, such as
779 warnings). In the case of bounce messages for non-SMTP mesages, there was
780 also a typo: it was using "Auto_submitted" (underscore instead of
781 hyphen). Since every message generated by Exim is necessarily in response
782 to another message, thes have all been changed to:
784 Auto-Submitted: auto-replied
786 in accordance with these statements in the RFC:
788 The auto-replied keyword:
790 - SHOULD be used on messages sent in direct response to another
791 message by an automatic process,
793 - MUST NOT be used on manually-generated messages,
795 - MAY be used on Delivery Status Notifications (DSNs) and Message
796 Disposition Notifications (MDNs),
798 - MUST NOT be used on messages generated by automatic or periodic
799 processes, except for messages which are automatic responses to
802 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
803 to the default Received: header definition.
805 PH/12 Added log selector acl_warn_skipped (default on).
807 PH/13 After a successful wildlsearch lookup, discard the values of numeric
808 variables because (a) they are in the wrong storage pool and (b) even if
809 they were copied, it wouldn't work properly because of the caching.
811 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
812 checking when decoding. Apparently there are clients that generate
813 overlong encoded strings. Why am I not surprised?
815 PH/15 If the first argument of "${if match_address" was not empty, but did not
816 contain an "@" character, Exim crashed. Now it writes a panic log message
817 and treats the condition as false.
819 PH/16 In autoreply, treat an empty string for "once" the same as unset.
821 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
822 extension "envelope-auth". The code is finished and in agreement with
823 other implementations, but there is no documentation so far and in fact,
824 nobody wrote the draft yet. This extension is currently #undef'ed, thus
825 not changing the active code.
827 Print executed "if" and "elsif" statements when debugging is used. This
828 helps a great deal to understand what a filter does.
830 Document more things not specified clearly in RFC3028. I had all this
831 sorted out, when out of a sudden new issues came to my mind. Oops."
833 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
836 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
837 canonical form (as documented). However, after a host name lookup from
838 the IP address, check_host() was doing a simple string comparison with
839 addresses acquired from the DNS when checking that the found name did
840 have the original IP as one of its addresses. Since any found IPv6
841 addresses are likely to be in abbreviated form, the comparison could
842 fail. Luckily, there already exists a function for doing the comparison
843 by converting both addresses to binary, so now that is used instead of
846 PH/20 There was another similar case to PH/19, when a complete host name was
847 given in a host list; looking up its IP address could give an abbreviated
848 form, whereas the current host's name might or might not be abbreviated.
849 The same fix has been applied.
855 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
856 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
859 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
860 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
862 PH/03 Typo: missing ".o" in src/pcre/Makefile.
864 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
865 header line, restrict the check to what is listed in RFCs 2369 and 2929.
866 Also, for "Auto-Submitted", treat anything other than "no" as
867 non-personal, in accordance with RFC 3834. (Previously it treated
868 anything starting "auto-" as non-personal.)
870 TF/01 The control=submission/name=... option had a problem with syntax
871 errors if the name included a slash character. The /name= option
872 now slurps the rest of the string, so it can include any characters
873 but it must come last in the list of options (after /sender_retain
876 PH/05 Some modifications to the interface to the fake nameserver for the new
884 TK/01 Added the "success_on_redirect" address verification option. See
885 NewStuff for rationale and an example.
887 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
889 PH/02 Patch to exigrep to allow it to work on syslog lines.
891 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
892 fread() to skip over the body file's header line, because in Cygwin the
893 header line is locked and is inaccessible.
895 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
896 co-exist for some time) to make it clear that it is the Exim ID that is
897 referenced, not the Message-ID: header line.
899 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
900 string_format() function, because snprintf() does not exist on all
903 PH/06 The use of forbid_filter_existstest now also locks out the use of the
904 ${stat: expansion item.
906 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
907 protocol synchronization error", to keep the pedants happy.
909 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
910 well as for IRIX systems, when gcc is being used. See the host.c source
913 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
915 PH/10 Named domain lists were not working if used in a queue_smtp_domains
918 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
919 transport and to the smtp transport in LMTP mode.
921 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
923 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
924 run a filter in a subprocess. This could lead to confusion in subsequent
925 lookups in the parent process. There should also be a search_tidyup() at
926 the end of the subprocess.
928 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
929 only if the host matched helo_try_verify_hosts, which caused the
930 verification to occur when the EHLO/HELO command was issued. The ACL just
931 tested the remembered result. Now, if a previous verification attempt has
932 not happened, "verify = helo" does it there and then.
934 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
936 TK/03 Fix log output including CR from clamd.
938 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
939 exist provoked a memory error which could cause a segfault.
941 PH/15 Installed PCRE 6.2
943 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
945 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
946 of the problem. Specifically, suggested +O2 rather than +O1 for the
949 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
951 PH/20 If a delivery was routed to a non-standard port by means of an SRV
952 record, the port was not correctly logged when the outgoing_port log
953 selector was set (it logged the transort's default port).
955 PH/21 Added support for host-specific ports to manualroute, queryprogram,
956 fallback_hosts, and "hosts" in the smtp transport.
958 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
959 host errors such as "Connection refused".
961 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
962 authentication with radiusclient 0.4.9:
964 - Error returned from rc_read_config was caught wrongly
965 - Username/password not passed on to radius server due to wrong length.
967 The presumption is that some radiusclient API changes for 4.51/PH/17
968 were not taken care of correctly. The code is still untested by me (my
969 Linux distribution still has 0.3.2 of radiusclient), but it was
970 contributed by a Radius user.
972 PH/24 When doing a callout, the value of $domain wasn't set correctly when
973 expanding the "port" option of the smtp transport.
975 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
976 while reading a MIME header. Thanks to Tom Hughes for a patch.
978 PH/24 Include config.h inside local_scan.h so that configuration settings are
981 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
982 that in an ACL for RCPT (for example), you can examine exactly what was
985 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
986 commands, but it was not correctly comparing the address with the actual
987 client host address. Thus, it would show the EHLO address in Received:
988 header lines when this was not necessary.
990 PH/27 Added the % operator to ${eval:}.
992 PH/28 Exim tries to create and chdir to its spool directory when it starts;
993 it should be ignoring failures (because with -C, for example, it has lost
994 privilege). It wasn't ignoring creation failures other than "already
997 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1000 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1001 because an input line was too long, either on its own, or by virtue of
1002 too many continuations, the temporary file was not being removed, and the
1003 return code was incorrect.
1005 PH/31 Missing "BOOL" in function definition in filtertest.c.
1007 PH/32 Applied Sieve patches from the maintainer.
1009 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1011 PH/33 Added "verify = not_blind".
1013 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1014 Local/Makefile (with some defaults set). These are used in built scripts
1015 such as exicyclog, but they have never been used in the exim_install
1016 script (though there are many overriding facilities there). I have
1017 arranged that the exim_install script now takes note of these two
1020 PH/35 Installed configuration files for Dragonfly.
1022 PH/36 When a locally submitted message by a trusted user did not contain a
1023 From: header, and the sender address was obtained from -f or from an SMTP
1024 MAIL command, and the trusted user did not use -F to supply a sender
1025 name, $originator_name was incorrectly used when constructing a From:
1026 header. Furthermore, $originator_name was used for submission mode
1027 messages from external hosts without From: headers in a similar way,
1028 which is clearly wrong.
1030 PH/37 Added control=suppress_local_fixups.
1032 PH/38 When log_selector = +received_sender was set, and the addition of the
1033 sender made the log line's construction buffer exactly full, or one byte
1034 less than full, an overflow happened when the terminating "\n" was
1037 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1038 when the result of a list match is failure because a DNS lookup failed.
1040 PH/40 RM_COMMAND is now used in the building process.
1042 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1043 the "build-* directories that it finds.
1045 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1046 address in a domain literal was a prefix of an interface address.
1048 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1049 when verifying a sender address, unless rewrite_headers is false.
1051 PH/44 Wrote a long comment about why errors_to addresses are verified as
1052 recipients, not senders.
1054 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1055 the ratelimit ACL was added.
1057 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1059 PH/46 Added extra information about PostgreSQL errors to the error string.
1061 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1062 avoiding the need to install special zones in a real server. This is
1063 backwards compatible; if it can't find the fake resolver, it drops back.
1064 Thus, both old and new test suites can be run.
1066 TF/02 Added util/ratelimit.pl
1068 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1069 clock is set back in time.
1071 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1072 Candler <B.Candler@pobox.com>.
1074 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1075 OK for addresses that are the result of redirections.
1077 PH/48 A number of further additions for the benefit of the new test suite,
1078 including a fake gethostbyname() that interfaces to the fake DNS resolver
1081 TF/06 The fix for widen_domains has also been applied to qualify_single and
1082 search_parents which are the other dnslookup options that can cause
1085 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1088 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1090 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1091 service" if it encountered an MX record with an empty target hostname.
1092 The message is now "an MX or SRV record indicated no SMTP service".
1094 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1095 if the DNS of the sending site is misconfigured. This is quite a
1096 common situation. This change restores the behaviour of treating a
1097 helo verification defer as a failure.
1099 PH/51 If self=fail was set on a router, the bounce message did not include the
1100 actual error message.
1106 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1108 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1109 command itself ended in error, the underlying message about the transport
1110 filter timeout was being overwritten with the pipe command error. Now the
1111 underlying error message should be appended to the second error message.
1113 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1115 PH/02 Reduce the amount of output that "make" produces by default. Full output
1116 can still be requested.
1118 PH/03 The warning log line about a condition test deferring for a "warn" verb
1119 was being output only once per connection, rather than after each
1120 occurrence (because it was using the same function as for successful
1121 "warn" verbs). This seems wrong, so I have changed it.
1123 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1124 it should not have, which might have caused a crash in the right
1125 circumstances, but probably never did.
1127 PH/04 Installed a modified version of Tony Finch's patch to make submission
1128 mode fix the return path as well as the Sender: header line, and to
1129 add a /name= option so that you can make the user's friendly name appear
1132 TF/03 Added the control = fakedefer ACL modifier.
1134 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1135 Mark Lowes for thorough testing.
1137 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1139 TK/03 Merged latest SRS patch from Miles Wilton.
1141 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1142 with the definition in sysexits.h (which is #included earlier).
1143 Fortunately, Exim does not actually use EX_OK. The code used to try to
1144 preserve the sysexits.h value, by assumimg that macro definitions were
1145 scanned for macro replacements. I have been disabused of this notion,
1146 so now the code just undefines EX_OK before #including unistd.h.
1148 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1149 in the smtp transport. When a block could not be written in a single
1150 write() function, the timeout was being re-applied to each part-write.
1151 This seems wrong - if the receiver was accepting one byte at a time it
1152 would take for ever. The timeout is now adjusted when this happens. It
1153 doesn't have to be particularly precise.
1155 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1156 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1158 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1159 without a domain if the check to <postmaster@domain> fails.
1161 SC/01 Eximstats: added -xls and the ability to specify output files
1162 (patch written by Frank Heydlauf).
1164 SC/02 Eximstats: use FileHandles for outputing results.
1166 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1168 SC/04 Eximstats: fixed display of large numbers with -nvr option
1170 SC/05 Eximstats: fixed merging of reports with empty tables.
1172 SC/06 Eximstats: added the -include_original_destination flag
1174 SC/07 Eximstats: removed tabs and trailing whitespace.
1176 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1178 TK/06 MBOX spool code: Add real "From " MBOX separator line
1179 so the .eml file is really in mbox format (even though
1180 most programs do not really care). Patch from Alex Miller.
1182 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1183 The latter is generated from $received_to and is only set if the
1184 message has one envelope recipient. SA can use these headers,
1185 obviously out-of-the-box. Patch from Alex Miller.
1187 PH/08 The ${def test on a variable was returning false if the variable's
1188 value was "0", contrary to what the specification has always said!
1189 The result should be true unless the variable is empty.
1191 PH/09 The syntax error of a character other than { following "${if
1192 def:variable_name" (after optional whitespace) was not being diagnosed.
1193 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1194 accidental colon was present, for example, could give incorrect results.
1196 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1197 result is used (not including appendfile, where other changes are about
1200 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1201 This involved changing a lot of size variables from int to off_t. It
1202 should work with maildirs and everything.
1204 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1205 spamd dying while we are connected to it.
1207 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1210 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1211 function for BDB 4.3.
1213 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1215 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1216 the sources for PCRE have been re-arranged and more files are now
1219 PH/15 The code I had for printing potentially long long variables in PH/11
1220 above was not the best (it lost precision). The length of off_t variables
1221 is now inspected at build time, and an appropriate printing format (%ld
1222 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1223 to be "long long int" or "long int". This is needed for the internal
1224 formatting function string_vformat().
1226 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1227 the configuration file to be ":syslog", then the script "guesses" where
1228 the logs files are, rather than using the compiled in default. In our
1229 case the guess is not the same as the compiled default, so the script
1230 suddenly stopped working when I started to use syslog. The patch checks
1231 to see if log_file_path is "". If so, it attempts to read it from exim
1232 with no configuration file to get the compiled in version, before it
1233 falls back to the previous guessing code."
1235 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1236 implementing BATV in an Exim configuration. See NewStuff for the gory
1239 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1240 Makefile that are specific to HP-UX.
1242 PH/18 If the "use_postmaster" option was set for a recipient callout together
1243 with the "random" option, the postmaster address was used as the MAIL
1244 FROM address for the random test, but not for the subsequent recipient
1245 test. It is now used for both.
1247 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1248 patch removes a few documentation additions to RFC 3028, because the
1249 latest draft now contains them. It adds the new en;ascii-case comparator
1250 and a new error check for 8bit text in MIME parts. Comparator and
1251 require names are now matched exactly. I enabled the subaddress
1252 extension, but it is not well tested yet (read: it works for me)."
1254 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1255 rework some of the code of TK/09 above to avoid the hardwired use of
1256 "%lld" and "long long". Replaced the call to snprintf() with a call to
1259 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1260 records point to non-existent hosts", "retry timeout exceeded", and
1261 "retry time not reached for any host after a long failure period".
1263 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1264 experimental DomainKeys support:
1266 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1267 (2) On an error such as an illegally used "control", the wrong name for
1268 the control was given.
1270 These problems did NOT occur unless DomainKeys support was compiled.
1272 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1274 PH/24 Added ${if match_ip condition.
1276 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1277 databases so that it will be absolutely obvious if a crash occurs in the
1278 DB library. This is a regular occurrence (often caused by mis-matched
1281 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1282 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1283 picked up on a user's system that detects such things. There doesn't seem
1284 to be a gcc warning option for this - only an attribute that has to be
1285 put on the function's prototype. It seems that in Fedora Core 4 they have
1286 set this on a number of new functions. No doubt there will be more in due
1289 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1290 specify a transport. However, if an address that was verified by such a
1291 router was the subject of a callout, Exim crashed because it tried to
1292 read the rcpt_include_affixes from the non-existent transport. Now it
1293 just assumes that the setting of that option is false. This bug was
1294 introduced by 4.51/PH/31.
1296 PH/28 Changed -d+all to exclude +memory, because that information is very
1297 rarely of interest, but it makes the output a lot bigger. People tend to
1298 do -d+all out of habit.
1300 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1301 code in os-type was giving problems when libc.so lives in lib64, like on
1304 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1305 aren't the modern standard, and it seems that some systems' include files
1306 don't always have them. Exim was already checking for some of the newer
1307 ones like T_AAAA, and defining it itself. I've added checks for all the
1308 record types that Exim uses.
1310 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1311 not automatically generating a new one, as it is supposed to. This
1312 prevented TLS from working. If the file did exist, but contained invalid
1313 data, a new version was generated, as expected. It was only the case of a
1314 non-existent file that was broken.
1316 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1317 with a change in libdomainkeys > 0.64.
1319 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1320 from DNS. If the selector record carries the flag, it now has
1321 precedence over the domain-wide flag.
1323 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1325 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1326 the use of a port name that isn't defined in /etc/services) occurred, the
1327 message was deferred as in a normal delivery, and thus remained on the
1328 spool, instead of being failed because of the mua_wrapper setting. This
1329 is now fixed, and I tidied up some of the mua_wrapper messages at the
1332 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1333 the messages in a hash of arrays rather than using individual hashes.
1334 This is a bit cleaner and results in dramatic memory savings, albeit
1335 at a slight CPU cost.
1337 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1338 as requested by Marc Sherman.
1340 SC/10 Eximstats: added histograms for user specified patterns as requested
1343 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1345 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1346 fopen() in the content-scanning modules that did not already have it.
1352 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1353 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1355 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1357 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1359 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1360 to test Sieve filters that use "vacation".
1362 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1363 that changes the way the GnuTLS parameters are stored in the cache file.
1364 The new format can be generated externally. For backward compatibility,
1365 if the data in the cache doesn't make sense, Exim assumes it has read an
1366 old-format file, and it generates new data and writes a new file. This
1367 means that you can't go back to an older release without removing the
1370 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1371 work if there are any delivery delays because "one_time" forces the
1372 parent to be marked "delivered", so its unseen clone is never tried
1373 again. For this reason, Exim now forbids the simultaneous setting of
1376 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1377 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1378 entry if you want to know the details. The fix, however, neglected to
1379 consider the case when local delivery batching is involved. The test for
1380 "previously delivered" was not happening when checking to see if an
1381 address could be batched with a previous (undelivered) one; under
1382 certain circumstances this could lead to multiple deliveries to the same
1385 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1386 in its include files, and this causes problems building Exim.
1388 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1389 header_syntax) but Exim was just ignoring anything given after a slash.
1390 In particular, this caused confusion with an attempt to use "verify =
1391 reverse_host_lookup/defer_ok". An error is now given when options are
1392 supplied for verify items that do not have them. (Maybe reverse_host_
1393 lookup should have a defer_ok option, but that's a different point.)
1395 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1396 defined by RFC 821) to 2048, because there were problems with some AUTH
1397 commands, and RFC 1869 says the size should be increased for extended
1398 SMTP commands that take arguments.
1400 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1403 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1404 "unknown" error; now it says that the functionality isn't in the binary.
1406 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1407 an address' error message when a string expansion fails (syntax or
1408 whatever). Otherwise the password may appear in the log. Following change
1409 PH/42 below, there is no longer a chance of it appearing in a bounce
1412 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1414 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1415 of its addresses was ever tried. (Bugzilla bug #2.)
1417 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1418 the result incorrectly in the debug output. (It correctly added a newline
1419 to what was transported.)
1421 TF/01 Added $received_time.
1423 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1424 commented out examples of how to interface to a virus scanner and to
1425 SpamAssassin. Also added commented examples of av_scanner and
1426 spamd_address settings.
1428 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1429 and controls are allowed in which ACLs. There were a couple of minor
1430 errors. Some of the entries in the conditions table (which is a table of
1431 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1432 negation of where the condition IS allowed.
1434 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1436 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1437 header file does not have a version number, so I've had to invent a new
1438 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1439 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1440 radiusclient), but it was contributed by a Radius user.
1442 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1443 files or directories, for OpenSSL.
1445 PH/19 When an Exim process that is running as root has to create an Exim log
1446 file, it does so in a subprocess that runs as exim:exim so as to get the
1447 ownership right at creation (otherwise, other Exim processes might see
1448 the file with the wrong ownership). There was no test for failure of this
1449 fork() call, which would lead to the process getting stuck as it waited
1450 for a non-existent subprocess. Forks do occasionally fail when resources
1451 run out. I reviewed all the other calls to fork(); they all seem to check
1454 PH/20 When checking for unexpected SMTP input at connect time (before writing
1455 the banner), Exim was not dealing correctly with a non-positive return
1456 from the read() function. If the client had disconnected by this time,
1457 the result was a log entry for a synchronization error with an empty
1458 string after "input=" when read() returned zero. If read() returned -1
1459 (an event I could not check), uninitialized data bytes were printed.
1460 There were reports of junk text (parts of files, etc) appearing after
1463 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1465 PH/22 Added support for macro redefinition, and (re)definition in between
1466 driver and ACL definitions.
1468 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1469 forgetting to use the resulting value; it was using the unexpanded value.
1471 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1472 hadn't been configured. The fix is from Juergen Kreileder, who
1473 understands it better than I do:
1475 "Here's what I see happening with three configured cyrus_sasl
1476 authenticators configured (plain, login, cram-md5):
1478 On startup auth_cyrus_sasl_init() gets called for each of these.
1479 This means three calls to sasl_listmech() without a specified mech_list.
1480 => SASL tests which mechs of all available mechs actually work
1481 => three warnings about OTP not working
1482 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1484 With the patch, sasl_listmech() also gets called three times. But now
1485 SASL's mech_list option is set to the server_mech specified in the the
1486 authenticator. Or in other words, the answer from sasl_listmech()
1487 gets limited to just the mech you're testing for (which is different
1489 => the return list contains just 'plain' or 'login', 'cram-md5' or
1490 nothing depending on the value of ob->server_mech.
1492 I've just tested the patch: Authentication still works fine,
1493 unavailable mechs specified in the exim configuration are still
1494 caught, and the auth.log warnings about OTP are gone."
1496 PH/25 When debugging is enabled, the contents of the command line are added
1497 to the debugging output, even when log_selector=+arguments is not
1500 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1501 answer is "GNU", and only if the return is "GNU/something" is the answer
1504 PH/27 $acl_verify_message is now set immediately after the failure of a
1505 verification in an ACL, and so is available in subsequent modifiers. In
1506 particular, the message can be preserved by coding like this:
1508 warn !verify = sender
1509 set acl_m0 = $acl_verify_message
1511 Previously, $acl_verify_message was set only while expanding "message"
1512 and "log_message" when a very denied access.
1514 PH/28 Modified OS/os.c-Linux with
1516 -#ifndef OS_LOAD_AVERAGE
1517 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1519 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1520 nomenclature these days.)
1522 PH/29 Installed patch from the Sieve maintainer that adds the options
1523 sieve_useraddress and sieve_subaddress to the redirect router.
1525 PH/30 In these circumstances:
1526 . Two addresses routed to the same list of hosts;
1527 . First host does not offer TLS;
1528 . First host accepts first address;
1529 . First host gives temporary error to second address;
1530 . Second host offers TLS and a TLS session is established;
1531 . Second host accepts second address.
1532 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1533 and peerdn, if requested) that were in fact used only for the second
1536 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1537 attention to any local part prefix or suffix that was matched by the
1538 router that accepted the address. It now behaves in the same way as it
1539 does for delivery: the affixes are removed from the local part unless
1540 rcpt_include_affixes is set on the transport.
1542 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1543 timeout during the DATA phase of an incoming message.
1545 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1546 applied a patch sanctioned by the Sieve maintainer.
1548 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1549 the uid or gid is negative. A case of a negative gid caused this to be
1550 noticed. The fix allows for either to be negative.
1552 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1553 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1556 PH/36 The change PH/12 above was broken. Fixed it.
1558 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1559 the grounds that routing the same address twice would always produce the
1560 same answer. This might have been true once, but it is certainly no
1561 longer true now. Routing a child address may depend on the previous
1562 routing that produced that child. Some complicated redirection strategies
1563 went wrong when messages had multiple recipients, and made Exim's
1564 behaviour dependent on the order in which the addresses were given.
1566 I have moved the duplicate checking until after the routing is complete.
1567 Exim scans the addresses that are assigned to local and remote
1568 transports, and removes any duplicates. This means that more work will be
1569 done, as duplicates will always all be routed, but duplicates are
1570 presumably rare, so I don't expect this is of any significance.
1572 For deliveries to pipes, files, and autoreplies, the duplicate checking
1573 still happens during the routing process, since they are not going to be
1576 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1577 It corrects a timeout issue with spamd. This is Ian's comment: "The
1578 background is that sometimes spamd either never reads data from a
1579 connection it has accepted, or it never writes response data. The exiscan
1580 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1581 blindly assumes that writes won't block so it may never time out."
1583 PH/39 Allow G after quota size as well as K and M.
1585 PH/40 The value set for $authenticated_id in an authenticator may not contain
1586 binary zeroes or newlines because the value is written to log lines and
1587 to spool files. There was no check on this. Now the value is run through
1588 the string_printing() function so that such characters are converted to
1589 printable escape sequences.
1591 PH/41 $message_linecount is a new variable that contains the total number of
1592 lines in the message. Compare $body_linecount, which is the count for the
1595 PH/42 Exim no longer gives details of delivery errors for specific addresses in
1596 bounce and delay warning messages, except in certain special cases, which
1599 (a) An SMTP error message from a remote host;
1600 (b) A message specified in a :fail: redirection;
1601 (c) A message specified in a "fail" command in a system filter;
1602 (d) A message specified in a FAIL return from the queryprogram router;
1603 (e) A message specified by the cannot_route_message router option.
1605 In these cases only, Exim does include the error details in bounce and
1606 warning messages. There are also a few cases where bland messages such
1607 as "unrouteable address" or "local delivery error" are given.
1609 PH/43 $value is now also set for the "else" part of a ${run expansion.
1611 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1612 being worked on, but at least Exim now implements the latest version to
1615 PH/45 In a pipe transport, although a timeout while waiting for the pipe
1616 process to complete was treated as a delivery failure, a timeout while
1617 writing the message to the pipe was logged, but erroneously treated as a
1618 successful delivery. Such timeouts include transport filter timeouts. For
1619 consistency with the overall process timeout, these timeouts are now
1620 treated as errors, giving rise to delivery failures by default. However,
1621 there is now a new Boolean option for the pipe transport called
1622 timeout_defer, which, if set TRUE, converts the failures into defers for
1623 both kinds of timeout. A transport filter timeout is now identified in
1626 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1627 systems where "make" and "gmake" are different, calling "gmake" at top
1628 level broke things. I've arranged for the value of $(MAKE) to be passed
1629 from the Makefile to this script so that it can call the same version of
1633 A note about Exim versions 4.44 and 4.50
1634 ----------------------------------------
1636 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1637 changes of various kinds. As a consequence, a big documentation update was
1638 needed. This delayed the release for rather longer than seemed good, especially
1639 in the light of a couple of (minor) security issues. Therefore, the changes
1640 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1641 release. So 4.44 and 4.50 are in effect two different branches that both start
1644 I have left the 4.50 change log unchanged; it contains all the changes since
1645 4.43. The change log for 4.44 is below; many of its items are identical to
1646 those for 4.50. This seems to be the most sensible way to preserve the
1647 historical information.
1653 1. Minor wording change to the doc/README.SIEVE file.
1655 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
1656 computation of the current number of files was incorrect.
1658 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1659 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1662 4. Give more explanation in the error message when the command for a transport
1663 filter fails to execute.
1665 5. There are several places where Exim runs a non-Exim command in a
1666 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1667 was being done only for the command run by the queryprogram router. It is
1668 now done for all such subprocesses. The other cases are: ${run, transport
1669 filters, and the commands run by the lmtp and pipe transports.
1671 6. Added CONFIGURE_GROUP build-time option.
1673 7. Some older OS have a limit of 256 on the maximum number of file
1674 descriptors. Exim was using setrlimit() to set 1000 as a large value
1675 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1676 systems. I've change it so that if it can't get 1000, it tries for 256.
1678 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1679 was an oversight, and furthermore, ever since the addition of extra
1680 controls (e.g. 4.43/32), the checks on when to allow different forms of
1681 "control" were broken. There should now be diagnostics for all cases when a
1682 control that does not make sense is encountered.
1684 9. Added the /retain_sender option to "control=submission".
1686 10. $recipients is now available in the predata ACL (oversight).
1688 11. Tidy the search cache before the fork to do a delivery from a message
1689 received from the command line. Otherwise the child will trigger a lookup
1690 failure and thereby defer the delivery if it tries to use (for example) a
1691 cached ldap connection that the parent has called unbind on.
1693 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1694 of $address_data from the recipient verification was clobbered by the
1695 sender verification.
1697 13. The value of address_data from a sender verification is now available in
1698 $sender_address_data in subsequent conditions in the ACL statement.
1700 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1702 15. Added a new option "connect=<time>" to callout options, to set a different
1705 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1706 was its contents. (It was OK if the option was not defined at all.)
1708 17. A "Completed" log line is now written for messages that are removed from
1709 the spool by the -Mrm option.
1711 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1712 information about exactly what failed.
1714 19. Added -dd to debug only the daemon process.
1716 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1717 handles timeouts, both on the server side and network timeouts. Renamed the
1718 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1720 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1721 It is now set to "smtps".
1723 22. $host_address is now set to the target address during the checking of
1724 ignore_target_hosts.
1726 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1727 being passed; this would have caused $sender_host_name to have been used if
1728 matching the list had actually called for a host name (not very likely,
1729 since this list is usually IP addresses). A host name is now passed as
1732 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1733 code that specifies a non-restarting handler (typically sigaction() in
1734 modern systems) in an attempt to fix a rare and obscure crash bug.
1736 25. Narrowed the window for a race in the daemon that could cause it to ignore
1737 SIGCHLD signals. This is not a major problem, because they are used only to
1738 wake it up if nothing else does.
1740 26. A malformed maildirsize file could cause Exim to calculate negative values
1741 for the mailbox size or file count. Odd effects could occur as a result.
1742 The maildirsize information is now recalculated if the size or filecount
1745 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1746 support for a long time. Removed HAVE_SYS_VFS_H.
1748 28. Installed the latest version of exipick from John Jetmore.
1750 29. In an address list, if the pattern was not a regular expression, an empty
1751 subject address (from a bounce message) matched only if the pattern was an
1752 empty string. Non-empty patterns were not even tested. This was the wrong
1753 because it is perfectly reasonable to use an empty address as part of a
1754 database query. An empty address is now tested by patterns that are
1755 lookups. However, all the other forms of pattern expect the subject to
1756 contain a local part and a domain, and therefore, for them, an empty
1757 address still always fails if the pattern is not itself empty.
1759 30. Exim went into a mad DNS loop when attempting to do a callout where the
1760 host was specified on an smtp transport, and looking it up yielded more
1761 than one IP address.
1763 31. Re-factored the code for checking spool and log partition space into a
1764 function that finds that data and another that does the check. The former
1765 is then used to implement four new variables: $spool_space, $log_space,
1766 $spool_inodes, and $log_inodes.
1768 32. The RFC2047 encoding function was originally intended for short strings
1769 such as real names; it was not keeping to the 75-character limit for
1770 encoded words that the RFC imposes. It now respects the limit, and
1771 generates multiple encoded words if necessary. To be on the safe side, I
1772 have increased the buffer size for the ${rfc2047: expansion operator from
1775 33. It is now permitted to omit both strings after an "if" condition; if the
1776 condition is true, the result is "true". As before, when the second string
1777 is omitted, a false condition yields an empty string. This makes it less
1778 cumbersome to write custom ACL and router conditions.
1780 34. Failure to deliver a bounce message always caused it to be frozen, even if
1781 there was an errors_to setting on the router. The errors_to setting is now
1784 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1785 canonical form (fully expanded) before being placed in
1786 $sender_host_address.
1788 36. The table in the code that translates DNS record types into text (T_A to
1789 "A" for instance) was missing entries for NS and CNAME. It is just possible
1790 that this could have caused confusion if both these types were looked up
1791 for the same domain, because the text type is used as part of Exim's
1792 per-process caching. But the chance of anyone hitting this buglet seems
1795 37. The dnsdb lookup has been extended in a number of ways.
1797 (1) There is a new type, "zns", which walks up the domain tree until it
1798 finds some nameserver records. It should be used with care.
1800 (2) There is a new type, "mxh", which is like "mx" except that it returns
1801 just the host names, not the priorities.
1803 (3) It is now possible to give a list of domains (or IP addresses) to be
1804 looked up. The behaviour when one of the lookups defers can be
1805 controlled by a keyword.
1807 (4) It is now possible to specify the separator character for use when
1808 multiple records are returned.
1810 38. The dnslists ACL condition has been extended: it is now possible to supply
1811 a list of IP addresses and/or domains to be looked up in a particular DNS
1814 39. Added log_selector=+queue_time_overall.
1816 40. When running the queue in the test harness, wait just a tad after forking a
1817 delivery process, to get repeatability of debugging output.
1819 41. Include certificate and key file names in error message when GnuTLS fails
1820 to set them up, because the GnuTLS error message doesn't include the name
1821 of the failing file when there is a problem reading it.
1823 42. Allow both -bf and -bF in the same test run.
1825 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1827 44. The "Exiscan patch" is now merged into the mainline Exim source.
1829 45. Sometimes the final signoff response after QUIT could fail to get
1830 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1831 < 0 before doing a fflush(). This bug looks as though it goes back to the
1832 introduction of TLS in release 3.20, but "sometimes" must have been rare
1833 because the tests only now provoked it.
1835 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1836 (this can affect the format of dates).
1838 47. exim_tidydb, when checking for the continued existence of a message for
1839 which it has found a message-specific retry record, was not finding
1840 messages that were in split spool directories. Consequently, it was
1841 deleting retry records that should have stayed in existence.
1843 48. Steve fixed some bugs in eximstats.
1845 49. The SPA authentication driver was not abandoning authentication and moving
1846 on to the next authenticator when an expansion was forced to fail,
1847 contradicting the general specification for all authenticators. Instead it
1848 was generating a temporary error. It now behaves as specified.
1850 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1851 (the order specifies the preference for clients). The order is now AES256,
1852 AES128, 3DES, ARCFOUR128.
1854 51. Small patch to Sieve code - explicitly set From: when generating an
1857 52. Exim crashed if a remote delivery caused a very long error message to be
1858 recorded - for instance if somebody sent an entire SpamAssassin report back
1859 as a large number of 550 error lines. This bug was coincidentally fixed by
1860 increasing the size of one of Exim's internal buffers (big_buffer) that
1861 happened as part of the Exiscan merge. However, to be on the safe side, I
1862 have made the code more robust (and fixed the comments that describe what
1865 53. Now that there can be additional text after "Completed" in log lines (if
1866 the queue_time_overall log selector is set), a one-byte patch to exigrep
1867 was needed to allow it to recognize "Completed" as not the last thing in
1870 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1871 patch that reportedly fixes this has been added. I am not expert enough to
1872 create a test for it. This is what the patch creator wrote:
1874 "I found a little strange behaviour of ldap code when working with
1875 Windows 2003 AD Domain, where users was placed in more than one
1876 Organization Units. When I tried to give exim partial DN, the exit code
1877 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1878 But simultaneously result of request was absolutely normal ldap result,
1879 so I produce this patch..."
1881 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1882 REFERENCE, so I have modified the code to exclude the patch when that macro
1885 55. Some experimental protocols are using DNS PTR records for new purposes. The
1886 keys for these records are domain names, not reversed IP addresses. The
1887 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1888 leaves it alone. Component reversal etc. now happens only for IP addresses.
1891 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1893 57. Double the size of the debug message buffer (to 2048) so that more of very
1894 long debug lines gets shown.
1896 58. The exicyclog utility now does better if the number of log files to keep
1897 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1899 59. Two changes related to the smtp_active_hostname option:
1901 (1) $smtp_active_hostname is now available as a variable.
1902 (2) The default for smtp_banner uses $smtp_active_hostname instead
1903 of $primary_hostname.
1905 60. The host_aton() function is supposed to be passed a string that is known
1906 to be a valid IP address. However, in the case of IPv6 addresses, it was
1907 not checking this. This is a hostage to fortune. Exim now panics and dies
1908 if the condition is not met. A case was found where this could be provoked
1909 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1910 components; fortuitously, this particular loophole had already been fixed
1911 by change 4.50/55 above.
1913 If there are any other similar loopholes, the new check in host_aton()
1914 itself should stop them being exploited. The report I received stated that
1915 data on the command line could provoke the exploit when Exim was running as
1916 exim, but did not say which command line option was involved. All I could
1917 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1918 running as the user.
1921 61. There was a buffer overflow vulnerability in the SPA authentication code
1922 (which came originally from the Samba project). I have added a test to the
1923 spa_base64_to_bits() function which I hope fixes it.
1926 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1927 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1929 63. The daemon start-up calls getloadavg() while still root for those OS that
1930 need the first call to be done as root, but it missed one case: when
1931 deliver_queue_load_max is set with deliver_drop_privilege. This is
1932 necessary for the benefit of the queue runner, because there is no re-exec
1933 when deliver_drop_privilege is set.
1935 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1936 This has been fixed.
1938 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1939 was in use, was not putting the data itself into the right store pool;
1940 consequently, it could be overwritten for a subsequent message in the same
1941 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1944 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1946 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1947 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1948 the function were treating the return as a boolean value, which happened to
1949 work because 0=false and not-0=true, but is not correct code.
1951 68. The host_aton() function was not handling scoped IPv6 addresses (those
1952 with, for example, "%eth0" on the end) correctly.
1954 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1955 negated items (that is, ~something) in unsigned ints. Some compilers
1956 apparently mutter when there is no cast.
1958 70. If an address verification called from an ACL failed, and did not produce a
1959 user-specific message (i.e. there was only a "system" message), nothing was
1960 put in $acl_verify_message. In this situation, it now puts the system
1963 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1964 session; change 4.31/43 added the unwanted input to the log line - except
1965 that it did not do this in the start of session case. It now does.
1967 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1968 This is wrong and can cause the other end to generate a synchronization
1969 error if it is another Exim or anything else that does the synchronization
1970 check. A QUIT command is no longer sent after a timeout.
1972 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1973 during host lookups.
1975 74. The defer_ok option of callout verification was not working if it was used
1976 when verifying addresses in header lines, that is, for this case:
1978 verify = header_sender/callout=defer_ok
1980 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1981 those file descriptors could be used for SMTP connections. If anything
1982 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1983 could be sent to the SMTP client, causing chaos. The daemon now opens
1984 stdin, stdout, and stderr to /dev/null when it puts itself into the
1987 76. Arrange for output from Perl's "warn" command to be written to Exim's main
1988 log by default. The user can override this with suitable Perl magic.
1990 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
1991 the log message when discard triggers, was not working for the DATA ACL or
1992 for the non-SMTP ACL.
1994 78. Error message wording change in sieve.c.
1996 79. If smtp_accept_max_per_host was set, the number of connections could be
1997 restricted to fewer than expected, because the daemon was trying to set up
1998 a new connection before checking whether the processes handling previous
1999 connections had finished. The check for completed processes is now done
2000 earlier. On busy systems, this bug wouldn't be noticed because something
2001 else would have woken the daemon, and it would have reaped the completed
2004 80. If a message was submitted locally by a user whose login name contained one
2005 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2006 It caused a spool format error. I have fixed the spool reading code. A
2007 related problem was that the "from" clause in the Received: line became
2008 illegal because of the space(s). It is now covered by ${quote_local_part.
2010 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2013 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2015 83. Patch from Sieve maintainer to fix unterminated string problem in
2016 "vacation" handling.
2018 84. Some minor changes to the Linux configuration files to help with other
2019 OS variants using glibc.
2021 85. One more patch for Sieve to update vacation handling to latest spec.
2024 ----------------------------------------------------
2025 See the note above about the 4.44 and 4.50 releases.
2026 ----------------------------------------------------
2032 1. Change 4.43/35 introduced a bug that caused file counts to be
2033 incorrectly computed when quota_filecount was set in an appendfile
2036 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2037 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2040 3. Give more explanation in the error message when the command for a transport
2041 filter fails to execute.
2043 4. There are several places where Exim runs a non-Exim command in a
2044 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2045 was being done only for the command run by the queryprogram router. It is
2046 now done for all such subprocesses. The other cases are: ${run, transport
2047 filters, and the commands run by the lmtp and pipe transports.
2049 5. Some older OS have a limit of 256 on the maximum number of file
2050 descriptors. Exim was using setrlimit() to set 1000 as a large value
2051 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2052 systems. I've change it so that if it can't get 1000, it tries for 256.
2054 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2055 was an oversight, and furthermore, ever since the addition of extra
2056 controls (e.g. 4.43/32), the checks on when to allow different forms of
2057 "control" were broken. There should now be diagnostics for all cases when a
2058 control that does not make sense is encountered.
2060 7. $recipients is now available in the predata ACL (oversight).
2062 8. Tidy the search cache before the fork to do a delivery from a message
2063 received from the command line. Otherwise the child will trigger a lookup
2064 failure and thereby defer the delivery if it tries to use (for example) a
2065 cached ldap connection that the parent has called unbind on.
2067 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2068 of $address_data from the recipient verification was clobbered by the
2069 sender verification.
2071 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2072 was its contents. (It was OK if the option was not defined at all.)
2074 11. A "Completed" log line is now written for messages that are removed from
2075 the spool by the -Mrm option.
2077 12. $host_address is now set to the target address during the checking of
2078 ignore_target_hosts.
2080 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2081 being passed; this would have caused $sender_host_name to have been used if
2082 matching the list had actually called for a host name (not very likely,
2083 since this list is usually IP addresses). A host name is now passed as
2086 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2087 code that specifies a non-restarting handler (typically sigaction() in
2088 modern systems) in an attempt to fix a rare and obscure crash bug.
2090 15. Narrowed the window for a race in the daemon that could cause it to ignore
2091 SIGCHLD signals. This is not a major problem, because they are used only to
2092 wake it up if nothing else does.
2094 16. A malformed maildirsize file could cause Exim to calculate negative values
2095 for the mailbox size or file count. Odd effects could occur as a result.
2096 The maildirsize information is now recalculated if the size or filecount
2099 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2100 support for a long time. Removed HAVE_SYS_VFS_H.
2102 18. Updated exipick to current release from John Jetmore.
2104 19. Allow an empty sender to be matched against a lookup in an address list.
2105 Previously the only cases considered were a regular expression, or an
2108 20. Exim went into a mad DNS lookup loop when doing a callout where the
2109 host was specified on the transport, if the DNS lookup yielded more than
2112 21. The RFC2047 encoding function was originally intended for short strings
2113 such as real names; it was not keeping to the 75-character limit for
2114 encoded words that the RFC imposes. It now respects the limit, and
2115 generates multiple encoded words if necessary. To be on the safe side, I
2116 have increased the buffer size for the ${rfc2047: expansion operator from
2119 22. Failure to deliver a bounce message always caused it to be frozen, even if
2120 there was an errors_to setting on the router. The errors_to setting is now
2123 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2124 canonical form (fully expanded) before being placed in
2125 $sender_host_address.
2127 24. Updated eximstats to version 1.33
2129 25. Include certificate and key file names in error message when GnuTLS fails
2130 to set them up, because the GnuTLS error message doesn't include the name
2131 of the failing file when there is a problem reading it.
2133 26. Expand error message when OpenSSL has problems setting up cert/key files.
2136 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2137 (this can affect the format of dates).
2139 28. exim_tidydb, when checking for the continued existence of a message for
2140 which it has found a message-specific retry record, was not finding
2141 messages that were in split spool directories. Consequently, it was
2142 deleting retry records that should have stayed in existence.
2144 29. eximstats updated to version 1.35
2145 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2146 1.35 - bugfix such that pie charts by volume are generated correctly
2148 30. The SPA authentication driver was not abandoning authentication and moving
2149 on to the next authenticator when an expansion was forced to fail,
2150 contradicting the general specification for all authenticators. Instead it
2151 was generating a temporary error. It now behaves as specified.
2153 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2154 (the order specifies the preference for clients). The order is now AES256,
2155 AES128, 3DES, ARCFOUR128.
2157 31. Small patch to Sieve code - explicitly set From: when generating an
2160 32. Exim crashed if a remote delivery caused a very long error message to be
2161 recorded - for instance if somebody sent an entire SpamAssassin report back
2162 as a large number of 550 error lines. This bug was coincidentally fixed by
2163 increasing the size of one of Exim's internal buffers (big_buffer) that
2164 happened as part of the Exiscan merge. However, to be on the safe side, I
2165 have made the code more robust (and fixed the comments that describe what
2168 33. Some experimental protocols are using DNS PTR records for new purposes. The
2169 keys for these records are domain names, not reversed IP addresses. The
2170 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2171 leaves it alone. Component reversal etc. now happens only for IP addresses.
2174 34. The host_aton() function is supposed to be passed a string that is known
2175 to be a valid IP address. However, in the case of IPv6 addresses, it was
2176 not checking this. This is a hostage to fortune. Exim now panics and dies
2177 if the condition is not met. A case was found where this could be provoked
2178 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2179 components; fortuitously, this particular loophole had already been fixed
2180 by change 4.50/55 or 4.44/33 above.
2182 If there are any other similar loopholes, the new check in host_aton()
2183 itself should stop them being exploited. The report I received stated that
2184 data on the command line could provoke the exploit when Exim was running as
2185 exim, but did not say which command line option was involved. All I could
2186 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2187 running as the user.
2190 35. There was a buffer overflow vulnerability in the SPA authentication code
2191 (which came originally from the Samba project). I have added a test to the
2192 spa_base64_to_bits() function which I hope fixes it.
2195 36. The daemon start-up calls getloadavg() while still root for those OS that
2196 need the first call to be done as root, but it missed one case: when
2197 deliver_queue_load_max is set with deliver_drop_privilege. This is
2198 necessary for the benefit of the queue runner, because there is no re-exec
2199 when deliver_drop_privilege is set.
2201 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2202 was in use, was not putting the data itself into the right store pool;
2203 consequently, it could be overwritten for a subsequent message in the same
2204 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2207 38. Sometimes the final signoff response after QUIT could fail to get
2208 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2209 < 0 before doing a fflush(). This bug looks as though it goes back to the
2210 introduction of TLS in release 3.20, but "sometimes" must have been rare
2211 because the tests only now provoked it.
2217 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2218 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2219 Now it returns an integer. A number of calls were still expecting a T/F
2220 return. Fortuitously, in all cases, the tests worked in OK situations,
2221 which is the norm. However, things would have gone wrong on any write
2222 failures on the smtp file descriptor. This function is used when sending
2223 messages over SMTP and also when doing verify callouts.
2225 2. When Exim is called to do synchronous delivery of a locally submitted
2226 message (the -odf or -odi options), it no longer closes stderr before doing
2229 3. Implemented the mua_wrapper option.
2231 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2233 5. Implemented the functions header_remove(), header_testname(),
2234 header_add_at_position(), and receive_remove_recipient(), and exported them
2237 6. If an ACL "warn" statement specified the addition of headers, Exim already
2238 inserted X-ACL-Warn: at the start if there was no header name. However, it
2239 was not making this test for the second and subsequent header lines if
2240 there were newlines in the string. This meant that an invalid header could
2241 be inserted if Exim was badly configured.
2243 7. Allow an ACL "warn" statement to add header lines at the start or after all
2244 the Received: headers, as well as at the end.
2246 8. Added the rcpt_4xx retry error code.
2248 9. Added postmaster_mailfrom=xxx to callout verification option.
2250 10. Added mailfrom=xxxx to the callout verification option, for verify=
2253 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2254 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2256 12. Inserted some casts to stop certain compilers warning when using pointer
2257 differences as field lengths or precisions in printf-type calls (mostly
2258 affecting debugging statements).
2260 13. Added optional readline() support for -be (dynamically loaded).
2262 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2263 same clock tick as a message's arrival, so that its received time was the
2264 same as the "first fail" time on the retry record, and that message
2265 remained on the queue past the ultimate address timeout, every queue runner
2266 would try a delivery (because it was past the ultimate address timeout) but
2267 after another failure, the ultimate address timeout, which should have then
2268 bounced the address, did not kick in. This was a "< instead of <=" error;
2269 in most cases the first failure would have been in the next clock tick
2270 after the received time, and all would be well.
2272 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2273 being recognized when the domain list was tested by the match_domain
2274 condition in an expansion string.
2276 16. Added the ${str2b64: operator.
2278 17. Exim was always calling setrlimit() to set a large limit for the number of
2279 processes, without checking whether the existing limit was already
2280 adequate. (It did check for the limit on file descriptors.) Furthermore,
2281 errors from getrlimit() and setrlimit() were being ignored. Now they are
2282 logged to the main and panic logs, but Exim does carry on, to try to do its
2283 job under whatever limits there are.
2285 18. Imported PCRE 5.0.
2287 19. Trivial typo in log message " temporarily refused connection" (the leading
2290 20. If the log selector return_path_on_delivery was set and an address was
2291 redirected to /dev/null, the delivery process crashed because it assumed
2292 that a return path would always be set for a "successful" delivery. In this
2293 case, the whole delivery is bypassed as an optimization, and therefore no
2296 21. Internal re-arrangement: the function for sending a challenge and reading
2297 a response while authentication was assuming a zero-terminated challenge
2298 string. It's now changed to take a pointer and a length, to allow for
2299 binary data in such strings.
2301 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2303 23. Exim was not respecting finduser_retries when seeking the login of the
2304 uid under which it was called; it was always trying 10 times. (The default
2305 setting of finduser_retries is zero.) Also, it was sleeping after the final
2306 failure, which is pointless.
2308 24. Implemented tls_on_connect_ports.
2310 25. Implemented acl_smtp_predata.
2312 26. If the domain in control=submission is set empty, Exim assumes that the
2313 authenticated id is a complete email address when it generates From: or
2314 Sender: header lines.
2316 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2317 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2318 chown and chgrp in /bin and hostname in /usr/bin.
2320 28. Exim was keeping the "process log" file open after each use, just as it
2321 does for the main log. This opens the possibility of it remaining open for
2322 long periods when the USR1 signal hits a daemon. Occasional processlog
2323 errors were reported, that could have been caused by this. Anyway, it seems
2324 much more sensible not to leave this file open at all, so that is what now
2327 29. The long-running daemon process does not normally write to the log once it
2328 has entered its main loop, and it closes the log before doing so. This is
2329 so that log files can straightforwardly be renamed and moved. However,
2330 there are a couple of unusual error situations where the daemon does write
2331 log entries, and I had neglected to close the log afterwards.
2333 30. The text of an SMTP error response that was received during a remote
2334 delivery was being truncated at 512 bytes. This is too short for some of
2335 the long messages that one sometimes sees. I've increased the limit to
2338 31. It is now possible to make retry rules that apply only when a message has a
2339 specific sender, in particular, an empty sender.
2341 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2342 it possible to be selective about when SMTP synchronization is enforced.
2344 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2346 32. Implemented hosts_connection_nolog.
2348 33. Added an ACL for QUIT.
2350 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2353 35. Added mailbox_size and mailbox_filecount to appendfile.
2355 36. Added control = no_multiline_responses to ACLs.
2357 37. There was a bug in the logic of the code that waits for the clock to tick
2358 in the case where the clock went backwards by a substantial amount such
2359 that the microsecond fraction of "now" was more than the microsecond
2360 fraction of "then" (but the whole seconds number was less).
2362 38. Added support for the libradius Radius client library this is found on
2363 FreeBSD (previously only the radiusclient library was supported).
2369 1. When certain lookups returned multiple values in the form name=value, the
2370 quoting of the values was not always being done properly. Specifically:
2371 (a) If the value started with a double quote, but contained no whitespace,
2373 (b) If the value contained whitespace other than a space character (i.e.
2374 tabs or newlines or carriage returns) it was not quoted.
2375 This fix has been applied to the mysql and pgsql lookups by writing a
2376 separate quoting function and calling it from the lookup code. The fix
2377 should probably also be applied to nisplus, ibase and oracle lookups, but
2378 since I cannot test any of those, I have not disturbed their existing code.
2380 2. A hit in the callout cache for a specific address caused a log line with no
2381 reason for rejecting RCPT. Now it says "Previous (cached) callout
2382 verification failure".
2384 3. There was an off-by-one bug in the queryprogram router. An over-long
2385 return line was truncated at 256 instead of 255 characters, thereby
2386 overflowing its buffer with the terminating zero. As well as fixing this, I
2387 have increased the buffer size to 1024 (and made a note to document this).
2389 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2390 when Exim is waiting for an SMTP response from a remote server, Exim
2391 restarts its select() call on the socket, thereby resetting its timeout.
2392 This is not a problem when such interrupts are rare. Somebody set up a cron
2393 job to run exiwhat every 2 minutes, which is less than the normal select()
2394 timeout (5 or 10 minutes). This meant that the select() timeout never
2395 kicked in because it was always reset. I have fixed this by comparing the
2396 time when an interrupt arrives with the time at the start of the first call
2397 to select(). If more time than the timeout has elapsed, the interrupt is
2398 treated as a timeout.
2400 5. Some internal re-factoring in preparation for the addition of Sieve
2401 extensions (by MH). In particular, the "personal" test is moved to a
2402 separate function, and given an option for scanning Cc: and Bcc: (which is
2403 not set for Exim filters).
2405 6. When Exim created an email address using the login of the caller as the
2406 local part (e.g. when creating a From: or Sender: header line), it was not
2407 quoting the local part when it contained special characters such as @.
2409 7. Installed new OpenBSD configuration files.
2411 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2412 try to make them clearer.
2414 9. Callout options, other than the timeout value, were being ignored when
2415 verifying sender addresses in header lines. For example, when using
2417 verify = header_sender/callout=no_cache
2419 the cache was (incorrectly) being used.
2421 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2422 only the Cygwin environment.
2424 11. When return_path_on_delivery was set as a log selector, if different remote
2425 addresses in the same message used different return paths and parallel
2426 remote delivery occurred, the wrong values would sometimes be logged.
2427 (Whenever a remote delivery process finished, the return path value from
2428 the most recently started remote delivery process was logged.)
2430 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2431 lines when AUTH and/or TLS are in use. This is the "received protocol"
2432 field. Exim used to use "asmtp" for authenticated SMTP, without any
2433 indication (in the protocol name) for TLS use. Now it follows the RFC and
2434 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2435 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2436 names appear in log lines as well as in Received: header lines.
2438 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2439 extensions, and comparison tests, and to fix some bugs.
2441 14. Changes to the "personal" filter test:
2443 (1) The test was buggy in that it was just doing the equivalent of
2444 "contains" tests on header lines. For example, if a user's address was
2445 anne@some.where, the "personal" test would incorrectly be true for
2447 To: susanne@some.where
2449 This test is now done by extracting each address from the header in turn,
2450 and checking the entire address. Other tests that are part of "personal"
2451 are now done using regular expressions (for example, to check local parts
2452 of addresses in From: header lines).
2454 (2) The list of non-personal local parts in From: addresses has been
2455 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2456 taken from the Sieve specification recommendations.
2458 (3) If the message contains any header line starting with "List-" it is
2459 treated as non-personal.
2461 (4) The test for "circular" in the Subject: header line has been removed
2462 because it now seems ill-conceived.
2464 15. Minor typos in src/EDITME comments corrected.
2466 16. Installed latest exipick from John Jetmore.
2468 17. If headers_add on a router specified a text string that was too long for
2469 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2470 of string_sprintf() is now avoided.
2472 18. $message_body_size was not set (it was always zero) when running the DATA
2473 ACL and the local_scan() function.
2475 19. For the "mail" command in an Exim filter, no default was being set for
2476 the once_repeat time, causing a random time value to be used if "once" was
2477 specified. (If the value happened to be <= 0, no repeat happened.) The
2478 default is now 0s, meaning "never repeat". The "vacation" command was OK
2479 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2480 (I found it when inspecting the code).
2482 20. There is now an overall timeout for performing a callout verification. It
2483 defaults to 4 times the callout timeout, which applies to individual SMTP
2484 commands during the callout. The overall timeout applies when there is more
2485 than one host that can be tried. The timeout is checked before trying the
2486 next host. This prevents very long delays if there are a large number of
2487 hosts and all are timing out (e.g. when the network connections are timing
2488 out). The value of the overall timeout can be changed by specifying an
2489 additional sub-option for "callout", called "maxwait". For example:
2491 verify = sender/callout=5s,maxwait=20s
2493 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2494 to the end before writing, but this should make it even safer).
2496 22. Exim was forgetting that it had advertised PIPELINING for the second and
2497 subsequent messages on an SMTP connection. It was also not resetting its
2498 memory on STARTTLS and an internal HELO.
2500 23. When Exim logs an SMTP synchronization error within a session, it now
2501 records whether PIPELINING has been advertised or not.
2503 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2504 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2505 rather than long int.
2507 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2509 26. Added the never_mail option to autoreply.
2515 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2516 crash if the getsockname() call failed; this can happen if a connection is
2517 closed very soon after it is established. The problem was simply in the
2518 order in which certain operations were done, causing Exim to try to write
2519 to the SMTP stream before it had set up the file descriptor. The bug has
2520 been fixed by making things happen in the correct order.
2526 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2527 before the connection was closed, thus losing the rejection response.
2529 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2530 some early Solaris releases, but causes trouble in current releases where
2531 socklen_t is defined.
2533 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2536 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2539 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2540 that releasing the top of it at the end releases what was used for sub-
2541 expansions (unless the block got too big). However, discard this block if
2542 the first thing is a variable or header, so that we can use its block when
2543 it is dynamic (useful for very large $message_headers, for example).
2545 6. Lookups now cache *every* query, not just the most recent. A new, separate
2546 store pool is used for this. It can be recovered when all lookup caches are
2547 flushed. Lookups now release memory at the end of their result strings.
2548 This has involved some general refactoring of the lookup sources.
2550 7. Some code has been added to the store_xxx() functions to reduce the amount
2551 of flapping under certain conditions.
2553 8. log_incoming_interface used to affect only the <= reception log lines. Now
2554 it causes the local interface and port to be added to several more SMTP log
2555 lines, for example "SMTP connection from", and rejection lines.
2557 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2559 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2561 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2562 could be overwritten at the end of the current message (or the start of a
2563 new message if it was set in a HELO ACL). The value is now preserved for
2564 the duration of the SMTP connection.
2566 12. If a transport had a headers_rewrite setting, and a matching header line
2567 contained an unqualified address, that address was qualified, even if it
2568 did not match any rewriting rules. The underlying bug was that the values
2569 of the flags that permit the existence of unqualified sender and recipient
2570 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2571 non-local messages, and by -bnq for local messages) were not being
2572 preserved with the message after it was received.
2574 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2575 "next input=" as part of the text comprising the host identity instead of
2576 the correct text. The code was using the same buffer for two different
2577 strings. However, depending on which order the printing function evaluated
2578 its arguments, the bug did not always show up. Under Linux, for example, my
2579 test suite worked just fine.
2581 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2582 doesn't work with some older versions of Perl. It has been changed to "my",
2583 which in any case is probably the better facility to use.
2585 15. A really picky compiler found some instances of statements for creating
2586 error messages that either had too many or two few arguments for the format
2589 16. The size of the buffer for calls to the DNS resolver has been increased
2590 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2591 for addresses that have a lot of PTR records. This alleviates a problem; it
2592 does not fully solve it.
2594 17. A dnsdb lookup for PTR records that receives more data than will fit in the
2595 buffer now truncates the list and logs the incident, which is the same
2596 action as happens when Exim is looking up a host name and its aliases.
2597 Previously in this situation something unpredictable would happen;
2598 sometimes it was "internal error: store_reset failed".
2600 18. If a server dropped the connection unexpectedly when an Exim client was
2601 using GnuTLS and trying to read a response, the client delivery process
2602 crashed while trying to generate an error log message.
2604 19. If a "warn" verb in an ACL added multiple headers to a message in a single
2605 string, for example:
2607 warn message = H1: something\nH2: something
2609 the text was added as a single header line from Exim's point of view
2610 though it ended up OK in the delivered message. However, searching for the
2611 second and subsequent header lines using $h_h2: did not work. This has been
2612 fixed. Similarly, if a system filter added multiple headers in this way,
2613 the routers could not see them.
2615 20. Expanded the error message when iplsearch is called with an invalid key to
2616 suggest using net-iplsearch in a host list.
2618 21. When running tests using -bh, any delays imposed by "delay" modifiers in
2619 ACLs are no longer actually imposed (and a message to that effect is
2622 22. If a "gecos" field in a passwd entry contained escaped characters, in
2623 particular, if it contained a \" sequence, Exim got it wrong when building
2624 a From: or a Sender: header from that name. A second bug also caused
2625 incorrect handling when an unquoted " was present following a character
2626 that needed quoting.
2628 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2629 was not being matched caselessly.
2631 24. Arranged for all hyphens in the exim.8 source to be escaped with
2634 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2635 properly. Recipient callout cache records were still being keyed to include
2636 the sender, even when use_sender was set false. This led to far more
2637 callouts that were necessary. The sender is no longer included in the key
2638 when use_sender is false.
2640 26. Added "control = submission" modifier to ACLs.
2642 27. Added the ${base62d: operator to decode base 62 numbers.
2644 28. dnsdb lookups can now access SRV records.
2646 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2647 the configuration file.
2649 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2650 (-v) mode. This makes the output for a verbose queue run more intelligible.
2652 31. Added a use_postmaster feature to recipient callouts.
2654 32. Added the $body_zerocount variable, containing the number of binary zero
2655 bytes in the message body.
2657 33. The time of last modification of the "new" subdirectory is now used as the
2658 "mailbox time last read" when there is a quota error for a maildir
2661 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2663 35. Added +ignore_unknown as a special item in host lists.
2665 36. Code for decoding IPv6 addresses in host lists is now included, even if
2666 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2667 address was recognized as an IP address, but was then not correctly decoded
2668 into binary, causing unexpected and incorrect effects when compared with
2675 1. Very minor rewording of debugging text in manualroute to say "list of
2676 hosts" instead of "hostlist".
2678 2. If verify=header_syntax was set, and a header line with an unqualified
2679 address (no domain) and a large number of spaces between the end of the
2680 name and the colon was received, the reception process suffered a buffer
2681 overflow, and (when I tested it) crashed. This was caused by some obsolete
2682 code that should have been removed. The fix is to remove it!
2684 3. When running in the test harness, delay a bit after writing a bounce
2685 message to get a bit more predictability in the log output.
2687 4. Added a call to search_tidyup() just before forking a reception process. In
2688 theory, someone could use a lookup in the expansion of smtp_accept_max_
2689 per_host which, without the tidyup, could leave open a database connection.
2691 5. Added the variables $recipient_data and $sender_data which get set from a
2692 lookup success in an ACL "recipients" or "senders" condition, or a router
2693 "senders" option, similar to $domain_data and $local_part_data.
2695 6. Moved the writing of debug_print from before to after the "senders" test
2698 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2699 problems for message scanning, either using a data ACL, or using
2700 local_scan() because the Received: header was not generated till after they
2701 were called (in order to set the time as the time of reception completion).
2702 I have revised the way this works. The header is now generated after the
2703 body is received, but before the ACL or local_scan() are called. After they
2704 are run, the timestamp in the header is updated.
2710 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2711 before starting a queue runner without re-exec. This happened only when
2712 deliver_drop_privilege was set or when the Exim user was set to root. The
2713 effect of the bug was that timeouts during subsequent deliveries caused
2714 crashes instead of being properly handled. The handler is now left at its
2715 default (and expected) setting.
2717 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2718 message, again when deliver_drop_privilege is set or Exim is run as root.
2719 The bug described in (1) was not present in this case, but the tidying up
2720 of the other signals was missing. I have made the two cases consistent.
2722 3. The ignore_target_hosts setting on a manualroute router was being ignored
2723 for hosts that were looked up using the /MX notation.
2725 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2728 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2729 operated on the sender address. After changing the $sender_address to <>
2730 for the sender address verify, Exim was re-instated it as the original
2731 (before rewriting) address, but remembering that it had rewritten it, so it
2732 wasn't rewriting it again. This bug also had the effect of breaking the
2733 sender address verification caching when the sender address was rewritten.
2735 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2736 This has been changed so that if the ip literal address matches
2737 ignore_target_hosts, the router declines.
2739 7. Added expansion conditions match_domain, match_address, and match_local_
2740 part (NOT match_host).
2742 8. The placeholder for the Received: header didn't have a length field set.
2744 9. Added code to Exim itself and to exim_lock to test for a specific race
2745 condition that could lead to file corruption when using MBX delivery. The
2746 issue is with the lockfile that is created in /tmp. If this file is removed
2747 after a process has opened it but before that process has acquired a lock,
2748 there is the potential for a second process to recreate the file and also
2749 acquire a lock. This could lead to two Exim processes writing to the file
2750 at the same time. The added code performs the same test as UW imapd; it
2751 checks after acquiring the lock that its file descriptor still refers to
2752 the same named file.
2754 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2755 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2758 11. Added the smtp_active_hostname option. If used, this will typically be made
2759 to depend on the incoming interface address. Because $interface_address is
2760 not set up until the daemon has forked a reception process, error responses
2761 that can happen earlier (such as "too many connections") no longer contain
2764 12. If an expansion in a condition on a "warn" statement fails because a lookup
2765 defers, the "warn" statement is abandoned, and the next ACL statement is
2766 processed. Previously this caused the whole ACL to be aborted.
2768 13. Added the iplsearch lookup type.
2770 14. Added ident_timeout as a log selector.
2772 15. Added tls_certificate_verified as a log selector.
2774 16. Added a global option tls_require_ciphers (compare the smtp transport
2775 option of the same name). This controls incoming TLS connections.
2777 17. I finally figured out how to make tls_require_ciphers do a similar thing
2778 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2779 before starting the TLS session.
2781 18. Tabs are now shown as \t in -bP output.
2783 19. If the log selector return_path_on_delivery was set, Exim crashed when
2784 bouncing a message because it had too many Received: header lines.
2786 20. If two routers both had headers_remove settings, and the first one included
2787 a superfluous trailing colon, the final name in the first list and the
2788 first name in the second list were incorrectly joined into one item (with a
2789 colon in the middle).
2795 1. Added -C and -D options to the exinext utility, mainly to make it easier
2796 to include in the automated testing, but these could be helpful when
2797 multiple configurations are in use.
2799 2. The exinext utility was not formatting the output nicely when there was
2800 an alternate port involved in the retry record key, nor when there was a
2801 message id as well (for retries that were specific to a specific message
2802 and a specific host). It was also confused by IPv6 addresses, because of
2803 the additional colons they contain. I have fixed the IPv4 problem, and
2804 patched it up to do a reasonable job for IPv6.
2806 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2807 delivery, the log line now contains "pipelined" if PIPELINING was used.
2809 4. An SMTP transport process used to panic and die if the bind() call to set
2810 an explicit outgoing interface failed. This has been changed; it is now
2811 treated in the same way as a connect() failure.
2813 5. A reference to $sender_host_name in the part of a conditional expansion
2814 that was being skipped was still causing a DNS lookup. This no longer
2817 6. The def: expansion condition was not recognizing references to header lines
2818 that used bh_ and bheader_.
2820 7. Added the _cache feature to named lists.
2822 8. The code for checking quota_filecount in the appendfile transport was
2823 allowing one more file than it should have been.
2825 9. For compatibility with Sendmail, the command line option
2833 and sets the incoming protocol and host name (for trusted callers). The
2834 host name and its colon can be omitted when only the protocol is to be set.
2835 Note the Exim already has two private options, -pd and -ps, that refer to
2836 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2837 "s", but I don't think that's a major issue.
2839 10. A number of refactoring changes to the code, none of which should affect
2842 (a) The number of logging options was getting close to filling up the
2843 32-bit word that was used as a bit map. I have split them into two classes:
2844 those that are passed in the argument to log_write(), and those that are
2845 only ever tested independently outside of that function. These are now in
2846 separate 32-bit words, so there is plenty of room for expansion again.
2847 There is no change in the user interface or the logging behaviour.
2849 (b) When building, for example, log lines, the code previously used a
2850 macro that called string_cat() twice, in order to add two strings. This is
2851 not really sufficiently general. Furthermore, there was one instance where
2852 it was actually wrong because one of the argument was used twice, and in
2853 one call a function was used. (As it happened, calling the function twice
2854 did not affect the overall behaviour.) The macro has been replaced by a
2855 function that can join an arbitrary number of extra strings onto a growing
2858 (c) The code for expansion conditions now uses a table and a binary chop
2859 instead of a serial search (which was left over from when there were very
2860 few conditions). Also, it now recognizes conditions like "pam" even when
2861 the relevant support is not compiled in: a suitably worded error message is
2862 given if an attempt is made to use such a condition.
2864 11. Added ${time_interval:xxxxx}.
2866 12. A bug was causing one of the ddress fields not to be passed back correctly
2867 from remote delivery subprocesses. The field in question was not being
2868 subsequently used, so this caused to problems in practice.
2870 13. Added new log selectors queue_time and deliver_time.
2872 14. Might have fixed a bug in maildirsizefile handling that threw up
2873 "unexpected character" debug warnings, and recalculated the data
2874 unnecessarily. In any case, I expanded the warning message to give more
2877 15. Added the message "Restricted characters in address" to the statements in
2878 the default ACL that block characters like @ and % in local parts.
2880 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2881 Three changes have been made:
2883 (a) There was a serious bug; a negative response to MAIL caused the whole
2884 recipient domain to be cached as invalid, thereby blocking all messages
2885 to all local parts at the same domain, from all senders. This bug has
2886 been fixed. The domain is no longer cached after a negative response to
2887 MAIL if the sender used is not empty.
2889 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2892 (c) A new callout option, "use_sender" has been added for people who want
2893 the modified behaviour.
2899 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2902 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2903 indeed breaks things for older releases.
2905 3. Added additional logging to the case where there is a problem reading data
2906 from a filter that is running in a subprocess using a pipe, in order to
2907 try to track down a specific problem.
2909 4. Testing facility fudge: when running in the test harness and attempting
2910 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2911 getting "No route to host". Convert this to a timeout.
2913 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2916 6. Some OS don't have socklen_t but use size_t instead. This affects the
2917 fifth argument of getsockopt() amongst other things. This is now
2918 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2919 can be set for individual OS. I have set it for SunOS5, OSF1, and
2920 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2921 some earlier ones do not.
2923 7. Change 4.30/15 was not doing the test caselessly.
2925 8. The standard form for an IPv6 address literal was being rejected by address
2926 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2927 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2928 this, as well as the form without the "IPv6" on the front (but only when
2929 address literals are enabled, of course).
2931 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2933 10. Exim crashed if a message with an empty sender address specified by -f
2934 encountered a router with an errors_to setting. This could be provoked only
2935 by a command such as
2939 where an empty string was supplied; "<>" did not hit this bug.
2941 11. Installed PCRE release 4.5.
2943 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2944 remained set. It is now erased.
2946 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2947 times from message ids (which are base 36 rather than the normal 62).
2949 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2950 were being counted as actual protocol errors, and logged if the log
2951 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2952 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2953 and DATA following a set of rejected RCPTs do not count as protocol errors.
2954 In other words, Exim assumes they were pipelined, though this may not
2955 actually be the case. Of course, in all cases the client gets an
2956 appropriate error code.
2958 15. If a lookup fails in an ACL condition, a message about the failure may
2959 be available; it is used if testing the ACL cannot continue, because most
2960 such messages specify what the cause of the deferral is. However, some
2961 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2962 that caused an old message to be retained and used if a later statement
2963 caused a defer, replacing the real cause of the deferral.
2965 16. If an IP address had so many PTR records that the DNS lookup buffer
2966 was not large enough to hold them, Exim could crash while trying to process
2967 the truncated data. It now detects and logs this case.
2969 17. Further to 4.21/58, another change has been made: if (and only if) the
2970 first line of a message (the first header line) ends with CRLF, a bare LF
2971 in a subsequent header line has a space inserted after it, so as not to
2972 terminate the header.
2974 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
2975 unnecessarily, used atoi() instead of strtol(), and didn't check the
2976 termination when getting file sizes from file names by regex.
2978 19. Completely re-implemented the support for maildirsize files, in the light
2979 of a number of problems with the previous contributed implementation
2980 (4.30/29). In particular:
2982 . If the quota is zero, the maildirsize file is maintained, but no quota is
2985 . If the maildir directory does not exist, it is created before any attempt
2986 to write a maildirsize file.
2988 . The quota value in the file is just a cache; if the quota is changed in
2989 the transport, the new value overrides.
2991 . A regular expression is available for excluding directories from the
2994 20. The autoreply transport checks the characters in options that define the
2995 message's headers; it allows continued headers, but it was checking with
2996 isspace() after an embedded newline instead of explicitly looking for a
2999 21. If all the "regular" hosts to which an address was routed had passed their
3000 expiry times, and had not reached their retry times, the address was
3001 bounced, even if fallback hosts were defined. Now Exim should go on to try
3004 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3005 equivalent code in the SMTP transport. Some hosts send humungous responses
3006 to HELO/EHLO, more than 1024 it seems.
3008 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3009 gives compiler warnings in some environments. I've now done it "properly",
3012 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3013 (because of problems with the built-in one) was declared to return uschar *
3014 instead of char *, causing compiler failure.
3016 25. Fixed a file descriptor leak when processing alias/forward files.
3018 26. Fixed a minor format string issue in dbfn.c.
3020 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3022 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3023 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3026 29. When a sender address is verified, it is cached, to save repeating the test
3027 when there is more than one recipient in a message. However, when the
3028 verification involves a callout, it is possible for different callout
3029 options to be set for different recipients. It is too complicated to keep
3030 track of this in the cache, so now Exim always runs a verification when a
3031 callout is required, relying on the callout cache for the optimization.
3032 The overhead is duplication of the address routing, but this should not be
3035 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3036 to be verified with callout=postmaster, and the main callout worked but the
3037 postmaster check failed, the verification correctly failed. However, if a
3038 subsequent RCPT command asked for sender verification *without* the
3039 postmaster check, incorrect caching caused this verification also to fail,
3042 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3043 it was not caching the DNS options (qualify_single, search_parents) that
3044 were used when the lookup failed. A subsequent lookup with different
3045 options therefore always gave the same answer, though there were cases
3046 where it should not have. (Example: a "domains = !$mx_any" option on a
3047 dnslookup router: the "domains" option is always processed without any
3048 widening, but the router might have qualify_single set.) Now Exim uses the
3049 cached value only when the same options are set.
3051 32. Added John Jetmore's "exipick" utility to the distribution.
3053 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3054 than a timeout (e.g. a certificate is required, and is not provided), an
3055 Exim server now closes the connection immediately. Previously it waited for
3056 the client to close - but if the client is SSL, it seems that they each
3057 wait for each other, leading to a delay before one of them times out.
3059 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3060 maintained 0.8.x compatibility because I don't think many are using it, and
3061 it is clearly obsolete.
3063 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3066 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3067 client certificate was expired. A simple patch fixes this, though I don't
3068 understand the full logic of why the verify callback is called multiple
3071 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3072 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3073 which causes problems with some clients (such as the Certicom SSL Plus
3074 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3075 disables the coutermeasure allowing Eudora to connect."
3077 38. Exim was not checking that a write() to a log file succeeded. This could
3078 lead to Bad Things if a log got too big, in particular if it hit a file
3079 size limit. Exim now panics and dies if it cannot write to a log file, just
3080 as it does if it cannot open a log file.
3082 39. Modified OS/Makefile-Linux so that it now contains
3084 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3086 The two -D definitions ensure that Exim is compiled with large file
3087 support, which makes it possible to handle log files that are bigger than
3090 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3091 instance) a domain was checked against a named list that involved a lookup,
3092 causing $domain_data to be set, then another domain was checked against the
3093 same list, then the first domain was re-checked, the value of $domain_data
3094 after the final check could be wrong. In particular, if the second check
3095 failed, it could be set empty. This bug probably also applied to
3098 41. The strip_trailing_dot option was not being applied to the address given
3099 with the -f command-line option.
3101 42. The code for reading a message's header from the spool was incrementing
3102 $received_count, but never initializing it. This meant that the value was
3103 incorrect (doubled) while delivering a message in the same process in which
3104 it was received. In the most common configuration of Exim, this never
3105 happens - a fresh exec is done - but it can happen when
3106 deliver_drop_privilege is set.
3108 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3109 it now includes up to 150 characters of the unexpected data in the log
3112 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3113 and building data strings. The size of both of these buffers was 10 000
3114 bytes - far larger than anybody would *ever* want, thought I. Needless to
3115 say, somebody hit the limit. I have increased the maximum line length to
3116 20 000 and the maximum data length of concatenated lines to 100 000. I have
3117 also fixed two bugs, because there was no checking on these buffers. Tsk,
3118 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3119 buffer is too small.
3121 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3122 lsearch lookups. Now it does.
3124 46. When parsing a route_list item in a manualroute router, a fixed-length
3125 buffer was used for the list of hosts. I made this 1024 bytes long,
3126 thinking that nobody would ever have a list of hosts that long. Wrong.
3127 Somebody had a whole pile of complicated expansion conditions, and the
3128 string was silently truncated, leading to an expansion error. It turns out
3129 that it is easier to change to an unlimited length (owing to other changes
3130 that have happened since this code was originally written) than to build
3131 structure for giving a limitation error. The length of the item that
3132 expands into the list of hosts is now unlimited.
3134 47. The lsearch lookup could not handle data where the length of text line was
3135 more than 4095 characters. Such lines were truncated, leading to shortened
3136 data being returned. It should now handle lines of any length.
3138 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3139 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3142 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3143 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3144 no longer happens in comment lines. A list of replacements is now placed
3145 at the head of all of the source files, except those whose only change is
3146 to replace PERL_COMMAND in the very first #! line.
3148 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3149 messages on the queue, with a bottom-up merge sort, using code contributed
3150 by Michael Haardt. This should make operations like -bp somewhat faster on
3151 large queues. It won't affect queue runners, except when queue_run_in_order
3154 51. Installed eximstats 1.31 in the distribution.
3156 52. Added support for SRV lookups to the dnslookup router.
3158 53. If an ACL referred to $message_body or $message_body_end, the value was not
3159 reset for any messages that followed in the same SMTP session.
3161 54. The store-handling optimization for building very long strings was not
3162 differentiating between the different store pools. I don't think this
3163 actually made any difference in practice, but I've tidied it.
3165 55. While running the routers to verify a sender address, $sender_address
3166 was still set to the sender address. This is wrong, because when routing to
3167 send a bounce to the sender, it would be empty. Therefore, I have changed
3168 it so that, while verifying a sender address, $sender_address is set to <>.
3169 (There is no change to what happens when verifying a recipient address.)
3171 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3172 target A or AAAA records (if not already returned) without resetting the
3173 qualify_single or search_parents options of the DNS resolver. These are
3174 inappropriate in this case because the targets of MX and SRV records must
3175 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3176 target that, when qualified, matched something in the local domain. These
3177 two options are now turned off when doing these lookups.
3179 57. It seems that at least some releases of Reiserfs (which does not have the
3180 concept of a fixed number of inodes) returns zero and not -1 for the
3181 number of available inodes. This interacted badly with check_spool_inodes,
3182 which assumed that -1 was the "no such thing" setting. What I have done is
3183 to check that the total number of inodes is greater than zero before doing
3184 the test of how many are available.
3186 58. When a "warn" ACL statement has a log_message modifier, the message is
3187 remembered, and not repeated. This is to avoid a lot of repetition when a
3188 message has many recipients that cause the same warning to be written.
3189 Howewer, Exim was preserving the list of already written lines for an
3190 entire SMTP session, which doesn't seem right. The memory is now reset if a
3191 new message is started.
3193 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3194 debugging output unless log_rewrite was also set.
3196 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3197 of (int)(handle) when we know that handle contains (void *)(-1).
3199 61. The Exim daemon panic-logs an error return when it closes the incoming
3200 connection. However "connection reset by peer" seems to be common, and
3201 isn't really an error worthy of noting specially, so that particular error
3204 62. When Exim is trying to find all the local interfaces, it used to panic and
3205 die if the ioctl to get the interface flags failed. However, it seems that
3206 on at least one OS (Solaris 9) it is possible to have an interface that is
3207 included in the list of interfaces, but for which you get a failure error
3208 for this call. This happens when the interface is not "plumbed" into a
3209 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3210 failure of the "get flags" call assumes that the interface is down.
3212 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3213 makes life easier for people who are doing arithmetic on fields extracted
3214 from dates, where you often get leading zeros that should not be
3215 interpreted as octal.
3217 64. Added qualify_domain to the redirect router, to override the global
3220 65. If a pathologically long header line contained very many addresses (the
3221 report of this problem mentioned 10 000) and each of them was rewritten,
3222 Exim could use up a very large amount of memory. (It kept on making new
3223 copies of the header line as it rewrote, and never released the old ones.)
3224 At the expense of a bit more processing, the header rewriting function has
3225 been changed so that it no longer eats memory in this way.
3227 66. The generation of the Received: header has been moved from the time that a
3228 message starts to be received, to the time that it finishes. The timestamp
3229 in the Received: header should now be very close to that of the <= log
3230 line. There are two side-effects of this change:
3232 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3233 logged header lines no longer include the local Received: line, because
3234 it has not yet been created. The same applies to a copy of the message
3235 that is returned to a non-SMTP sender when a message is rejected.
3237 (b) When a filter file is tested using -bf, no additional Received: header
3238 is added to the test message. After some thought, I decided that this
3241 This change does not affect the value of $received_for. It is still set
3242 after address rewriting, but before local_scan() is called.
3244 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3246 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3247 gave an unhelpful panic error message, and a defer error. I have managed to
3248 change this behaviour so that it now rejects any supplied certificate,
3249 which seems right, as the list of acceptable certificates is empty.
3251 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3252 gave an unhelpful defer error. I have not managed to make this reject any
3253 supplied certificates, but the error message it gives is "no certificate
3254 supplied", which is not helpful.
3256 70. exigrep's output now also includes lines that are not associated with any
3257 message, but which match the given pattern. Implemented by a patch from
3258 Martin Sluka, which also tidied up the Perl a bit.
3260 71. Recipient callout verification, like sender verification, was using <> in
3261 the MAIL FROM command. This isn't really the right thing, since the actual
3262 sender may affect whether the remote host accepts the recipient or not. I
3263 have changed it to use the actual sender in the callout; this means that
3264 the cache record is now keyed on a recipient/sender pair, not just the
3265 recipient address. There doesn't seem to be a real danger of callout loops,
3266 since a callout by the remote host to check the sender would use <>.
3267 [SEE ABOVE: changed after hitting problems.]
3269 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3270 temporary errors. However, in the case of such a code being given after
3271 the end of a data transmission (i.e. after ".") Exim was failing to write
3272 a retry record for the message. (Yes, there was some broken host that was
3273 actually sending 8xx at this point.)
3275 73. An unknown lookup type in a host list could cause Exim to panic-die when
3276 the list was checked. (An example that provoked this was putting <; in the
3277 middle of a list instead of at the start.) If this happened during a DATA
3278 ACL check, a -D file could be left lying around. This kind of configuration
3279 error no longer causes Exim to die; instead it causes a defer errror. The
3280 incident is still logged to the main and panic logs.
3282 74. Buglet left over from Exim 3 conversion. The message "too many messages
3283 in one connection" was written to the rejectlog but not the mainlog, except
3284 when address rewriting (yes!) was being logged.
3286 75. Added write_rejectlog option.
3288 76. When a system filter was run not as root (that is, when system_filter_user
3289 was set), the values of the $n variables were not being returned to the
3290 main process; thus, they were not subsequently available in the $sn
3293 77. Added +return_path_on_delivery log selector.
3295 78. A connection timeout was being treated differently from recipients deferred
3296 when testing hosts_max_try with a message that was older than the host's
3297 retry timeout. (The host should not be counted, thus allowing all hosts to
3298 be tried at least once before bouncing.) This may have been the cause of an
3299 occasionally reported bug whereby a message would remain on the queue
3300 longer than the retry timeout, but would be bounced if a delivery was
3301 forced. I say "may" because I never totally pinned down the problem;
3302 setting up timeout/retry tests is difficult. See also the next item.
3304 79. The ultimate address timeout was not being applied to errors that involved
3305 a combination of host plus message (for example, a timeout on a MAIL
3306 command). When an address resolved to a number of possible hosts, and they
3307 were not all tried for each delivery (e.g. because of hosts_max_try), a
3308 message could remain on the queue longer than the retry timeout.
3310 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3313 81. Fixed an obscure SMTP outgoing bug which required at least the following
3314 conditions: (a) there was another message waiting for the same server;
3315 (b) the server returned 5xx to all RCPT commands in the first message so
3316 that the message was not completed; (c) the server dropped the connection
3317 or gave a negative response to the RSET that Exim sends to abort the
3318 transaction. The observed case was a dropped connection after DATA that had
3319 been sent in pipelining mode. That is, the server had advertised PIPELINING
3320 but was not implementing it correctly. The effect of the bug was incorrect
3321 behaviour, such as trying another host, and this could lead to a crash.
3327 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3328 and daemon.c were passed as pointers to ints; they should have been
3329 pointers to socklen_t variables (which are typically unsigned ints).
3331 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3334 3. Fixed a really odd bug that affected only the testing scheme; patching a
3335 certain fixed string in the binary changed the value of another string that
3336 happened to be identical to the end of the original first string.
3338 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3339 name", it returns that address as the IP address. On some operating
3340 systems (e.g. Solaris), it also passes back the IP address string as the
3341 "host name". However, on others (e.g. Linux), it passes back an empty
3342 string. Exim wasn't checking for this, and was changing the host name to an
3343 empty string, assuming it had been canonicized.
3345 5. Although rare, it is permitted to have more than one PTR record for a given
3346 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3347 all the names associated with an address, because they do in Solaris.
3348 However, it seems that they do not in Linux for data that comes from the
3349 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3350 I found this out when I moved to a new Linux workstation and tried to run
3351 the Exim test suite.
3353 To get round this problem I have changed the code so that it now does its
3354 own call to the DNS to look up PTR records when searching for a host name.
3355 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3356 addresses that are only in /etc/hosts are still found.
3358 This behaviour is, however, controlled by an option called host_lookup_
3359 order, which defaults to "bydns:byaddr". If people want to use the other
3360 order, or indeed, just use one or the other means of lookup, they can
3361 specify it in this variable.
3363 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3364 some operating systems, this comes back from gethostbyaddr() as an empty
3365 string, and this is what Exim used to test for. However, it seems that in
3366 other systems, "." is yielded. Exim now tests for this case too.
3368 7. The values of check_spool_space and check_log_space are now held internally
3369 as a number of kilobytes instead of an absolute number of bytes. If a
3370 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3371 kilobyte. This means that much larger values can be stored.
3373 8. Exim monitor: an attempt to get the action menu when not actually pointing
3374 at a message produces an empty menu entitled "No message selected". This
3375 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3376 no entries in it ("Shell widget menu has zero width and/or height"). So I
3377 have added a single, blank menu entry in this case.
3379 9. Added ${quote_local_part.
3381 10. MIME decoding is now applied to the contents of Subject: header lines when
3384 11. Now that a reference to $sender_host_address automatically causes a reverse
3385 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3386 host lookup before query-style lookups in lists that might use this
3387 variable. This has therefore been abolished, and the "net-" prefix is no
3388 longer necessary for query-style lookups.
3390 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3391 have been a typo for LFLAGS, so it has been changed.
3393 13. The install script calls Exim with "-C /dev/null" in order to find the
3394 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3395 to be output. Howeve, since Exim outputs its version number before the
3396 error, it didn't break the script. It just looked ugly. I fixed this by
3397 always allowing "-C /dev/null" if the caller is root.
3399 14. Ignore overlarge ACL variable number when reading spool file - insurance
3400 against a later release with more variables having written the file.
3402 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3403 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3404 this, as well as the form without the "IPv6" on the front.
3406 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3407 OS/Makefile-Darwin file.
3409 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3410 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3412 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3413 variable. However, some OS use ints for inodes. I've added cast to long int
3414 to get rid of the compiler warning.
3416 19. I had forgotten to lock out "/../" in configuration file names when
3417 ALT_CONFIG_PREFIX was set.
3419 20. Routers used for verification do not need to specify transports. However,
3420 if such a router generated a host list, and callout was configured, Exim
3421 crashed, because it could not find a port number from the (non-existent)
3422 transport. It now assumes port 25 in this circumstance.
3424 21. Added the -t option to exigrep.
3426 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3427 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3428 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3429 them is set, LOOKUP_LSEARCH is forced.
3431 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3433 24. Added sender and host information to the "rejected by local_scan()" log
3434 line; previously there was no indication of these.
3436 25. Added .include_if_exists.
3438 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3439 Linux. It turns out that not all file systems support this. Apparently some
3440 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3441 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3442 not supported on the file descriptor, is now ignored when Exim is trying to
3443 sync a directory. This applies only to Linux.
3445 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3447 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3448 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3451 29. Added support for maildirsize files from supplied patch (modified a bit).
3453 30. The use of :fail: followed by an empty string could lead Exim to respond to
3454 sender verification failures with (e.g.):
3456 550 Verification failed for <xxx>
3457 550 Sender verify failed
3459 where the first response line was missing the '-' that indicates it is not
3460 the final line of the response.
3462 31. The loop for finding the name of the user that called Exim had a hardwired
3463 limit of 10; it now uses the value of finduser_retries, which is used for
3464 all other user lookups.
3466 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3469 33. Exim was neglecting to zero errno before one call of strtol() when
3470 expanding a string and expecting an integer value. On some systems this
3471 resulted in spurious "integer overflow" errors. Also, it was casting the
3472 result into an int without checking.
3474 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3475 did not work. The code looks as if it has *never* worked, though it appears
3476 to have been documented since at least releast 1.62. I have made it work.
3478 35. The "timeout_DNS" error in retry rules, also documented since at least
3479 1.62, also never worked. As it isn't clear exactly what this means, and
3480 clearly it isn't a major issue, I have abolished the feature by treating it
3481 as "timeout", and writing a warning to the main and panic logs.
3483 36. The display of retry rules for -brt wasn't always showing the error code
3486 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3487 timeout_connect_A, timeout_connect_MX.
3489 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3490 to the empty sender.
3492 39. The daemon was not analysing the content of -oX till after it had closed
3493 stderr and disconnected from the controlling terminal. This meant that any
3494 syntax errors were only noted on the panic log, and the return code from
3495 the command was 0. By re-arranging the code a little, I've made the
3496 decoding happen first, so such errors now appear on stderr, and the return
3497 code is 1. However, the actual setting up of the sockets still happens in
3498 the disconnected process, so errors there are still only recorded on the
3501 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3502 connections (as happens on some IP stacks) was logged at start up time as
3503 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3504 it from "IPv6 and IPv4", which means that two separate sockets are being
3507 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3508 says whether AF_INET or AF_INET6 was passed as an argument.
3510 42. Exiwhat output was messed up when time zones were included in log
3513 43. Exiwhat now gives more information about the daemon's listening ports,
3514 and whether -tls-on-connect was used.
3516 44. The "port" option of the smtp transport is now expanded.
3518 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3519 silently ignored. Now an error message is written to the main and panic
3522 46. There's a new ACL modifier called "logwrite" which writes to a log file
3523 as soon as it is encountered.
3525 47. Added $local_user_uid and $local_user_gid at routing time.
3527 48. Exim crashed when trying to verify a sender address that was being
3530 49. Exim was recognizing only a space character after ".include". It now also
3531 recognizes a tab character.
3533 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3534 extracting the relevant information from the specification. The man page no
3535 longer contains scrambled data for the -d option, and I've added a section
3536 at the front about calling Exim under different names.
3538 51. Added "extra_headers" argument to the "mail" command in filter files.
3540 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3543 53. Installed eximstats 1.29.
3545 54. Added transport_filter_timeout as a generic transport option.
3547 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3548 Cc: header lines. This was required by RFC 822, but it not required by RFC
3551 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3552 incoming messages that did not have them. Now it does so only if the
3553 message originates locally, that is, if there is no associated remote host
3554 address. When Resent- header lines are present, this applies to the Resent-
3555 lines rather than the non-Resent- lines.
3557 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3558 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3560 58. Messages for configuration errors now include the name of the main
3561 configuration file - useful now that there may be more than one file in a
3562 list (.included file names were always shown).
3564 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3565 for those rare installations that do not start the daemon as root or run it
3566 setuid root. I've cut out the call to initgroups() if the daemon is not
3569 60. The Exim user and group can now be bound into the binary as text strings
3570 that are looked up at the start of Exim's processing.
3572 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3574 62. Added $mailstore_basename variable.
3576 63. Installed patch to sieve.c from Michael Haardt.
3578 64. When Exim failed to open the panic log after failing to open the main log,
3579 the original message it was trying to log was written to stderr and debug
3580 output, but if they were not available (the usual case in production), it
3581 was lost. Now it is written to syslog before the two lines that record the
3582 failures to open the logs.
3584 65. Users' Exim filters run in subprocesses under the user's uid. It is
3585 possible for a "deliver" command or an alias in a "personal" command to
3586 provoke an address rewrite. If logging of address rewriting is configured,
3587 this fails because the process is not running as root or exim. There may be
3588 a better way of dealing with this, but for the moment (because 4.30 needs
3589 to be released), I have disabled address rewrite logging when running a
3590 filter in a non-root, non-exim process.
3596 1. The buildconfig auxiliary program wasn't quoting the value set for
3597 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3598 not defined. This bug was masked in 4.22 by the effect that was fixed in
3601 2. Some messages that were rejected after a message id was allocated were
3602 shown as "incomplete" by exigrep. It no longer does this for messages that
3603 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3605 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3606 have allow_domain_literals set, the ID did not get logged in the <= line.
3607 Domain literals are now always recognized in Message-ID: header lines.
3609 4. The first argument for a ${extract expansion item is the key name or field
3610 number. Leading and trailing spaces in this item were not being ignored,
3611 causing some misleading effects.
3613 5. When deliver_drop_privilege was set, single queue runner processes started
3614 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3615 same command in the process it spins off) were not dropping privilege.
3617 6. When the daemon running as "exim" started a queue runner, it always
3618 re-executed Exim in the spun-off process. This is a waste of effort when
3619 deliver_drop_privilege is set. The new process now just calls the
3620 queue-runner function directly.
3626 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3629 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3630 ignored. Though the use of -oP was forcing the writing of a pid file, it
3631 was always written to the default place.
3633 3. If the message "no IP address found for host xxxx" is generated during
3634 incoming verification, it is now followed by identification of the incoming
3635 connection (so you can more easily find what provoked it).
3637 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3639 5. Added some features to "harden" Exim a bit more against certain attacks:
3641 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3642 be put in Local/Makefile. This is like the never_users runtime option,
3643 but it cannot be overridden. The default setting is "root".
3645 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3646 prefix string with which any file named in a -C command line option
3649 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3650 is retained for -C and -D only if the caller of Exim is root. Without
3651 it, the exim user may also use -C and -D and retain privilege.
3653 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3654 command line option is disabled.
3656 6. Macro names set by the -D option must start with an upper case letter, just
3657 like macro names defined in the configuration file.
3659 7. Added "dereference=" facility to LDAP.
3661 8. Two instances of the typo "uknown" in the source files are fixed.
3663 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3664 the Configure-Makefile script screwed up while processing it.
3666 10. Incorporated PCRE 4.4.
3668 11. The SMTP synchronization check was not operating right at the start of an
3669 SMTP session. For example, it could not catch a HELO sent before the client
3670 waited for the greeting. There is now a check for outstanding input at the
3671 point when the greeting is written. Because of the duplex, asynchronous
3672 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3673 way, but not yet received, when the check is performed.
3675 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3676 on TCP/IP sockets, because this apparently causes some broken clients to
3679 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3680 unchanged) from the Cygwin maintainer.
3682 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
3683 was supported instead of testing for mbx. Effectively a typo.
3685 15. The spa authenticator server code was not checking that the input it
3686 received was valid base64.
3688 16. The debug output line for the "set" modifier in ACLs was not showing the
3689 name of the variable that was being set.
3691 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
3693 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
3694 lookup on its own. Something else was needed to trigger the lookup. For
3695 example, a match in host_lookup or the need for a host name in a host list.
3696 Now, if $sender_host_name is referenced and the host name has not yet been
3697 looked up, a lookup is performed. If the lookup fails, the variable remains
3698 empty, and $host_lookup_failed is set to "1".
3700 19. Added "eqi" as a case-independent comparison operator.
3702 20. The saslauthd authentication condition could segfault if neither service
3703 nor realm was specified.
3705 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3706 error message that was logged was misleading, and incoming SMTP
3707 connections were dropped. The message is now more accurate, and temporary
3708 errors are given to SMTP connections.
3710 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3711 (except RSET) with a 421 error, until QUIT is received. However, it was
3712 failing to send a response to QUIT.
3714 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3715 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3716 ACL is now run after the helo_try_verify_hosts code.
3718 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3719 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3720 case-independent, so other case variants are also recognized). Apparently
3721 some systems use these upper case variants.
3723 25. If more than two messages were waiting for the same host, and a transport
3724 filter was specified for the transport, Exim sent two messages over the
3725 same TCP/IP connection, and then failed with "socket operation on non-
3726 socket" when it tried to send the third.
3728 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3730 27. The extern definition of crypt16() in expand.c was not being excluded when
3731 the OS had its own crypt16() function.
3733 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3734 as a preferred synonym for return_size_limit, both as an option and as an
3737 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3739 30. Changed the default configuration ACL to relax the local part checking rule
3740 for addresses that are not in any local domains. For these addresses,
3741 slashes and pipe symbols are allowed within local parts, but the sequence
3742 /../ is explicitly forbidden.
3744 31. SPA server authentication was not clearing the challenge buffer before
3747 32. log_message in a "warn" ACL statement was writing to the reject log as
3748 well as to the main log, which contradicts the documentation and doesn't
3749 seem right (because no rejection is happening). So I have stopped it.
3751 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3752 However, I am unable to do any testing of this.
3754 34. Fixed an infelicity in the appendfile transport. When checking directories
3755 for a mailbox, to see if any needed to be created, it was accidentally
3756 using path names with one or more superfluous leading slashes; tracing
3757 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3759 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3760 contents are added to the log line that is written for every discarded
3761 recipient. (Previously a log_message setting was ignored.)
3763 36. The ${quote: operator now quotes the string if it is empty.
3765 37. The install script runs exim in order to find its version number. If for
3766 some reason other than non-existence or emptiness, which it checks, it
3767 could not run './exim', it was installing it with an empty version number,
3768 i.e. as "exim-". This error state is now caught, and the installation is
3771 38. An argument was missing from the function that creates an error message
3772 when Exim fails to connect to the socket for saslauthd authentication.
3773 This could cause Exim to crash, or give a corrupted message.
3775 39. Added isip, isip4, and isip6 to ${if conditions.
3777 40. The ACL variables $acl_xx are now saved with the message, and can be
3778 accessed later in routers, transports, and filters.
3780 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3781 strings in the file are not string-expanded.
3783 42. If a MAIL command specified a SIZE value that was too large to fit into an
3784 int variable, the check against message_size_limit failed. Such values are
3785 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3786 day this will have to be increased, but I don't think I want to be around
3787 when emails are that large.
3794 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3795 iconv() is not standard in FreeBSD.
3797 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3798 IPv6 enabled. The observed symptom was a segmentation fault on return from
3799 the function os_common_find_running_interfaces() in src/os.c.
3801 3. In the check_special_case() function in daemon.c I had used "errno" as an
3802 argument name, which causes warnings on some systems. This was basically a
3803 typo, since it was named "eno" in the comments!
3805 4. The code that waits for the clock to tick (at a resolution of some fraction
3806 of a second) so as to ensure message-id uniqueness was always waiting for
3807 at least one whole tick, when it could have waited for less. [This is
3808 almost certainly not relevant at current processor speeds, where it is
3809 unlikely to ever wait at all. But we try to future-proof.]
3811 5. The function that sleeps for a time interval that includes fractions of a
3812 second contained a race. It did not block SIGALRM between setting the
3813 timer, and suspending (a couple of lines later). If the interval was short
3814 and the sigsuspend() was delayed until after it had expired, the suspension
3815 never ended. On busy systems this could lead to processes getting stuck for
3818 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3819 process, before it forks any delivery processes. The open lookup caching
3820 mechanism meant that the open file or database connection was passed into
3821 the delivery process. The problem was that delivery processes always tidy
3822 up cached lookup data. This could cause a problem for the next delivery
3823 process started by the queue runner, because the external queue runner
3824 process does not know about the closure. So the next delivery process
3825 still has data in the lookup cache. In the case of a file lookup, there was
3826 no problem because closing a file descriptor in a subprocess doesn't affect
3827 the parent. However, if the lookup was caching a connection to a database,
3828 the connection was closed, and the second delivery process was likely to
3829 see errors such as "PGSQL: query failed: server closed the connection
3830 unexpectedly". The problem has been fixed by closing all cached lookups
3831 in a queue runner before running a delivery process.
3833 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3834 seem to have the "const" qualifier which it has on other OS. I've
3837 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3838 *of the same type* (client or server) with the same public name that an
3839 error should be diagnosed.
3841 9. When Exim looked up a host name for an IP address, but failed to find the
3842 original IP address when looking up the host name (a safety check), it
3843 output the message "<ip address> does not match any IP for NULL", which was
3844 confusing, to say the least. The bug was that the host name should have
3845 appeared instead of "NULL".
3847 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3848 user that is built into the binary, and the -C or -D options is used, root
3849 privilege is dropped before the configuration file is read. In addition,
3850 logging is switched to stderr instead of the normal log files. If the
3851 configuration then re-defines the Exim user, the unprivileged environment
3852 is probably not what is expected, so Exim logs a panic warning message (but
3855 However, if deliver_drop_privilege is set, the unprivileged state may well
3856 be exactly what is intended, so the warning has been cut out in that case,
3857 and Exim is allowed to try to write to its normal log files.
3863 1. smtp_return_error_details was not giving details for temporary sender
3864 or receiver verification errors.
3866 2. Diagnose a configuration error if two authenticators have the same public
3869 3. Exim used not to create the message log file for a message until the first
3870 delivery attempt. This could be confusing when incoming messages were held
3871 for policy or load reasons. The message log file is now created at the time
3872 the message is received, and an initial "Received" line is written to it.
3874 4. The automatically generated man page for command line options had a minor
3875 bug that caused no ill effects; however, a more serious problem was that
3876 the procedure for building the man page automatically didn't always
3877 operate. Consequently, release 4.20 contains an out-of-date version. This
3878 shouldn't happen again.
3880 5. When building Exim with embedded Perl support, the script that builds the
3881 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3882 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3884 6. The freeze_tell option was not being used for messages that were frozen on
3885 arrival, either by an ACL or by local_scan().
3887 7. Added the smtp_incomplete_transaction log selector.
3889 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3890 was accepting AUTH without a new EHLO.
3892 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3893 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3896 10. Logging of TCP/IP connections (when configured) now happens in the main
3897 daemon process instead of the child process, so that the TCP/IP connection
3898 count is more accurate (but it can never be perfect).
3900 11. The use of "drop" in a nested ACL was not being handled correctly in the
3901 outer ACL. Now, if condition failure induced by the nested "drop" causes
3902 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3903 or "require"), the connection is dropped.
3905 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3906 that yield "discard" can now be used with an "accept" or a "discard" verb,
3907 but an error is generated for any others (because I can't see a useful way
3908 to define what should happen).
3910 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3911 are now processed in the same way as lines in the Exim configuration file.
3912 In particular, continuation lines are supported.
3914 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3916 15. Added -ti meaning -t -i.
3918 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3919 domains, and warn by logging if others are found.
3921 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3922 returned from GIFCONF when Exim is trying to find the list of interfaces on
3923 a host. The code in os.c has been modified to copy each ifreq to an aligned
3924 structure in all cases.
3926 Also, in some cases, the returned ifreq's were being copied to a 'struct
3927 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3928 means the last couple of bytes of an IPv6 address could be chopped if the
3929 ifreq contained only a normal sockaddr (14 bytes storage).
3931 18. Named domain lists were not supported in the hosts_treat_as_local option.
3932 An entry such as +xxxx was not recognized, and was treated as a literal
3935 19. Ensure that header lines added by a DATA ACL are included in the reject log
3936 if the ACL subsequently rejects the message.
3938 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3939 MD5 (which is deprecated).
3941 21. When testing a filter file using -bf, Exim was writing a message when it
3942 took the sender from a "From " line in the message, but it was not doing so
3943 when it took $return_path from a Return-Path: header line. It now does.
3945 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3946 with a valid header line field name (a series of printing characters
3947 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3949 23. Changed "disc" in the source to "disk" to conform to the documentation and
3950 the book and for uniformity.
3952 24. Ignore Sendmail's -Ooption=value command line item.
3954 25. When execve() failed while trying to run a command in a pipe transport,
3955 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3956 could be confused with a return value of 69 from the command itself. This
3957 has been changed to 127, the value the shell returns if it is asked to run
3958 a non-existent command. The wording for the related log line suggests a
3959 non-existent command as the problem.
3961 26. If received_header_text expands to an empty string, do not add a Received:
3962 header line to the message. (Well, it adds a token one on the spool, but
3963 marks it "old" so that it doesn't get used or transmitted.)
3965 27. Installed eximstats 1.28 (addition of -nt option).
3967 28. There was no check for failure on the call to getsockname() in the daemon
3968 code. This can fail if there is a shortage of resources on the system, with
3969 ENOMEM, for example. A temporary error is now given on failure.
3971 29. Contrary to the C standard, it seems that in some environments, the
3972 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3973 program. Exim now does this explicitly; it affects the formatting of
3974 timestamps using strftime().
3976 30. If exiqsumm was given junk data, it threw up some uninitialized variable
3977 complaints. I've now initialized all the variables, to avoid this.
3979 32. Header lines added by a system filter were not being "seen" during
3980 transport-time rewrites.
3982 33. The info_callback() function passed to OpenSSL is set up with type void
3983 (*)(SSL *, int, int), as described somewhere. However, when calling the
3984 function (actually a macro) that sets it up, the type void(*)() is
3985 expected. I've put in a cast to prevent warnings from picky compilers.
3987 34. If a DNS black list lookup found a CNAME record, but there were no A
3988 records associated with the domain it pointed at, Exim crashed.
3990 35. If a DNS black list lookup returned more than one A record, Exim ignored
3991 all but the first. It now scans all returned addresses if a particular IP
3992 value is being sought. In this situation, the contents of the
3993 $dnslist_value variable are a list of all the addresses, separated by a
3996 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
3997 to accept a host name and all its aliases if the forward lookup for any of
3998 them yielded the IP address of the incoming connection. Now it accepts only
3999 those names whose forward lookup yields the correct IP address. Any other
4000 names are discarded. This closes a loophole whereby a rogue DNS
4001 administrator could create reverse DNS records to break through a
4002 wildcarded host restriction in an ACL.
4004 37. If a user filter or a system filter that ran in a subprocess used any of
4005 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4006 the wrong values were passed to the pipe command ($thisaddress had the
4007 value of $0, $0 had the value of $1, etc). This bug was introduced by
4008 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4010 38. Improved the line breaking for long SMTP error messages from ACLs.
4011 Previously, if there was no break point between 40 and 75 characters, Exim
4012 left the rest of the message alone. Two changes have been made: (a) I've
4013 reduced the minimum length to 35 characters; (b) if it can't find a break
4014 point between 35 and 75 characters, it looks ahead and uses the first one
4015 that it finds. This may give the occasional overlong line, but at least the
4016 remaining text gets split now.
4018 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4019 file descriptors might be low, and that setting 1000 would always raise it.
4020 It turns out that in some environments, the limit is already over 1000 and
4021 that lowering it causes trouble. So now Exim takes care not to decrease it.
4023 40. When delivering a message, the value of $return_path is set to $sender_
4024 address at the start of routing (routers may change the value). By an
4025 oversight, this default was not being set up when an address was tested by
4026 -bt or -bv, which affected the outcome if any router or filter referred to
4029 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4030 log only when "message" or "log_message" are set. However, if one of the
4031 conditions was an address verification, or a call to a nested ACL, the
4032 messages generated by the underlying test were being passed through. This
4033 no longer happens. The underlying message is available in $acl_verify_
4034 message for both "message" and "log_message" expansions, so it can be
4035 passed through if needed.
4037 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4038 new expansion $bh_ to give the encoded byte string without charset
4039 translation. Translation happens only if iconv() is available; HAVE_ICONV
4040 indicates this at build time. HEADERS_CHARSET gives the charset to
4041 translate to; headers_charset can change it in the configuration, and
4042 "headers charset" can change it in an individual filter file.
4044 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4045 that creates RFC 2047 encoded "words" labels them as that charset instead
4046 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4047 expansion operator; (ii) when Exim creates a From: line for a local
4048 message; (iii) when a header line is rewritten to include a "phrase" part.
4050 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4051 buggy, causing it to skip the first lines of messages whose message ID
4052 ended in 'D'. This would not have bitten before Exim release 4.14, because
4053 message IDs were unlikely to end in 'D' before then. The effect was to have
4054 incorrect size information for certain domains.
4056 45. #include "config.h" was missing at the start of the crypt16.c module. This
4057 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4060 46. If there was a timeout during a "random" callout check, Exim treated it as
4061 a failure of the random address, and carried on sending RSET and the real
4062 address. If the delay was just some slowness somewhere, the response to the
4063 original RCPT would be taken as a response to RSET and so on, causing
4064 mayhem of various kinds.
4066 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4067 when I implemented it. It didn't allow for the fact that some option values
4068 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4069 the right test for positive values.
4071 48. Domain names in DNS records are case-independent. Exim always looks them up
4072 in lower case. Some resolvers return domain names in exactly the case they
4073 appear in the zone file, that is, they may contain uppercase letters. Not
4074 all resolvers do this - some return always lower case. Exim was treating a
4075 change of case by a resolver as a change of domain, similar to a widening
4076 of a domain abbreviation. This triggered its re-routing code and so it was
4077 trying to route what was effectively the same domain again. This normally
4078 caused routing to fail (because the router wouldn't handle the domain
4079 twice). Now Exim checks for this case specially, and just changes the
4080 casing of the domain that it ultimately uses when it transmits the message
4083 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4086 50. If a filter generated a file delivery with a non-absolute name (possible if
4087 no home directory exists for the router), the forbid_file option was not
4090 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4091 the existing equality matching.
4093 52. Exim was using ints instead of ino_t variables in some places where it was
4094 dealing with inode numbers.
4096 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4097 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4098 called TMPDIR, and if it finds it is different, it changes its value.
4100 54. The smtp_printf() function is now made available to local_scan() so
4101 additional output lines can be written before returning. There is also an
4102 smtp_fflush() function to enable the detection of a dropped connection.
4103 The variables smtp_input and smtp_batched_input are exported to
4106 55. Changed the default runtime configuration: the message "Unknown user"
4107 has been removed from the ACL, and instead placed on the localuser router,
4108 using the cannot_route_message feature. This means that any verification
4109 failures that generate their own messages won't get overridden. Similarly,
4110 the "Unrouteable address" message that was in the ACL for unverifiable
4111 relay addresses has also been removed.
4113 56. Added hosts_avoid_esmtp to the smtp transport.
4115 57. The exicyclog script was not checking for the esoteric option
4116 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4117 will work only if exicyclog is run under the appropriate euid.
4119 58. Following a discussion on the list, the rules by which Exim recognises line
4120 endings on incoming messages have been changed. The -dropcr and drop_cr
4121 options are now no-ops, retained only for backwards compatibility. The
4122 following line terminators are recognized: LF CRLF CR. However, special
4123 processing applies to CR:
4125 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4126 nor a local message in the state where . is a terminator.
4128 (ii) If a bare CR is encountered in a header line, an extra space is added
4129 after the line terminator so as not to end the header. The reasoning
4130 behind this is that bare CRs in header lines are most likely either
4131 to be mistakes, or people trying to play silly games.
4133 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4134 was being incorrectly given as 18 bytes larger than it should have been.
4135 This is a VOB (very old bug).
4137 60. This may never have affected anything current, but just in case it has:
4138 When the local host is found other than at the start of a list of hosts,
4139 the local host, those with the same MX, and any that follow, are discarded.
4140 When the list in question was part of a longer list of hosts, the following
4141 hosts (not currently being processed) were also being discarded. This no
4142 longer happens. I'm not sure if this situation could ever has previously
4145 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4148 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4149 header. This is something that is recommended in a new Internet Draft, and
4150 is something that is documented as being done by Sendmail. There are two
4151 possible values. For messages generated by the autoreply transport, Exim
4154 Auto-Submitted: auto-replied
4156 whereas for all other generated messages (e.g. bounces) it adds
4158 Auto-Submitted: auto-generated
4160 63. The "personal" condition in filters now includes a test for the
4161 Auto-Submitted: header. If it contains the string "auto-" the message it
4162 not considered personal.
4164 64. Added rcpt_include_affixes as a generic transport option.
4166 65. Added queue_only_override (default true).
4168 66. Added the syslog_duplication option.
4170 67. If what should have been the first header line of a message consisted of
4171 a space followed by a colon, Exim was mis-interpreting it as a header line.
4172 It isn't of course - it is syntactically invalid and should therefore be
4173 treated as the start of the message body. The misbehaviour could have
4174 caused a number of strange effects, including loss of data in subsequent
4175 header lines, and spool format errors.
4177 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4178 client host had authenticated. This control can now be exercised by an ACL
4179 for more flexibility.
4181 69. By default, callouts do not happen when testing with -bh. There is now a
4182 variant, -bhc, which does actually run the callout code, including
4183 consulting and updating the callout cache.
4185 70. Added support for saslauthd authentication, courtesy of Alexander
4188 71. If statvfs() failed on the spool or log directories while checking their
4189 size for availability, Exim confusingly gave the error "space shortage".
4190 Furthermore, in debugging mode it crashed with a floating point exception.
4191 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4192 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4193 serious problem, Exim now writes to the main and panic logs when this
4194 happens, with details of the failure. It then refuses to accept the
4195 incoming message, giving the message "spool directory problem" or "log
4196 directory problem" with a 421 code for SMTP messages.
4198 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4199 0, 1, and 2 exist, because some OS complain for execs without them (see
4200 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4201 descriptors. However, the code omitted to check that the open succeeded,
4202 causing mysterious errors if for some reason the permissions on /dev/null
4203 got screwed. Now Exim writes a message to the main and panic logs, and
4204 bombs out if it can't open /dev/null.
4206 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4207 interact so that it is all more flexible. It is supposed to remain
4208 backwards compatible. Also added extra_local_interfaces.
4210 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4211 to bomb out with an assertion failure - to the client this appears as a
4212 connection drop. This problem occurs in the part of the code that was taken
4213 from the Samba project. Fortunately, the assertion is in a very simple
4214 function, so I have fixed this by reproducing the function inline in the
4215 one place where it is called, and arranging for authentication to fail
4216 instead of killing the process with assert().
4218 75. The SPA client code was not working when the server requested OEM rather
4219 than Unicode encoding.
4221 76. Added code to make require_files with a specific uid setting more usable in
4222 the case where statting the file as root fails - usually a non-root-mounted
4223 NFS file system. When this happens and the failure is EACCES, Exim now
4224 forks a subprocess and does the per-uid checking as the relevant uid.
4226 77. Added process_log_path.
4228 78. If log_file_path was not explicitly set, a setting of check_log_space or
4229 check_log_inodes was ignored.
4231 79. If a space check for the spool or log partitions fails, the incident is now
4232 logged. Of course, in the latter case the data may get lost...
4234 80. Added the %p formatting code to string_format() so that it can be used to
4235 print addresses in debug_print(). Adjusted all the address printing in the
4236 debugging in store.c to use %p rather than %d.
4238 81. There was a concern that a line of code in smtp_in.c could overflow a
4239 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4240 initially expressed, the concern was not well-founded, because trailing
4241 spaces are removed early. However, if the trailing spaces were followed by
4242 a NULL, they did not get removed, so the overflow was possible. Two fixes
4245 (a) I re-wrote the offending code in a cleaner fashion.
4246 (b) If an incoming SMTP command contains a NULL character, it is rejected
4249 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4250 runs initgroups(), so that if the Exim user is in any additional groups,
4251 they will be used during message reception.
4257 The change log for 4.20 and earlier releases has been archived.