1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.416 2006/10/24 15:01:25 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 (see also 21 below):
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.
137 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
138 feature that was used to support insecure browsers during the U.S. crypto
139 embargo. It requires special client support, and Exim is probably the
140 only MTA that supported it -- and would never use it because real RSA is
141 always available. This code has been removed, because it had the bad
142 effect of slowing Exim down by computing (never used) parameters for the
143 RSA_EXPORT functionality.
145 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
146 authenticator to fail if there's a tab character in the incoming data
147 (there should never be unless someone is messing about, as it's supposed
148 to be base64-encoded). Also added, on Timo's advice, the "secured" option
149 if the connection is using TLS or if the remote IP is the same as the
150 local IP, and the "valid-client-cert option" if a client certificate has
153 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
154 authenticators. This can be used for authorization after authentication
155 succeeds. (In the case of plaintext, it servers for both authentication
158 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
159 if any retry times were supplied.
161 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
162 connection, where there is no client IP address to check. In this
163 situation, the verify now always succeeds.
165 PH/25 Applied John Jetmore's -Mset patch.
167 PH/26 Added -bem to be like -Mset, but loading a message from a file.
169 PH/27 In a string expansion for a processed (not raw) header when multiple
170 headers of the same name were present, leading whitespace was being
171 removed from all of them, but trailing whitespace was being removed only
172 from the last one. Now trailing whitespace is removed from each header
173 before concatenation.
175 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
176 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
182 SC/01 Use a glob alias rather than an array ref in eximstats generated
183 parser. This improves both readability and performance.
185 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
186 Don't display local sender or destination tables in eximstats unless
187 there is data to show.
188 Added average volumes into the eximstats top table text output.
190 SC/03 Collect data on the number of addresses (recipients) as well
191 as the number of messages in eximstats.
193 TF/01 Correct an error in the documentation for the redirect router. Exim
194 does (usually) call initgroups() when daemonizing.
196 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
197 with consistent privilege compared to when running as a daemon.
199 TF/03 Note in the spec that $authenticated_id is not set for local
200 submissions from trusted users.
202 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
203 Thanks to Dean Brooks <dean@iglou.com> for the patch.
205 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
206 by adding some example configuration directives to the default
207 configuration file. A little bit of work is required to uncomment the
208 directives and define how usernames and passwords are checked, but
209 there is now a framework to start from.
211 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
212 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
213 without this. I don't know how relevant this is to other LDAP libraries.
215 PH/02 Add the verb name to the "unknown ACL verb" error.
217 PH/03 Magnus Holmgren's patch for filter_prepend_home.
219 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
221 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
222 directory not expanded when it should be if an expanded home directory
223 was set for the address (which is overridden by the transport).
225 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
228 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
229 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
230 because it is too late at that time, and has no effect.
232 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
233 security issue with \' (bugzilla #107). I could not use the
234 PQescapeStringConn() function, because it needs a PGconn value as one of
237 PH/08 When testing addresses using -bt, indicate those final addresses that
238 are duplicates that would not cause an additional delivery. At least one
239 person was confused, thinking that -bt output corresponded to deliveries.
240 (Suppressing duplicates isn't a good idea as you lose the information
241 about possibly different redirections that led to the duplicates.)
243 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
244 systems where poll() doesn't work, in particular OS X.
246 PH/10 Added more information to debugging output for retry time not reached.
248 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
249 operations in malware.c.
251 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
254 PH/13 If write_rejectlog was set false when logging was sent to syslog with
255 syslog_duplication set false, log lines that would normally be written
256 both the the main log and to the reject log were not written to syslog at
259 PH/14 In the default configuration, change the use of "message" in ACL warn
260 statements to "add_header".
262 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
263 not followed by a command (e.g. "seen endif").
265 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
266 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
269 PH/17 Added extra conditions to the default value of delay_warning_condition
273 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
274 { match{$h_precedence:}{(?i)bulk|list|junk} } \
275 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
278 The Auto-Submitted: and various List- headers are standardised, whereas I
279 don't think Precedence: ever was.
281 PH/18 Refactored debugging code in route_finduser() to show more information,
282 in particular, the error code if getpwnam() issues one.
284 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
285 This is apparently needed in addition to the PH/07 change above to avoid
286 any possible encoding problems.
288 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
289 but not after initializing Perl.
291 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
292 output them only if debugging. By default they are written stderr,
293 apparently, which is not desirable.
295 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
298 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
301 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
303 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
304 authenticated or an ident call has been made. Suppress the default
305 values for $authenticated_id and $authenticated_sender (but permit -oMai
306 and -oMas) when testing with -bh.
308 PH/24 Re-jigged the order of the tests in the default configuration so that the
309 tests for valid domains and recipients precede the DNS black list and CSA
310 tests, on the grounds that those ones are more expensive.
312 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
313 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
314 This bug exists in every version of Exim that I still have, right back to
317 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
318 However, an attempt to turn on case-sensitivity in a regex key by
319 including (?-i) didn't work because the subject string was already
320 lowercased, and the effects were non-intuitive. It turns out that a
321 one-line patch can be used to allow (?-i) to work as expected.
327 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
328 other effects) broke the use of negated acl sub-conditions.
330 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
333 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
334 "Deny" causes Exim to reject the incoming connection with a 554 error.
335 Unfortunately, if there is a major crisis, such as a disk failure,
336 tcp-wrappers gives "deny", whereas what one would like would be some
337 kind of temporary error. A kludge has been added to help with this.
338 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
339 554 error is used if errno is still zero or contains ENOENT (which occurs
340 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
343 PH/03 Add -lutil to the default FreeBSD LIBS setting.
345 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
346 errors. Otherwise a message that provokes a temporary error (when other
347 messages do not) can cause a whole host to time out.
349 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
350 addresses were routed directly to files or pipes from a redirect router.
351 File deliveries just didn't batch; pipe deliveries might have suffered
354 PH/06 A failure to get a lock for a hints database would erroneously always say
355 "Failed to get write lock", even when it was really a read lock.
357 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
358 of 0600. This has been changed to use the value of the lockfile_mode
359 option (which defaults to 0600).
361 PH/08 Applied small patch from the Sieve maintainer.
363 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
364 folder from quota calculations, a direct delivery into this folder messed
365 up the contents of the maildirsize file. This was because the regex was
366 used only to exclude .Trash (or whatever) when the size of the mailbox
367 was calculated. There was no check that a delivery was happening into an
368 excluded directory. This bug has been fixed by ignoring all quota
369 processing for deliveries into excluded directories.
371 PH/10 Added the maildirfolder_create_regex option to appendfile.
377 PH/01 The code for finding all the local interface addresses on a FreeBSD
378 system running IPv6 was broken. This may well have applied to all BSD
379 systems, as well as to others that have similar system calls. The broken
380 code found IPv4 interfaces correctly, but gave incorrect values for the
381 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
382 that it would not match correctly against @[] and not recognize the IPv6
385 PH/02 The ipliteral router was not recognizing addresses of the form user@
386 [ipv6:....] because it didn't know about the "ipv6:" prefix.
388 PH/03 Added disable_ipv6.
390 PH/04 Changed $reply_address to use the raw form of the headers instead of the
391 decoded form, because it is most often used to construct To: headers
392 lines in autoreplies, and the decoded form may well be syntactically
393 invalid. However, $reply_address has leading white space removed, and all
394 newlines turned into spaces so that the autoreply transport does not
397 PH/05 If group was specified without a user on a router, and no group or user
398 was specified on a transport, the group from the router was ignored.
400 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
401 for visible compile-time settings that can be used to change these
402 numbers, for those that want even more. Backwards compatibility with old
403 spool files has been maintained. However, going back to a previous Exim
404 release will lost any variables that are in spool files.
406 PH/07 Two small changes when running in the test harness: increase delay when
407 passing a TCP/IP connection to a new process, in case the original
408 process has to generate a bounce, and remove special handling of
409 127.0.0.2 (sic), which is no longer necessary.
411 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
412 be the same on different OS.
414 PH/09 Moved a debug statement in filter processing to avoid a race problem when
417 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
418 whether --show-vars was specified or not
420 JJ/02 exipick: Added support for new ACL variable spool format introduced
423 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
424 syntactically invalid From: or Reply-to: line, and a filter used this to
425 generate an autoreply, and therefore failed to obtain an address for the
426 autoreply, Exim could try to deliver to a non-existent relative file
427 name, causing unrelated and misleading errors. What now happens is that
428 it logs this as a hard delivery error, but does not attempt to create a
431 PH/11 The exinext utility has a -C option for testing purposes, but although
432 the given file was scanned by exinext itself; it wasn't being passed on
433 when Exim was called.
435 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
436 an end-of-file indication when reading a command response.
438 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
439 compiled. In many other places in Exim, IPv6 addresses are always
440 recognized, so I have changed this. It also means that IPv4 domain
441 literals of the form [IPV4:n.n.n.n] are now always recognized.
443 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
444 used if the router is not running as root, for example, when verifying at
445 ACL time, or when using -bh. The debugging output from this situation was
446 non-existent - all you got was a failure to exec. I have made two
449 (a) Failures to set uid/gid, the current directory, or a process leader
450 in a subprocess such as that created by queryprogram now generate
451 suitable debugging ouput when -d is set.
453 (b) The queryprogram router detects when it is not running as root,
454 outputs suitable debugging information if -d is set, and then runs
455 the subprocess without attempting to change uid/gid.
457 PH/15 Minor change to Makefile for building test_host (undocumented testing
460 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
461 additional section of a DNS packet that returns MX or SRV records.
462 Instead, it always explicitly searches for A/AAAA records. This avoids
463 major problems that occur when a DNS server includes only records of one
464 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
465 fixed another bug: if SRV records were looked up and the corresponding
466 address records were *not* found in the additional section, the port
467 values from the SRV records were lost.
469 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
470 using the correct key (the original address) when searching the retry
471 rules in order to find which one to use for generating the retry hint.
473 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
474 adding the default one. Similarly, if it contains a Reply-To: header, the
475 errors_reply_to option, if set, is not used.
477 PH/19 When calculating a retry time, Exim used to measure the "time since
478 failure" by looking at the "first failed" field in the retry record. Now
479 it does not use this if it is later than than the arrival time of the
480 message. Instead it uses the arrival time. This makes for better
481 behaviour in cases where some deliveries succeed, thus re-setting the
482 "first failed" field. An example is a quota failure for a huge message
483 when small messages continue to be delivered. Without this change, the
484 "time since failure" will always be short, possible causing more frequent
485 delivery attempts for the huge message than are intended.
486 [Note: This change was subsequently modified - see PH/04 for 4.62.]
488 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
489 $1, $2, $3) because the numerical variables can be reset during some
490 expansion items (e.g. "match"), thereby losing the authentication data.
492 PH/21 Make -bV show the size of off_t variables so that the test suite can
493 decide whether to run tests for quotas > 2G.
495 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
496 mailbox_size, and mailbox_filecount in the appendfile transport. If a
497 filecount value is greater than 2G or if a quota value is greater than 2G
498 on a system where the size of off_t is not greater than 4, a panic error
501 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
502 never match. The debug and -bh output now contains an explicit error
503 message indicating a malformed IPv4 address or mask.
505 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
506 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
509 PH/25 Do not write to syslog when running in the test harness. The only
510 occasion when this arises is a failure to open the main or panic logs
511 (for which there is an explicit test).
513 PH/26 Added the /no_tell option to "control=freeze".
515 PH/27 If a host name lookup failed very early in a connection, for example, if
516 the IP address matched host_lookup and the reverse lookup yielded a name
517 that did not have a forward lookup, an error message of the form "no IP
518 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
519 could be logged. Now it outputs the IP address instead of "NULL".
521 PH/28 An enabling patch from MH: add new function child_open_exim2() which
522 allows the sender and the authenticated sender to be set when
523 submitting a message from within Exim. Since child_open_exim() is
524 documented for local_scan(), the new function should be too.
526 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
527 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
528 results in an empty string is now treated as unset.
530 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
532 PH/31 Added sender_verify_fail logging option.
534 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
535 needed by RFC 822 but not by RFC 2822 was commented out. I have now
536 tidied the source and removed it altogether.
538 PH/33 When a queue run was abandoned because the load average was too high, a
539 log line was always written; now it is written only if the queue_run log
540 selector is set. In addition, the log line for abandonment now contains
541 information about the queue run such as the pid. This is always present
542 in "start" and "stop" lines but was omitted from the "abandon" line.
544 PH/34 Omit spaces between a header name and the colon in the error message that
545 is given when verify = headers_syntax fails (if there are lots of them,
546 the message gets confusing).
548 PH/35 Change the default for dns_check_names_pattern to allow slashes within
549 names, as there are now some PTR records that contain slashes. This check
550 is only to protect against broken name servers that fall over on strange
551 characters, so the fact that it applies to all lookups doesn't matter.
553 PH/36 Now that the new test suite is complete, we can remove some of the
554 special code in Exim that was needed for the old test suite. For example,
555 sorting DNS records because real resolvers return them in an arbitrary
556 order. The new test suite's fake resolver always returns records in the
559 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
560 bounces) except when queue_only is set, to avoid logging races between
561 the different processes.
563 PH/38 Panic-die if .include specifies a non-absolute path.
565 PH/39 A tweak to the "H" retry rule from its user.
567 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
568 a label. They prevented compilation on older perls.
570 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
571 a warning to be raised on newish perls.
573 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
574 on queue. Changes to match documented behaviour of showing count of
575 messages matching specified criteria.
577 PH/40 Changed the default ident timeout from 30s to 5s.
579 PH/41 Added support for the use of login_cap features, on those BSD systems
580 that have them, for controlling the resources used by pipe deliveries.
582 PH/42 The content-scanning code uses fopen() to create files in which to put
583 message data. Previously it was not paying any attention to the mode of
584 the files. Exim runs with umask(0) because the rest of the code creates
585 files with open(), and sets the required mode explicitly. Thus, these
586 files were ending up world-writeable. This was not a big issue, because,
587 being within the spool directory, they were not world-accessible. I have
588 created a function called modefopen, which takes an additional mode
589 argument. It sets umask(777), creates the file, chmods it to the required
590 mode, then resets the umask. All the relevant calls to fopen() in the
591 content scanning code have been changed to use this function.
593 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
594 to 24 hours. This avoids potential overflow problems when processing G
595 and H retry rules. I suspect nobody ever tinkers with this value.
597 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
599 PH/45 When the plaintext authenticator is running as a client, the server's
600 challenges are checked to ensure they are valid base64 strings. By
601 default, the authentication attempt is cancelled if an invalid string is
602 received. Setting client_ignore_invalid_base64 true ignores these errors.
603 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
604 they are received. Thus, the responses can be made to depend on the
605 challenges. If an invalid string is ignored, an empty string is placed in
608 PH/46 Messages that are created by the autoreply transport now contains a
609 References: header, in accordance with RFCs 2822 and 3834.
611 PH/47 Added authenticated_sender_force to the smtp transport.
613 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
615 PH/49 Installed latest patch from the Sieve maintainer.
617 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
618 was also set, the appendfile transport was unnecessarily scanning a
619 directory of message files (e.g. for maildir delivery) to find the count
620 of files (along with the size), even though it did not need this
621 information. It now does the scan only if it needs to find either the
622 size of the count of files.
624 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
626 PH/52 Two bugs concerned with error handling when the smtp transport is
629 (i) Exim was not creating retry information for temporary errors given
630 for individual recipients after the DATA command when the smtp transport
631 was used in LMTP mode. This meant that they could be retried too
632 frequently, and not timed out correctly.
634 (ii) Exim was setting the flag that allows error details to be returned
635 for LMTP errors on RCPT commands, but not for LMTP errors for individual
636 recipients that were returned after the DATA command.
638 PH/53 This is related to PH/52, but is more general: for any failing address,
639 when detailed error information was permitted to be returned to the
640 sender, but the error was temporary, then after the final timeout, only
641 "retry timeout exceeded" was returned. Now it returns the full error as
642 well as "retry timeout exceeded".
644 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
645 do this, and (what is worse) MTAs that accept it.
647 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
648 will now be deprecated.
650 PH/56 New os.c-cygwin from the Cygwin maintainer.
652 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
653 formats (previously only available in exim formats via -bpr, -bpru,
654 and -bpra. Now also available in native and exiqgrep formats)
656 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
657 with very large, slow to parse queues
659 JJ/08 exipick: added ! as generic prefix to negate any criteria format
661 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
663 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
664 responses to authentication challenges, though it was showing the
665 challenges; (ii) I've removed the CR characters from the debug output for
668 PH/58 Allow for the insertion of a newline as well as a space when a string
669 is turned into more than one encoded-word during RFC 2047 encoding. The
670 Sieve code now uses this.
672 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
673 data_4xx, lost_connection, tls_required.
675 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
676 message was being sent as an SMTP response.
678 PH/61 Add -l and -k options to exicyclog.
680 PH/62 When verifying, if an address was redirected to one new address, so that
681 verification continued, and the new address failed or deferred after
682 having set something in $address_data, the value of $address_data was not
683 passed back to the ACL. This was different to the case when no
684 redirection occurred. The value is now passed back in both cases.
686 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
687 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
688 use login_cap.h, so on its own it isn't the distinguishing feature. The
689 new name refers directly to the setclassresources() function.
691 PH/65 Added configuration files for NetBSD3.
693 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
695 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
696 is preferred over IPv4.
698 PH/68 The bounce_return_message and bounce_return_body options were not being
699 honoured for bounces generated during the reception of non-SMTP messages.
700 In particular, this applied to messages rejected by the ACL. This bug has
701 been fixed. However, if bounce_return_message is true and bounce_return_
702 body is false, the headers that are returned for a non-SMTP message
703 include only those that have been read before the error was detected.
704 (In the case of an ACL rejection, they have all been read.)
706 PH/69 The HTML version of the specification is now built in a directory called
707 spec_html instead of spec.html, because the latter looks like a path with
708 a MIME-type, and this confuses some software.
710 PH/70 Catch two compiler warnings in sieve.c.
712 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
713 function verify_get_ident() calls ip_connect() to connect a socket, but
714 if the "connect()" function timed out, ip_connect() used to close the
715 socket. However, verify_get_ident() also closes the socket later, and in
716 between Exim writes to the log, which may get opened at this point. When
717 the socket was closed in ip_connect(), the log could get the same file
718 descriptor number as the socket. This naturally causes chaos. The fix is
719 not to close the socket in ip_connect(); the socket should be closed by
720 the function that creates it. There was only one place in the code where
721 this was missing, in the iplookup router, which I don't think anybody now
722 uses, but I've fixed it anyway.
724 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
725 well as to direct DNS lookups. Otherwise the handling of names in host
726 lists is inconsistent and therefore confusing.
732 PH/01 Two changes to the default runtime configuration:
734 (1) Move the checks for relay_from_hosts and authenticated clients from
735 after to before the (commented out) DNS black list checks.
737 (2) Add control=submission to the relay_from_hosts and authenticated
738 clients checks, on the grounds that messages accepted by these
739 statements are most likely to be submissions.
741 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
743 (1) Generate an error if the third argument for the ${prvs expansion is
746 (2) Treat a missing third argument of ${prvscheck as if it were an empty
749 (3) Reset the variables that are obtained from the first argument of
750 ${prvscheck and used in the second argument before leaving the code,
751 because their memory is reclaimed, so using them afterwards may do
754 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
755 one (it's much easier than Tom thought :-).
757 (5) Because of (4), we can now allow for the use of $prvscheck_result
758 inside the third argument.
760 PH/03 For some reason, the default setting of PATH when running a command from
761 a pipe transport was just "/usr/bin". I have changed it to
764 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
765 anything to be listed in the output from -bV.
767 PH/05 When a filter generated an autoreply, the entire To: header line was
768 quoted in the delivery log line, like this:
770 => >A.N.Other <ano@some.domain> <original@ddress> ...
772 This has been changed so that it extracts the operative address. There
773 may be more than one such address. If so, they are comma-separated, like
776 => >ano@some.domain,ona@other.domain <original@ddress> ...
778 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
779 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
780 not being looked up in the rDNS to get a host name, Exim was showing the
781 IP address twice in Received: lines, even though the IP addresses were
782 identical. For example:
784 Received: from [1.2.3.4] (helo=[1.2.3.4])
786 However, if the real host name was known, it was omitting the HELO data
787 if it matched the actual IP address. This has been tidied up so that it
788 doesn't show the same IP address twice.
790 PH/07 When both +timestamp and +memory debugging was on, the value given by
791 $tod_xxx expansions could be wrong, because the tod_stamp() function was
792 called by the debug printing, thereby overwriting the timestamp buffer.
793 Debugging no longer uses the tod_stamp() function when +timestamp is set.
795 PH/08 When the original message was included in an autoreply transport, it
796 always said "this is a copy of the message, including all the headers",
797 even if body_only or headers_only was set. It now gives an appropriate
800 PH/09 Applied a patch from the Sieve maintainer which:
802 o fixes some comments
803 o adds the (disabled) notify extension core
804 o adds some debug output for the result of if/elsif tests
805 o points to the current vacation draft in the documentation
806 and documents the missing references header update
810 o fixes a bug in processing the envelope test (when testing
811 multiple envelope elements, the last element determinted the
814 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
815 Electronic Mail") by including:
817 Auto-submitted: auto-generated
819 in the messages that it generates (bounce messages and others, such as
820 warnings). In the case of bounce messages for non-SMTP mesages, there was
821 also a typo: it was using "Auto_submitted" (underscore instead of
822 hyphen). Since every message generated by Exim is necessarily in response
823 to another message, thes have all been changed to:
825 Auto-Submitted: auto-replied
827 in accordance with these statements in the RFC:
829 The auto-replied keyword:
831 - SHOULD be used on messages sent in direct response to another
832 message by an automatic process,
834 - MUST NOT be used on manually-generated messages,
836 - MAY be used on Delivery Status Notifications (DSNs) and Message
837 Disposition Notifications (MDNs),
839 - MUST NOT be used on messages generated by automatic or periodic
840 processes, except for messages which are automatic responses to
843 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
844 to the default Received: header definition.
846 PH/12 Added log selector acl_warn_skipped (default on).
848 PH/13 After a successful wildlsearch lookup, discard the values of numeric
849 variables because (a) they are in the wrong storage pool and (b) even if
850 they were copied, it wouldn't work properly because of the caching.
852 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
853 checking when decoding. Apparently there are clients that generate
854 overlong encoded strings. Why am I not surprised?
856 PH/15 If the first argument of "${if match_address" was not empty, but did not
857 contain an "@" character, Exim crashed. Now it writes a panic log message
858 and treats the condition as false.
860 PH/16 In autoreply, treat an empty string for "once" the same as unset.
862 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
863 extension "envelope-auth". The code is finished and in agreement with
864 other implementations, but there is no documentation so far and in fact,
865 nobody wrote the draft yet. This extension is currently #undef'ed, thus
866 not changing the active code.
868 Print executed "if" and "elsif" statements when debugging is used. This
869 helps a great deal to understand what a filter does.
871 Document more things not specified clearly in RFC3028. I had all this
872 sorted out, when out of a sudden new issues came to my mind. Oops."
874 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
877 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
878 canonical form (as documented). However, after a host name lookup from
879 the IP address, check_host() was doing a simple string comparison with
880 addresses acquired from the DNS when checking that the found name did
881 have the original IP as one of its addresses. Since any found IPv6
882 addresses are likely to be in abbreviated form, the comparison could
883 fail. Luckily, there already exists a function for doing the comparison
884 by converting both addresses to binary, so now that is used instead of
887 PH/20 There was another similar case to PH/19, when a complete host name was
888 given in a host list; looking up its IP address could give an abbreviated
889 form, whereas the current host's name might or might not be abbreviated.
890 The same fix has been applied.
896 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
897 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
900 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
901 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
903 PH/03 Typo: missing ".o" in src/pcre/Makefile.
905 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
906 header line, restrict the check to what is listed in RFCs 2369 and 2929.
907 Also, for "Auto-Submitted", treat anything other than "no" as
908 non-personal, in accordance with RFC 3834. (Previously it treated
909 anything starting "auto-" as non-personal.)
911 TF/01 The control=submission/name=... option had a problem with syntax
912 errors if the name included a slash character. The /name= option
913 now slurps the rest of the string, so it can include any characters
914 but it must come last in the list of options (after /sender_retain
917 PH/05 Some modifications to the interface to the fake nameserver for the new
925 TK/01 Added the "success_on_redirect" address verification option. See
926 NewStuff for rationale and an example.
928 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
930 PH/02 Patch to exigrep to allow it to work on syslog lines.
932 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
933 fread() to skip over the body file's header line, because in Cygwin the
934 header line is locked and is inaccessible.
936 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
937 co-exist for some time) to make it clear that it is the Exim ID that is
938 referenced, not the Message-ID: header line.
940 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
941 string_format() function, because snprintf() does not exist on all
944 PH/06 The use of forbid_filter_existstest now also locks out the use of the
945 ${stat: expansion item.
947 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
948 protocol synchronization error", to keep the pedants happy.
950 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
951 well as for IRIX systems, when gcc is being used. See the host.c source
954 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
956 PH/10 Named domain lists were not working if used in a queue_smtp_domains
959 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
960 transport and to the smtp transport in LMTP mode.
962 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
964 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
965 run a filter in a subprocess. This could lead to confusion in subsequent
966 lookups in the parent process. There should also be a search_tidyup() at
967 the end of the subprocess.
969 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
970 only if the host matched helo_try_verify_hosts, which caused the
971 verification to occur when the EHLO/HELO command was issued. The ACL just
972 tested the remembered result. Now, if a previous verification attempt has
973 not happened, "verify = helo" does it there and then.
975 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
977 TK/03 Fix log output including CR from clamd.
979 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
980 exist provoked a memory error which could cause a segfault.
982 PH/15 Installed PCRE 6.2
984 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
986 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
987 of the problem. Specifically, suggested +O2 rather than +O1 for the
990 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
992 PH/20 If a delivery was routed to a non-standard port by means of an SRV
993 record, the port was not correctly logged when the outgoing_port log
994 selector was set (it logged the transort's default port).
996 PH/21 Added support for host-specific ports to manualroute, queryprogram,
997 fallback_hosts, and "hosts" in the smtp transport.
999 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1000 host errors such as "Connection refused".
1002 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1003 authentication with radiusclient 0.4.9:
1005 - Error returned from rc_read_config was caught wrongly
1006 - Username/password not passed on to radius server due to wrong length.
1008 The presumption is that some radiusclient API changes for 4.51/PH/17
1009 were not taken care of correctly. The code is still untested by me (my
1010 Linux distribution still has 0.3.2 of radiusclient), but it was
1011 contributed by a Radius user.
1013 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1014 expanding the "port" option of the smtp transport.
1016 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1017 while reading a MIME header. Thanks to Tom Hughes for a patch.
1019 PH/24 Include config.h inside local_scan.h so that configuration settings are
1022 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1023 that in an ACL for RCPT (for example), you can examine exactly what was
1026 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1027 commands, but it was not correctly comparing the address with the actual
1028 client host address. Thus, it would show the EHLO address in Received:
1029 header lines when this was not necessary.
1031 PH/27 Added the % operator to ${eval:}.
1033 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1034 it should be ignoring failures (because with -C, for example, it has lost
1035 privilege). It wasn't ignoring creation failures other than "already
1038 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1041 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1042 because an input line was too long, either on its own, or by virtue of
1043 too many continuations, the temporary file was not being removed, and the
1044 return code was incorrect.
1046 PH/31 Missing "BOOL" in function definition in filtertest.c.
1048 PH/32 Applied Sieve patches from the maintainer.
1050 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1052 PH/33 Added "verify = not_blind".
1054 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1055 Local/Makefile (with some defaults set). These are used in built scripts
1056 such as exicyclog, but they have never been used in the exim_install
1057 script (though there are many overriding facilities there). I have
1058 arranged that the exim_install script now takes note of these two
1061 PH/35 Installed configuration files for Dragonfly.
1063 PH/36 When a locally submitted message by a trusted user did not contain a
1064 From: header, and the sender address was obtained from -f or from an SMTP
1065 MAIL command, and the trusted user did not use -F to supply a sender
1066 name, $originator_name was incorrectly used when constructing a From:
1067 header. Furthermore, $originator_name was used for submission mode
1068 messages from external hosts without From: headers in a similar way,
1069 which is clearly wrong.
1071 PH/37 Added control=suppress_local_fixups.
1073 PH/38 When log_selector = +received_sender was set, and the addition of the
1074 sender made the log line's construction buffer exactly full, or one byte
1075 less than full, an overflow happened when the terminating "\n" was
1078 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1079 when the result of a list match is failure because a DNS lookup failed.
1081 PH/40 RM_COMMAND is now used in the building process.
1083 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1084 the "build-* directories that it finds.
1086 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1087 address in a domain literal was a prefix of an interface address.
1089 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1090 when verifying a sender address, unless rewrite_headers is false.
1092 PH/44 Wrote a long comment about why errors_to addresses are verified as
1093 recipients, not senders.
1095 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1096 the ratelimit ACL was added.
1098 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1100 PH/46 Added extra information about PostgreSQL errors to the error string.
1102 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1103 avoiding the need to install special zones in a real server. This is
1104 backwards compatible; if it can't find the fake resolver, it drops back.
1105 Thus, both old and new test suites can be run.
1107 TF/02 Added util/ratelimit.pl
1109 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1110 clock is set back in time.
1112 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1113 Candler <B.Candler@pobox.com>.
1115 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1116 OK for addresses that are the result of redirections.
1118 PH/48 A number of further additions for the benefit of the new test suite,
1119 including a fake gethostbyname() that interfaces to the fake DNS resolver
1122 TF/06 The fix for widen_domains has also been applied to qualify_single and
1123 search_parents which are the other dnslookup options that can cause
1126 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1129 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1131 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1132 service" if it encountered an MX record with an empty target hostname.
1133 The message is now "an MX or SRV record indicated no SMTP service".
1135 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1136 if the DNS of the sending site is misconfigured. This is quite a
1137 common situation. This change restores the behaviour of treating a
1138 helo verification defer as a failure.
1140 PH/51 If self=fail was set on a router, the bounce message did not include the
1141 actual error message.
1147 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1149 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1150 command itself ended in error, the underlying message about the transport
1151 filter timeout was being overwritten with the pipe command error. Now the
1152 underlying error message should be appended to the second error message.
1154 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1156 PH/02 Reduce the amount of output that "make" produces by default. Full output
1157 can still be requested.
1159 PH/03 The warning log line about a condition test deferring for a "warn" verb
1160 was being output only once per connection, rather than after each
1161 occurrence (because it was using the same function as for successful
1162 "warn" verbs). This seems wrong, so I have changed it.
1164 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1165 it should not have, which might have caused a crash in the right
1166 circumstances, but probably never did.
1168 PH/04 Installed a modified version of Tony Finch's patch to make submission
1169 mode fix the return path as well as the Sender: header line, and to
1170 add a /name= option so that you can make the user's friendly name appear
1173 TF/03 Added the control = fakedefer ACL modifier.
1175 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1176 Mark Lowes for thorough testing.
1178 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1180 TK/03 Merged latest SRS patch from Miles Wilton.
1182 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1183 with the definition in sysexits.h (which is #included earlier).
1184 Fortunately, Exim does not actually use EX_OK. The code used to try to
1185 preserve the sysexits.h value, by assumimg that macro definitions were
1186 scanned for macro replacements. I have been disabused of this notion,
1187 so now the code just undefines EX_OK before #including unistd.h.
1189 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1190 in the smtp transport. When a block could not be written in a single
1191 write() function, the timeout was being re-applied to each part-write.
1192 This seems wrong - if the receiver was accepting one byte at a time it
1193 would take for ever. The timeout is now adjusted when this happens. It
1194 doesn't have to be particularly precise.
1196 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1197 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1199 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1200 without a domain if the check to <postmaster@domain> fails.
1202 SC/01 Eximstats: added -xls and the ability to specify output files
1203 (patch written by Frank Heydlauf).
1205 SC/02 Eximstats: use FileHandles for outputing results.
1207 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1209 SC/04 Eximstats: fixed display of large numbers with -nvr option
1211 SC/05 Eximstats: fixed merging of reports with empty tables.
1213 SC/06 Eximstats: added the -include_original_destination flag
1215 SC/07 Eximstats: removed tabs and trailing whitespace.
1217 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1219 TK/06 MBOX spool code: Add real "From " MBOX separator line
1220 so the .eml file is really in mbox format (even though
1221 most programs do not really care). Patch from Alex Miller.
1223 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1224 The latter is generated from $received_to and is only set if the
1225 message has one envelope recipient. SA can use these headers,
1226 obviously out-of-the-box. Patch from Alex Miller.
1228 PH/08 The ${def test on a variable was returning false if the variable's
1229 value was "0", contrary to what the specification has always said!
1230 The result should be true unless the variable is empty.
1232 PH/09 The syntax error of a character other than { following "${if
1233 def:variable_name" (after optional whitespace) was not being diagnosed.
1234 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1235 accidental colon was present, for example, could give incorrect results.
1237 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1238 result is used (not including appendfile, where other changes are about
1241 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1242 This involved changing a lot of size variables from int to off_t. It
1243 should work with maildirs and everything.
1245 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1246 spamd dying while we are connected to it.
1248 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1251 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1252 function for BDB 4.3.
1254 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1256 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1257 the sources for PCRE have been re-arranged and more files are now
1260 PH/15 The code I had for printing potentially long long variables in PH/11
1261 above was not the best (it lost precision). The length of off_t variables
1262 is now inspected at build time, and an appropriate printing format (%ld
1263 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1264 to be "long long int" or "long int". This is needed for the internal
1265 formatting function string_vformat().
1267 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1268 the configuration file to be ":syslog", then the script "guesses" where
1269 the logs files are, rather than using the compiled in default. In our
1270 case the guess is not the same as the compiled default, so the script
1271 suddenly stopped working when I started to use syslog. The patch checks
1272 to see if log_file_path is "". If so, it attempts to read it from exim
1273 with no configuration file to get the compiled in version, before it
1274 falls back to the previous guessing code."
1276 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1277 implementing BATV in an Exim configuration. See NewStuff for the gory
1280 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1281 Makefile that are specific to HP-UX.
1283 PH/18 If the "use_postmaster" option was set for a recipient callout together
1284 with the "random" option, the postmaster address was used as the MAIL
1285 FROM address for the random test, but not for the subsequent recipient
1286 test. It is now used for both.
1288 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1289 patch removes a few documentation additions to RFC 3028, because the
1290 latest draft now contains them. It adds the new en;ascii-case comparator
1291 and a new error check for 8bit text in MIME parts. Comparator and
1292 require names are now matched exactly. I enabled the subaddress
1293 extension, but it is not well tested yet (read: it works for me)."
1295 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1296 rework some of the code of TK/09 above to avoid the hardwired use of
1297 "%lld" and "long long". Replaced the call to snprintf() with a call to
1300 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1301 records point to non-existent hosts", "retry timeout exceeded", and
1302 "retry time not reached for any host after a long failure period".
1304 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1305 experimental DomainKeys support:
1307 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1308 (2) On an error such as an illegally used "control", the wrong name for
1309 the control was given.
1311 These problems did NOT occur unless DomainKeys support was compiled.
1313 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1315 PH/24 Added ${if match_ip condition.
1317 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1318 databases so that it will be absolutely obvious if a crash occurs in the
1319 DB library. This is a regular occurrence (often caused by mis-matched
1322 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1323 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1324 picked up on a user's system that detects such things. There doesn't seem
1325 to be a gcc warning option for this - only an attribute that has to be
1326 put on the function's prototype. It seems that in Fedora Core 4 they have
1327 set this on a number of new functions. No doubt there will be more in due
1330 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1331 specify a transport. However, if an address that was verified by such a
1332 router was the subject of a callout, Exim crashed because it tried to
1333 read the rcpt_include_affixes from the non-existent transport. Now it
1334 just assumes that the setting of that option is false. This bug was
1335 introduced by 4.51/PH/31.
1337 PH/28 Changed -d+all to exclude +memory, because that information is very
1338 rarely of interest, but it makes the output a lot bigger. People tend to
1339 do -d+all out of habit.
1341 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1342 code in os-type was giving problems when libc.so lives in lib64, like on
1345 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1346 aren't the modern standard, and it seems that some systems' include files
1347 don't always have them. Exim was already checking for some of the newer
1348 ones like T_AAAA, and defining it itself. I've added checks for all the
1349 record types that Exim uses.
1351 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1352 not automatically generating a new one, as it is supposed to. This
1353 prevented TLS from working. If the file did exist, but contained invalid
1354 data, a new version was generated, as expected. It was only the case of a
1355 non-existent file that was broken.
1357 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1358 with a change in libdomainkeys > 0.64.
1360 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1361 from DNS. If the selector record carries the flag, it now has
1362 precedence over the domain-wide flag.
1364 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1366 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1367 the use of a port name that isn't defined in /etc/services) occurred, the
1368 message was deferred as in a normal delivery, and thus remained on the
1369 spool, instead of being failed because of the mua_wrapper setting. This
1370 is now fixed, and I tidied up some of the mua_wrapper messages at the
1373 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1374 the messages in a hash of arrays rather than using individual hashes.
1375 This is a bit cleaner and results in dramatic memory savings, albeit
1376 at a slight CPU cost.
1378 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1379 as requested by Marc Sherman.
1381 SC/10 Eximstats: added histograms for user specified patterns as requested
1384 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1386 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1387 fopen() in the content-scanning modules that did not already have it.
1393 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1394 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1396 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1398 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1400 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1401 to test Sieve filters that use "vacation".
1403 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1404 that changes the way the GnuTLS parameters are stored in the cache file.
1405 The new format can be generated externally. For backward compatibility,
1406 if the data in the cache doesn't make sense, Exim assumes it has read an
1407 old-format file, and it generates new data and writes a new file. This
1408 means that you can't go back to an older release without removing the
1411 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1412 work if there are any delivery delays because "one_time" forces the
1413 parent to be marked "delivered", so its unseen clone is never tried
1414 again. For this reason, Exim now forbids the simultaneous setting of
1417 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1418 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1419 entry if you want to know the details. The fix, however, neglected to
1420 consider the case when local delivery batching is involved. The test for
1421 "previously delivered" was not happening when checking to see if an
1422 address could be batched with a previous (undelivered) one; under
1423 certain circumstances this could lead to multiple deliveries to the same
1426 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1427 in its include files, and this causes problems building Exim.
1429 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1430 header_syntax) but Exim was just ignoring anything given after a slash.
1431 In particular, this caused confusion with an attempt to use "verify =
1432 reverse_host_lookup/defer_ok". An error is now given when options are
1433 supplied for verify items that do not have them. (Maybe reverse_host_
1434 lookup should have a defer_ok option, but that's a different point.)
1436 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1437 defined by RFC 821) to 2048, because there were problems with some AUTH
1438 commands, and RFC 1869 says the size should be increased for extended
1439 SMTP commands that take arguments.
1441 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1444 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1445 "unknown" error; now it says that the functionality isn't in the binary.
1447 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1448 an address' error message when a string expansion fails (syntax or
1449 whatever). Otherwise the password may appear in the log. Following change
1450 PH/42 below, there is no longer a chance of it appearing in a bounce
1453 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1455 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1456 of its addresses was ever tried. (Bugzilla bug #2.)
1458 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1459 the result incorrectly in the debug output. (It correctly added a newline
1460 to what was transported.)
1462 TF/01 Added $received_time.
1464 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1465 commented out examples of how to interface to a virus scanner and to
1466 SpamAssassin. Also added commented examples of av_scanner and
1467 spamd_address settings.
1469 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1470 and controls are allowed in which ACLs. There were a couple of minor
1471 errors. Some of the entries in the conditions table (which is a table of
1472 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1473 negation of where the condition IS allowed.
1475 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1477 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1478 header file does not have a version number, so I've had to invent a new
1479 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1480 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1481 radiusclient), but it was contributed by a Radius user.
1483 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1484 files or directories, for OpenSSL.
1486 PH/19 When an Exim process that is running as root has to create an Exim log
1487 file, it does so in a subprocess that runs as exim:exim so as to get the
1488 ownership right at creation (otherwise, other Exim processes might see
1489 the file with the wrong ownership). There was no test for failure of this
1490 fork() call, which would lead to the process getting stuck as it waited
1491 for a non-existent subprocess. Forks do occasionally fail when resources
1492 run out. I reviewed all the other calls to fork(); they all seem to check
1495 PH/20 When checking for unexpected SMTP input at connect time (before writing
1496 the banner), Exim was not dealing correctly with a non-positive return
1497 from the read() function. If the client had disconnected by this time,
1498 the result was a log entry for a synchronization error with an empty
1499 string after "input=" when read() returned zero. If read() returned -1
1500 (an event I could not check), uninitialized data bytes were printed.
1501 There were reports of junk text (parts of files, etc) appearing after
1504 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1506 PH/22 Added support for macro redefinition, and (re)definition in between
1507 driver and ACL definitions.
1509 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1510 forgetting to use the resulting value; it was using the unexpanded value.
1512 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1513 hadn't been configured. The fix is from Juergen Kreileder, who
1514 understands it better than I do:
1516 "Here's what I see happening with three configured cyrus_sasl
1517 authenticators configured (plain, login, cram-md5):
1519 On startup auth_cyrus_sasl_init() gets called for each of these.
1520 This means three calls to sasl_listmech() without a specified mech_list.
1521 => SASL tests which mechs of all available mechs actually work
1522 => three warnings about OTP not working
1523 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1525 With the patch, sasl_listmech() also gets called three times. But now
1526 SASL's mech_list option is set to the server_mech specified in the the
1527 authenticator. Or in other words, the answer from sasl_listmech()
1528 gets limited to just the mech you're testing for (which is different
1530 => the return list contains just 'plain' or 'login', 'cram-md5' or
1531 nothing depending on the value of ob->server_mech.
1533 I've just tested the patch: Authentication still works fine,
1534 unavailable mechs specified in the exim configuration are still
1535 caught, and the auth.log warnings about OTP are gone."
1537 PH/25 When debugging is enabled, the contents of the command line are added
1538 to the debugging output, even when log_selector=+arguments is not
1541 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1542 answer is "GNU", and only if the return is "GNU/something" is the answer
1545 PH/27 $acl_verify_message is now set immediately after the failure of a
1546 verification in an ACL, and so is available in subsequent modifiers. In
1547 particular, the message can be preserved by coding like this:
1549 warn !verify = sender
1550 set acl_m0 = $acl_verify_message
1552 Previously, $acl_verify_message was set only while expanding "message"
1553 and "log_message" when a very denied access.
1555 PH/28 Modified OS/os.c-Linux with
1557 -#ifndef OS_LOAD_AVERAGE
1558 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1560 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1561 nomenclature these days.)
1563 PH/29 Installed patch from the Sieve maintainer that adds the options
1564 sieve_useraddress and sieve_subaddress to the redirect router.
1566 PH/30 In these circumstances:
1567 . Two addresses routed to the same list of hosts;
1568 . First host does not offer TLS;
1569 . First host accepts first address;
1570 . First host gives temporary error to second address;
1571 . Second host offers TLS and a TLS session is established;
1572 . Second host accepts second address.
1573 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1574 and peerdn, if requested) that were in fact used only for the second
1577 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1578 attention to any local part prefix or suffix that was matched by the
1579 router that accepted the address. It now behaves in the same way as it
1580 does for delivery: the affixes are removed from the local part unless
1581 rcpt_include_affixes is set on the transport.
1583 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1584 timeout during the DATA phase of an incoming message.
1586 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1587 applied a patch sanctioned by the Sieve maintainer.
1589 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1590 the uid or gid is negative. A case of a negative gid caused this to be
1591 noticed. The fix allows for either to be negative.
1593 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1594 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1597 PH/36 The change PH/12 above was broken. Fixed it.
1599 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1600 the grounds that routing the same address twice would always produce the
1601 same answer. This might have been true once, but it is certainly no
1602 longer true now. Routing a child address may depend on the previous
1603 routing that produced that child. Some complicated redirection strategies
1604 went wrong when messages had multiple recipients, and made Exim's
1605 behaviour dependent on the order in which the addresses were given.
1607 I have moved the duplicate checking until after the routing is complete.
1608 Exim scans the addresses that are assigned to local and remote
1609 transports, and removes any duplicates. This means that more work will be
1610 done, as duplicates will always all be routed, but duplicates are
1611 presumably rare, so I don't expect this is of any significance.
1613 For deliveries to pipes, files, and autoreplies, the duplicate checking
1614 still happens during the routing process, since they are not going to be
1617 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1618 It corrects a timeout issue with spamd. This is Ian's comment: "The
1619 background is that sometimes spamd either never reads data from a
1620 connection it has accepted, or it never writes response data. The exiscan
1621 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1622 blindly assumes that writes won't block so it may never time out."
1624 PH/39 Allow G after quota size as well as K and M.
1626 PH/40 The value set for $authenticated_id in an authenticator may not contain
1627 binary zeroes or newlines because the value is written to log lines and
1628 to spool files. There was no check on this. Now the value is run through
1629 the string_printing() function so that such characters are converted to
1630 printable escape sequences.
1632 PH/41 $message_linecount is a new variable that contains the total number of
1633 lines in the message. Compare $body_linecount, which is the count for the
1636 PH/42 Exim no longer gives details of delivery errors for specific addresses in
1637 bounce and delay warning messages, except in certain special cases, which
1640 (a) An SMTP error message from a remote host;
1641 (b) A message specified in a :fail: redirection;
1642 (c) A message specified in a "fail" command in a system filter;
1643 (d) A message specified in a FAIL return from the queryprogram router;
1644 (e) A message specified by the cannot_route_message router option.
1646 In these cases only, Exim does include the error details in bounce and
1647 warning messages. There are also a few cases where bland messages such
1648 as "unrouteable address" or "local delivery error" are given.
1650 PH/43 $value is now also set for the "else" part of a ${run expansion.
1652 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1653 being worked on, but at least Exim now implements the latest version to
1656 PH/45 In a pipe transport, although a timeout while waiting for the pipe
1657 process to complete was treated as a delivery failure, a timeout while
1658 writing the message to the pipe was logged, but erroneously treated as a
1659 successful delivery. Such timeouts include transport filter timeouts. For
1660 consistency with the overall process timeout, these timeouts are now
1661 treated as errors, giving rise to delivery failures by default. However,
1662 there is now a new Boolean option for the pipe transport called
1663 timeout_defer, which, if set TRUE, converts the failures into defers for
1664 both kinds of timeout. A transport filter timeout is now identified in
1667 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1668 systems where "make" and "gmake" are different, calling "gmake" at top
1669 level broke things. I've arranged for the value of $(MAKE) to be passed
1670 from the Makefile to this script so that it can call the same version of
1674 A note about Exim versions 4.44 and 4.50
1675 ----------------------------------------
1677 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1678 changes of various kinds. As a consequence, a big documentation update was
1679 needed. This delayed the release for rather longer than seemed good, especially
1680 in the light of a couple of (minor) security issues. Therefore, the changes
1681 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1682 release. So 4.44 and 4.50 are in effect two different branches that both start
1685 I have left the 4.50 change log unchanged; it contains all the changes since
1686 4.43. The change log for 4.44 is below; many of its items are identical to
1687 those for 4.50. This seems to be the most sensible way to preserve the
1688 historical information.
1694 1. Minor wording change to the doc/README.SIEVE file.
1696 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
1697 computation of the current number of files was incorrect.
1699 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1700 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1703 4. Give more explanation in the error message when the command for a transport
1704 filter fails to execute.
1706 5. There are several places where Exim runs a non-Exim command in a
1707 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1708 was being done only for the command run by the queryprogram router. It is
1709 now done for all such subprocesses. The other cases are: ${run, transport
1710 filters, and the commands run by the lmtp and pipe transports.
1712 6. Added CONFIGURE_GROUP build-time option.
1714 7. Some older OS have a limit of 256 on the maximum number of file
1715 descriptors. Exim was using setrlimit() to set 1000 as a large value
1716 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1717 systems. I've change it so that if it can't get 1000, it tries for 256.
1719 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1720 was an oversight, and furthermore, ever since the addition of extra
1721 controls (e.g. 4.43/32), the checks on when to allow different forms of
1722 "control" were broken. There should now be diagnostics for all cases when a
1723 control that does not make sense is encountered.
1725 9. Added the /retain_sender option to "control=submission".
1727 10. $recipients is now available in the predata ACL (oversight).
1729 11. Tidy the search cache before the fork to do a delivery from a message
1730 received from the command line. Otherwise the child will trigger a lookup
1731 failure and thereby defer the delivery if it tries to use (for example) a
1732 cached ldap connection that the parent has called unbind on.
1734 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1735 of $address_data from the recipient verification was clobbered by the
1736 sender verification.
1738 13. The value of address_data from a sender verification is now available in
1739 $sender_address_data in subsequent conditions in the ACL statement.
1741 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1743 15. Added a new option "connect=<time>" to callout options, to set a different
1746 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1747 was its contents. (It was OK if the option was not defined at all.)
1749 17. A "Completed" log line is now written for messages that are removed from
1750 the spool by the -Mrm option.
1752 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1753 information about exactly what failed.
1755 19. Added -dd to debug only the daemon process.
1757 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1758 handles timeouts, both on the server side and network timeouts. Renamed the
1759 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1761 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1762 It is now set to "smtps".
1764 22. $host_address is now set to the target address during the checking of
1765 ignore_target_hosts.
1767 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1768 being passed; this would have caused $sender_host_name to have been used if
1769 matching the list had actually called for a host name (not very likely,
1770 since this list is usually IP addresses). A host name is now passed as
1773 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1774 code that specifies a non-restarting handler (typically sigaction() in
1775 modern systems) in an attempt to fix a rare and obscure crash bug.
1777 25. Narrowed the window for a race in the daemon that could cause it to ignore
1778 SIGCHLD signals. This is not a major problem, because they are used only to
1779 wake it up if nothing else does.
1781 26. A malformed maildirsize file could cause Exim to calculate negative values
1782 for the mailbox size or file count. Odd effects could occur as a result.
1783 The maildirsize information is now recalculated if the size or filecount
1786 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1787 support for a long time. Removed HAVE_SYS_VFS_H.
1789 28. Installed the latest version of exipick from John Jetmore.
1791 29. In an address list, if the pattern was not a regular expression, an empty
1792 subject address (from a bounce message) matched only if the pattern was an
1793 empty string. Non-empty patterns were not even tested. This was the wrong
1794 because it is perfectly reasonable to use an empty address as part of a
1795 database query. An empty address is now tested by patterns that are
1796 lookups. However, all the other forms of pattern expect the subject to
1797 contain a local part and a domain, and therefore, for them, an empty
1798 address still always fails if the pattern is not itself empty.
1800 30. Exim went into a mad DNS loop when attempting to do a callout where the
1801 host was specified on an smtp transport, and looking it up yielded more
1802 than one IP address.
1804 31. Re-factored the code for checking spool and log partition space into a
1805 function that finds that data and another that does the check. The former
1806 is then used to implement four new variables: $spool_space, $log_space,
1807 $spool_inodes, and $log_inodes.
1809 32. The RFC2047 encoding function was originally intended for short strings
1810 such as real names; it was not keeping to the 75-character limit for
1811 encoded words that the RFC imposes. It now respects the limit, and
1812 generates multiple encoded words if necessary. To be on the safe side, I
1813 have increased the buffer size for the ${rfc2047: expansion operator from
1816 33. It is now permitted to omit both strings after an "if" condition; if the
1817 condition is true, the result is "true". As before, when the second string
1818 is omitted, a false condition yields an empty string. This makes it less
1819 cumbersome to write custom ACL and router conditions.
1821 34. Failure to deliver a bounce message always caused it to be frozen, even if
1822 there was an errors_to setting on the router. The errors_to setting is now
1825 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1826 canonical form (fully expanded) before being placed in
1827 $sender_host_address.
1829 36. The table in the code that translates DNS record types into text (T_A to
1830 "A" for instance) was missing entries for NS and CNAME. It is just possible
1831 that this could have caused confusion if both these types were looked up
1832 for the same domain, because the text type is used as part of Exim's
1833 per-process caching. But the chance of anyone hitting this buglet seems
1836 37. The dnsdb lookup has been extended in a number of ways.
1838 (1) There is a new type, "zns", which walks up the domain tree until it
1839 finds some nameserver records. It should be used with care.
1841 (2) There is a new type, "mxh", which is like "mx" except that it returns
1842 just the host names, not the priorities.
1844 (3) It is now possible to give a list of domains (or IP addresses) to be
1845 looked up. The behaviour when one of the lookups defers can be
1846 controlled by a keyword.
1848 (4) It is now possible to specify the separator character for use when
1849 multiple records are returned.
1851 38. The dnslists ACL condition has been extended: it is now possible to supply
1852 a list of IP addresses and/or domains to be looked up in a particular DNS
1855 39. Added log_selector=+queue_time_overall.
1857 40. When running the queue in the test harness, wait just a tad after forking a
1858 delivery process, to get repeatability of debugging output.
1860 41. Include certificate and key file names in error message when GnuTLS fails
1861 to set them up, because the GnuTLS error message doesn't include the name
1862 of the failing file when there is a problem reading it.
1864 42. Allow both -bf and -bF in the same test run.
1866 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1868 44. The "Exiscan patch" is now merged into the mainline Exim source.
1870 45. Sometimes the final signoff response after QUIT could fail to get
1871 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1872 < 0 before doing a fflush(). This bug looks as though it goes back to the
1873 introduction of TLS in release 3.20, but "sometimes" must have been rare
1874 because the tests only now provoked it.
1876 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1877 (this can affect the format of dates).
1879 47. exim_tidydb, when checking for the continued existence of a message for
1880 which it has found a message-specific retry record, was not finding
1881 messages that were in split spool directories. Consequently, it was
1882 deleting retry records that should have stayed in existence.
1884 48. Steve fixed some bugs in eximstats.
1886 49. The SPA authentication driver was not abandoning authentication and moving
1887 on to the next authenticator when an expansion was forced to fail,
1888 contradicting the general specification for all authenticators. Instead it
1889 was generating a temporary error. It now behaves as specified.
1891 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1892 (the order specifies the preference for clients). The order is now AES256,
1893 AES128, 3DES, ARCFOUR128.
1895 51. Small patch to Sieve code - explicitly set From: when generating an
1898 52. Exim crashed if a remote delivery caused a very long error message to be
1899 recorded - for instance if somebody sent an entire SpamAssassin report back
1900 as a large number of 550 error lines. This bug was coincidentally fixed by
1901 increasing the size of one of Exim's internal buffers (big_buffer) that
1902 happened as part of the Exiscan merge. However, to be on the safe side, I
1903 have made the code more robust (and fixed the comments that describe what
1906 53. Now that there can be additional text after "Completed" in log lines (if
1907 the queue_time_overall log selector is set), a one-byte patch to exigrep
1908 was needed to allow it to recognize "Completed" as not the last thing in
1911 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1912 patch that reportedly fixes this has been added. I am not expert enough to
1913 create a test for it. This is what the patch creator wrote:
1915 "I found a little strange behaviour of ldap code when working with
1916 Windows 2003 AD Domain, where users was placed in more than one
1917 Organization Units. When I tried to give exim partial DN, the exit code
1918 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1919 But simultaneously result of request was absolutely normal ldap result,
1920 so I produce this patch..."
1922 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1923 REFERENCE, so I have modified the code to exclude the patch when that macro
1926 55. Some experimental protocols are using DNS PTR records for new purposes. The
1927 keys for these records are domain names, not reversed IP addresses. The
1928 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1929 leaves it alone. Component reversal etc. now happens only for IP addresses.
1932 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1934 57. Double the size of the debug message buffer (to 2048) so that more of very
1935 long debug lines gets shown.
1937 58. The exicyclog utility now does better if the number of log files to keep
1938 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1940 59. Two changes related to the smtp_active_hostname option:
1942 (1) $smtp_active_hostname is now available as a variable.
1943 (2) The default for smtp_banner uses $smtp_active_hostname instead
1944 of $primary_hostname.
1946 60. The host_aton() function is supposed to be passed a string that is known
1947 to be a valid IP address. However, in the case of IPv6 addresses, it was
1948 not checking this. This is a hostage to fortune. Exim now panics and dies
1949 if the condition is not met. A case was found where this could be provoked
1950 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1951 components; fortuitously, this particular loophole had already been fixed
1952 by change 4.50/55 above.
1954 If there are any other similar loopholes, the new check in host_aton()
1955 itself should stop them being exploited. The report I received stated that
1956 data on the command line could provoke the exploit when Exim was running as
1957 exim, but did not say which command line option was involved. All I could
1958 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1959 running as the user.
1962 61. There was a buffer overflow vulnerability in the SPA authentication code
1963 (which came originally from the Samba project). I have added a test to the
1964 spa_base64_to_bits() function which I hope fixes it.
1967 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1968 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1970 63. The daemon start-up calls getloadavg() while still root for those OS that
1971 need the first call to be done as root, but it missed one case: when
1972 deliver_queue_load_max is set with deliver_drop_privilege. This is
1973 necessary for the benefit of the queue runner, because there is no re-exec
1974 when deliver_drop_privilege is set.
1976 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1977 This has been fixed.
1979 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1980 was in use, was not putting the data itself into the right store pool;
1981 consequently, it could be overwritten for a subsequent message in the same
1982 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1985 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1987 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1988 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1989 the function were treating the return as a boolean value, which happened to
1990 work because 0=false and not-0=true, but is not correct code.
1992 68. The host_aton() function was not handling scoped IPv6 addresses (those
1993 with, for example, "%eth0" on the end) correctly.
1995 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1996 negated items (that is, ~something) in unsigned ints. Some compilers
1997 apparently mutter when there is no cast.
1999 70. If an address verification called from an ACL failed, and did not produce a
2000 user-specific message (i.e. there was only a "system" message), nothing was
2001 put in $acl_verify_message. In this situation, it now puts the system
2004 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2005 session; change 4.31/43 added the unwanted input to the log line - except
2006 that it did not do this in the start of session case. It now does.
2008 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2009 This is wrong and can cause the other end to generate a synchronization
2010 error if it is another Exim or anything else that does the synchronization
2011 check. A QUIT command is no longer sent after a timeout.
2013 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2014 during host lookups.
2016 74. The defer_ok option of callout verification was not working if it was used
2017 when verifying addresses in header lines, that is, for this case:
2019 verify = header_sender/callout=defer_ok
2021 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2022 those file descriptors could be used for SMTP connections. If anything
2023 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2024 could be sent to the SMTP client, causing chaos. The daemon now opens
2025 stdin, stdout, and stderr to /dev/null when it puts itself into the
2028 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2029 log by default. The user can override this with suitable Perl magic.
2031 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2032 the log message when discard triggers, was not working for the DATA ACL or
2033 for the non-SMTP ACL.
2035 78. Error message wording change in sieve.c.
2037 79. If smtp_accept_max_per_host was set, the number of connections could be
2038 restricted to fewer than expected, because the daemon was trying to set up
2039 a new connection before checking whether the processes handling previous
2040 connections had finished. The check for completed processes is now done
2041 earlier. On busy systems, this bug wouldn't be noticed because something
2042 else would have woken the daemon, and it would have reaped the completed
2045 80. If a message was submitted locally by a user whose login name contained one
2046 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2047 It caused a spool format error. I have fixed the spool reading code. A
2048 related problem was that the "from" clause in the Received: line became
2049 illegal because of the space(s). It is now covered by ${quote_local_part.
2051 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2054 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2056 83. Patch from Sieve maintainer to fix unterminated string problem in
2057 "vacation" handling.
2059 84. Some minor changes to the Linux configuration files to help with other
2060 OS variants using glibc.
2062 85. One more patch for Sieve to update vacation handling to latest spec.
2065 ----------------------------------------------------
2066 See the note above about the 4.44 and 4.50 releases.
2067 ----------------------------------------------------
2073 1. Change 4.43/35 introduced a bug that caused file counts to be
2074 incorrectly computed when quota_filecount was set in an appendfile
2077 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2078 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2081 3. Give more explanation in the error message when the command for a transport
2082 filter fails to execute.
2084 4. There are several places where Exim runs a non-Exim command in a
2085 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2086 was being done only for the command run by the queryprogram router. It is
2087 now done for all such subprocesses. The other cases are: ${run, transport
2088 filters, and the commands run by the lmtp and pipe transports.
2090 5. Some older OS have a limit of 256 on the maximum number of file
2091 descriptors. Exim was using setrlimit() to set 1000 as a large value
2092 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2093 systems. I've change it so that if it can't get 1000, it tries for 256.
2095 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2096 was an oversight, and furthermore, ever since the addition of extra
2097 controls (e.g. 4.43/32), the checks on when to allow different forms of
2098 "control" were broken. There should now be diagnostics for all cases when a
2099 control that does not make sense is encountered.
2101 7. $recipients is now available in the predata ACL (oversight).
2103 8. Tidy the search cache before the fork to do a delivery from a message
2104 received from the command line. Otherwise the child will trigger a lookup
2105 failure and thereby defer the delivery if it tries to use (for example) a
2106 cached ldap connection that the parent has called unbind on.
2108 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2109 of $address_data from the recipient verification was clobbered by the
2110 sender verification.
2112 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2113 was its contents. (It was OK if the option was not defined at all.)
2115 11. A "Completed" log line is now written for messages that are removed from
2116 the spool by the -Mrm option.
2118 12. $host_address is now set to the target address during the checking of
2119 ignore_target_hosts.
2121 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2122 being passed; this would have caused $sender_host_name to have been used if
2123 matching the list had actually called for a host name (not very likely,
2124 since this list is usually IP addresses). A host name is now passed as
2127 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2128 code that specifies a non-restarting handler (typically sigaction() in
2129 modern systems) in an attempt to fix a rare and obscure crash bug.
2131 15. Narrowed the window for a race in the daemon that could cause it to ignore
2132 SIGCHLD signals. This is not a major problem, because they are used only to
2133 wake it up if nothing else does.
2135 16. A malformed maildirsize file could cause Exim to calculate negative values
2136 for the mailbox size or file count. Odd effects could occur as a result.
2137 The maildirsize information is now recalculated if the size or filecount
2140 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2141 support for a long time. Removed HAVE_SYS_VFS_H.
2143 18. Updated exipick to current release from John Jetmore.
2145 19. Allow an empty sender to be matched against a lookup in an address list.
2146 Previously the only cases considered were a regular expression, or an
2149 20. Exim went into a mad DNS lookup loop when doing a callout where the
2150 host was specified on the transport, if the DNS lookup yielded more than
2153 21. The RFC2047 encoding function was originally intended for short strings
2154 such as real names; it was not keeping to the 75-character limit for
2155 encoded words that the RFC imposes. It now respects the limit, and
2156 generates multiple encoded words if necessary. To be on the safe side, I
2157 have increased the buffer size for the ${rfc2047: expansion operator from
2160 22. Failure to deliver a bounce message always caused it to be frozen, even if
2161 there was an errors_to setting on the router. The errors_to setting is now
2164 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2165 canonical form (fully expanded) before being placed in
2166 $sender_host_address.
2168 24. Updated eximstats to version 1.33
2170 25. Include certificate and key file names in error message when GnuTLS fails
2171 to set them up, because the GnuTLS error message doesn't include the name
2172 of the failing file when there is a problem reading it.
2174 26. Expand error message when OpenSSL has problems setting up cert/key files.
2177 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2178 (this can affect the format of dates).
2180 28. exim_tidydb, when checking for the continued existence of a message for
2181 which it has found a message-specific retry record, was not finding
2182 messages that were in split spool directories. Consequently, it was
2183 deleting retry records that should have stayed in existence.
2185 29. eximstats updated to version 1.35
2186 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2187 1.35 - bugfix such that pie charts by volume are generated correctly
2189 30. The SPA authentication driver was not abandoning authentication and moving
2190 on to the next authenticator when an expansion was forced to fail,
2191 contradicting the general specification for all authenticators. Instead it
2192 was generating a temporary error. It now behaves as specified.
2194 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2195 (the order specifies the preference for clients). The order is now AES256,
2196 AES128, 3DES, ARCFOUR128.
2198 31. Small patch to Sieve code - explicitly set From: when generating an
2201 32. Exim crashed if a remote delivery caused a very long error message to be
2202 recorded - for instance if somebody sent an entire SpamAssassin report back
2203 as a large number of 550 error lines. This bug was coincidentally fixed by
2204 increasing the size of one of Exim's internal buffers (big_buffer) that
2205 happened as part of the Exiscan merge. However, to be on the safe side, I
2206 have made the code more robust (and fixed the comments that describe what
2209 33. Some experimental protocols are using DNS PTR records for new purposes. The
2210 keys for these records are domain names, not reversed IP addresses. The
2211 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2212 leaves it alone. Component reversal etc. now happens only for IP addresses.
2215 34. The host_aton() function is supposed to be passed a string that is known
2216 to be a valid IP address. However, in the case of IPv6 addresses, it was
2217 not checking this. This is a hostage to fortune. Exim now panics and dies
2218 if the condition is not met. A case was found where this could be provoked
2219 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2220 components; fortuitously, this particular loophole had already been fixed
2221 by change 4.50/55 or 4.44/33 above.
2223 If there are any other similar loopholes, the new check in host_aton()
2224 itself should stop them being exploited. The report I received stated that
2225 data on the command line could provoke the exploit when Exim was running as
2226 exim, but did not say which command line option was involved. All I could
2227 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2228 running as the user.
2231 35. There was a buffer overflow vulnerability in the SPA authentication code
2232 (which came originally from the Samba project). I have added a test to the
2233 spa_base64_to_bits() function which I hope fixes it.
2236 36. The daemon start-up calls getloadavg() while still root for those OS that
2237 need the first call to be done as root, but it missed one case: when
2238 deliver_queue_load_max is set with deliver_drop_privilege. This is
2239 necessary for the benefit of the queue runner, because there is no re-exec
2240 when deliver_drop_privilege is set.
2242 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2243 was in use, was not putting the data itself into the right store pool;
2244 consequently, it could be overwritten for a subsequent message in the same
2245 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2248 38. Sometimes the final signoff response after QUIT could fail to get
2249 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2250 < 0 before doing a fflush(). This bug looks as though it goes back to the
2251 introduction of TLS in release 3.20, but "sometimes" must have been rare
2252 because the tests only now provoked it.
2258 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2259 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2260 Now it returns an integer. A number of calls were still expecting a T/F
2261 return. Fortuitously, in all cases, the tests worked in OK situations,
2262 which is the norm. However, things would have gone wrong on any write
2263 failures on the smtp file descriptor. This function is used when sending
2264 messages over SMTP and also when doing verify callouts.
2266 2. When Exim is called to do synchronous delivery of a locally submitted
2267 message (the -odf or -odi options), it no longer closes stderr before doing
2270 3. Implemented the mua_wrapper option.
2272 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2274 5. Implemented the functions header_remove(), header_testname(),
2275 header_add_at_position(), and receive_remove_recipient(), and exported them
2278 6. If an ACL "warn" statement specified the addition of headers, Exim already
2279 inserted X-ACL-Warn: at the start if there was no header name. However, it
2280 was not making this test for the second and subsequent header lines if
2281 there were newlines in the string. This meant that an invalid header could
2282 be inserted if Exim was badly configured.
2284 7. Allow an ACL "warn" statement to add header lines at the start or after all
2285 the Received: headers, as well as at the end.
2287 8. Added the rcpt_4xx retry error code.
2289 9. Added postmaster_mailfrom=xxx to callout verification option.
2291 10. Added mailfrom=xxxx to the callout verification option, for verify=
2294 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2295 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2297 12. Inserted some casts to stop certain compilers warning when using pointer
2298 differences as field lengths or precisions in printf-type calls (mostly
2299 affecting debugging statements).
2301 13. Added optional readline() support for -be (dynamically loaded).
2303 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2304 same clock tick as a message's arrival, so that its received time was the
2305 same as the "first fail" time on the retry record, and that message
2306 remained on the queue past the ultimate address timeout, every queue runner
2307 would try a delivery (because it was past the ultimate address timeout) but
2308 after another failure, the ultimate address timeout, which should have then
2309 bounced the address, did not kick in. This was a "< instead of <=" error;
2310 in most cases the first failure would have been in the next clock tick
2311 after the received time, and all would be well.
2313 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2314 being recognized when the domain list was tested by the match_domain
2315 condition in an expansion string.
2317 16. Added the ${str2b64: operator.
2319 17. Exim was always calling setrlimit() to set a large limit for the number of
2320 processes, without checking whether the existing limit was already
2321 adequate. (It did check for the limit on file descriptors.) Furthermore,
2322 errors from getrlimit() and setrlimit() were being ignored. Now they are
2323 logged to the main and panic logs, but Exim does carry on, to try to do its
2324 job under whatever limits there are.
2326 18. Imported PCRE 5.0.
2328 19. Trivial typo in log message " temporarily refused connection" (the leading
2331 20. If the log selector return_path_on_delivery was set and an address was
2332 redirected to /dev/null, the delivery process crashed because it assumed
2333 that a return path would always be set for a "successful" delivery. In this
2334 case, the whole delivery is bypassed as an optimization, and therefore no
2337 21. Internal re-arrangement: the function for sending a challenge and reading
2338 a response while authentication was assuming a zero-terminated challenge
2339 string. It's now changed to take a pointer and a length, to allow for
2340 binary data in such strings.
2342 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2344 23. Exim was not respecting finduser_retries when seeking the login of the
2345 uid under which it was called; it was always trying 10 times. (The default
2346 setting of finduser_retries is zero.) Also, it was sleeping after the final
2347 failure, which is pointless.
2349 24. Implemented tls_on_connect_ports.
2351 25. Implemented acl_smtp_predata.
2353 26. If the domain in control=submission is set empty, Exim assumes that the
2354 authenticated id is a complete email address when it generates From: or
2355 Sender: header lines.
2357 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2358 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2359 chown and chgrp in /bin and hostname in /usr/bin.
2361 28. Exim was keeping the "process log" file open after each use, just as it
2362 does for the main log. This opens the possibility of it remaining open for
2363 long periods when the USR1 signal hits a daemon. Occasional processlog
2364 errors were reported, that could have been caused by this. Anyway, it seems
2365 much more sensible not to leave this file open at all, so that is what now
2368 29. The long-running daemon process does not normally write to the log once it
2369 has entered its main loop, and it closes the log before doing so. This is
2370 so that log files can straightforwardly be renamed and moved. However,
2371 there are a couple of unusual error situations where the daemon does write
2372 log entries, and I had neglected to close the log afterwards.
2374 30. The text of an SMTP error response that was received during a remote
2375 delivery was being truncated at 512 bytes. This is too short for some of
2376 the long messages that one sometimes sees. I've increased the limit to
2379 31. It is now possible to make retry rules that apply only when a message has a
2380 specific sender, in particular, an empty sender.
2382 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2383 it possible to be selective about when SMTP synchronization is enforced.
2385 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2387 32. Implemented hosts_connection_nolog.
2389 33. Added an ACL for QUIT.
2391 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2394 35. Added mailbox_size and mailbox_filecount to appendfile.
2396 36. Added control = no_multiline_responses to ACLs.
2398 37. There was a bug in the logic of the code that waits for the clock to tick
2399 in the case where the clock went backwards by a substantial amount such
2400 that the microsecond fraction of "now" was more than the microsecond
2401 fraction of "then" (but the whole seconds number was less).
2403 38. Added support for the libradius Radius client library this is found on
2404 FreeBSD (previously only the radiusclient library was supported).
2410 1. When certain lookups returned multiple values in the form name=value, the
2411 quoting of the values was not always being done properly. Specifically:
2412 (a) If the value started with a double quote, but contained no whitespace,
2414 (b) If the value contained whitespace other than a space character (i.e.
2415 tabs or newlines or carriage returns) it was not quoted.
2416 This fix has been applied to the mysql and pgsql lookups by writing a
2417 separate quoting function and calling it from the lookup code. The fix
2418 should probably also be applied to nisplus, ibase and oracle lookups, but
2419 since I cannot test any of those, I have not disturbed their existing code.
2421 2. A hit in the callout cache for a specific address caused a log line with no
2422 reason for rejecting RCPT. Now it says "Previous (cached) callout
2423 verification failure".
2425 3. There was an off-by-one bug in the queryprogram router. An over-long
2426 return line was truncated at 256 instead of 255 characters, thereby
2427 overflowing its buffer with the terminating zero. As well as fixing this, I
2428 have increased the buffer size to 1024 (and made a note to document this).
2430 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2431 when Exim is waiting for an SMTP response from a remote server, Exim
2432 restarts its select() call on the socket, thereby resetting its timeout.
2433 This is not a problem when such interrupts are rare. Somebody set up a cron
2434 job to run exiwhat every 2 minutes, which is less than the normal select()
2435 timeout (5 or 10 minutes). This meant that the select() timeout never
2436 kicked in because it was always reset. I have fixed this by comparing the
2437 time when an interrupt arrives with the time at the start of the first call
2438 to select(). If more time than the timeout has elapsed, the interrupt is
2439 treated as a timeout.
2441 5. Some internal re-factoring in preparation for the addition of Sieve
2442 extensions (by MH). In particular, the "personal" test is moved to a
2443 separate function, and given an option for scanning Cc: and Bcc: (which is
2444 not set for Exim filters).
2446 6. When Exim created an email address using the login of the caller as the
2447 local part (e.g. when creating a From: or Sender: header line), it was not
2448 quoting the local part when it contained special characters such as @.
2450 7. Installed new OpenBSD configuration files.
2452 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2453 try to make them clearer.
2455 9. Callout options, other than the timeout value, were being ignored when
2456 verifying sender addresses in header lines. For example, when using
2458 verify = header_sender/callout=no_cache
2460 the cache was (incorrectly) being used.
2462 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2463 only the Cygwin environment.
2465 11. When return_path_on_delivery was set as a log selector, if different remote
2466 addresses in the same message used different return paths and parallel
2467 remote delivery occurred, the wrong values would sometimes be logged.
2468 (Whenever a remote delivery process finished, the return path value from
2469 the most recently started remote delivery process was logged.)
2471 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2472 lines when AUTH and/or TLS are in use. This is the "received protocol"
2473 field. Exim used to use "asmtp" for authenticated SMTP, without any
2474 indication (in the protocol name) for TLS use. Now it follows the RFC and
2475 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2476 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2477 names appear in log lines as well as in Received: header lines.
2479 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2480 extensions, and comparison tests, and to fix some bugs.
2482 14. Changes to the "personal" filter test:
2484 (1) The test was buggy in that it was just doing the equivalent of
2485 "contains" tests on header lines. For example, if a user's address was
2486 anne@some.where, the "personal" test would incorrectly be true for
2488 To: susanne@some.where
2490 This test is now done by extracting each address from the header in turn,
2491 and checking the entire address. Other tests that are part of "personal"
2492 are now done using regular expressions (for example, to check local parts
2493 of addresses in From: header lines).
2495 (2) The list of non-personal local parts in From: addresses has been
2496 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2497 taken from the Sieve specification recommendations.
2499 (3) If the message contains any header line starting with "List-" it is
2500 treated as non-personal.
2502 (4) The test for "circular" in the Subject: header line has been removed
2503 because it now seems ill-conceived.
2505 15. Minor typos in src/EDITME comments corrected.
2507 16. Installed latest exipick from John Jetmore.
2509 17. If headers_add on a router specified a text string that was too long for
2510 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2511 of string_sprintf() is now avoided.
2513 18. $message_body_size was not set (it was always zero) when running the DATA
2514 ACL and the local_scan() function.
2516 19. For the "mail" command in an Exim filter, no default was being set for
2517 the once_repeat time, causing a random time value to be used if "once" was
2518 specified. (If the value happened to be <= 0, no repeat happened.) The
2519 default is now 0s, meaning "never repeat". The "vacation" command was OK
2520 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2521 (I found it when inspecting the code).
2523 20. There is now an overall timeout for performing a callout verification. It
2524 defaults to 4 times the callout timeout, which applies to individual SMTP
2525 commands during the callout. The overall timeout applies when there is more
2526 than one host that can be tried. The timeout is checked before trying the
2527 next host. This prevents very long delays if there are a large number of
2528 hosts and all are timing out (e.g. when the network connections are timing
2529 out). The value of the overall timeout can be changed by specifying an
2530 additional sub-option for "callout", called "maxwait". For example:
2532 verify = sender/callout=5s,maxwait=20s
2534 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2535 to the end before writing, but this should make it even safer).
2537 22. Exim was forgetting that it had advertised PIPELINING for the second and
2538 subsequent messages on an SMTP connection. It was also not resetting its
2539 memory on STARTTLS and an internal HELO.
2541 23. When Exim logs an SMTP synchronization error within a session, it now
2542 records whether PIPELINING has been advertised or not.
2544 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2545 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2546 rather than long int.
2548 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2550 26. Added the never_mail option to autoreply.
2556 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2557 crash if the getsockname() call failed; this can happen if a connection is
2558 closed very soon after it is established. The problem was simply in the
2559 order in which certain operations were done, causing Exim to try to write
2560 to the SMTP stream before it had set up the file descriptor. The bug has
2561 been fixed by making things happen in the correct order.
2567 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2568 before the connection was closed, thus losing the rejection response.
2570 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2571 some early Solaris releases, but causes trouble in current releases where
2572 socklen_t is defined.
2574 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2577 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2580 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2581 that releasing the top of it at the end releases what was used for sub-
2582 expansions (unless the block got too big). However, discard this block if
2583 the first thing is a variable or header, so that we can use its block when
2584 it is dynamic (useful for very large $message_headers, for example).
2586 6. Lookups now cache *every* query, not just the most recent. A new, separate
2587 store pool is used for this. It can be recovered when all lookup caches are
2588 flushed. Lookups now release memory at the end of their result strings.
2589 This has involved some general refactoring of the lookup sources.
2591 7. Some code has been added to the store_xxx() functions to reduce the amount
2592 of flapping under certain conditions.
2594 8. log_incoming_interface used to affect only the <= reception log lines. Now
2595 it causes the local interface and port to be added to several more SMTP log
2596 lines, for example "SMTP connection from", and rejection lines.
2598 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2600 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2602 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2603 could be overwritten at the end of the current message (or the start of a
2604 new message if it was set in a HELO ACL). The value is now preserved for
2605 the duration of the SMTP connection.
2607 12. If a transport had a headers_rewrite setting, and a matching header line
2608 contained an unqualified address, that address was qualified, even if it
2609 did not match any rewriting rules. The underlying bug was that the values
2610 of the flags that permit the existence of unqualified sender and recipient
2611 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2612 non-local messages, and by -bnq for local messages) were not being
2613 preserved with the message after it was received.
2615 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2616 "next input=" as part of the text comprising the host identity instead of
2617 the correct text. The code was using the same buffer for two different
2618 strings. However, depending on which order the printing function evaluated
2619 its arguments, the bug did not always show up. Under Linux, for example, my
2620 test suite worked just fine.
2622 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2623 doesn't work with some older versions of Perl. It has been changed to "my",
2624 which in any case is probably the better facility to use.
2626 15. A really picky compiler found some instances of statements for creating
2627 error messages that either had too many or two few arguments for the format
2630 16. The size of the buffer for calls to the DNS resolver has been increased
2631 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2632 for addresses that have a lot of PTR records. This alleviates a problem; it
2633 does not fully solve it.
2635 17. A dnsdb lookup for PTR records that receives more data than will fit in the
2636 buffer now truncates the list and logs the incident, which is the same
2637 action as happens when Exim is looking up a host name and its aliases.
2638 Previously in this situation something unpredictable would happen;
2639 sometimes it was "internal error: store_reset failed".
2641 18. If a server dropped the connection unexpectedly when an Exim client was
2642 using GnuTLS and trying to read a response, the client delivery process
2643 crashed while trying to generate an error log message.
2645 19. If a "warn" verb in an ACL added multiple headers to a message in a single
2646 string, for example:
2648 warn message = H1: something\nH2: something
2650 the text was added as a single header line from Exim's point of view
2651 though it ended up OK in the delivered message. However, searching for the
2652 second and subsequent header lines using $h_h2: did not work. This has been
2653 fixed. Similarly, if a system filter added multiple headers in this way,
2654 the routers could not see them.
2656 20. Expanded the error message when iplsearch is called with an invalid key to
2657 suggest using net-iplsearch in a host list.
2659 21. When running tests using -bh, any delays imposed by "delay" modifiers in
2660 ACLs are no longer actually imposed (and a message to that effect is
2663 22. If a "gecos" field in a passwd entry contained escaped characters, in
2664 particular, if it contained a \" sequence, Exim got it wrong when building
2665 a From: or a Sender: header from that name. A second bug also caused
2666 incorrect handling when an unquoted " was present following a character
2667 that needed quoting.
2669 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2670 was not being matched caselessly.
2672 24. Arranged for all hyphens in the exim.8 source to be escaped with
2675 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2676 properly. Recipient callout cache records were still being keyed to include
2677 the sender, even when use_sender was set false. This led to far more
2678 callouts that were necessary. The sender is no longer included in the key
2679 when use_sender is false.
2681 26. Added "control = submission" modifier to ACLs.
2683 27. Added the ${base62d: operator to decode base 62 numbers.
2685 28. dnsdb lookups can now access SRV records.
2687 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2688 the configuration file.
2690 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2691 (-v) mode. This makes the output for a verbose queue run more intelligible.
2693 31. Added a use_postmaster feature to recipient callouts.
2695 32. Added the $body_zerocount variable, containing the number of binary zero
2696 bytes in the message body.
2698 33. The time of last modification of the "new" subdirectory is now used as the
2699 "mailbox time last read" when there is a quota error for a maildir
2702 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2704 35. Added +ignore_unknown as a special item in host lists.
2706 36. Code for decoding IPv6 addresses in host lists is now included, even if
2707 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2708 address was recognized as an IP address, but was then not correctly decoded
2709 into binary, causing unexpected and incorrect effects when compared with
2716 1. Very minor rewording of debugging text in manualroute to say "list of
2717 hosts" instead of "hostlist".
2719 2. If verify=header_syntax was set, and a header line with an unqualified
2720 address (no domain) and a large number of spaces between the end of the
2721 name and the colon was received, the reception process suffered a buffer
2722 overflow, and (when I tested it) crashed. This was caused by some obsolete
2723 code that should have been removed. The fix is to remove it!
2725 3. When running in the test harness, delay a bit after writing a bounce
2726 message to get a bit more predictability in the log output.
2728 4. Added a call to search_tidyup() just before forking a reception process. In
2729 theory, someone could use a lookup in the expansion of smtp_accept_max_
2730 per_host which, without the tidyup, could leave open a database connection.
2732 5. Added the variables $recipient_data and $sender_data which get set from a
2733 lookup success in an ACL "recipients" or "senders" condition, or a router
2734 "senders" option, similar to $domain_data and $local_part_data.
2736 6. Moved the writing of debug_print from before to after the "senders" test
2739 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2740 problems for message scanning, either using a data ACL, or using
2741 local_scan() because the Received: header was not generated till after they
2742 were called (in order to set the time as the time of reception completion).
2743 I have revised the way this works. The header is now generated after the
2744 body is received, but before the ACL or local_scan() are called. After they
2745 are run, the timestamp in the header is updated.
2751 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2752 before starting a queue runner without re-exec. This happened only when
2753 deliver_drop_privilege was set or when the Exim user was set to root. The
2754 effect of the bug was that timeouts during subsequent deliveries caused
2755 crashes instead of being properly handled. The handler is now left at its
2756 default (and expected) setting.
2758 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2759 message, again when deliver_drop_privilege is set or Exim is run as root.
2760 The bug described in (1) was not present in this case, but the tidying up
2761 of the other signals was missing. I have made the two cases consistent.
2763 3. The ignore_target_hosts setting on a manualroute router was being ignored
2764 for hosts that were looked up using the /MX notation.
2766 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2769 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2770 operated on the sender address. After changing the $sender_address to <>
2771 for the sender address verify, Exim was re-instated it as the original
2772 (before rewriting) address, but remembering that it had rewritten it, so it
2773 wasn't rewriting it again. This bug also had the effect of breaking the
2774 sender address verification caching when the sender address was rewritten.
2776 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2777 This has been changed so that if the ip literal address matches
2778 ignore_target_hosts, the router declines.
2780 7. Added expansion conditions match_domain, match_address, and match_local_
2781 part (NOT match_host).
2783 8. The placeholder for the Received: header didn't have a length field set.
2785 9. Added code to Exim itself and to exim_lock to test for a specific race
2786 condition that could lead to file corruption when using MBX delivery. The
2787 issue is with the lockfile that is created in /tmp. If this file is removed
2788 after a process has opened it but before that process has acquired a lock,
2789 there is the potential for a second process to recreate the file and also
2790 acquire a lock. This could lead to two Exim processes writing to the file
2791 at the same time. The added code performs the same test as UW imapd; it
2792 checks after acquiring the lock that its file descriptor still refers to
2793 the same named file.
2795 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2796 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2799 11. Added the smtp_active_hostname option. If used, this will typically be made
2800 to depend on the incoming interface address. Because $interface_address is
2801 not set up until the daemon has forked a reception process, error responses
2802 that can happen earlier (such as "too many connections") no longer contain
2805 12. If an expansion in a condition on a "warn" statement fails because a lookup
2806 defers, the "warn" statement is abandoned, and the next ACL statement is
2807 processed. Previously this caused the whole ACL to be aborted.
2809 13. Added the iplsearch lookup type.
2811 14. Added ident_timeout as a log selector.
2813 15. Added tls_certificate_verified as a log selector.
2815 16. Added a global option tls_require_ciphers (compare the smtp transport
2816 option of the same name). This controls incoming TLS connections.
2818 17. I finally figured out how to make tls_require_ciphers do a similar thing
2819 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2820 before starting the TLS session.
2822 18. Tabs are now shown as \t in -bP output.
2824 19. If the log selector return_path_on_delivery was set, Exim crashed when
2825 bouncing a message because it had too many Received: header lines.
2827 20. If two routers both had headers_remove settings, and the first one included
2828 a superfluous trailing colon, the final name in the first list and the
2829 first name in the second list were incorrectly joined into one item (with a
2830 colon in the middle).
2836 1. Added -C and -D options to the exinext utility, mainly to make it easier
2837 to include in the automated testing, but these could be helpful when
2838 multiple configurations are in use.
2840 2. The exinext utility was not formatting the output nicely when there was
2841 an alternate port involved in the retry record key, nor when there was a
2842 message id as well (for retries that were specific to a specific message
2843 and a specific host). It was also confused by IPv6 addresses, because of
2844 the additional colons they contain. I have fixed the IPv4 problem, and
2845 patched it up to do a reasonable job for IPv6.
2847 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2848 delivery, the log line now contains "pipelined" if PIPELINING was used.
2850 4. An SMTP transport process used to panic and die if the bind() call to set
2851 an explicit outgoing interface failed. This has been changed; it is now
2852 treated in the same way as a connect() failure.
2854 5. A reference to $sender_host_name in the part of a conditional expansion
2855 that was being skipped was still causing a DNS lookup. This no longer
2858 6. The def: expansion condition was not recognizing references to header lines
2859 that used bh_ and bheader_.
2861 7. Added the _cache feature to named lists.
2863 8. The code for checking quota_filecount in the appendfile transport was
2864 allowing one more file than it should have been.
2866 9. For compatibility with Sendmail, the command line option
2874 and sets the incoming protocol and host name (for trusted callers). The
2875 host name and its colon can be omitted when only the protocol is to be set.
2876 Note the Exim already has two private options, -pd and -ps, that refer to
2877 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2878 "s", but I don't think that's a major issue.
2880 10. A number of refactoring changes to the code, none of which should affect
2883 (a) The number of logging options was getting close to filling up the
2884 32-bit word that was used as a bit map. I have split them into two classes:
2885 those that are passed in the argument to log_write(), and those that are
2886 only ever tested independently outside of that function. These are now in
2887 separate 32-bit words, so there is plenty of room for expansion again.
2888 There is no change in the user interface or the logging behaviour.
2890 (b) When building, for example, log lines, the code previously used a
2891 macro that called string_cat() twice, in order to add two strings. This is
2892 not really sufficiently general. Furthermore, there was one instance where
2893 it was actually wrong because one of the argument was used twice, and in
2894 one call a function was used. (As it happened, calling the function twice
2895 did not affect the overall behaviour.) The macro has been replaced by a
2896 function that can join an arbitrary number of extra strings onto a growing
2899 (c) The code for expansion conditions now uses a table and a binary chop
2900 instead of a serial search (which was left over from when there were very
2901 few conditions). Also, it now recognizes conditions like "pam" even when
2902 the relevant support is not compiled in: a suitably worded error message is
2903 given if an attempt is made to use such a condition.
2905 11. Added ${time_interval:xxxxx}.
2907 12. A bug was causing one of the ddress fields not to be passed back correctly
2908 from remote delivery subprocesses. The field in question was not being
2909 subsequently used, so this caused to problems in practice.
2911 13. Added new log selectors queue_time and deliver_time.
2913 14. Might have fixed a bug in maildirsizefile handling that threw up
2914 "unexpected character" debug warnings, and recalculated the data
2915 unnecessarily. In any case, I expanded the warning message to give more
2918 15. Added the message "Restricted characters in address" to the statements in
2919 the default ACL that block characters like @ and % in local parts.
2921 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2922 Three changes have been made:
2924 (a) There was a serious bug; a negative response to MAIL caused the whole
2925 recipient domain to be cached as invalid, thereby blocking all messages
2926 to all local parts at the same domain, from all senders. This bug has
2927 been fixed. The domain is no longer cached after a negative response to
2928 MAIL if the sender used is not empty.
2930 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2933 (c) A new callout option, "use_sender" has been added for people who want
2934 the modified behaviour.
2940 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2943 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2944 indeed breaks things for older releases.
2946 3. Added additional logging to the case where there is a problem reading data
2947 from a filter that is running in a subprocess using a pipe, in order to
2948 try to track down a specific problem.
2950 4. Testing facility fudge: when running in the test harness and attempting
2951 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2952 getting "No route to host". Convert this to a timeout.
2954 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2957 6. Some OS don't have socklen_t but use size_t instead. This affects the
2958 fifth argument of getsockopt() amongst other things. This is now
2959 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2960 can be set for individual OS. I have set it for SunOS5, OSF1, and
2961 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2962 some earlier ones do not.
2964 7. Change 4.30/15 was not doing the test caselessly.
2966 8. The standard form for an IPv6 address literal was being rejected by address
2967 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2968 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2969 this, as well as the form without the "IPv6" on the front (but only when
2970 address literals are enabled, of course).
2972 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2974 10. Exim crashed if a message with an empty sender address specified by -f
2975 encountered a router with an errors_to setting. This could be provoked only
2976 by a command such as
2980 where an empty string was supplied; "<>" did not hit this bug.
2982 11. Installed PCRE release 4.5.
2984 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2985 remained set. It is now erased.
2987 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2988 times from message ids (which are base 36 rather than the normal 62).
2990 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2991 were being counted as actual protocol errors, and logged if the log
2992 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2993 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2994 and DATA following a set of rejected RCPTs do not count as protocol errors.
2995 In other words, Exim assumes they were pipelined, though this may not
2996 actually be the case. Of course, in all cases the client gets an
2997 appropriate error code.
2999 15. If a lookup fails in an ACL condition, a message about the failure may
3000 be available; it is used if testing the ACL cannot continue, because most
3001 such messages specify what the cause of the deferral is. However, some
3002 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3003 that caused an old message to be retained and used if a later statement
3004 caused a defer, replacing the real cause of the deferral.
3006 16. If an IP address had so many PTR records that the DNS lookup buffer
3007 was not large enough to hold them, Exim could crash while trying to process
3008 the truncated data. It now detects and logs this case.
3010 17. Further to 4.21/58, another change has been made: if (and only if) the
3011 first line of a message (the first header line) ends with CRLF, a bare LF
3012 in a subsequent header line has a space inserted after it, so as not to
3013 terminate the header.
3015 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3016 unnecessarily, used atoi() instead of strtol(), and didn't check the
3017 termination when getting file sizes from file names by regex.
3019 19. Completely re-implemented the support for maildirsize files, in the light
3020 of a number of problems with the previous contributed implementation
3021 (4.30/29). In particular:
3023 . If the quota is zero, the maildirsize file is maintained, but no quota is
3026 . If the maildir directory does not exist, it is created before any attempt
3027 to write a maildirsize file.
3029 . The quota value in the file is just a cache; if the quota is changed in
3030 the transport, the new value overrides.
3032 . A regular expression is available for excluding directories from the
3035 20. The autoreply transport checks the characters in options that define the
3036 message's headers; it allows continued headers, but it was checking with
3037 isspace() after an embedded newline instead of explicitly looking for a
3040 21. If all the "regular" hosts to which an address was routed had passed their
3041 expiry times, and had not reached their retry times, the address was
3042 bounced, even if fallback hosts were defined. Now Exim should go on to try
3045 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3046 equivalent code in the SMTP transport. Some hosts send humungous responses
3047 to HELO/EHLO, more than 1024 it seems.
3049 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3050 gives compiler warnings in some environments. I've now done it "properly",
3053 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3054 (because of problems with the built-in one) was declared to return uschar *
3055 instead of char *, causing compiler failure.
3057 25. Fixed a file descriptor leak when processing alias/forward files.
3059 26. Fixed a minor format string issue in dbfn.c.
3061 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3063 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3064 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3067 29. When a sender address is verified, it is cached, to save repeating the test
3068 when there is more than one recipient in a message. However, when the
3069 verification involves a callout, it is possible for different callout
3070 options to be set for different recipients. It is too complicated to keep
3071 track of this in the cache, so now Exim always runs a verification when a
3072 callout is required, relying on the callout cache for the optimization.
3073 The overhead is duplication of the address routing, but this should not be
3076 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3077 to be verified with callout=postmaster, and the main callout worked but the
3078 postmaster check failed, the verification correctly failed. However, if a
3079 subsequent RCPT command asked for sender verification *without* the
3080 postmaster check, incorrect caching caused this verification also to fail,
3083 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3084 it was not caching the DNS options (qualify_single, search_parents) that
3085 were used when the lookup failed. A subsequent lookup with different
3086 options therefore always gave the same answer, though there were cases
3087 where it should not have. (Example: a "domains = !$mx_any" option on a
3088 dnslookup router: the "domains" option is always processed without any
3089 widening, but the router might have qualify_single set.) Now Exim uses the
3090 cached value only when the same options are set.
3092 32. Added John Jetmore's "exipick" utility to the distribution.
3094 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3095 than a timeout (e.g. a certificate is required, and is not provided), an
3096 Exim server now closes the connection immediately. Previously it waited for
3097 the client to close - but if the client is SSL, it seems that they each
3098 wait for each other, leading to a delay before one of them times out.
3100 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3101 maintained 0.8.x compatibility because I don't think many are using it, and
3102 it is clearly obsolete.
3104 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3107 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3108 client certificate was expired. A simple patch fixes this, though I don't
3109 understand the full logic of why the verify callback is called multiple
3112 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3113 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3114 which causes problems with some clients (such as the Certicom SSL Plus
3115 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3116 disables the coutermeasure allowing Eudora to connect."
3118 38. Exim was not checking that a write() to a log file succeeded. This could
3119 lead to Bad Things if a log got too big, in particular if it hit a file
3120 size limit. Exim now panics and dies if it cannot write to a log file, just
3121 as it does if it cannot open a log file.
3123 39. Modified OS/Makefile-Linux so that it now contains
3125 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3127 The two -D definitions ensure that Exim is compiled with large file
3128 support, which makes it possible to handle log files that are bigger than
3131 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3132 instance) a domain was checked against a named list that involved a lookup,
3133 causing $domain_data to be set, then another domain was checked against the
3134 same list, then the first domain was re-checked, the value of $domain_data
3135 after the final check could be wrong. In particular, if the second check
3136 failed, it could be set empty. This bug probably also applied to
3139 41. The strip_trailing_dot option was not being applied to the address given
3140 with the -f command-line option.
3142 42. The code for reading a message's header from the spool was incrementing
3143 $received_count, but never initializing it. This meant that the value was
3144 incorrect (doubled) while delivering a message in the same process in which
3145 it was received. In the most common configuration of Exim, this never
3146 happens - a fresh exec is done - but it can happen when
3147 deliver_drop_privilege is set.
3149 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3150 it now includes up to 150 characters of the unexpected data in the log
3153 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3154 and building data strings. The size of both of these buffers was 10 000
3155 bytes - far larger than anybody would *ever* want, thought I. Needless to
3156 say, somebody hit the limit. I have increased the maximum line length to
3157 20 000 and the maximum data length of concatenated lines to 100 000. I have
3158 also fixed two bugs, because there was no checking on these buffers. Tsk,
3159 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3160 buffer is too small.
3162 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3163 lsearch lookups. Now it does.
3165 46. When parsing a route_list item in a manualroute router, a fixed-length
3166 buffer was used for the list of hosts. I made this 1024 bytes long,
3167 thinking that nobody would ever have a list of hosts that long. Wrong.
3168 Somebody had a whole pile of complicated expansion conditions, and the
3169 string was silently truncated, leading to an expansion error. It turns out
3170 that it is easier to change to an unlimited length (owing to other changes
3171 that have happened since this code was originally written) than to build
3172 structure for giving a limitation error. The length of the item that
3173 expands into the list of hosts is now unlimited.
3175 47. The lsearch lookup could not handle data where the length of text line was
3176 more than 4095 characters. Such lines were truncated, leading to shortened
3177 data being returned. It should now handle lines of any length.
3179 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3180 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3183 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3184 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3185 no longer happens in comment lines. A list of replacements is now placed
3186 at the head of all of the source files, except those whose only change is
3187 to replace PERL_COMMAND in the very first #! line.
3189 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3190 messages on the queue, with a bottom-up merge sort, using code contributed
3191 by Michael Haardt. This should make operations like -bp somewhat faster on
3192 large queues. It won't affect queue runners, except when queue_run_in_order
3195 51. Installed eximstats 1.31 in the distribution.
3197 52. Added support for SRV lookups to the dnslookup router.
3199 53. If an ACL referred to $message_body or $message_body_end, the value was not
3200 reset for any messages that followed in the same SMTP session.
3202 54. The store-handling optimization for building very long strings was not
3203 differentiating between the different store pools. I don't think this
3204 actually made any difference in practice, but I've tidied it.
3206 55. While running the routers to verify a sender address, $sender_address
3207 was still set to the sender address. This is wrong, because when routing to
3208 send a bounce to the sender, it would be empty. Therefore, I have changed
3209 it so that, while verifying a sender address, $sender_address is set to <>.
3210 (There is no change to what happens when verifying a recipient address.)
3212 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3213 target A or AAAA records (if not already returned) without resetting the
3214 qualify_single or search_parents options of the DNS resolver. These are
3215 inappropriate in this case because the targets of MX and SRV records must
3216 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3217 target that, when qualified, matched something in the local domain. These
3218 two options are now turned off when doing these lookups.
3220 57. It seems that at least some releases of Reiserfs (which does not have the
3221 concept of a fixed number of inodes) returns zero and not -1 for the
3222 number of available inodes. This interacted badly with check_spool_inodes,
3223 which assumed that -1 was the "no such thing" setting. What I have done is
3224 to check that the total number of inodes is greater than zero before doing
3225 the test of how many are available.
3227 58. When a "warn" ACL statement has a log_message modifier, the message is
3228 remembered, and not repeated. This is to avoid a lot of repetition when a
3229 message has many recipients that cause the same warning to be written.
3230 Howewer, Exim was preserving the list of already written lines for an
3231 entire SMTP session, which doesn't seem right. The memory is now reset if a
3232 new message is started.
3234 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3235 debugging output unless log_rewrite was also set.
3237 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3238 of (int)(handle) when we know that handle contains (void *)(-1).
3240 61. The Exim daemon panic-logs an error return when it closes the incoming
3241 connection. However "connection reset by peer" seems to be common, and
3242 isn't really an error worthy of noting specially, so that particular error
3245 62. When Exim is trying to find all the local interfaces, it used to panic and
3246 die if the ioctl to get the interface flags failed. However, it seems that
3247 on at least one OS (Solaris 9) it is possible to have an interface that is
3248 included in the list of interfaces, but for which you get a failure error
3249 for this call. This happens when the interface is not "plumbed" into a
3250 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3251 failure of the "get flags" call assumes that the interface is down.
3253 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3254 makes life easier for people who are doing arithmetic on fields extracted
3255 from dates, where you often get leading zeros that should not be
3256 interpreted as octal.
3258 64. Added qualify_domain to the redirect router, to override the global
3261 65. If a pathologically long header line contained very many addresses (the
3262 report of this problem mentioned 10 000) and each of them was rewritten,
3263 Exim could use up a very large amount of memory. (It kept on making new
3264 copies of the header line as it rewrote, and never released the old ones.)
3265 At the expense of a bit more processing, the header rewriting function has
3266 been changed so that it no longer eats memory in this way.
3268 66. The generation of the Received: header has been moved from the time that a
3269 message starts to be received, to the time that it finishes. The timestamp
3270 in the Received: header should now be very close to that of the <= log
3271 line. There are two side-effects of this change:
3273 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3274 logged header lines no longer include the local Received: line, because
3275 it has not yet been created. The same applies to a copy of the message
3276 that is returned to a non-SMTP sender when a message is rejected.
3278 (b) When a filter file is tested using -bf, no additional Received: header
3279 is added to the test message. After some thought, I decided that this
3282 This change does not affect the value of $received_for. It is still set
3283 after address rewriting, but before local_scan() is called.
3285 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3287 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3288 gave an unhelpful panic error message, and a defer error. I have managed to
3289 change this behaviour so that it now rejects any supplied certificate,
3290 which seems right, as the list of acceptable certificates is empty.
3292 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3293 gave an unhelpful defer error. I have not managed to make this reject any
3294 supplied certificates, but the error message it gives is "no certificate
3295 supplied", which is not helpful.
3297 70. exigrep's output now also includes lines that are not associated with any
3298 message, but which match the given pattern. Implemented by a patch from
3299 Martin Sluka, which also tidied up the Perl a bit.
3301 71. Recipient callout verification, like sender verification, was using <> in
3302 the MAIL FROM command. This isn't really the right thing, since the actual
3303 sender may affect whether the remote host accepts the recipient or not. I
3304 have changed it to use the actual sender in the callout; this means that
3305 the cache record is now keyed on a recipient/sender pair, not just the
3306 recipient address. There doesn't seem to be a real danger of callout loops,
3307 since a callout by the remote host to check the sender would use <>.
3308 [SEE ABOVE: changed after hitting problems.]
3310 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3311 temporary errors. However, in the case of such a code being given after
3312 the end of a data transmission (i.e. after ".") Exim was failing to write
3313 a retry record for the message. (Yes, there was some broken host that was
3314 actually sending 8xx at this point.)
3316 73. An unknown lookup type in a host list could cause Exim to panic-die when
3317 the list was checked. (An example that provoked this was putting <; in the
3318 middle of a list instead of at the start.) If this happened during a DATA
3319 ACL check, a -D file could be left lying around. This kind of configuration
3320 error no longer causes Exim to die; instead it causes a defer errror. The
3321 incident is still logged to the main and panic logs.
3323 74. Buglet left over from Exim 3 conversion. The message "too many messages
3324 in one connection" was written to the rejectlog but not the mainlog, except
3325 when address rewriting (yes!) was being logged.
3327 75. Added write_rejectlog option.
3329 76. When a system filter was run not as root (that is, when system_filter_user
3330 was set), the values of the $n variables were not being returned to the
3331 main process; thus, they were not subsequently available in the $sn
3334 77. Added +return_path_on_delivery log selector.
3336 78. A connection timeout was being treated differently from recipients deferred
3337 when testing hosts_max_try with a message that was older than the host's
3338 retry timeout. (The host should not be counted, thus allowing all hosts to
3339 be tried at least once before bouncing.) This may have been the cause of an
3340 occasionally reported bug whereby a message would remain on the queue
3341 longer than the retry timeout, but would be bounced if a delivery was
3342 forced. I say "may" because I never totally pinned down the problem;
3343 setting up timeout/retry tests is difficult. See also the next item.
3345 79. The ultimate address timeout was not being applied to errors that involved
3346 a combination of host plus message (for example, a timeout on a MAIL
3347 command). When an address resolved to a number of possible hosts, and they
3348 were not all tried for each delivery (e.g. because of hosts_max_try), a
3349 message could remain on the queue longer than the retry timeout.
3351 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3354 81. Fixed an obscure SMTP outgoing bug which required at least the following
3355 conditions: (a) there was another message waiting for the same server;
3356 (b) the server returned 5xx to all RCPT commands in the first message so
3357 that the message was not completed; (c) the server dropped the connection
3358 or gave a negative response to the RSET that Exim sends to abort the
3359 transaction. The observed case was a dropped connection after DATA that had
3360 been sent in pipelining mode. That is, the server had advertised PIPELINING
3361 but was not implementing it correctly. The effect of the bug was incorrect
3362 behaviour, such as trying another host, and this could lead to a crash.
3368 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3369 and daemon.c were passed as pointers to ints; they should have been
3370 pointers to socklen_t variables (which are typically unsigned ints).
3372 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3375 3. Fixed a really odd bug that affected only the testing scheme; patching a
3376 certain fixed string in the binary changed the value of another string that
3377 happened to be identical to the end of the original first string.
3379 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3380 name", it returns that address as the IP address. On some operating
3381 systems (e.g. Solaris), it also passes back the IP address string as the
3382 "host name". However, on others (e.g. Linux), it passes back an empty
3383 string. Exim wasn't checking for this, and was changing the host name to an
3384 empty string, assuming it had been canonicized.
3386 5. Although rare, it is permitted to have more than one PTR record for a given
3387 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3388 all the names associated with an address, because they do in Solaris.
3389 However, it seems that they do not in Linux for data that comes from the
3390 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3391 I found this out when I moved to a new Linux workstation and tried to run
3392 the Exim test suite.
3394 To get round this problem I have changed the code so that it now does its
3395 own call to the DNS to look up PTR records when searching for a host name.
3396 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3397 addresses that are only in /etc/hosts are still found.
3399 This behaviour is, however, controlled by an option called host_lookup_
3400 order, which defaults to "bydns:byaddr". If people want to use the other
3401 order, or indeed, just use one or the other means of lookup, they can
3402 specify it in this variable.
3404 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3405 some operating systems, this comes back from gethostbyaddr() as an empty
3406 string, and this is what Exim used to test for. However, it seems that in
3407 other systems, "." is yielded. Exim now tests for this case too.
3409 7. The values of check_spool_space and check_log_space are now held internally
3410 as a number of kilobytes instead of an absolute number of bytes. If a
3411 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3412 kilobyte. This means that much larger values can be stored.
3414 8. Exim monitor: an attempt to get the action menu when not actually pointing
3415 at a message produces an empty menu entitled "No message selected". This
3416 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3417 no entries in it ("Shell widget menu has zero width and/or height"). So I
3418 have added a single, blank menu entry in this case.
3420 9. Added ${quote_local_part.
3422 10. MIME decoding is now applied to the contents of Subject: header lines when
3425 11. Now that a reference to $sender_host_address automatically causes a reverse
3426 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3427 host lookup before query-style lookups in lists that might use this
3428 variable. This has therefore been abolished, and the "net-" prefix is no
3429 longer necessary for query-style lookups.
3431 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3432 have been a typo for LFLAGS, so it has been changed.
3434 13. The install script calls Exim with "-C /dev/null" in order to find the
3435 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3436 to be output. Howeve, since Exim outputs its version number before the
3437 error, it didn't break the script. It just looked ugly. I fixed this by
3438 always allowing "-C /dev/null" if the caller is root.
3440 14. Ignore overlarge ACL variable number when reading spool file - insurance
3441 against a later release with more variables having written the file.
3443 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3444 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3445 this, as well as the form without the "IPv6" on the front.
3447 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3448 OS/Makefile-Darwin file.
3450 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3451 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3453 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3454 variable. However, some OS use ints for inodes. I've added cast to long int
3455 to get rid of the compiler warning.
3457 19. I had forgotten to lock out "/../" in configuration file names when
3458 ALT_CONFIG_PREFIX was set.
3460 20. Routers used for verification do not need to specify transports. However,
3461 if such a router generated a host list, and callout was configured, Exim
3462 crashed, because it could not find a port number from the (non-existent)
3463 transport. It now assumes port 25 in this circumstance.
3465 21. Added the -t option to exigrep.
3467 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3468 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3469 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3470 them is set, LOOKUP_LSEARCH is forced.
3472 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3474 24. Added sender and host information to the "rejected by local_scan()" log
3475 line; previously there was no indication of these.
3477 25. Added .include_if_exists.
3479 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3480 Linux. It turns out that not all file systems support this. Apparently some
3481 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3482 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3483 not supported on the file descriptor, is now ignored when Exim is trying to
3484 sync a directory. This applies only to Linux.
3486 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3488 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3489 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3492 29. Added support for maildirsize files from supplied patch (modified a bit).
3494 30. The use of :fail: followed by an empty string could lead Exim to respond to
3495 sender verification failures with (e.g.):
3497 550 Verification failed for <xxx>
3498 550 Sender verify failed
3500 where the first response line was missing the '-' that indicates it is not
3501 the final line of the response.
3503 31. The loop for finding the name of the user that called Exim had a hardwired
3504 limit of 10; it now uses the value of finduser_retries, which is used for
3505 all other user lookups.
3507 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3510 33. Exim was neglecting to zero errno before one call of strtol() when
3511 expanding a string and expecting an integer value. On some systems this
3512 resulted in spurious "integer overflow" errors. Also, it was casting the
3513 result into an int without checking.
3515 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3516 did not work. The code looks as if it has *never* worked, though it appears
3517 to have been documented since at least releast 1.62. I have made it work.
3519 35. The "timeout_DNS" error in retry rules, also documented since at least
3520 1.62, also never worked. As it isn't clear exactly what this means, and
3521 clearly it isn't a major issue, I have abolished the feature by treating it
3522 as "timeout", and writing a warning to the main and panic logs.
3524 36. The display of retry rules for -brt wasn't always showing the error code
3527 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3528 timeout_connect_A, timeout_connect_MX.
3530 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3531 to the empty sender.
3533 39. The daemon was not analysing the content of -oX till after it had closed
3534 stderr and disconnected from the controlling terminal. This meant that any
3535 syntax errors were only noted on the panic log, and the return code from
3536 the command was 0. By re-arranging the code a little, I've made the
3537 decoding happen first, so such errors now appear on stderr, and the return
3538 code is 1. However, the actual setting up of the sockets still happens in
3539 the disconnected process, so errors there are still only recorded on the
3542 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3543 connections (as happens on some IP stacks) was logged at start up time as
3544 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3545 it from "IPv6 and IPv4", which means that two separate sockets are being
3548 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3549 says whether AF_INET or AF_INET6 was passed as an argument.
3551 42. Exiwhat output was messed up when time zones were included in log
3554 43. Exiwhat now gives more information about the daemon's listening ports,
3555 and whether -tls-on-connect was used.
3557 44. The "port" option of the smtp transport is now expanded.
3559 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3560 silently ignored. Now an error message is written to the main and panic
3563 46. There's a new ACL modifier called "logwrite" which writes to a log file
3564 as soon as it is encountered.
3566 47. Added $local_user_uid and $local_user_gid at routing time.
3568 48. Exim crashed when trying to verify a sender address that was being
3571 49. Exim was recognizing only a space character after ".include". It now also
3572 recognizes a tab character.
3574 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3575 extracting the relevant information from the specification. The man page no
3576 longer contains scrambled data for the -d option, and I've added a section
3577 at the front about calling Exim under different names.
3579 51. Added "extra_headers" argument to the "mail" command in filter files.
3581 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3584 53. Installed eximstats 1.29.
3586 54. Added transport_filter_timeout as a generic transport option.
3588 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3589 Cc: header lines. This was required by RFC 822, but it not required by RFC
3592 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3593 incoming messages that did not have them. Now it does so only if the
3594 message originates locally, that is, if there is no associated remote host
3595 address. When Resent- header lines are present, this applies to the Resent-
3596 lines rather than the non-Resent- lines.
3598 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3599 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3601 58. Messages for configuration errors now include the name of the main
3602 configuration file - useful now that there may be more than one file in a
3603 list (.included file names were always shown).
3605 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3606 for those rare installations that do not start the daemon as root or run it
3607 setuid root. I've cut out the call to initgroups() if the daemon is not
3610 60. The Exim user and group can now be bound into the binary as text strings
3611 that are looked up at the start of Exim's processing.
3613 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3615 62. Added $mailstore_basename variable.
3617 63. Installed patch to sieve.c from Michael Haardt.
3619 64. When Exim failed to open the panic log after failing to open the main log,
3620 the original message it was trying to log was written to stderr and debug
3621 output, but if they were not available (the usual case in production), it
3622 was lost. Now it is written to syslog before the two lines that record the
3623 failures to open the logs.
3625 65. Users' Exim filters run in subprocesses under the user's uid. It is
3626 possible for a "deliver" command or an alias in a "personal" command to
3627 provoke an address rewrite. If logging of address rewriting is configured,
3628 this fails because the process is not running as root or exim. There may be
3629 a better way of dealing with this, but for the moment (because 4.30 needs
3630 to be released), I have disabled address rewrite logging when running a
3631 filter in a non-root, non-exim process.
3637 1. The buildconfig auxiliary program wasn't quoting the value set for
3638 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3639 not defined. This bug was masked in 4.22 by the effect that was fixed in
3642 2. Some messages that were rejected after a message id was allocated were
3643 shown as "incomplete" by exigrep. It no longer does this for messages that
3644 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3646 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3647 have allow_domain_literals set, the ID did not get logged in the <= line.
3648 Domain literals are now always recognized in Message-ID: header lines.
3650 4. The first argument for a ${extract expansion item is the key name or field
3651 number. Leading and trailing spaces in this item were not being ignored,
3652 causing some misleading effects.
3654 5. When deliver_drop_privilege was set, single queue runner processes started
3655 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3656 same command in the process it spins off) were not dropping privilege.
3658 6. When the daemon running as "exim" started a queue runner, it always
3659 re-executed Exim in the spun-off process. This is a waste of effort when
3660 deliver_drop_privilege is set. The new process now just calls the
3661 queue-runner function directly.
3667 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3670 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3671 ignored. Though the use of -oP was forcing the writing of a pid file, it
3672 was always written to the default place.
3674 3. If the message "no IP address found for host xxxx" is generated during
3675 incoming verification, it is now followed by identification of the incoming
3676 connection (so you can more easily find what provoked it).
3678 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3680 5. Added some features to "harden" Exim a bit more against certain attacks:
3682 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3683 be put in Local/Makefile. This is like the never_users runtime option,
3684 but it cannot be overridden. The default setting is "root".
3686 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3687 prefix string with which any file named in a -C command line option
3690 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3691 is retained for -C and -D only if the caller of Exim is root. Without
3692 it, the exim user may also use -C and -D and retain privilege.
3694 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3695 command line option is disabled.
3697 6. Macro names set by the -D option must start with an upper case letter, just
3698 like macro names defined in the configuration file.
3700 7. Added "dereference=" facility to LDAP.
3702 8. Two instances of the typo "uknown" in the source files are fixed.
3704 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3705 the Configure-Makefile script screwed up while processing it.
3707 10. Incorporated PCRE 4.4.
3709 11. The SMTP synchronization check was not operating right at the start of an
3710 SMTP session. For example, it could not catch a HELO sent before the client
3711 waited for the greeting. There is now a check for outstanding input at the
3712 point when the greeting is written. Because of the duplex, asynchronous
3713 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3714 way, but not yet received, when the check is performed.
3716 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3717 on TCP/IP sockets, because this apparently causes some broken clients to
3720 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3721 unchanged) from the Cygwin maintainer.
3723 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
3724 was supported instead of testing for mbx. Effectively a typo.
3726 15. The spa authenticator server code was not checking that the input it
3727 received was valid base64.
3729 16. The debug output line for the "set" modifier in ACLs was not showing the
3730 name of the variable that was being set.
3732 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
3734 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
3735 lookup on its own. Something else was needed to trigger the lookup. For
3736 example, a match in host_lookup or the need for a host name in a host list.
3737 Now, if $sender_host_name is referenced and the host name has not yet been
3738 looked up, a lookup is performed. If the lookup fails, the variable remains
3739 empty, and $host_lookup_failed is set to "1".
3741 19. Added "eqi" as a case-independent comparison operator.
3743 20. The saslauthd authentication condition could segfault if neither service
3744 nor realm was specified.
3746 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3747 error message that was logged was misleading, and incoming SMTP
3748 connections were dropped. The message is now more accurate, and temporary
3749 errors are given to SMTP connections.
3751 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3752 (except RSET) with a 421 error, until QUIT is received. However, it was
3753 failing to send a response to QUIT.
3755 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3756 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3757 ACL is now run after the helo_try_verify_hosts code.
3759 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3760 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3761 case-independent, so other case variants are also recognized). Apparently
3762 some systems use these upper case variants.
3764 25. If more than two messages were waiting for the same host, and a transport
3765 filter was specified for the transport, Exim sent two messages over the
3766 same TCP/IP connection, and then failed with "socket operation on non-
3767 socket" when it tried to send the third.
3769 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3771 27. The extern definition of crypt16() in expand.c was not being excluded when
3772 the OS had its own crypt16() function.
3774 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3775 as a preferred synonym for return_size_limit, both as an option and as an
3778 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3780 30. Changed the default configuration ACL to relax the local part checking rule
3781 for addresses that are not in any local domains. For these addresses,
3782 slashes and pipe symbols are allowed within local parts, but the sequence
3783 /../ is explicitly forbidden.
3785 31. SPA server authentication was not clearing the challenge buffer before
3788 32. log_message in a "warn" ACL statement was writing to the reject log as
3789 well as to the main log, which contradicts the documentation and doesn't
3790 seem right (because no rejection is happening). So I have stopped it.
3792 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3793 However, I am unable to do any testing of this.
3795 34. Fixed an infelicity in the appendfile transport. When checking directories
3796 for a mailbox, to see if any needed to be created, it was accidentally
3797 using path names with one or more superfluous leading slashes; tracing
3798 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3800 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3801 contents are added to the log line that is written for every discarded
3802 recipient. (Previously a log_message setting was ignored.)
3804 36. The ${quote: operator now quotes the string if it is empty.
3806 37. The install script runs exim in order to find its version number. If for
3807 some reason other than non-existence or emptiness, which it checks, it
3808 could not run './exim', it was installing it with an empty version number,
3809 i.e. as "exim-". This error state is now caught, and the installation is
3812 38. An argument was missing from the function that creates an error message
3813 when Exim fails to connect to the socket for saslauthd authentication.
3814 This could cause Exim to crash, or give a corrupted message.
3816 39. Added isip, isip4, and isip6 to ${if conditions.
3818 40. The ACL variables $acl_xx are now saved with the message, and can be
3819 accessed later in routers, transports, and filters.
3821 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3822 strings in the file are not string-expanded.
3824 42. If a MAIL command specified a SIZE value that was too large to fit into an
3825 int variable, the check against message_size_limit failed. Such values are
3826 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3827 day this will have to be increased, but I don't think I want to be around
3828 when emails are that large.
3835 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3836 iconv() is not standard in FreeBSD.
3838 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3839 IPv6 enabled. The observed symptom was a segmentation fault on return from
3840 the function os_common_find_running_interfaces() in src/os.c.
3842 3. In the check_special_case() function in daemon.c I had used "errno" as an
3843 argument name, which causes warnings on some systems. This was basically a
3844 typo, since it was named "eno" in the comments!
3846 4. The code that waits for the clock to tick (at a resolution of some fraction
3847 of a second) so as to ensure message-id uniqueness was always waiting for
3848 at least one whole tick, when it could have waited for less. [This is
3849 almost certainly not relevant at current processor speeds, where it is
3850 unlikely to ever wait at all. But we try to future-proof.]
3852 5. The function that sleeps for a time interval that includes fractions of a
3853 second contained a race. It did not block SIGALRM between setting the
3854 timer, and suspending (a couple of lines later). If the interval was short
3855 and the sigsuspend() was delayed until after it had expired, the suspension
3856 never ended. On busy systems this could lead to processes getting stuck for
3859 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3860 process, before it forks any delivery processes. The open lookup caching
3861 mechanism meant that the open file or database connection was passed into
3862 the delivery process. The problem was that delivery processes always tidy
3863 up cached lookup data. This could cause a problem for the next delivery
3864 process started by the queue runner, because the external queue runner
3865 process does not know about the closure. So the next delivery process
3866 still has data in the lookup cache. In the case of a file lookup, there was
3867 no problem because closing a file descriptor in a subprocess doesn't affect
3868 the parent. However, if the lookup was caching a connection to a database,
3869 the connection was closed, and the second delivery process was likely to
3870 see errors such as "PGSQL: query failed: server closed the connection
3871 unexpectedly". The problem has been fixed by closing all cached lookups
3872 in a queue runner before running a delivery process.
3874 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3875 seem to have the "const" qualifier which it has on other OS. I've
3878 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3879 *of the same type* (client or server) with the same public name that an
3880 error should be diagnosed.
3882 9. When Exim looked up a host name for an IP address, but failed to find the
3883 original IP address when looking up the host name (a safety check), it
3884 output the message "<ip address> does not match any IP for NULL", which was
3885 confusing, to say the least. The bug was that the host name should have
3886 appeared instead of "NULL".
3888 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3889 user that is built into the binary, and the -C or -D options is used, root
3890 privilege is dropped before the configuration file is read. In addition,
3891 logging is switched to stderr instead of the normal log files. If the
3892 configuration then re-defines the Exim user, the unprivileged environment
3893 is probably not what is expected, so Exim logs a panic warning message (but
3896 However, if deliver_drop_privilege is set, the unprivileged state may well
3897 be exactly what is intended, so the warning has been cut out in that case,
3898 and Exim is allowed to try to write to its normal log files.
3904 1. smtp_return_error_details was not giving details for temporary sender
3905 or receiver verification errors.
3907 2. Diagnose a configuration error if two authenticators have the same public
3910 3. Exim used not to create the message log file for a message until the first
3911 delivery attempt. This could be confusing when incoming messages were held
3912 for policy or load reasons. The message log file is now created at the time
3913 the message is received, and an initial "Received" line is written to it.
3915 4. The automatically generated man page for command line options had a minor
3916 bug that caused no ill effects; however, a more serious problem was that
3917 the procedure for building the man page automatically didn't always
3918 operate. Consequently, release 4.20 contains an out-of-date version. This
3919 shouldn't happen again.
3921 5. When building Exim with embedded Perl support, the script that builds the
3922 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3923 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3925 6. The freeze_tell option was not being used for messages that were frozen on
3926 arrival, either by an ACL or by local_scan().
3928 7. Added the smtp_incomplete_transaction log selector.
3930 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3931 was accepting AUTH without a new EHLO.
3933 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3934 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3937 10. Logging of TCP/IP connections (when configured) now happens in the main
3938 daemon process instead of the child process, so that the TCP/IP connection
3939 count is more accurate (but it can never be perfect).
3941 11. The use of "drop" in a nested ACL was not being handled correctly in the
3942 outer ACL. Now, if condition failure induced by the nested "drop" causes
3943 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3944 or "require"), the connection is dropped.
3946 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3947 that yield "discard" can now be used with an "accept" or a "discard" verb,
3948 but an error is generated for any others (because I can't see a useful way
3949 to define what should happen).
3951 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3952 are now processed in the same way as lines in the Exim configuration file.
3953 In particular, continuation lines are supported.
3955 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3957 15. Added -ti meaning -t -i.
3959 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3960 domains, and warn by logging if others are found.
3962 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3963 returned from GIFCONF when Exim is trying to find the list of interfaces on
3964 a host. The code in os.c has been modified to copy each ifreq to an aligned
3965 structure in all cases.
3967 Also, in some cases, the returned ifreq's were being copied to a 'struct
3968 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3969 means the last couple of bytes of an IPv6 address could be chopped if the
3970 ifreq contained only a normal sockaddr (14 bytes storage).
3972 18. Named domain lists were not supported in the hosts_treat_as_local option.
3973 An entry such as +xxxx was not recognized, and was treated as a literal
3976 19. Ensure that header lines added by a DATA ACL are included in the reject log
3977 if the ACL subsequently rejects the message.
3979 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3980 MD5 (which is deprecated).
3982 21. When testing a filter file using -bf, Exim was writing a message when it
3983 took the sender from a "From " line in the message, but it was not doing so
3984 when it took $return_path from a Return-Path: header line. It now does.
3986 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3987 with a valid header line field name (a series of printing characters
3988 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3990 23. Changed "disc" in the source to "disk" to conform to the documentation and
3991 the book and for uniformity.
3993 24. Ignore Sendmail's -Ooption=value command line item.
3995 25. When execve() failed while trying to run a command in a pipe transport,
3996 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3997 could be confused with a return value of 69 from the command itself. This
3998 has been changed to 127, the value the shell returns if it is asked to run
3999 a non-existent command. The wording for the related log line suggests a
4000 non-existent command as the problem.
4002 26. If received_header_text expands to an empty string, do not add a Received:
4003 header line to the message. (Well, it adds a token one on the spool, but
4004 marks it "old" so that it doesn't get used or transmitted.)
4006 27. Installed eximstats 1.28 (addition of -nt option).
4008 28. There was no check for failure on the call to getsockname() in the daemon
4009 code. This can fail if there is a shortage of resources on the system, with
4010 ENOMEM, for example. A temporary error is now given on failure.
4012 29. Contrary to the C standard, it seems that in some environments, the
4013 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4014 program. Exim now does this explicitly; it affects the formatting of
4015 timestamps using strftime().
4017 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4018 complaints. I've now initialized all the variables, to avoid this.
4020 32. Header lines added by a system filter were not being "seen" during
4021 transport-time rewrites.
4023 33. The info_callback() function passed to OpenSSL is set up with type void
4024 (*)(SSL *, int, int), as described somewhere. However, when calling the
4025 function (actually a macro) that sets it up, the type void(*)() is
4026 expected. I've put in a cast to prevent warnings from picky compilers.
4028 34. If a DNS black list lookup found a CNAME record, but there were no A
4029 records associated with the domain it pointed at, Exim crashed.
4031 35. If a DNS black list lookup returned more than one A record, Exim ignored
4032 all but the first. It now scans all returned addresses if a particular IP
4033 value is being sought. In this situation, the contents of the
4034 $dnslist_value variable are a list of all the addresses, separated by a
4037 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4038 to accept a host name and all its aliases if the forward lookup for any of
4039 them yielded the IP address of the incoming connection. Now it accepts only
4040 those names whose forward lookup yields the correct IP address. Any other
4041 names are discarded. This closes a loophole whereby a rogue DNS
4042 administrator could create reverse DNS records to break through a
4043 wildcarded host restriction in an ACL.
4045 37. If a user filter or a system filter that ran in a subprocess used any of
4046 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4047 the wrong values were passed to the pipe command ($thisaddress had the
4048 value of $0, $0 had the value of $1, etc). This bug was introduced by
4049 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4051 38. Improved the line breaking for long SMTP error messages from ACLs.
4052 Previously, if there was no break point between 40 and 75 characters, Exim
4053 left the rest of the message alone. Two changes have been made: (a) I've
4054 reduced the minimum length to 35 characters; (b) if it can't find a break
4055 point between 35 and 75 characters, it looks ahead and uses the first one
4056 that it finds. This may give the occasional overlong line, but at least the
4057 remaining text gets split now.
4059 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4060 file descriptors might be low, and that setting 1000 would always raise it.
4061 It turns out that in some environments, the limit is already over 1000 and
4062 that lowering it causes trouble. So now Exim takes care not to decrease it.
4064 40. When delivering a message, the value of $return_path is set to $sender_
4065 address at the start of routing (routers may change the value). By an
4066 oversight, this default was not being set up when an address was tested by
4067 -bt or -bv, which affected the outcome if any router or filter referred to
4070 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4071 log only when "message" or "log_message" are set. However, if one of the
4072 conditions was an address verification, or a call to a nested ACL, the
4073 messages generated by the underlying test were being passed through. This
4074 no longer happens. The underlying message is available in $acl_verify_
4075 message for both "message" and "log_message" expansions, so it can be
4076 passed through if needed.
4078 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4079 new expansion $bh_ to give the encoded byte string without charset
4080 translation. Translation happens only if iconv() is available; HAVE_ICONV
4081 indicates this at build time. HEADERS_CHARSET gives the charset to
4082 translate to; headers_charset can change it in the configuration, and
4083 "headers charset" can change it in an individual filter file.
4085 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4086 that creates RFC 2047 encoded "words" labels them as that charset instead
4087 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4088 expansion operator; (ii) when Exim creates a From: line for a local
4089 message; (iii) when a header line is rewritten to include a "phrase" part.
4091 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4092 buggy, causing it to skip the first lines of messages whose message ID
4093 ended in 'D'. This would not have bitten before Exim release 4.14, because
4094 message IDs were unlikely to end in 'D' before then. The effect was to have
4095 incorrect size information for certain domains.
4097 45. #include "config.h" was missing at the start of the crypt16.c module. This
4098 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4101 46. If there was a timeout during a "random" callout check, Exim treated it as
4102 a failure of the random address, and carried on sending RSET and the real
4103 address. If the delay was just some slowness somewhere, the response to the
4104 original RCPT would be taken as a response to RSET and so on, causing
4105 mayhem of various kinds.
4107 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4108 when I implemented it. It didn't allow for the fact that some option values
4109 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4110 the right test for positive values.
4112 48. Domain names in DNS records are case-independent. Exim always looks them up
4113 in lower case. Some resolvers return domain names in exactly the case they
4114 appear in the zone file, that is, they may contain uppercase letters. Not
4115 all resolvers do this - some return always lower case. Exim was treating a
4116 change of case by a resolver as a change of domain, similar to a widening
4117 of a domain abbreviation. This triggered its re-routing code and so it was
4118 trying to route what was effectively the same domain again. This normally
4119 caused routing to fail (because the router wouldn't handle the domain
4120 twice). Now Exim checks for this case specially, and just changes the
4121 casing of the domain that it ultimately uses when it transmits the message
4124 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4127 50. If a filter generated a file delivery with a non-absolute name (possible if
4128 no home directory exists for the router), the forbid_file option was not
4131 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4132 the existing equality matching.
4134 52. Exim was using ints instead of ino_t variables in some places where it was
4135 dealing with inode numbers.
4137 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4138 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4139 called TMPDIR, and if it finds it is different, it changes its value.
4141 54. The smtp_printf() function is now made available to local_scan() so
4142 additional output lines can be written before returning. There is also an
4143 smtp_fflush() function to enable the detection of a dropped connection.
4144 The variables smtp_input and smtp_batched_input are exported to
4147 55. Changed the default runtime configuration: the message "Unknown user"
4148 has been removed from the ACL, and instead placed on the localuser router,
4149 using the cannot_route_message feature. This means that any verification
4150 failures that generate their own messages won't get overridden. Similarly,
4151 the "Unrouteable address" message that was in the ACL for unverifiable
4152 relay addresses has also been removed.
4154 56. Added hosts_avoid_esmtp to the smtp transport.
4156 57. The exicyclog script was not checking for the esoteric option
4157 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4158 will work only if exicyclog is run under the appropriate euid.
4160 58. Following a discussion on the list, the rules by which Exim recognises line
4161 endings on incoming messages have been changed. The -dropcr and drop_cr
4162 options are now no-ops, retained only for backwards compatibility. The
4163 following line terminators are recognized: LF CRLF CR. However, special
4164 processing applies to CR:
4166 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4167 nor a local message in the state where . is a terminator.
4169 (ii) If a bare CR is encountered in a header line, an extra space is added
4170 after the line terminator so as not to end the header. The reasoning
4171 behind this is that bare CRs in header lines are most likely either
4172 to be mistakes, or people trying to play silly games.
4174 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4175 was being incorrectly given as 18 bytes larger than it should have been.
4176 This is a VOB (very old bug).
4178 60. This may never have affected anything current, but just in case it has:
4179 When the local host is found other than at the start of a list of hosts,
4180 the local host, those with the same MX, and any that follow, are discarded.
4181 When the list in question was part of a longer list of hosts, the following
4182 hosts (not currently being processed) were also being discarded. This no
4183 longer happens. I'm not sure if this situation could ever has previously
4186 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4189 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4190 header. This is something that is recommended in a new Internet Draft, and
4191 is something that is documented as being done by Sendmail. There are two
4192 possible values. For messages generated by the autoreply transport, Exim
4195 Auto-Submitted: auto-replied
4197 whereas for all other generated messages (e.g. bounces) it adds
4199 Auto-Submitted: auto-generated
4201 63. The "personal" condition in filters now includes a test for the
4202 Auto-Submitted: header. If it contains the string "auto-" the message it
4203 not considered personal.
4205 64. Added rcpt_include_affixes as a generic transport option.
4207 65. Added queue_only_override (default true).
4209 66. Added the syslog_duplication option.
4211 67. If what should have been the first header line of a message consisted of
4212 a space followed by a colon, Exim was mis-interpreting it as a header line.
4213 It isn't of course - it is syntactically invalid and should therefore be
4214 treated as the start of the message body. The misbehaviour could have
4215 caused a number of strange effects, including loss of data in subsequent
4216 header lines, and spool format errors.
4218 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4219 client host had authenticated. This control can now be exercised by an ACL
4220 for more flexibility.
4222 69. By default, callouts do not happen when testing with -bh. There is now a
4223 variant, -bhc, which does actually run the callout code, including
4224 consulting and updating the callout cache.
4226 70. Added support for saslauthd authentication, courtesy of Alexander
4229 71. If statvfs() failed on the spool or log directories while checking their
4230 size for availability, Exim confusingly gave the error "space shortage".
4231 Furthermore, in debugging mode it crashed with a floating point exception.
4232 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4233 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4234 serious problem, Exim now writes to the main and panic logs when this
4235 happens, with details of the failure. It then refuses to accept the
4236 incoming message, giving the message "spool directory problem" or "log
4237 directory problem" with a 421 code for SMTP messages.
4239 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4240 0, 1, and 2 exist, because some OS complain for execs without them (see
4241 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4242 descriptors. However, the code omitted to check that the open succeeded,
4243 causing mysterious errors if for some reason the permissions on /dev/null
4244 got screwed. Now Exim writes a message to the main and panic logs, and
4245 bombs out if it can't open /dev/null.
4247 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4248 interact so that it is all more flexible. It is supposed to remain
4249 backwards compatible. Also added extra_local_interfaces.
4251 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4252 to bomb out with an assertion failure - to the client this appears as a
4253 connection drop. This problem occurs in the part of the code that was taken
4254 from the Samba project. Fortunately, the assertion is in a very simple
4255 function, so I have fixed this by reproducing the function inline in the
4256 one place where it is called, and arranging for authentication to fail
4257 instead of killing the process with assert().
4259 75. The SPA client code was not working when the server requested OEM rather
4260 than Unicode encoding.
4262 76. Added code to make require_files with a specific uid setting more usable in
4263 the case where statting the file as root fails - usually a non-root-mounted
4264 NFS file system. When this happens and the failure is EACCES, Exim now
4265 forks a subprocess and does the per-uid checking as the relevant uid.
4267 77. Added process_log_path.
4269 78. If log_file_path was not explicitly set, a setting of check_log_space or
4270 check_log_inodes was ignored.
4272 79. If a space check for the spool or log partitions fails, the incident is now
4273 logged. Of course, in the latter case the data may get lost...
4275 80. Added the %p formatting code to string_format() so that it can be used to
4276 print addresses in debug_print(). Adjusted all the address printing in the
4277 debugging in store.c to use %p rather than %d.
4279 81. There was a concern that a line of code in smtp_in.c could overflow a
4280 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4281 initially expressed, the concern was not well-founded, because trailing
4282 spaces are removed early. However, if the trailing spaces were followed by
4283 a NULL, they did not get removed, so the overflow was possible. Two fixes
4286 (a) I re-wrote the offending code in a cleaner fashion.
4287 (b) If an incoming SMTP command contains a NULL character, it is rejected
4290 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4291 runs initgroups(), so that if the Exim user is in any additional groups,
4292 they will be used during message reception.
4298 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob