Added some extra debugging when updating a wait-xxx database.
[exim.git] / doc / doc-txt / ChangeLog
CommitLineData
7a0743eb 1$Cambridge: exim/doc/doc-txt/ChangeLog,v 1.421 2006/10/31 12:16:26 ph10 Exp $
495ae4b0
PH
2
3Change log file for Exim from version 4.21
4-------------------------------------------
5
944e9e9c
TF
6Exim version 4.64
7-----------------
21d74bd9
TK
8TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
9 leftover -K file (the existence of which was triggered by #402).
10 While we were at it, introduced process PID as part of the -K
11 filename. This should rule out race conditions when creating
12 these files.
13
14TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
15 processing considerably. Previous code took too long for large mails,
16 triggering a timeout which in turn triggers #401.
17
18TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
19 in the DK code in transports.c. sendfile() is not really portable,
20 hence the _LINUX specificness.
944e9e9c
TF
21
22TF/01 In the add_headers option to the mail command in an Exim filter,
23 there was a bug that Exim would claim a syntax error in any
24 header after the first one which had an odd number of characters
25 in the field name.
26
2b1c6e3a
PH
27PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
28 callout verification, Exim cached a "reject" for the entire domain. This
29 is correct for most verifications, but it is not correct for a recipient
30 verification with use_sender or use_postmaster set, because in that case
31 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
32 case of MAIL FROM:<> rejection from other early rejections (e.g.
33 rejection of HELO). When verifying a recipient using a non-null MAIL
34 address, the cache is ignored if it shows MAIL FROM:<> rejection.
35 Whatever the result of the callout, the value of the domain cache is
36 left unchanged (for any other kind of callout, getting as far as trying
37 RCPT means that the domain itself is ok).
38
1f872c80
PH
39PH/02 Tidied a number of unused variable and signed/unsigned warnings that
40 gcc 4.1.1 threw up.
41
42PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
43 manifest itself as EPIPE rather than ECONNECT. When tidying away a
44 session, the daemon ignores ECONNECT errors and logs others; it now
45 ignores EPIPE as well.
46
d203e649
PH
47PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
48 (quoted-printable decoding).
49
cc2ed8f7 50PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
21a04aa3 51 later the small subsequent patch to fix an introduced bug.
f951fd57 52
ddfcd446
PH
53PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
54
d45b1de8
PH
55PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
56
57PH/08 An error is now given if message_size_limit is specified negative.
58
38a0a95f 59PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
641cb756 60 to be given (somewhat) arbitrary names.
38a0a95f 61
a2405d83
JJ
62JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
63 in 4.64-PH/09.
64
65JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
66 miscellaneous code fixes
67
6ea85e9a
PH
68PH/10 Added the log_reject_target ACL modifier to specify where to log
69 rejections.
70
26da7e20
PH
71PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
72 hostname. This is wrong, because it relates to the incoming message (and
73 probably the interface on which it is arriving) and not to the outgoing
74 callout (which could be using a different interface). This has been
75 changed to use the value of the helo_data option from the smtp transport
76 instead - this is what is used when a message is actually being sent. If
77 there is no remote transport (possible with a router that sets up host
78 addresses), $smtp_active_hostname is used.
6ea85e9a 79
14aa5a05 80PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
7befa435 81 tweaks were necessary in order to get it to work (see also 21 below):
14aa5a05
PH
82 (a) The code assumed that strncpy() returns a negative number on buffer
83 overflow, which isn't the case. Replaced with Exim's string_format()
84 function.
85 (b) There were several signed/unsigned issues. I just did the minimum
86 hacking in of casts. There is scope for a larger refactoring.
87 (c) The code used strcasecmp() which is not a standard C function.
88 Replaced with Exim's strcmpic() function.
89 (d) The code set only $1; it now sets $auth1 as well.
90 (e) A simple test gave the error "authentication client didn't specify
91 service in request". It would seem that Dovecot has changed its
92 interface. Fortunately there's a specification; I followed it and
93 changed what the client sends and it appears to be working now.
94
ff75a1f7
PH
95PH/13 Added $message_headers_raw to provide the headers without RFC 2047
96 decoding.
97
e6f6568e
PH
98PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
99 address A is aliased to B and C, where B exists and C does not. Without
100 -v the output is "A verified" because verification stops after a
101 successful redirection if more than one address is generated. However,
102 with -v the child addresses are also verified. Exim was outputting "A
103 failed to verify" and then showing the successful verification for C,
104 with its parentage. It now outputs "B failed to verify", showing B's
105 parentage before showing the successful verification of C.
106
d6f6e0dc
PH
107PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
108 look up a TXT record in a specific list after matching in a combined
109 list.
110
322050c2
PH
111PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
112 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
113 they consult the DNS. I had assumed they would set it the way they
114 wanted; and indeed my experiments on Linux seem to show that in some
115 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
116 To be on the safe side, however, I have now made the interface to
117 host_find_byname() similar to host_find_bydns(), with an argument
118 containing the DNS resolver options. The host_find_byname() function now
119 sets these options at its start, just as host_find_bydns() does. The smtp
120 transport options dns_qualify_single and dns_search_parents are passed to
121 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
122 of host_find_byname() use the default settings of RES_DEFNAMES
123 (qualify_single) but not RES_DNSRCH (search_parents).
124
08955dd3
PH
125PH/17 Applied (a modified version of) Nico Erfurth's patch to make
126 spool_read_header() do less string testing, by means of a preliminary
127 switch on the second character of optional "-foo" lines. (This is
128 overdue, caused by the large number of possibilities that now exist.
129 Originally there were few.) While I was there, I also converted the
130 str(n)cmp tests so they don't re-test the leading "-" and the first
131 character, in the hope this might squeeze out yet more improvement.
132
1eccaa59
PH
133PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
134 flag allowing group syntax was set by the header_syntax check but not
135 turned off, possible causing trouble later; (2) The flag was not being
136 set at all for the header_verify test, causing "group"-style headers to
137 be rejected. I have now set it in this case, and also caused header_
138 verify to ignore an empty address taken from a group. While doing this, I
139 came across some other cases where the code for allowing group syntax
140 while scanning a header line wasn't quite right (mostly, not resetting
141 the flag correctly in the right place). These bugs could have caused
142 trouble for malformed header lines. I hope it is now all correct.
143
602e59e5
PH
144PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
145 with the "reply" argument non-NULL. The code, however (which originally
146 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
147 but it didn't always do it. This confused somebody who was copying the
148 code for some other use. I have removed all the tests.
149
411ef850
PH
150PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
151 feature that was used to support insecure browsers during the U.S. crypto
152 embargo. It requires special client support, and Exim is probably the
153 only MTA that supported it -- and would never use it because real RSA is
154 always available. This code has been removed, because it had the bad
155 effect of slowing Exim down by computing (never used) parameters for the
156 RSA_EXPORT functionality.
157
7befa435
PH
158PH/21 On the advice of Timo Sirainen, added a check to the dovecot
159 authenticator to fail if there's a tab character in the incoming data
160 (there should never be unless someone is messing about, as it's supposed
161 to be base64-encoded). Also added, on Timo's advice, the "secured" option
162 if the connection is using TLS or if the remote IP is the same as the
163 local IP, and the "valid-client-cert option" if a client certificate has
164 been verified.
165
48da4259 166PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
16ff981e
PH
167 authenticators. This can be used for authorization after authentication
168 succeeds. (In the case of plaintext, it servers for both authentication
169 and authorization.)
170
48da4259
PH
171PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
172 if any retry times were supplied.
173
d1d5595c
PH
174PH/24 Exim crashed if verify=helo was activated during an incoming -bs
175 connection, where there is no client IP address to check. In this
176 situation, the verify now always succeeds.
177
0ef732d9
PH
178PH/25 Applied John Jetmore's -Mset patch.
179
328895cc
PH
180PH/26 Added -bem to be like -Mset, but loading a message from a file.
181
fd700877
PH
182PH/27 In a string expansion for a processed (not raw) header when multiple
183 headers of the same name were present, leading whitespace was being
184 removed from all of them, but trailing whitespace was being removed only
185 from the last one. Now trailing whitespace is removed from each header
f6c332bd
PH
186 before concatenation. Completely empty headers in a concatenation (as
187 before) are ignored.
fd700877 188
8dce1a6f
PH
189PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
190 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
191
17af4a17
PH
192PH/29 [Removed. This was a change that I later backed out, and forgot to
193 correct the ChangeLog entry (that I had efficiently created) before
194 committing the later change.]
f6c332bd
PH
195
196PH/30 Exim was sometimes attempting to deliver messages that had suffered
197 address errors (4xx response to RCPT) over the same connection as other
198 messages routed to the same hosts. Such deliveries are always "forced",
199 so retry times are not inspected. This resulted in far too many retries
200 for the affected addresses. The effect occurred only when there were more
201 hosts than the hosts_max_try setting in the smtp transport when it had
202 the 4xx errors. Those hosts that it had tried were not added to the list
203 of hosts for which the message was waiting, so if all were tried, there
204 was no problem. Two fixes have been applied:
205
206 (i) If there are any address or message errors in an SMTP delivery, none
207 of the hosts (tried or untried) are now added to the list of hosts
208 for which the message is waiting, so the message should not be a
209 candidate for sending over the same connection that was used for a
210 successful delivery of some other message. This seems entirely
211 reasonable: after all the message is NOT "waiting for some host".
212 This is so "obvious" that I'm not sure why it wasn't done
213 previously. Hope I haven't missed anything, but it can't do any
214 harm, as the worst effect is to miss an optimization.
215
216 (ii) If, despite (i), such a delivery is accidentally attempted, the
217 routing retry time is respected, so at least it doesn't keep
218 hammering the server.
219
c1114884
PH
220PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
221 in ${readsocket because some servers need this prod.
222
7a0743eb
PH
223PH/32 Added some extra debug output when updating a wait-xxx database.
224
944e9e9c 225
478be7b0
SC
226Exim version 4.63
227-----------------
228
229SC/01 Use a glob alias rather than an array ref in eximstats generated
230 parser. This improves both readability and performance.
231
232SC/02 Collect SpamAssassin and rejection statistics in eximstats.
233 Don't display local sender or destination tables in eximstats unless
234 there is data to show.
235 Added average volumes into the eximstats top table text output.
236
237SC/03 Collect data on the number of addresses (recipients) as well
238 as the number of messages in eximstats.
239
2b965a65
TF
240TF/01 Correct an error in the documentation for the redirect router. Exim
241 does (usually) call initgroups() when daemonizing.
478be7b0 242
45b91596
PH
243TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
244 with consistent privilege compared to when running as a daemon.
478be7b0 245
c59f5781
TF
246TF/03 Note in the spec that $authenticated_id is not set for local
247 submissions from trusted users.
248
90fc3069
TF
249TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
250 Thanks to Dean Brooks <dean@iglou.com> for the patch.
251
6083aca0
TF
252TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
253 by adding some example configuration directives to the default
254 configuration file. A little bit of work is required to uncomment the
255 directives and define how usernames and passwords are checked, but
256 there is now a framework to start from.
257
765b530f
PH
258PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
259 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
260 without this. I don't know how relevant this is to other LDAP libraries.
261
4e167a8c
PH
262PH/02 Add the verb name to the "unknown ACL verb" error.
263
4608d683
PH
264PH/03 Magnus Holmgren's patch for filter_prepend_home.
265
b8dc3e4a
PH
266PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
267
5418e93b
PH
268PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
269 directory not expanded when it should be if an expanded home directory
270 was set for the address (which is overridden by the transport).
271
b4a9bda2
PH
272PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
273 libradius.
274
45b91596
PH
275PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
276 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
277 because it is too late at that time, and has no effect.
278
5547e2c5
PH
279PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
280 security issue with \' (bugzilla #107). I could not use the
281 PQescapeStringConn() function, because it needs a PGconn value as one of
282 its arguments.
283
dbcef0ea
PH
284PH/08 When testing addresses using -bt, indicate those final addresses that
285 are duplicates that would not cause an additional delivery. At least one
286 person was confused, thinking that -bt output corresponded to deliveries.
287 (Suppressing duplicates isn't a good idea as you lose the information
288 about possibly different redirections that led to the duplicates.)
289
25257489
PH
290PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
291 systems where poll() doesn't work, in particular OS X.
292
c816d124
PH
293PH/10 Added more information to debugging output for retry time not reached.
294
a9ccd69a
PH
295PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
296 operations in malware.c.
297
75fa1910
PH
298PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
299 signatures.
300
a7d7aa58
PH
301PH/13 If write_rejectlog was set false when logging was sent to syslog with
302 syslog_duplication set false, log lines that would normally be written
303 both the the main log and to the reject log were not written to syslog at
304 all.
305
42119b09
PH
306PH/14 In the default configuration, change the use of "message" in ACL warn
307 statements to "add_header".
308
41609df5
PH
309PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
310 not followed by a command (e.g. "seen endif").
311
a5bd321b
PH
312PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
313 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
314 latter.
315
e85a7ad5 316PH/17 Added extra conditions to the default value of delay_warning_condition
5dff5817
PH
317 so that it is now:
318
e85a7ad5
PH
319 ${if or { \
320 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
321 { match{$h_precedence:}{(?i)bulk|list|junk} } \
322 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
5dff5817
PH
323 }{no}{yes}}
324
e85a7ad5
PH
325 The Auto-Submitted: and various List- headers are standardised, whereas I
326 don't think Precedence: ever was.
5dff5817 327
d8fe1c03
PH
328PH/18 Refactored debugging code in route_finduser() to show more information,
329 in particular, the error code if getpwnam() issues one.
330
16282d2b
PH
331PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
332 This is apparently needed in addition to the PH/07 change above to avoid
333 any possible encoding problems.
334
35d40a98
PH
335PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
336 but not after initializing Perl.
337
034d99ab
PH
338PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
339 output them only if debugging. By default they are written stderr,
340 apparently, which is not desirable.
341
6ec97b1b
PH
342PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
343 queries.
344
e22ca4ac
JJ
345JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
346 --not options
347
348JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
349
33d73e3b
PH
350PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
351 authenticated or an ident call has been made. Suppress the default
352 values for $authenticated_id and $authenticated_sender (but permit -oMai
353 and -oMas) when testing with -bh.
354
9ecb03f3
PH
355PH/24 Re-jigged the order of the tests in the default configuration so that the
356 tests for valid domains and recipients precede the DNS black list and CSA
357 tests, on the grounds that those ones are more expensive.
358
084efe8d
PH
359PH/25 Exim was not testing for a space following SMTP commands such as EHLO
360 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
361 This bug exists in every version of Exim that I still have, right back to
362 0.12.
363
366fc9f0
PH
364PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
365 However, an attempt to turn on case-sensitivity in a regex key by
366 including (?-i) didn't work because the subject string was already
367 lowercased, and the effects were non-intuitive. It turns out that a
368 one-line patch can be used to allow (?-i) to work as expected.
369
c59f5781 370
c887c79e
TF
371Exim version 4.62
372-----------------
373
374TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
375 other effects) broke the use of negated acl sub-conditions.
376
1cce3af8
PH
377PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
378 patch).
379
afb3eaaf
PH
380PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
381 "Deny" causes Exim to reject the incoming connection with a 554 error.
382 Unfortunately, if there is a major crisis, such as a disk failure,
383 tcp-wrappers gives "deny", whereas what one would like would be some
384 kind of temporary error. A kludge has been added to help with this.
385 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
386 554 error is used if errno is still zero or contains ENOENT (which occurs
387 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
388 451 error is used.
389
e173618b
PH
390PH/03 Add -lutil to the default FreeBSD LIBS setting.
391
dd16e114
PH
392PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
393 errors. Otherwise a message that provokes a temporary error (when other
394 messages do not) can cause a whole host to time out.
395
f7fd3850
PH
396PH/05 Batch deliveries by appendfile and pipe transports did not work when the
397 addresses were routed directly to files or pipes from a redirect router.
398 File deliveries just didn't batch; pipe deliveries might have suffered
399 odd errors.
400
d87df92c
PH
401PH/06 A failure to get a lock for a hints database would erroneously always say
402 "Failed to get write lock", even when it was really a read lock.
403
7e9f683d
PH
404PH/07 The appendfile transport was creating MBX lock files with a fixed mode
405 of 0600. This has been changed to use the value of the lockfile_mode
406 option (which defaults to 0600).
407
bfad5236
PH
408PH/08 Applied small patch from the Sieve maintainer.
409
01c490df
PH
410PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
411 folder from quota calculations, a direct delivery into this folder messed
412 up the contents of the maildirsize file. This was because the regex was
413 used only to exclude .Trash (or whatever) when the size of the mailbox
414 was calculated. There was no check that a delivery was happening into an
415 excluded directory. This bug has been fixed by ignoring all quota
416 processing for deliveries into excluded directories.
417
d6629cdc
PH
418PH/10 Added the maildirfolder_create_regex option to appendfile.
419
1cce3af8 420
214e2000
PH
421Exim version 4.61
422-----------------
423
424PH/01 The code for finding all the local interface addresses on a FreeBSD
425 system running IPv6 was broken. This may well have applied to all BSD
426 systems, as well as to others that have similar system calls. The broken
427 code found IPv4 interfaces correctly, but gave incorrect values for the
428 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
429 that it would not match correctly against @[] and not recognize the IPv6
430 addresses as local.
431
f9daeae0
PH
432PH/02 The ipliteral router was not recognizing addresses of the form user@
433 [ipv6:....] because it didn't know about the "ipv6:" prefix.
434
7e66e54d
PH
435PH/03 Added disable_ipv6.
436
c8ea1597
PH
437PH/04 Changed $reply_address to use the raw form of the headers instead of the
438 decoded form, because it is most often used to construct To: headers
439 lines in autoreplies, and the decoded form may well be syntactically
440 invalid. However, $reply_address has leading white space removed, and all
441 newlines turned into spaces so that the autoreply transport does not
442 grumble.
443
911f6fde
PH
444PH/05 If group was specified without a user on a router, and no group or user
445 was specified on a transport, the group from the router was ignored.
446
47ca6d6c
PH
447PH/06 Increased the number of ACL variables to 20 of each type, and arranged
448 for visible compile-time settings that can be used to change these
449 numbers, for those that want even more. Backwards compatibility with old
450 spool files has been maintained. However, going back to a previous Exim
451 release will lost any variables that are in spool files.
452
ed0e9820
PH
453PH/07 Two small changes when running in the test harness: increase delay when
454 passing a TCP/IP connection to a new process, in case the original
455 process has to generate a bounce, and remove special handling of
456 127.0.0.2 (sic), which is no longer necessary.
457
eff37e47
PH
458PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
459 be the same on different OS.
460
1921d2ea
PH
461PH/09 Moved a debug statement in filter processing to avoid a race problem when
462 testing.
463
b3f69ca8
JJ
464JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
465 whether --show-vars was specified or not
466
467JJ/02 exipick: Added support for new ACL variable spool format introduced
468 in 4.61-PH/06
469
424a1c63
PH
470PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
471 syntactically invalid From: or Reply-to: line, and a filter used this to
472 generate an autoreply, and therefore failed to obtain an address for the
473 autoreply, Exim could try to deliver to a non-existent relative file
474 name, causing unrelated and misleading errors. What now happens is that
475 it logs this as a hard delivery error, but does not attempt to create a
476 bounce message.
477
7a100415
PH
478PH/11 The exinext utility has a -C option for testing purposes, but although
479 the given file was scanned by exinext itself; it wasn't being passed on
480 when Exim was called.
481
19b9dc85
PH
482PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
483 an end-of-file indication when reading a command response.
484
309bd837
PH
485PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
486 compiled. In many other places in Exim, IPv6 addresses are always
487 recognized, so I have changed this. It also means that IPv4 domain
488 literals of the form [IPV4:n.n.n.n] are now always recognized.
489
59e82a2a
PH
490PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
491 used if the router is not running as root, for example, when verifying at
492 ACL time, or when using -bh. The debugging output from this situation was
493 non-existent - all you got was a failure to exec. I have made two
494 changes:
495
496 (a) Failures to set uid/gid, the current directory, or a process leader
497 in a subprocess such as that created by queryprogram now generate
498 suitable debugging ouput when -d is set.
499
500 (b) The queryprogram router detects when it is not running as root,
501 outputs suitable debugging information if -d is set, and then runs
502 the subprocess without attempting to change uid/gid.
503
9edc04ce
PH
504PH/15 Minor change to Makefile for building test_host (undocumented testing
505 feature).
506
1349e1e5
PH
507PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
508 additional section of a DNS packet that returns MX or SRV records.
509 Instead, it always explicitly searches for A/AAAA records. This avoids
510 major problems that occur when a DNS server includes only records of one
511 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
512 fixed another bug: if SRV records were looked up and the corresponding
513 address records were *not* found in the additional section, the port
514 values from the SRV records were lost.
515
ea49d0e1
PH
516PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
517 using the correct key (the original address) when searching the retry
518 rules in order to find which one to use for generating the retry hint.
519
064a94c9
PH
520PH/18 If quota_warn_message contains a From: header, Exim now refrains from
521 adding the default one. Similarly, if it contains a Reply-To: header, the
522 errors_reply_to option, if set, is not used.
523
727071f8
PH
524PH/19 When calculating a retry time, Exim used to measure the "time since
525 failure" by looking at the "first failed" field in the retry record. Now
526 it does not use this if it is later than than the arrival time of the
527 message. Instead it uses the arrival time. This makes for better
528 behaviour in cases where some deliveries succeed, thus re-setting the
529 "first failed" field. An example is a quota failure for a huge message
530 when small messages continue to be delivered. Without this change, the
531 "time since failure" will always be short, possible causing more frequent
532 delivery attempts for the huge message than are intended.
dd16e114 533 [Note: This change was subsequently modified - see PH/04 for 4.62.]
727071f8 534
f78eb7c6
PH
535PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
536 $1, $2, $3) because the numerical variables can be reset during some
537 expansion items (e.g. "match"), thereby losing the authentication data.
538
21c28500
PH
539PH/21 Make -bV show the size of off_t variables so that the test suite can
540 decide whether to run tests for quotas > 2G.
541
542PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
543 mailbox_size, and mailbox_filecount in the appendfile transport. If a
544 filecount value is greater than 2G or if a quota value is greater than 2G
545 on a system where the size of off_t is not greater than 4, a panic error
546 is given.
547
1688f43b
PH
548PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
549 never match. The debug and -bh output now contains an explicit error
550 message indicating a malformed IPv4 address or mask.
551
552PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
553 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
554 PH/23 above applies.
555
9675b384
PH
556PH/25 Do not write to syslog when running in the test harness. The only
557 occasion when this arises is a failure to open the main or panic logs
558 (for which there is an explicit test).
559
6a3f1455
PH
560PH/26 Added the /no_tell option to "control=freeze".
561
dac79d3e
PH
562PH/27 If a host name lookup failed very early in a connection, for example, if
563 the IP address matched host_lookup and the reverse lookup yielded a name
564 that did not have a forward lookup, an error message of the form "no IP
565 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
566 could be logged. Now it outputs the IP address instead of "NULL".
1349e1e5 567
5977a0b3
PH
568PH/28 An enabling patch from MH: add new function child_open_exim2() which
569 allows the sender and the authenticated sender to be set when
570 submitting a message from within Exim. Since child_open_exim() is
571 documented for local_scan(), the new function should be too.
572
c91535f3
PH
573PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
574 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
575 results in an empty string is now treated as unset.
576
0d46a8c8
PH
577PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
578
278c6e6c
PH
579PH/31 Added sender_verify_fail logging option.
580
2cbb4081
PH
581PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
582 needed by RFC 822 but not by RFC 2822 was commented out. I have now
583 tidied the source and removed it altogether.
584
3eef829e
PH
585PH/33 When a queue run was abandoned because the load average was too high, a
586 log line was always written; now it is written only if the queue_run log
587 selector is set. In addition, the log line for abandonment now contains
588 information about the queue run such as the pid. This is always present
589 in "start" and "stop" lines but was omitted from the "abandon" line.
590
1ab95fa6
PH
591PH/34 Omit spaces between a header name and the colon in the error message that
592 is given when verify = headers_syntax fails (if there are lots of them,
593 the message gets confusing).
594
230205fc
PH
595PH/35 Change the default for dns_check_names_pattern to allow slashes within
596 names, as there are now some PTR records that contain slashes. This check
597 is only to protect against broken name servers that fall over on strange
598 characters, so the fact that it applies to all lookups doesn't matter.
599
75e0e026
PH
600PH/36 Now that the new test suite is complete, we can remove some of the
601 special code in Exim that was needed for the old test suite. For example,
602 sorting DNS records because real resolvers return them in an arbitrary
603 order. The new test suite's fake resolver always returns records in the
604 same order.
605
606PH/37 When running in the test harness, use -odi for submitted messages (e.g.
607 bounces) except when queue_only is set, to avoid logging races between
608 the different processes.
609
145396a6
PH
610PH/38 Panic-die if .include specifies a non-absolute path.
611
3cd34f13
PH
612PH/39 A tweak to the "H" retry rule from its user.
613
11121d3d
JJ
614JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
615 a label. They prevented compilation on older perls.
616
617JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
618 a warning to be raised on newish perls.
619
620JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
621 on queue. Changes to match documented behaviour of showing count of
622 messages matching specified criteria.
623
8def5aaf
PH
624PH/40 Changed the default ident timeout from 30s to 5s.
625
929ba01c
PH
626PH/41 Added support for the use of login_cap features, on those BSD systems
627 that have them, for controlling the resources used by pipe deliveries.
628
2632889e
PH
629PH/42 The content-scanning code uses fopen() to create files in which to put
630 message data. Previously it was not paying any attention to the mode of
631 the files. Exim runs with umask(0) because the rest of the code creates
632 files with open(), and sets the required mode explicitly. Thus, these
633 files were ending up world-writeable. This was not a big issue, because,
634 being within the spool directory, they were not world-accessible. I have
635 created a function called modefopen, which takes an additional mode
636 argument. It sets umask(777), creates the file, chmods it to the required
637 mode, then resets the umask. All the relevant calls to fopen() in the
638 content scanning code have been changed to use this function.
639
944a9c55
PH
640PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
641 to 24 hours. This avoids potential overflow problems when processing G
642 and H retry rules. I suspect nobody ever tinkers with this value.
643
4a23603b
PH
644PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
645
4730f942
PH
646PH/45 When the plaintext authenticator is running as a client, the server's
647 challenges are checked to ensure they are valid base64 strings. By
648 default, the authentication attempt is cancelled if an invalid string is
649 received. Setting client_ignore_invalid_base64 true ignores these errors.
650 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
651 they are received. Thus, the responses can be made to depend on the
652 challenges. If an invalid string is ignored, an empty string is placed in
653 the variable.
654
30dba1e6
PH
655PH/46 Messages that are created by the autoreply transport now contains a
656 References: header, in accordance with RFCs 2822 and 3834.
657
382afc6b
PH
658PH/47 Added authenticated_sender_force to the smtp transport.
659
a86229cf
PH
660PH/48 The ${prvs expansion was broken on systems where time_t was long long.
661
50c99ba6
PH
662PH/49 Installed latest patch from the Sieve maintainer.
663
d35e429d
PH
664PH/50 When an Exim quota was set without a file count quota, and mailbox_size
665 was also set, the appendfile transport was unnecessarily scanning a
666 directory of message files (e.g. for maildir delivery) to find the count
667 of files (along with the size), even though it did not need this
668 information. It now does the scan only if it needs to find either the
669 size of the count of files.
670
f90d018c
PH
671PH/51 Added ${time_eval: to convert Exim time strings into seconds.
672
75def545
PH
673PH/52 Two bugs concerned with error handling when the smtp transport is
674 used in LMTP mode:
675
676 (i) Exim was not creating retry information for temporary errors given
677 for individual recipients after the DATA command when the smtp transport
678 was used in LMTP mode. This meant that they could be retried too
679 frequently, and not timed out correctly.
680
681 (ii) Exim was setting the flag that allows error details to be returned
682 for LMTP errors on RCPT commands, but not for LMTP errors for individual
683 recipients that were returned after the DATA command.
684
685PH/53 This is related to PH/52, but is more general: for any failing address,
686 when detailed error information was permitted to be returned to the
687 sender, but the error was temporary, then after the final timeout, only
688 "retry timeout exceeded" was returned. Now it returns the full error as
689 well as "retry timeout exceeded".
690
c46782ef
PH
691PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
692 do this, and (what is worse) MTAs that accept it.
693
71fafd95
PH
694PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
695 will now be deprecated.
696
2c5db4fd
PH
697PH/56 New os.c-cygwin from the Cygwin maintainer.
698
9cf6b11a
JJ
699JJ/06 exipick: added --unsorted option to allow unsorted output in all output
700 formats (previously only available in exim formats via -bpr, -bpru,
701 and -bpra. Now also available in native and exiqgrep formats)
702
703JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
704 with very large, slow to parse queues
705
706JJ/08 exipick: added ! as generic prefix to negate any criteria format
707
708JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
709
898d150f
PH
710PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
711 responses to authentication challenges, though it was showing the
712 challenges; (ii) I've removed the CR characters from the debug output for
713 SMTP output lines.
714
46218253
PH
715PH/58 Allow for the insertion of a newline as well as a space when a string
716 is turned into more than one encoded-word during RFC 2047 encoding. The
717 Sieve code now uses this.
718
e97957bc
PH
719PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
720 data_4xx, lost_connection, tls_required.
721
81e509d7
PH
722PH/60 When a VRFY deferred or FAILED, the log message rather than the user
723 message was being sent as an SMTP response.
724
3d240ff7
PH
725PH/61 Add -l and -k options to exicyclog.
726
b37c4101
PH
727PH/62 When verifying, if an address was redirected to one new address, so that
728 verification continued, and the new address failed or deferred after
729 having set something in $address_data, the value of $address_data was not
730 passed back to the ACL. This was different to the case when no
731 redirection occurred. The value is now passed back in both cases.
732
79378e0f
PH
733PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
734 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
735 use login_cap.h, so on its own it isn't the distinguishing feature. The
736 new name refers directly to the setclassresources() function.
737
e49c7bb4
PH
738PH/65 Added configuration files for NetBSD3.
739
d114ec46
PH
740PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
741
f3d7df6c
PH
742PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
743 is preferred over IPv4.
744
715ab376
PH
745PH/68 The bounce_return_message and bounce_return_body options were not being
746 honoured for bounces generated during the reception of non-SMTP messages.
747 In particular, this applied to messages rejected by the ACL. This bug has
748 been fixed. However, if bounce_return_message is true and bounce_return_
749 body is false, the headers that are returned for a non-SMTP message
750 include only those that have been read before the error was detected.
751 (In the case of an ACL rejection, they have all been read.)
752
6b31b150
PH
753PH/69 The HTML version of the specification is now built in a directory called
754 spec_html instead of spec.html, because the latter looks like a path with
755 a MIME-type, and this confuses some software.
756
757PH/70 Catch two compiler warnings in sieve.c.
758
d515a917
PH
759PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
760 function verify_get_ident() calls ip_connect() to connect a socket, but
761 if the "connect()" function timed out, ip_connect() used to close the
762 socket. However, verify_get_ident() also closes the socket later, and in
763 between Exim writes to the log, which may get opened at this point. When
764 the socket was closed in ip_connect(), the log could get the same file
765 descriptor number as the socket. This naturally causes chaos. The fix is
766 not to close the socket in ip_connect(); the socket should be closed by
767 the function that creates it. There was only one place in the code where
768 this was missing, in the iplookup router, which I don't think anybody now
769 uses, but I've fixed it anyway.
770
9b8fadde
PH
771PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
772 well as to direct DNS lookups. Otherwise the handling of names in host
773 lists is inconsistent and therefore confusing.
774
214e2000 775
5de37277
PH
776Exim version 4.60
777-----------------
778
cc38ddbf
PH
779PH/01 Two changes to the default runtime configuration:
780
781 (1) Move the checks for relay_from_hosts and authenticated clients from
782 after to before the (commented out) DNS black list checks.
783
784 (2) Add control=submission to the relay_from_hosts and authenticated
785 clients checks, on the grounds that messages accepted by these
786 statements are most likely to be submissions.
5de37277 787
72fdd6ae
PH
788PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
789
790 (1) Generate an error if the third argument for the ${prvs expansion is
791 not a single digit.
792
793 (2) Treat a missing third argument of ${prvscheck as if it were an empty
794 string.
795
796 (3) Reset the variables that are obtained from the first argument of
797 ${prvscheck and used in the second argument before leaving the code,
798 because their memory is reclaimed, so using them afterwards may do
799 silly things.
800
801 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
802 one (it's much easier than Tom thought :-).
803
804 (5) Because of (4), we can now allow for the use of $prvscheck_result
805 inside the third argument.
cb9328de 806
cb741023
PH
807PH/03 For some reason, the default setting of PATH when running a command from
808 a pipe transport was just "/usr/bin". I have changed it to
809 "/bin:/usr/bin".
810
f174f16e
PH
811PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
812 anything to be listed in the output from -bV.
b2f5a032 813
c25242d7
PH
814PH/05 When a filter generated an autoreply, the entire To: header line was
815 quoted in the delivery log line, like this:
816
817 => >A.N.Other <ano@some.domain> <original@ddress> ...
818
819 This has been changed so that it extracts the operative address. There
820 may be more than one such address. If so, they are comma-separated, like
821 this:
822
823 => >ano@some.domain,ona@other.domain <original@ddress> ...
824
82c19f95
PH
825PH/06 When a client host used a correct literal IP address in a HELO or EHLO
826 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
827 not being looked up in the rDNS to get a host name, Exim was showing the
828 IP address twice in Received: lines, even though the IP addresses were
829 identical. For example:
830
831 Received: from [1.2.3.4] (helo=[1.2.3.4])
832
833 However, if the real host name was known, it was omitting the HELO data
834 if it matched the actual IP address. This has been tidied up so that it
835 doesn't show the same IP address twice.
836
d7ffbc12
PH
837PH/07 When both +timestamp and +memory debugging was on, the value given by
838 $tod_xxx expansions could be wrong, because the tod_stamp() function was
839 called by the debug printing, thereby overwriting the timestamp buffer.
840 Debugging no longer uses the tod_stamp() function when +timestamp is set.
841
9f526266
PH
842PH/08 When the original message was included in an autoreply transport, it
843 always said "this is a copy of the message, including all the headers",
844 even if body_only or headers_only was set. It now gives an appropriate
845 message.
846
87fcc8b9
PH
847PH/09 Applied a patch from the Sieve maintainer which:
848
849 o fixes some comments
850 o adds the (disabled) notify extension core
851 o adds some debug output for the result of if/elsif tests
852 o points to the current vacation draft in the documentation
853 and documents the missing references header update
854
855 and most important:
856
857 o fixes a bug in processing the envelope test (when testing
858 multiple envelope elements, the last element determinted the
859 result)
860
456682f5
PH
861PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
862 Electronic Mail") by including:
863
864 Auto-submitted: auto-generated
865
866 in the messages that it generates (bounce messages and others, such as
867 warnings). In the case of bounce messages for non-SMTP mesages, there was
868 also a typo: it was using "Auto_submitted" (underscore instead of
869 hyphen). Since every message generated by Exim is necessarily in response
870 to another message, thes have all been changed to:
871
872 Auto-Submitted: auto-replied
873
874 in accordance with these statements in the RFC:
875
876 The auto-replied keyword:
877
878 - SHOULD be used on messages sent in direct response to another
879 message by an automatic process,
880
881 - MUST NOT be used on manually-generated messages,
882
883 - MAY be used on Delivery Status Notifications (DSNs) and Message
884 Disposition Notifications (MDNs),
885
886 - MUST NOT be used on messages generated by automatic or periodic
887 processes, except for messages which are automatic responses to
888 other messages.
889
3e46c1aa
PH
890PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
891 to the default Received: header definition.
456682f5 892
49826d12
PH
893PH/12 Added log selector acl_warn_skipped (default on).
894
eba0c039
PH
895PH/13 After a successful wildlsearch lookup, discard the values of numeric
896 variables because (a) they are in the wrong storage pool and (b) even if
897 they were copied, it wouldn't work properly because of the caching.
898
a0d6ba8a
PH
899PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
900 checking when decoding. Apparently there are clients that generate
901 overlong encoded strings. Why am I not surprised?
902
f0917727
PH
903PH/15 If the first argument of "${if match_address" was not empty, but did not
904 contain an "@" character, Exim crashed. Now it writes a panic log message
905 and treats the condition as false.
906
096fee00
PH
907PH/16 In autoreply, treat an empty string for "once" the same as unset.
908
024bd3c2
PH
909PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
910 extension "envelope-auth". The code is finished and in agreement with
911 other implementations, but there is no documentation so far and in fact,
912 nobody wrote the draft yet. This extension is currently #undef'ed, thus
913 not changing the active code.
914
915 Print executed "if" and "elsif" statements when debugging is used. This
916 helps a great deal to understand what a filter does.
917
918 Document more things not specified clearly in RFC3028. I had all this
919 sorted out, when out of a sudden new issues came to my mind. Oops."
920
df199fec
PH
921PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
922 (Bugzilla #53).
923
d27f1df3
PH
924PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
925 canonical form (as documented). However, after a host name lookup from
926 the IP address, check_host() was doing a simple string comparison with
927 addresses acquired from the DNS when checking that the found name did
928 have the original IP as one of its addresses. Since any found IPv6
929 addresses are likely to be in abbreviated form, the comparison could
930 fail. Luckily, there already exists a function for doing the comparison
931 by converting both addresses to binary, so now that is used instead of
932 the text comparison.
933
96776534
PH
934PH/20 There was another similar case to PH/19, when a complete host name was
935 given in a host list; looking up its IP address could give an abbreviated
936 form, whereas the current host's name might or might not be abbreviated.
937 The same fix has been applied.
938
5de37277 939
9a799bc0
PH
940Exim version 4.54
941-----------------
942
943PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
944 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
945 It now does.
946
99a4b039
PH
947PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
948 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
949
950PH/03 Typo: missing ".o" in src/pcre/Makefile.
951
4b233853
PH
952PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
953 header line, restrict the check to what is listed in RFCs 2369 and 2929.
954 Also, for "Auto-Submitted", treat anything other than "no" as
955 non-personal, in accordance with RFC 3834. (Previously it treated
956 anything starting "auto-" as non-personal.)
957
8857ccfd
PH
958TF/01 The control=submission/name=... option had a problem with syntax
959 errors if the name included a slash character. The /name= option
960 now slurps the rest of the string, so it can include any characters
961 but it must come last in the list of options (after /sender_retain
962 or /domain=).
963
433a2980
PH
964PH/05 Some modifications to the interface to the fake nameserver for the new
965 testing suite.
966
3e46c1aa 967
9a799bc0 968
e3a311ba
TK
969Exim version 4.53
970-----------------
971
972TK/01 Added the "success_on_redirect" address verification option. See
973 NewStuff for rationale and an example.
974
13b685f9
PH
975PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
976
395ff96d
PH
977PH/02 Patch to exigrep to allow it to work on syslog lines.
978
5b68f6e4
PH
979PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
980 fread() to skip over the body file's header line, because in Cygwin the
981 header line is locked and is inaccessible.
982
1ab52c69
PH
983PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
984 co-exist for some time) to make it clear that it is the Exim ID that is
985 referenced, not the Message-ID: header line.
986
b07e6aa3
PH
987PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
988 string_format() function, because snprintf() does not exist on all
989 operating systems.
990
254e032f
PH
991PH/06 The use of forbid_filter_existstest now also locks out the use of the
992 ${stat: expansion item.
993
3af76a81
PH
994PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
995 protocol synchronization error", to keep the pedants happy.
996
2548ba04
PH
997PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
998 well as for IRIX systems, when gcc is being used. See the host.c source
999 file for comments.
1000
b6c6011d
PH
1001PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1002
cf39cf57
PH
1003PH/10 Named domain lists were not working if used in a queue_smtp_domains
1004 setting.
1005
f1513293
PH
1006PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1007 transport and to the smtp transport in LMTP mode.
1008
727549a4
PH
1009TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1010
af46795e
PH
1011PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1012 run a filter in a subprocess. This could lead to confusion in subsequent
1013 lookups in the parent process. There should also be a search_tidyup() at
1014 the end of the subprocess.
1015
d7b47fd0
PH
1016PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1017 only if the host matched helo_try_verify_hosts, which caused the
1018 verification to occur when the EHLO/HELO command was issued. The ACL just
1019 tested the remembered result. Now, if a previous verification attempt has
1020 not happened, "verify = helo" does it there and then.
1021
ee744174
JJ
1022JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1023
b582ab87
PH
1024TK/03 Fix log output including CR from clamd.
1025
41a13e0a
PH
1026PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1027 exist provoked a memory error which could cause a segfault.
1028
f625cc5a
PH
1029PH/15 Installed PCRE 6.2
1030
1031PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1032
21f7af35
PH
1033PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1034 of the problem. Specifically, suggested +O2 rather than +O1 for the
1035 HP-UX compiler.
1036
31480e42
PH
1037PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1038
2d280592
PH
1039PH/20 If a delivery was routed to a non-standard port by means of an SRV
1040 record, the port was not correctly logged when the outgoing_port log
1041 selector was set (it logged the transort's default port).
1042
7cd1141b
PH
1043PH/21 Added support for host-specific ports to manualroute, queryprogram,
1044 fallback_hosts, and "hosts" in the smtp transport.
1045
1046PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1047 host errors such as "Connection refused".
1048
750af86e
PH
1049PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1050 authentication with radiusclient 0.4.9:
1051
1052 - Error returned from rc_read_config was caught wrongly
1053 - Username/password not passed on to radius server due to wrong length.
1054
1055 The presumption is that some radiusclient API changes for 4.51/PH/17
1056 were not taken care of correctly. The code is still untested by me (my
1057 Linux distribution still has 0.3.2 of radiusclient), but it was
1058 contributed by a Radius user.
1059
1060PH/24 When doing a callout, the value of $domain wasn't set correctly when
1061 expanding the "port" option of the smtp transport.
1062
4304270b
TK
1063TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1064 while reading a MIME header. Thanks to Tom Hughes for a patch.
1065
750af86e
PH
1066PH/24 Include config.h inside local_scan.h so that configuration settings are
1067 available.
1068
64ffc24f
PH
1069PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1070 that in an ACL for RCPT (for example), you can examine exactly what was
1071 received.
1072
5dd9625b
PH
1073PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1074 commands, but it was not correctly comparing the address with the actual
1075 client host address. Thus, it would show the EHLO address in Received:
1076 header lines when this was not necessary.
1077
5591031b
PH
1078PH/27 Added the % operator to ${eval:}.
1079
ba18e66a
PH
1080PH/28 Exim tries to create and chdir to its spool directory when it starts;
1081 it should be ignoring failures (because with -C, for example, it has lost
1082 privilege). It wasn't ignoring creation failures other than "already
1083 exists".
1084
9cec981f
PH
1085PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1086 -bV or -d is used.
1087
aa2b5c79
PH
1088PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1089 because an input line was too long, either on its own, or by virtue of
1509d3a8
PH
1090 too many continuations, the temporary file was not being removed, and the
1091 return code was incorrect.
aa2b5c79 1092
48a53b7f
PH
1093PH/31 Missing "BOOL" in function definition in filtertest.c.
1094
1c59d63b
PH
1095PH/32 Applied Sieve patches from the maintainer.
1096
671012da
TK
1097TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1098
1509d3a8
PH
1099PH/33 Added "verify = not_blind".
1100
1101PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1102 Local/Makefile (with some defaults set). These are used in built scripts
1103 such as exicyclog, but they have never been used in the exim_install
1104 script (though there are many overriding facilities there). I have
1105 arranged that the exim_install script now takes note of these two
1106 settings.
1107
1108PH/35 Installed configuration files for Dragonfly.
1109
2fe1a124
PH
1110PH/36 When a locally submitted message by a trusted user did not contain a
1111 From: header, and the sender address was obtained from -f or from an SMTP
1112 MAIL command, and the trusted user did not use -F to supply a sender
1113 name, $originator_name was incorrectly used when constructing a From:
1114 header. Furthermore, $originator_name was used for submission mode
1115 messages from external hosts without From: headers in a similar way,
1116 which is clearly wrong.
1117
8800895a
PH
1118PH/37 Added control=suppress_local_fixups.
1119
ccfdb010
PH
1120PH/38 When log_selector = +received_sender was set, and the addition of the
1121 sender made the log line's construction buffer exactly full, or one byte
1122 less than full, an overflow happened when the terminating "\n" was
1123 subsequently added.
1124
1130bfb0
PH
1125PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1126 when the result of a list match is failure because a DNS lookup failed.
1127
ebcb507f
PH
1128PH/40 RM_COMMAND is now used in the building process.
1129
c35e155c
PH
1130PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1131 the "build-* directories that it finds.
1132
95d1f782
PH
1133PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1134 address in a domain literal was a prefix of an interface address.
1135
fd6de02e
PH
1136PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1137 when verifying a sender address, unless rewrite_headers is false.
1138
58de37c5
PH
1139PH/44 Wrote a long comment about why errors_to addresses are verified as
1140 recipients, not senders.
1141
261cf466
TF
1142TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1143 the ratelimit ACL was added.
1144
3ee512ff
PH
1145PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1146
e08c430f
PH
1147PH/46 Added extra information about PostgreSQL errors to the error string.
1148
bef5a11f
PH
1149PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1150 avoiding the need to install special zones in a real server. This is
1151 backwards compatible; if it can't find the fake resolver, it drops back.
1152 Thus, both old and new test suites can be run.
1153
7546de58
TF
1154TF/02 Added util/ratelimit.pl
1155
e5d5a95f
TF
1156TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1157 clock is set back in time.
1158
2e88a017
TF
1159TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1160 Candler <B.Candler@pobox.com>.
1161
a5f65aa4
TF
1162TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1163 OK for addresses that are the result of redirections.
1164
e7726cbf
PH
1165PH/48 A number of further additions for the benefit of the new test suite,
1166 including a fake gethostbyname() that interfaces to the fake DNS resolver
1167 (see PH/47 above).
1168
a7fdad5b
TF
1169TF/06 The fix for widen_domains has also been applied to qualify_single and
1170 search_parents which are the other dnslookup options that can cause
1171 header rewrites.
1172
6af56900
PH
1173PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1174 type ("H").
1175
0925ede6
PH
1176PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1177
66afa403
TF
1178TF/07 Exim produced the error message "an SRV record indicated no SMTP
1179 service" if it encountered an MX record with an empty target hostname.
1180 The message is now "an MX or SRV record indicated no SMTP service".
1181
0154e85a
TF
1182TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1183 if the DNS of the sending site is misconfigured. This is quite a
1184 common situation. This change restores the behaviour of treating a
1185 helo verification defer as a failure.
1186
16f12c76
PH
1187PH/51 If self=fail was set on a router, the bounce message did not include the
1188 actual error message.
1189
bbe902f0 1190
e5a9dba6
PH
1191Exim version 4.52
1192-----------------
1193
1194TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1195
22c3b60b
PH
1196PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1197 command itself ended in error, the underlying message about the transport
1198 filter timeout was being overwritten with the pipe command error. Now the
1199 underlying error message should be appended to the second error message.
1200
06a9b4b5
PH
1201TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1202
c1ac6996
PH
1203PH/02 Reduce the amount of output that "make" produces by default. Full output
1204 can still be requested.
1205
9c7a242c
PH
1206PH/03 The warning log line about a condition test deferring for a "warn" verb
1207 was being output only once per connection, rather than after each
1208 occurrence (because it was using the same function as for successful
1209 "warn" verbs). This seems wrong, so I have changed it.
1210
87ba3f5f
PH
1211TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1212 it should not have, which might have caused a crash in the right
1213 circumstances, but probably never did.
1214
1215PH/04 Installed a modified version of Tony Finch's patch to make submission
1216 mode fix the return path as well as the Sender: header line, and to
1217 add a /name= option so that you can make the user's friendly name appear
1218 in the header line.
1219
29aba418
TF
1220TF/03 Added the control = fakedefer ACL modifier.
1221
fe0dab11
TF
1222TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1223 Mark Lowes for thorough testing.
870f6ba8 1224
11d337a4
TK
1225TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1226
1227TK/03 Merged latest SRS patch from Miles Wilton.
1228
415c8f3b
PH
1229PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1230 with the definition in sysexits.h (which is #included earlier).
1231 Fortunately, Exim does not actually use EX_OK. The code used to try to
1232 preserve the sysexits.h value, by assumimg that macro definitions were
1233 scanned for macro replacements. I have been disabused of this notion,
1234 so now the code just undefines EX_OK before #including unistd.h.
11d337a4 1235
958541e9
PH
1236PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1237 in the smtp transport. When a block could not be written in a single
1238 write() function, the timeout was being re-applied to each part-write.
1239 This seems wrong - if the receiver was accepting one byte at a time it
1240 would take for ever. The timeout is now adjusted when this happens. It
1241 doesn't have to be particularly precise.
1242
c206415f
TK
1243TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1244 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1245
2a4be8f9
PH
1246PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1247 without a domain if the check to <postmaster@domain> fails.
1248
1cba11c5
SC
1249SC/01 Eximstats: added -xls and the ability to specify output files
1250 (patch written by Frank Heydlauf).
1251
1252SC/02 Eximstats: use FileHandles for outputing results.
1253
1254SC/03 Eximstats: allow any combination of xls, txt, and html output.
1255
1256SC/04 Eximstats: fixed display of large numbers with -nvr option
1257
1258SC/05 Eximstats: fixed merging of reports with empty tables.
1259
1260SC/06 Eximstats: added the -include_original_destination flag
1261
1262SC/07 Eximstats: removed tabs and trailing whitespace.
1263
1005d00e
TK
1264TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1265
1266TK/06 MBOX spool code: Add real "From " MBOX separator line
1267 so the .eml file is really in mbox format (even though
1268 most programs do not really care). Patch from Alex Miller.
1269
1270TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1271 The latter is generated from $received_to and is only set if the
1272 message has one envelope recipient. SA can use these headers,
1273 obviously out-of-the-box. Patch from Alex Miller.
1274
9b4768fa
PH
1275PH/08 The ${def test on a variable was returning false if the variable's
1276 value was "0", contrary to what the specification has always said!
1277 The result should be true unless the variable is empty.
1278
1279PH/09 The syntax error of a character other than { following "${if
1280 def:variable_name" (after optional whitespace) was not being diagnosed.
1281 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1282 accidental colon was present, for example, could give incorrect results.
1283
0d7eb84a
PH
1284PH/10 Tidied the code in a number of places where the st_size field of a stat()
1285 result is used (not including appendfile, where other changes are about
1286 to be made).
1287
1288PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1289 This involved changing a lot of size variables from int to off_t. It
1290 should work with maildirs and everything.
1291
40727bee
TK
1292TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1293 spamd dying while we are connected to it.
1294
554d2369
TF
1295TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1296 <jgh@wizmail.org>
1297
1f922db1
PH
1298PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1299 function for BDB 4.3.
1300
ef213c3b
PH
1301PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1302
8ac170f3
PH
1303PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1304 the sources for PCRE have been re-arranged and more files are now
1305 involved.
1306
b1c749bb
PH
1307PH/15 The code I had for printing potentially long long variables in PH/11
1308 above was not the best (it lost precision). The length of off_t variables
1309 is now inspected at build time, and an appropriate printing format (%ld
c6c2dc1d
PH
1310 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1311 to be "long long int" or "long int". This is needed for the internal
1312 formatting function string_vformat().
b1c749bb 1313
4aac9b49
PH
1314PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1315 the configuration file to be ":syslog", then the script "guesses" where
1316 the logs files are, rather than using the compiled in default. In our
1317 case the guess is not the same as the compiled default, so the script
1318 suddenly stopped working when I started to use syslog. The patch checks
1319 to see if log_file_path is "". If so, it attempts to read it from exim
1320 with no configuration file to get the compiled in version, before it
1321 falls back to the previous guessing code."
1322
294520c8
TK
1323TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1324 implementing BATV in an Exim configuration. See NewStuff for the gory
1325 details.
1326
5bd022fe
PH
1327PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1328 Makefile that are specific to HP-UX.
1329
90e9ce59
PH
1330PH/18 If the "use_postmaster" option was set for a recipient callout together
1331 with the "random" option, the postmaster address was used as the MAIL
1332 FROM address for the random test, but not for the subsequent recipient
1333 test. It is now used for both.
1334
5ea81592
PH
1335PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1336 patch removes a few documentation additions to RFC 3028, because the
1337 latest draft now contains them. It adds the new en;ascii-case comparator
1338 and a new error check for 8bit text in MIME parts. Comparator and
1339 require names are now matched exactly. I enabled the subaddress
1340 extension, but it is not well tested yet (read: it works for me)."
1341
c6c2dc1d
PH
1342PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1343 rework some of the code of TK/09 above to avoid the hardwired use of
1344 "%lld" and "long long". Replaced the call to snprintf() with a call to
1345 string_vformat().
1346
fffffe4c
PH
1347PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1348 records point to non-existent hosts", "retry timeout exceeded", and
1349 "retry time not reached for any host after a long failure period".
ca02eafb 1350
9a26b6b2
PH
1351PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1352 experimental DomainKeys support:
1353
1354 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1355 (2) On an error such as an illegally used "control", the wrong name for
1356 the control was given.
1357
1358 These problems did NOT occur unless DomainKeys support was compiled.
1359
4aee0225
PH
1360PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1361
32d668a5
PH
1362PH/24 Added ${if match_ip condition.
1363
8187c3f3
PH
1364PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1365 databases so that it will be absolutely obvious if a crash occurs in the
1366 DB library. This is a regular occurrence (often caused by mis-matched
1367 db.h files).
1368
ff790e47 1369PH/26 Insert a lot of missing (void) casts for functions such as chown(),
f1e894f3
PH
1370 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1371 picked up on a user's system that detects such things. There doesn't seem
1372 to be a gcc warning option for this - only an attribute that has to be
1373 put on the function's prototype. It seems that in Fedora Core 4 they have
1374 set this on a number of new functions. No doubt there will be more in due
1375 course.
ff790e47 1376
5417f6d1
PH
1377PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1378 specify a transport. However, if an address that was verified by such a
1379 router was the subject of a callout, Exim crashed because it tried to
1380 read the rcpt_include_affixes from the non-existent transport. Now it
1381 just assumes that the setting of that option is false. This bug was
1382 introduced by 4.51/PH/31.
1383
59cf8544
PH
1384PH/28 Changed -d+all to exclude +memory, because that information is very
1385 rarely of interest, but it makes the output a lot bigger. People tend to
1386 do -d+all out of habit.
1387
e7ad8a65
PH
1388PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1389 code in os-type was giving problems when libc.so lives in lib64, like on
1390 x86_64 Fedora Core.
1391
ade42478
PH
1392PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1393 aren't the modern standard, and it seems that some systems' include files
1394 don't always have them. Exim was already checking for some of the newer
1395 ones like T_AAAA, and defining it itself. I've added checks for all the
1396 record types that Exim uses.
1397
182ad5cf
PH
1398PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1399 not automatically generating a new one, as it is supposed to. This
1400 prevented TLS from working. If the file did exist, but contained invalid
1401 data, a new version was generated, as expected. It was only the case of a
1402 non-existent file that was broken.
1403
b0d9fc80
TK
1404TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1405 with a change in libdomainkeys > 0.64.
1406
1407TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1408 from DNS. If the selector record carries the flag, it now has
1409 precedence over the domain-wide flag.
1410
1411TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1412
47c7a64a
PH
1413PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1414 the use of a port name that isn't defined in /etc/services) occurred, the
1415 message was deferred as in a normal delivery, and thus remained on the
1416 spool, instead of being failed because of the mua_wrapper setting. This
1417 is now fixed, and I tidied up some of the mua_wrapper messages at the
1418 same time.
1419
a388bce4
SC
1420SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1421 the messages in a hash of arrays rather than using individual hashes.
1422 This is a bit cleaner and results in dramatic memory savings, albeit
1423 at a slight CPU cost.
1424
1425SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1426 as requested by Marc Sherman.
1427
1428SC/10 Eximstats: added histograms for user specified patterns as requested
1429 by Marc Sherman.
1430
0793e4ed
SC
1431SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1432
c58b88df
PH
1433PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1434 fopen() in the content-scanning modules that did not already have it.
1435
e7ad8a65 1436
7982096b
PH
1437Exim version 4.51
1438-----------------
1439
1a46a8c5
PH
1440TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1441 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1442
2f079f46 1443TK/02 Fix ACL "control" statement not being available in MIME ACL.
1a46a8c5
PH
1444
1445TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1446
1447PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1448 to test Sieve filters that use "vacation".
1449
1450PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1451 that changes the way the GnuTLS parameters are stored in the cache file.
1452 The new format can be generated externally. For backward compatibility,
1453 if the data in the cache doesn't make sense, Exim assumes it has read an
1454 old-format file, and it generates new data and writes a new file. This
1455 means that you can't go back to an older release without removing the
1456 file.
1457
1458PH/03 A redirect router that has both "unseen" and "one_time" set does not
1459 work if there are any delivery delays because "one_time" forces the
1460 parent to be marked "delivered", so its unseen clone is never tried
1461 again. For this reason, Exim now forbids the simultaneous setting of
1462 these two options.
1463
1464PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1465 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1466 entry if you want to know the details. The fix, however, neglected to
1467 consider the case when local delivery batching is involved. The test for
1468 "previously delivered" was not happening when checking to see if an
1469 address could be batched with a previous (undelivered) one; under
1470 certain circumstances this could lead to multiple deliveries to the same
c2c19e9d 1471 address.
1a46a8c5
PH
1472
1473PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1474 in its include files, and this causes problems building Exim.
1475
1476PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1477 header_syntax) but Exim was just ignoring anything given after a slash.
1478 In particular, this caused confusion with an attempt to use "verify =
1479 reverse_host_lookup/defer_ok". An error is now given when options are
1480 supplied for verify items that do not have them. (Maybe reverse_host_
1481 lookup should have a defer_ok option, but that's a different point.)
1482
1483PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1484 defined by RFC 821) to 2048, because there were problems with some AUTH
1485 commands, and RFC 1869 says the size should be increased for extended
1486 SMTP commands that take arguments.
1487
1488PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1489 Finch).
1490
1491PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1492 "unknown" error; now it says that the functionality isn't in the binary.
8d67ada3 1493
49c2d5ea
PH
1494PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1495 an address' error message when a string expansion fails (syntax or
f331f3b6
PH
1496 whatever). Otherwise the password may appear in the log. Following change
1497 PH/42 below, there is no longer a chance of it appearing in a bounce
1498 message.
49c2d5ea 1499
bf759a8b
PH
1500PH/11 Installed exipick version 20050225.0 from John Jetmore.
1501
83364d30
PH
1502PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1503 of its addresses was ever tried. (Bugzilla bug #2.)
1504
7999bbd7
PH
1505PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1506 the result incorrectly in the debug output. (It correctly added a newline
1507 to what was transported.)
1508
7dbf77c9
PH
1509TF/01 Added $received_time.
1510
74e0617f
PH
1511PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1512 commented out examples of how to interface to a virus scanner and to
1513 SpamAssassin. Also added commented examples of av_scanner and
1514 spamd_address settings.
1515
2f079f46
PH
1516PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1517 and controls are allowed in which ACLs. There were a couple of minor
1518 errors. Some of the entries in the conditions table (which is a table of
1519 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1520 negation of where the condition IS allowed.
1521
8c841523
PH
1522PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1523
7766a4f0
PH
1524PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1525 header file does not have a version number, so I've had to invent a new
1526 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1527 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1528 radiusclient), but it was contributed by a Radius user.
1529
8b417f2c
PH
1530PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1531 files or directories, for OpenSSL.
1532
901f42cb
PH
1533PH/19 When an Exim process that is running as root has to create an Exim log
1534 file, it does so in a subprocess that runs as exim:exim so as to get the
1535 ownership right at creation (otherwise, other Exim processes might see
1536 the file with the wrong ownership). There was no test for failure of this
1537 fork() call, which would lead to the process getting stuck as it waited
1538 for a non-existent subprocess. Forks do occasionally fail when resources
1539 run out. I reviewed all the other calls to fork(); they all seem to check
1540 for failure.
1541
f9b9210e
PH
1542PH/20 When checking for unexpected SMTP input at connect time (before writing
1543 the banner), Exim was not dealing correctly with a non-positive return
1544 from the read() function. If the client had disconnected by this time,
1545 the result was a log entry for a synchronization error with an empty
1546 string after "input=" when read() returned zero. If read() returned -1
1547 (an event I could not check), uninitialized data bytes were printed.
1548 There were reports of junk text (parts of files, etc) appearing after
1549 "input=".
1550
54cdb463
PH
1551PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1552
cf00dad6
PH
1553PH/22 Added support for macro redefinition, and (re)definition in between
1554 driver and ACL definitions.
1555
acb1b346
PH
1556PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1557 forgetting to use the resulting value; it was using the unexpanded value.
1558
c5ddb310
PH
1559PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1560 hadn't been configured. The fix is from Juergen Kreileder, who
1561 understands it better than I do:
1562
1563 "Here's what I see happening with three configured cyrus_sasl
1564 authenticators configured (plain, login, cram-md5):
1565
1566 On startup auth_cyrus_sasl_init() gets called for each of these.
1567 This means three calls to sasl_listmech() without a specified mech_list.
1568 => SASL tests which mechs of all available mechs actually work
1569 => three warnings about OTP not working
1570 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1571
1572 With the patch, sasl_listmech() also gets called three times. But now
1573 SASL's mech_list option is set to the server_mech specified in the the
1574 authenticator. Or in other words, the answer from sasl_listmech()
1575 gets limited to just the mech you're testing for (which is different
1576 for each call.)
1577 => the return list contains just 'plain' or 'login', 'cram-md5' or
1578 nothing depending on the value of ob->server_mech.
1579
1580 I've just tested the patch: Authentication still works fine,
1581 unavailable mechs specified in the exim configuration are still
1582 caught, and the auth.log warnings about OTP are gone."
1583
31619da6
PH
1584PH/25 When debugging is enabled, the contents of the command line are added
1585 to the debugging output, even when log_selector=+arguments is not
1586 specified.
1587
bebaf0fc
PH
1588PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1589 answer is "GNU", and only if the return is "GNU/something" is the answer
1590 "Linux".
1591
475fe28a
PH
1592PH/27 $acl_verify_message is now set immediately after the failure of a
1593 verification in an ACL, and so is available in subsequent modifiers. In
1594 particular, the message can be preserved by coding like this:
1595
1596 warn !verify = sender
1597 set acl_m0 = $acl_verify_message
1598
1599 Previously, $acl_verify_message was set only while expanding "message"
1600 and "log_message" when a very denied access.
1601
7e8bec7a
PH
1602PH/28 Modified OS/os.c-Linux with
1603
1604 -#ifndef OS_LOAD_AVERAGE
1605 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1606
1607 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1608 nomenclature these days.)
1609
e4a89c47
PH
1610PH/29 Installed patch from the Sieve maintainer that adds the options
1611 sieve_useraddress and sieve_subaddress to the redirect router.
1612
5ca2a9a1
PH
1613PH/30 In these circumstances:
1614 . Two addresses routed to the same list of hosts;
1615 . First host does not offer TLS;
1616 . First host accepts first address;
1617 . First host gives temporary error to second address;
1618 . Second host offers TLS and a TLS session is established;
1619 . Second host accepts second address.
1620 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1621 and peerdn, if requested) that were in fact used only for the second
1622 address.
7e8bec7a 1623
c688b954
PH
1624PH/31 When doing a callout as part of verifying an address, Exim was not paying
1625 attention to any local part prefix or suffix that was matched by the
1626 router that accepted the address. It now behaves in the same way as it
1627 does for delivery: the affixes are removed from the local part unless
1628 rcpt_include_affixes is set on the transport.
1629
fed77020
PH
1630PH/32 Add the sender address, as F=<...>, to the log line when logging a
1631 timeout during the DATA phase of an incoming message.
1632
7fe1560f
PH
1633PH/33 Sieve envelope tests were broken for match types other than :is. I have
1634 applied a patch sanctioned by the Sieve maintainer.
c688b954 1635
ebb6e6d5
PH
1636PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1637 the uid or gid is negative. A case of a negative gid caused this to be
1638 noticed. The fix allows for either to be negative.
1639
9c4e8f60
PH
1640PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1641 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1642 overlooked.
1643
1644PH/36 The change PH/12 above was broken. Fixed it.
1645
d7174846
PH
1646PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1647 the grounds that routing the same address twice would always produce the
1648 same answer. This might have been true once, but it is certainly no
1649 longer true now. Routing a child address may depend on the previous
1650 routing that produced that child. Some complicated redirection strategies
1651 went wrong when messages had multiple recipients, and made Exim's
1652 behaviour dependent on the order in which the addresses were given.
1653
1654 I have moved the duplicate checking until after the routing is complete.
1655 Exim scans the addresses that are assigned to local and remote
1656 transports, and removes any duplicates. This means that more work will be
1657 done, as duplicates will always all be routed, but duplicates are
1658 presumably rare, so I don't expect this is of any significance.
1659
1660 For deliveries to pipes, files, and autoreplies, the duplicate checking
1661 still happens during the routing process, since they are not going to be
1662 routed further.
1663
cfe75fc3
PH
1664PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1665 It corrects a timeout issue with spamd. This is Ian's comment: "The
1666 background is that sometimes spamd either never reads data from a
1667 connection it has accepted, or it never writes response data. The exiscan
1668 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1669 blindly assumes that writes won't block so it may never time out."
1670
be22d70e
PH
1671PH/39 Allow G after quota size as well as K and M.
1672
0612b098
PH
1673PH/40 The value set for $authenticated_id in an authenticator may not contain
1674 binary zeroes or newlines because the value is written to log lines and
1675 to spool files. There was no check on this. Now the value is run through
1676 the string_printing() function so that such characters are converted to
1677 printable escape sequences.
1678
2e0c1448
PH
1679PH/41 $message_linecount is a new variable that contains the total number of
1680 lines in the message. Compare $body_linecount, which is the count for the
1681 body only.
1682
447d236c
PH
1683PH/42 Exim no longer gives details of delivery errors for specific addresses in
1684 bounce and delay warning messages, except in certain special cases, which
1685 are as follows:
1686
1687 (a) An SMTP error message from a remote host;
1688 (b) A message specified in a :fail: redirection;
1689 (c) A message specified in a "fail" command in a system filter;
1690 (d) A message specified in a FAIL return from the queryprogram router;
1691 (e) A message specified by the cannot_route_message router option.
1692
1693 In these cases only, Exim does include the error details in bounce and
1694 warning messages. There are also a few cases where bland messages such
1695 as "unrouteable address" or "local delivery error" are given.
1696
d20976dc
PH
1697PH/43 $value is now also set for the "else" part of a ${run expansion.
1698
f656d135
PH
1699PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1700 being worked on, but at least Exim now implements the latest version to
1701 play with."
1702
2e2a30b4
PH
1703PH/45 In a pipe transport, although a timeout while waiting for the pipe
1704 process to complete was treated as a delivery failure, a timeout while
1705 writing the message to the pipe was logged, but erroneously treated as a
1706 successful delivery. Such timeouts include transport filter timeouts. For
1707 consistency with the overall process timeout, these timeouts are now
1708 treated as errors, giving rise to delivery failures by default. However,
1709 there is now a new Boolean option for the pipe transport called
1710 timeout_defer, which, if set TRUE, converts the failures into defers for
1711 both kinds of timeout. A transport filter timeout is now identified in
1712 the log output.
1713
9176e9f0
PH
1714PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1715 systems where "make" and "gmake" are different, calling "gmake" at top
1716 level broke things. I've arranged for the value of $(MAKE) to be passed
1717 from the Makefile to this script so that it can call the same version of
1718 "make".
1719
7982096b 1720
bbe902f0
PH
1721A note about Exim versions 4.44 and 4.50
1722----------------------------------------
1723
1724Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1725changes of various kinds. As a consequence, a big documentation update was
1726needed. This delayed the release for rather longer than seemed good, especially
1727in the light of a couple of (minor) security issues. Therefore, the changes
1728that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1729release. So 4.44 and 4.50 are in effect two different branches that both start
1730from 4.43.
1731
1732I have left the 4.50 change log unchanged; it contains all the changes since
17334.43. The change log for 4.44 is below; many of its items are identical to
1734those for 4.50. This seems to be the most sensible way to preserve the
1735historical information.
1736
1737
f7b63901 1738Exim version 4.50
495ae4b0
PH
1739-----------------
1740
5fe762f6
PH
1741 1. Minor wording change to the doc/README.SIEVE file.
1742
139059f6 1743 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
5fe762f6 1744 computation of the current number of files was incorrect.
495ae4b0 1745
7086e875
PH
1746 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1747 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1748 place.
1749
35af9f61
PH
1750 4. Give more explanation in the error message when the command for a transport
1751 filter fails to execute.
1752
b668c215
PH
1753 5. There are several places where Exim runs a non-Exim command in a
1754 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1755 was being done only for the command run by the queryprogram router. It is
1756 now done for all such subprocesses. The other cases are: ${run, transport
1757 filters, and the commands run by the lmtp and pipe transports.
1758
a494b1e1
PH
1759 6. Added CONFIGURE_GROUP build-time option.
1760
1761 7. Some older OS have a limit of 256 on the maximum number of file
1762 descriptors. Exim was using setrlimit() to set 1000 as a large value
1763 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1764 systems. I've change it so that if it can't get 1000, it tries for 256.
35edf2ff 1765
c5fcb476
PH
1766 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1767 was an oversight, and furthermore, ever since the addition of extra
1768 controls (e.g. 4.43/32), the checks on when to allow different forms of
1769 "control" were broken. There should now be diagnostics for all cases when a
1770 control that does not make sense is encountered.
1771
69358f02
PH
1772 9. Added the /retain_sender option to "control=submission".
1773
5be20824
PH
177410. $recipients is now available in the predata ACL (oversight).
1775
eb2c0248
PH
177611. Tidy the search cache before the fork to do a delivery from a message
1777 received from the command line. Otherwise the child will trigger a lookup
1778 failure and thereby defer the delivery if it tries to use (for example) a
1779 cached ldap connection that the parent has called unbind on.
1780
2a3eea10
PH
178112. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1782 of $address_data from the recipient verification was clobbered by the
1783 sender verification.
1784
178513. The value of address_data from a sender verification is now available in
1786 $sender_address_data in subsequent conditions in the ACL statement.
1787
23c7ff99
PH
178814. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1789
4deaf07d
PH
179015. Added a new option "connect=<time>" to callout options, to set a different
1791 connection timeout.
1792
926e1192
PH
179316. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1794 was its contents. (It was OK if the option was not defined at all.)
1795
650edc6f
PH
179617. A "Completed" log line is now written for messages that are removed from
1797 the spool by the -Mrm option.
1798
2c7db3f5
PH
179918. New variables $sender_verify_failure and $recipient_verify_failure contain
1800 information about exactly what failed.
1801
3d235903
PH
180219. Added -dd to debug only the daemon process.
1803
7c7ad977
PH
180420. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1805 handles timeouts, both on the server side and network timeouts. Renamed the
1806 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1807
981756db
PH
180821. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1809 It is now set to "smtps".
1810
d4eb88df
PH
181122. $host_address is now set to the target address during the checking of
1812 ignore_target_hosts.
1813
181423. When checking ignore_target_hosts for an ipliteral router, no host name was
1815 being passed; this would have caused $sender_host_name to have been used if
1816 matching the list had actually called for a host name (not very likely,
1817 since this list is usually IP addresses). A host name is now passed as
1818 "[x.x.x.x]".
1819
7d468ab8
PH
182024. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1821 code that specifies a non-restarting handler (typically sigaction() in
1822 modern systems) in an attempt to fix a rare and obscure crash bug.
1823
182425. Narrowed the window for a race in the daemon that could cause it to ignore
1825 SIGCHLD signals. This is not a major problem, because they are used only to
1826 wake it up if nothing else does.
1827
62c0818f
PH
182826. A malformed maildirsize file could cause Exim to calculate negative values
1829 for the mailbox size or file count. Odd effects could occur as a result.
1830 The maildirsize information is now recalculated if the size or filecount
1831 end up negative.
1832
26034054
PH
183327. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1834 support for a long time. Removed HAVE_SYS_VFS_H.
1835
af66f652
PH
183628. Installed the latest version of exipick from John Jetmore.
1837
90af77f4
PH
183829. In an address list, if the pattern was not a regular expression, an empty
1839 subject address (from a bounce message) matched only if the pattern was an
1840 empty string. Non-empty patterns were not even tested. This was the wrong
1841 because it is perfectly reasonable to use an empty address as part of a
1842 database query. An empty address is now tested by patterns that are
1843 lookups. However, all the other forms of pattern expect the subject to
1844 contain a local part and a domain, and therefore, for them, an empty
1845 address still always fails if the pattern is not itself empty.
1846
d8ef3577
PH
184730. Exim went into a mad DNS loop when attempting to do a callout where the
1848 host was specified on an smtp transport, and looking it up yielded more
1849 than one IP address.
1850
5cb8cbc6
PH
185131. Re-factored the code for checking spool and log partition space into a
1852 function that finds that data and another that does the check. The former
1853 is then used to implement four new variables: $spool_space, $log_space,
1854 $spool_inodes, and $log_inodes.
1855
14702f5b
PH
185632. The RFC2047 encoding function was originally intended for short strings
1857 such as real names; it was not keeping to the 75-character limit for
1858 encoded words that the RFC imposes. It now respects the limit, and
1859 generates multiple encoded words if necessary. To be on the safe side, I
1860 have increased the buffer size for the ${rfc2047: expansion operator from
1861 1024 to 2048 bytes.
1862
063b1e99
PH
186333. It is now permitted to omit both strings after an "if" condition; if the
1864 condition is true, the result is "true". As before, when the second string
1865 is omitted, a false condition yields an empty string. This makes it less
1866 cumbersome to write custom ACL and router conditions.
1867
652e1b65
PH
186834. Failure to deliver a bounce message always caused it to be frozen, even if
1869 there was an errors_to setting on the router. The errors_to setting is now
1870 respected.
1871
6f0c9a4f
PH
187235. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1873 canonical form (fully expanded) before being placed in
1874 $sender_host_address.
1875
33397d19
PH
187636. The table in the code that translates DNS record types into text (T_A to
1877 "A" for instance) was missing entries for NS and CNAME. It is just possible
1878 that this could have caused confusion if both these types were looked up
1879 for the same domain, because the text type is used as part of Exim's
1880 per-process caching. But the chance of anyone hitting this buglet seems
1881 very small.
1882
7bb56e1f
PH
188337. The dnsdb lookup has been extended in a number of ways.
1884
1885 (1) There is a new type, "zns", which walks up the domain tree until it
1886 finds some nameserver records. It should be used with care.
1887
ea3bc19b
PH
1888 (2) There is a new type, "mxh", which is like "mx" except that it returns
1889 just the host names, not the priorities.
1890
1891 (3) It is now possible to give a list of domains (or IP addresses) to be
ff4dbb19
PH
1892 looked up. The behaviour when one of the lookups defers can be
1893 controlled by a keyword.
7bb56e1f 1894
ea3bc19b 1895 (4) It is now possible to specify the separator character for use when
7bb56e1f 1896 multiple records are returned.
33397d19 1897
0bcb2a0e
PH
189838. The dnslists ACL condition has been extended: it is now possible to supply
1899 a list of IP addresses and/or domains to be looked up in a particular DNS
1900 domain.
1901
2ac0e484
PH
190239. Added log_selector=+queue_time_overall.
1903
4e1fde53
PH
190440. When running the queue in the test harness, wait just a tad after forking a
1905 delivery process, to get repeatability of debugging output.
1906
de365ded
PH
190741. Include certificate and key file names in error message when GnuTLS fails
1908 to set them up, because the GnuTLS error message doesn't include the name
1909 of the failing file when there is a problem reading it.
1910
f05da2e8
PH
191142. Allow both -bf and -bF in the same test run.
1912
d6453af2
PH
191343. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1914
f7b63901
PH
191544. The "Exiscan patch" is now merged into the mainline Exim source.
1916
191745. Sometimes the final signoff response after QUIT could fail to get
1918 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1919 < 0 before doing a fflush(). This bug looks as though it goes back to the
1920 introduction of TLS in release 3.20, but "sometimes" must have been rare
1921 because the tests only now provoked it.
1922
a444213a
PH
192346. Reset the locale to "C" after calling embedded Perl, in case it was changed
1924 (this can affect the format of dates).
1925
0ec020ea
PH
192647. exim_tidydb, when checking for the continued existence of a message for
1927 which it has found a message-specific retry record, was not finding
1928 messages that were in split spool directories. Consequently, it was
1929 deleting retry records that should have stayed in existence.
1930
b1206957
PH
193148. Steve fixed some bugs in eximstats.
1932
193349. The SPA authentication driver was not abandoning authentication and moving
1934 on to the next authenticator when an expansion was forced to fail,
1935 contradicting the general specification for all authenticators. Instead it
1936 was generating a temporary error. It now behaves as specified.
1937
26dd5a95
PH
193850. The default ordering of permitted cipher suites for GnuTLS was pessimal
1939 (the order specifies the preference for clients). The order is now AES256,
1940 AES128, 3DES, ARCFOUR128.
1941
343b2385
PH
194251. Small patch to Sieve code - explicitly set From: when generating an
1943 autoreply.
1944
1c5466b9
PH
194552. Exim crashed if a remote delivery caused a very long error message to be
1946 recorded - for instance if somebody sent an entire SpamAssassin report back
1947 as a large number of 550 error lines. This bug was coincidentally fixed by
1948 increasing the size of one of Exim's internal buffers (big_buffer) that
1949 happened as part of the Exiscan merge. However, to be on the safe side, I
1950 have made the code more robust (and fixed the comments that describe what
1951 is going on).
1952
55ee9ee3
PH
195353. Now that there can be additional text after "Completed" in log lines (if
1954 the queue_time_overall log selector is set), a one-byte patch to exigrep
1955 was needed to allow it to recognize "Completed" as not the last thing in
1956 the line.
1957
d38f8232
PH
195854. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1959 patch that reportedly fixes this has been added. I am not expert enough to
1960 create a test for it. This is what the patch creator wrote:
1961
1962 "I found a little strange behaviour of ldap code when working with
1963 Windows 2003 AD Domain, where users was placed in more than one
1964 Organization Units. When I tried to give exim partial DN, the exit code
1965 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1966 But simultaneously result of request was absolutely normal ldap result,
1967 so I produce this patch..."
1968
3295e65b
PH
1969 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1970 REFERENCE, so I have modified the code to exclude the patch when that macro
1971 is not defined.
1972
7102e136
PH
197355. Some experimental protocols are using DNS PTR records for new purposes. The
1974 keys for these records are domain names, not reversed IP addresses. The
b975ba52
PH
1975 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1976 leaves it alone. Component reversal etc. now happens only for IP addresses.
ea3a6f44 1977 CAN-2005-0021
7102e136 1978
3ca0ba97
PH
197956. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1980
c2bcbe20
PH
198157. Double the size of the debug message buffer (to 2048) so that more of very
1982 long debug lines gets shown.
1983
18ce445d
PH
198458. The exicyclog utility now does better if the number of log files to keep
1985 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1986
1f5b4c3d
PH
198759. Two changes related to the smtp_active_hostname option:
1988
1989 (1) $smtp_active_hostname is now available as a variable.
1990 (2) The default for smtp_banner uses $smtp_active_hostname instead
1991 of $primary_hostname.
1992
b975ba52
PH
199360. The host_aton() function is supposed to be passed a string that is known
1994 to be a valid IP address. However, in the case of IPv6 addresses, it was
1995 not checking this. This is a hostage to fortune. Exim now panics and dies
1996 if the condition is not met. A case was found where this could be provoked
85b87bc2
PH
1997 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1998 components; fortuitously, this particular loophole had already been fixed
1999 by change 4.50/55 above.
2000
2001 If there are any other similar loopholes, the new check in host_aton()
2002 itself should stop them being exploited. The report I received stated that
2003 data on the command line could provoke the exploit when Exim was running as
2004 exim, but did not say which command line option was involved. All I could
2005 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2006 running as the user.
ea3a6f44 2007 CAN-2005-0021
85b87bc2
PH
2008
200961. There was a buffer overflow vulnerability in the SPA authentication code
2010 (which came originally from the Samba project). I have added a test to the
2011 spa_base64_to_bits() function which I hope fixes it.
ea3a6f44 2012 CAN-2005-0022
b975ba52 2013
17ffcae7
PH
201462. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2015 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2016
d95f9fdb
PH
201763. The daemon start-up calls getloadavg() while still root for those OS that
2018 need the first call to be done as root, but it missed one case: when
2019 deliver_queue_load_max is set with deliver_drop_privilege. This is
2020 necessary for the benefit of the queue runner, because there is no re-exec
2021 when deliver_drop_privilege is set.
2022
86b8287f
PH
202364. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2024 This has been fixed.
2025
60dc5e56
PH
202665. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2027 was in use, was not putting the data itself into the right store pool;
2028 consequently, it could be overwritten for a subsequent message in the same
2029 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2030 the caching.)
2031
533244af
PH
203266. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2033
a5a28604
PH
203467. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2035 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2036 the function were treating the return as a boolean value, which happened to
2037 work because 0=false and not-0=true, but is not correct code.
2038
7e634d24
PH
203968. The host_aton() function was not handling scoped IPv6 addresses (those
2040 with, for example, "%eth0" on the end) correctly.
2041
3e11c26b
PH
204269. Fixed some compiler warnings in acl.c for the bitmaps specified with
2043 negated items (that is, ~something) in unsigned ints. Some compilers
2044 apparently mutter when there is no cast.
2045
6729cf78
PH
204670. If an address verification called from an ACL failed, and did not produce a
2047 user-specific message (i.e. there was only a "system" message), nothing was
2048 put in $acl_verify_message. In this situation, it now puts the system
2049 message there.
2050
00f00ca5
PH
205171. Change 4.23/11 added synchronization checking at the start of an SMTP
2052 session; change 4.31/43 added the unwanted input to the log line - except
2053 that it did not do this in the start of session case. It now does.
2054
c9bdd01c
PH
205572. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2056 This is wrong and can cause the other end to generate a synchronization
2057 error if it is another Exim or anything else that does the synchronization
2058 check. A QUIT command is no longer sent after a timeout.
2059
d43194df
PH
206073. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2061 during host lookups.
2062
fe5b5d0b
PH
206374. The defer_ok option of callout verification was not working if it was used
2064 when verifying addresses in header lines, that is, for this case:
2065
2066 verify = header_sender/callout=defer_ok
2067
76a2d7ba
PH
206875. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2069 those file descriptors could be used for SMTP connections. If anything
2070 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2071 could be sent to the SMTP client, causing chaos. The daemon now opens
2072 stdin, stdout, and stderr to /dev/null when it puts itself into the
2073 background.
2074
207576. Arrange for output from Perl's "warn" command to be written to Exim's main
2076 log by default. The user can override this with suitable Perl magic.
2077
04f7d5b9
PH
207877. The use of log_message on a "discard" ACL verb, which is supposed to add to
2079 the log message when discard triggers, was not working for the DATA ACL or
2080 for the non-SMTP ACL.
2081
bc60667e
PH
208278. Error message wording change in sieve.c.
2083
bb6e88ff
PH
208479. If smtp_accept_max_per_host was set, the number of connections could be
2085 restricted to fewer than expected, because the daemon was trying to set up
2086 a new connection before checking whether the processes handling previous
2087 connections had finished. The check for completed processes is now done
2088 earlier. On busy systems, this bug wouldn't be noticed because something
2089 else would have woken the daemon, and it would have reaped the completed
2090 process earlier.
2091
1e70f85b
PH
209280. If a message was submitted locally by a user whose login name contained one
2093 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2094 It caused a spool format error. I have fixed the spool reading code. A
2095 related problem was that the "from" clause in the Received: line became
2096 illegal because of the space(s). It is now covered by ${quote_local_part.
2097
209881. Included the latest eximstats from Steve (adds average sizes to HTML Top
2099 tables).
2100
4e01f9d6
PH
210182. Updated OS/Makefile-AIX as per message from Mike Meredith.
2102
1ee1cef2
PH
210383. Patch from Sieve maintainer to fix unterminated string problem in
2104 "vacation" handling.
2105
6e2b4ccc
PH
210684. Some minor changes to the Linux configuration files to help with other
2107 OS variants using glibc.
2108
8e669ac1
PH
210985. One more patch for Sieve to update vacation handling to latest spec.
2110
495ae4b0 2111
bbe902f0
PH
2112----------------------------------------------------
2113See the note above about the 4.44 and 4.50 releases.
2114----------------------------------------------------
2115
2116
2117Exim version 4.44
2118-----------------
2119
2120 1. Change 4.43/35 introduced a bug that caused file counts to be
2121 incorrectly computed when quota_filecount was set in an appendfile
2122 transport
2123
2124 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2125 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2126 place.
2127
2128 3. Give more explanation in the error message when the command for a transport
2129 filter fails to execute.
2130
2131 4. There are several places where Exim runs a non-Exim command in a
2132 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2133 was being done only for the command run by the queryprogram router. It is
2134 now done for all such subprocesses. The other cases are: ${run, transport
2135 filters, and the commands run by the lmtp and pipe transports.
2136
2137 5. Some older OS have a limit of 256 on the maximum number of file
2138 descriptors. Exim was using setrlimit() to set 1000 as a large value
2139 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2140 systems. I've change it so that if it can't get 1000, it tries for 256.
2141
2142 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2143 was an oversight, and furthermore, ever since the addition of extra
2144 controls (e.g. 4.43/32), the checks on when to allow different forms of
2145 "control" were broken. There should now be diagnostics for all cases when a
2146 control that does not make sense is encountered.
2147
2148 7. $recipients is now available in the predata ACL (oversight).
2149
2150 8. Tidy the search cache before the fork to do a delivery from a message
2151 received from the command line. Otherwise the child will trigger a lookup
2152 failure and thereby defer the delivery if it tries to use (for example) a
2153 cached ldap connection that the parent has called unbind on.
2154
2155 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2156 of $address_data from the recipient verification was clobbered by the
2157 sender verification.
2158
215910. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2160 was its contents. (It was OK if the option was not defined at all.)
2161
216211. A "Completed" log line is now written for messages that are removed from
2163 the spool by the -Mrm option.
2164
216512. $host_address is now set to the target address during the checking of
2166 ignore_target_hosts.
2167
216813. When checking ignore_target_hosts for an ipliteral router, no host name was
2169 being passed; this would have caused $sender_host_name to have been used if
2170 matching the list had actually called for a host name (not very likely,
2171 since this list is usually IP addresses). A host name is now passed as
2172 "[x.x.x.x]".
2173
217414. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2175 code that specifies a non-restarting handler (typically sigaction() in
2176 modern systems) in an attempt to fix a rare and obscure crash bug.
2177
217815. Narrowed the window for a race in the daemon that could cause it to ignore
2179 SIGCHLD signals. This is not a major problem, because they are used only to
2180 wake it up if nothing else does.
2181
218216. A malformed maildirsize file could cause Exim to calculate negative values
2183 for the mailbox size or file count. Odd effects could occur as a result.
2184 The maildirsize information is now recalculated if the size or filecount
2185 end up negative.
2186
218717. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2188 support for a long time. Removed HAVE_SYS_VFS_H.
2189
ea3a6f44 219018. Updated exipick to current release from John Jetmore.
bbe902f0
PH
2191
219219. Allow an empty sender to be matched against a lookup in an address list.
2193 Previously the only cases considered were a regular expression, or an
2194 empty pattern.
2195
219620. Exim went into a mad DNS lookup loop when doing a callout where the
2197 host was specified on the transport, if the DNS lookup yielded more than
2198 one IP address.
2199
ea3a6f44
NM
220021. The RFC2047 encoding function was originally intended for short strings
2201 such as real names; it was not keeping to the 75-character limit for
2202 encoded words that the RFC imposes. It now respects the limit, and
2203 generates multiple encoded words if necessary. To be on the safe side, I
2204 have increased the buffer size for the ${rfc2047: expansion operator from
2205 1024 to 2048 bytes.
bbe902f0 2206
ea3a6f44
NM
220722. Failure to deliver a bounce message always caused it to be frozen, even if
2208 there was an errors_to setting on the router. The errors_to setting is now
2209 respected.
bbe902f0
PH
2210
221123. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2212 canonical form (fully expanded) before being placed in
2213 $sender_host_address.
2214
221524. Updated eximstats to version 1.33
2216
ea3a6f44
NM
221725. Include certificate and key file names in error message when GnuTLS fails
2218 to set them up, because the GnuTLS error message doesn't include the name
2219 of the failing file when there is a problem reading it.
bbe902f0
PH
2220
222126. Expand error message when OpenSSL has problems setting up cert/key files.
ea3a6f44 2222 As per change 25.
bbe902f0 2223
ea3a6f44
NM
222427. Reset the locale to "C" after calling embedded Perl, in case it was changed
2225 (this can affect the format of dates).
bbe902f0 2226
ea3a6f44
NM
222728. exim_tidydb, when checking for the continued existence of a message for
2228 which it has found a message-specific retry record, was not finding
2229 messages that were in split spool directories. Consequently, it was
2230 deleting retry records that should have stayed in existence.
bbe902f0
PH
2231
223229. eximstats updated to version 1.35
2233 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2234 1.35 - bugfix such that pie charts by volume are generated correctly
2235
ea3a6f44
NM
223630. The SPA authentication driver was not abandoning authentication and moving
2237 on to the next authenticator when an expansion was forced to fail,
2238 contradicting the general specification for all authenticators. Instead it
2239 was generating a temporary error. It now behaves as specified.
bbe902f0 2240
ea3a6f44
NM
224131. The default ordering of permitted cipher suites for GnuTLS was pessimal
2242 (the order specifies the preference for clients). The order is now AES256,
2243 AES128, 3DES, ARCFOUR128.
bbe902f0 2244
ea3a6f44
NM
224531. Small patch to Sieve code - explicitly set From: when generating an
2246 autoreply.
bbe902f0 2247
ea3a6f44
NM
224832. Exim crashed if a remote delivery caused a very long error message to be
2249 recorded - for instance if somebody sent an entire SpamAssassin report back
2250 as a large number of 550 error lines. This bug was coincidentally fixed by
2251 increasing the size of one of Exim's internal buffers (big_buffer) that
2252 happened as part of the Exiscan merge. However, to be on the safe side, I
2253 have made the code more robust (and fixed the comments that describe what
2254 is going on).
bbe902f0 2255
ea3a6f44
NM
225633. Some experimental protocols are using DNS PTR records for new purposes. The
2257 keys for these records are domain names, not reversed IP addresses. The
2258 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2259 leaves it alone. Component reversal etc. now happens only for IP addresses.
bbe902f0
PH
2260 CAN-2005-0021
2261
ea3a6f44
NM
226234. The host_aton() function is supposed to be passed a string that is known
2263 to be a valid IP address. However, in the case of IPv6 addresses, it was
2264 not checking this. This is a hostage to fortune. Exim now panics and dies
2265 if the condition is not met. A case was found where this could be provoked
2266 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2267 components; fortuitously, this particular loophole had already been fixed
2268 by change 4.50/55 or 4.44/33 above.
2269
2270 If there are any other similar loopholes, the new check in host_aton()
2271 itself should stop them being exploited. The report I received stated that
2272 data on the command line could provoke the exploit when Exim was running as
2273 exim, but did not say which command line option was involved. All I could
2274 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2275 running as the user.
bbe902f0
PH
2276 CAN-2005-0021
2277
ea3a6f44
NM
227835. There was a buffer overflow vulnerability in the SPA authentication code
2279 (which came originally from the Samba project). I have added a test to the
2280 spa_base64_to_bits() function which I hope fixes it.
bbe902f0
PH
2281 CAN-2005-0022
2282
ea3a6f44
NM
228336. The daemon start-up calls getloadavg() while still root for those OS that
2284 need the first call to be done as root, but it missed one case: when
2285 deliver_queue_load_max is set with deliver_drop_privilege. This is
2286 necessary for the benefit of the queue runner, because there is no re-exec
2287 when deliver_drop_privilege is set.
bbe902f0 2288
ea3a6f44
NM
228937. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2290 was in use, was not putting the data itself into the right store pool;
2291 consequently, it could be overwritten for a subsequent message in the same
2292 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2293 the caching.)
bbe902f0 2294
ea3a6f44
NM
229538. Sometimes the final signoff response after QUIT could fail to get
2296 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2297 < 0 before doing a fflush(). This bug looks as though it goes back to the
2298 introduction of TLS in release 3.20, but "sometimes" must have been rare
2299 because the tests only now provoked it.
bbe902f0
PH
2300
2301
495ae4b0
PH
2302Exim version 4.43
2303-----------------
2304
2305 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2306 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2307 Now it returns an integer. A number of calls were still expecting a T/F
2308 return. Fortuitously, in all cases, the tests worked in OK situations,
2309 which is the norm. However, things would have gone wrong on any write
2310 failures on the smtp file descriptor. This function is used when sending
2311 messages over SMTP and also when doing verify callouts.
2312
2313 2. When Exim is called to do synchronous delivery of a locally submitted
2314 message (the -odf or -odi options), it no longer closes stderr before doing
2315 the delivery.
2316
2317 3. Implemented the mua_wrapper option.
2318
2319 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2320
2321 5. Implemented the functions header_remove(), header_testname(),
2322 header_add_at_position(), and receive_remove_recipient(), and exported them
2323 to local_scan().
2324
2325 6. If an ACL "warn" statement specified the addition of headers, Exim already
2326 inserted X-ACL-Warn: at the start if there was no header name. However, it
2327 was not making this test for the second and subsequent header lines if
2328 there were newlines in the string. This meant that an invalid header could
2329 be inserted if Exim was badly configured.
2330
2331 7. Allow an ACL "warn" statement to add header lines at the start or after all
2332 the Received: headers, as well as at the end.
2333
2334 8. Added the rcpt_4xx retry error code.
2335
2336 9. Added postmaster_mailfrom=xxx to callout verification option.
2337
233810. Added mailfrom=xxxx to the callout verification option, for verify=
2339 header_sender only.
2340
234111. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2342 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2343
234412. Inserted some casts to stop certain compilers warning when using pointer
2345 differences as field lengths or precisions in printf-type calls (mostly
2346 affecting debugging statements).
2347
234813. Added optional readline() support for -be (dynamically loaded).
2349
235014. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2351 same clock tick as a message's arrival, so that its received time was the
2352 same as the "first fail" time on the retry record, and that message
2353 remained on the queue past the ultimate address timeout, every queue runner
2354 would try a delivery (because it was past the ultimate address timeout) but
2355 after another failure, the ultimate address timeout, which should have then
2356 bounced the address, did not kick in. This was a "< instead of <=" error;
2357 in most cases the first failure would have been in the next clock tick
2358 after the received time, and all would be well.
2359
236015. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2361 being recognized when the domain list was tested by the match_domain
2362 condition in an expansion string.
2363
236416. Added the ${str2b64: operator.
2365
236617. Exim was always calling setrlimit() to set a large limit for the number of
2367 processes, without checking whether the existing limit was already
2368 adequate. (It did check for the limit on file descriptors.) Furthermore,
2369 errors from getrlimit() and setrlimit() were being ignored. Now they are
2370 logged to the main and panic logs, but Exim does carry on, to try to do its
2371 job under whatever limits there are.
2372
237318. Imported PCRE 5.0.
2374
237519. Trivial typo in log message " temporarily refused connection" (the leading
2376 space).
2377
237820. If the log selector return_path_on_delivery was set and an address was
2379 redirected to /dev/null, the delivery process crashed because it assumed
2380 that a return path would always be set for a "successful" delivery. In this
2381 case, the whole delivery is bypassed as an optimization, and therefore no
2382 return path is set.
2383
238421. Internal re-arrangement: the function for sending a challenge and reading
2385 a response while authentication was assuming a zero-terminated challenge
2386 string. It's now changed to take a pointer and a length, to allow for
2387 binary data in such strings.
2388
238922. Added the cyrus_sasl authenticator (code supplied by MBM).
2390
239123. Exim was not respecting finduser_retries when seeking the login of the
2392 uid under which it was called; it was always trying 10 times. (The default
2393 setting of finduser_retries is zero.) Also, it was sleeping after the final
2394 failure, which is pointless.
2395
239624. Implemented tls_on_connect_ports.
2397
239825. Implemented acl_smtp_predata.
2399
240026. If the domain in control=submission is set empty, Exim assumes that the
2401 authenticated id is a complete email address when it generates From: or
2402 Sender: header lines.
2403
240427. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2405 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2406 chown and chgrp in /bin and hostname in /usr/bin.
2407
240828. Exim was keeping the "process log" file open after each use, just as it
2409 does for the main log. This opens the possibility of it remaining open for
2410 long periods when the USR1 signal hits a daemon. Occasional processlog
2411 errors were reported, that could have been caused by this. Anyway, it seems
2412 much more sensible not to leave this file open at all, so that is what now
2413 happens.
2414
241529. The long-running daemon process does not normally write to the log once it
2416 has entered its main loop, and it closes the log before doing so. This is
2417 so that log files can straightforwardly be renamed and moved. However,
2418 there are a couple of unusual error situations where the daemon does write
2419 log entries, and I had neglected to close the log afterwards.
2420
242130. The text of an SMTP error response that was received during a remote
2422 delivery was being truncated at 512 bytes. This is too short for some of
2423 the long messages that one sometimes sees. I've increased the limit to
2424 1024.
2425
242631. It is now possible to make retry rules that apply only when a message has a
2427 specific sender, in particular, an empty sender.
2428
242932. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2430 it possible to be selective about when SMTP synchronization is enforced.
2431
243233. Added "control = caseful_local_part" and "control = "caselower_local_part".
2433
243432. Implemented hosts_connection_nolog.
2435
243633. Added an ACL for QUIT.
2437
243834. Setting "delay_warning=" to disable warnings was not working; it gave a
2439 syntax error.
2440
244135. Added mailbox_size and mailbox_filecount to appendfile.
2442
244336. Added control = no_multiline_responses to ACLs.
2444
244537. There was a bug in the logic of the code that waits for the clock to tick
2446 in the case where the clock went backwards by a substantial amount such
2447 that the microsecond fraction of "now" was more than the microsecond
2448 fraction of "then" (but the whole seconds number was less).
2449
245038. Added support for the libradius Radius client library this is found on
2451 FreeBSD (previously only the radiusclient library was supported).
2452
2453
2454Exim version 4.42
2455-----------------
2456
2457 1. When certain lookups returned multiple values in the form name=value, the
2458 quoting of the values was not always being done properly. Specifically:
2459 (a) If the value started with a double quote, but contained no whitespace,
2460 it was not quoted.
2461 (b) If the value contained whitespace other than a space character (i.e.
2462 tabs or newlines or carriage returns) it was not quoted.
2463 This fix has been applied to the mysql and pgsql lookups by writing a
2464 separate quoting function and calling it from the lookup code. The fix
2465 should probably also be applied to nisplus, ibase and oracle lookups, but
2466 since I cannot test any of those, I have not disturbed their existing code.
2467
2468 2. A hit in the callout cache for a specific address caused a log line with no
2469 reason for rejecting RCPT. Now it says "Previous (cached) callout
2470 verification failure".
2471
2472 3. There was an off-by-one bug in the queryprogram router. An over-long
2473 return line was truncated at 256 instead of 255 characters, thereby
2474 overflowing its buffer with the terminating zero. As well as fixing this, I
2475 have increased the buffer size to 1024 (and made a note to document this).
2476
2477 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2478 when Exim is waiting for an SMTP response from a remote server, Exim
2479 restarts its select() call on the socket, thereby resetting its timeout.
2480 This is not a problem when such interrupts are rare. Somebody set up a cron
2481 job to run exiwhat every 2 minutes, which is less than the normal select()
2482 timeout (5 or 10 minutes). This meant that the select() timeout never
2483 kicked in because it was always reset. I have fixed this by comparing the
2484 time when an interrupt arrives with the time at the start of the first call
2485 to select(). If more time than the timeout has elapsed, the interrupt is
2486 treated as a timeout.
2487
2488 5. Some internal re-factoring in preparation for the addition of Sieve
2489 extensions (by MH). In particular, the "personal" test is moved to a
2490 separate function, and given an option for scanning Cc: and Bcc: (which is
2491 not set for Exim filters).
2492
2493 6. When Exim created an email address using the login of the caller as the
2494 local part (e.g. when creating a From: or Sender: header line), it was not
2495 quoting the local part when it contained special characters such as @.
2496
2497 7. Installed new OpenBSD configuration files.
2498
2499 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2500 try to make them clearer.
2501
2502 9. Callout options, other than the timeout value, were being ignored when
2503 verifying sender addresses in header lines. For example, when using
2504
2505 verify = header_sender/callout=no_cache
2506
2507 the cache was (incorrectly) being used.
2508
250910. Added a missing instance of ${EXE} to the exim_install script; this affects
2510 only the Cygwin environment.
2511
251211. When return_path_on_delivery was set as a log selector, if different remote
2513 addresses in the same message used different return paths and parallel
2514 remote delivery occurred, the wrong values would sometimes be logged.
2515 (Whenever a remote delivery process finished, the return path value from
2516 the most recently started remote delivery process was logged.)
2517
251812. RFC 3848 specifies standard names for the "with" phrase in Received: header
2519 lines when AUTH and/or TLS are in use. This is the "received protocol"
2520 field. Exim used to use "asmtp" for authenticated SMTP, without any
2521 indication (in the protocol name) for TLS use. Now it follows the RFC and
2522 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2523 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2524 names appear in log lines as well as in Received: header lines.
2525
252613. Installed MH's patches for Sieve to add the "copy" and "vacation"
2527 extensions, and comparison tests, and to fix some bugs.
2528
252914. Changes to the "personal" filter test:
2530
2531 (1) The test was buggy in that it was just doing the equivalent of
2532 "contains" tests on header lines. For example, if a user's address was
2533 anne@some.where, the "personal" test would incorrectly be true for
2534
2535 To: susanne@some.where
2536
2537 This test is now done by extracting each address from the header in turn,
2538 and checking the entire address. Other tests that are part of "personal"
2539 are now done using regular expressions (for example, to check local parts
2540 of addresses in From: header lines).
2541
2542 (2) The list of non-personal local parts in From: addresses has been
2543 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2544 taken from the Sieve specification recommendations.
2545
2546 (3) If the message contains any header line starting with "List-" it is
2547 treated as non-personal.
2548
2549 (4) The test for "circular" in the Subject: header line has been removed
2550 because it now seems ill-conceived.
2551
255215. Minor typos in src/EDITME comments corrected.
2553
255416. Installed latest exipick from John Jetmore.
2555
255617. If headers_add on a router specified a text string that was too long for
2557 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2558 of string_sprintf() is now avoided.
2559
256018. $message_body_size was not set (it was always zero) when running the DATA
2561 ACL and the local_scan() function.
2562
256319. For the "mail" command in an Exim filter, no default was being set for
2564 the once_repeat time, causing a random time value to be used if "once" was
2565 specified. (If the value happened to be <= 0, no repeat happened.) The
2566 default is now 0s, meaning "never repeat". The "vacation" command was OK
2567 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2568 (I found it when inspecting the code).
2569
257020. There is now an overall timeout for performing a callout verification. It
2571 defaults to 4 times the callout timeout, which applies to individual SMTP
2572 commands during the callout. The overall timeout applies when there is more
2573 than one host that can be tried. The timeout is checked before trying the
2574 next host. This prevents very long delays if there are a large number of
2575 hosts and all are timing out (e.g. when the network connections are timing
2576 out). The value of the overall timeout can be changed by specifying an
2577 additional sub-option for "callout", called "maxwait". For example:
2578
2579 verify = sender/callout=5s,maxwait=20s
2580
258121. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2582 to the end before writing, but this should make it even safer).
2583
258422. Exim was forgetting that it had advertised PIPELINING for the second and
2585 subsequent messages on an SMTP connection. It was also not resetting its
2586 memory on STARTTLS and an internal HELO.
2587
258823. When Exim logs an SMTP synchronization error within a session, it now
2589 records whether PIPELINING has been advertised or not.
2590
259124. Added 3 instances of "(long int)" casts to time_t variables that were being
2592 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2593 rather than long int.
2594
259525. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2596
259726. Added the never_mail option to autoreply.
2598
2599
2600Exim version 4.41
2601-----------------
2602
2603 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2604 crash if the getsockname() call failed; this can happen if a connection is
2605 closed very soon after it is established. The problem was simply in the
2606 order in which certain operations were done, causing Exim to try to write
2607 to the SMTP stream before it had set up the file descriptor. The bug has
2608 been fixed by making things happen in the correct order.
2609
2610
2611Exim version 4.40
2612-----------------
2613
2614 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2615 before the connection was closed, thus losing the rejection response.
2616
2617 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2618 some early Solaris releases, but causes trouble in current releases where
2619 socklen_t is defined.
2620
2621 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2622 always exist.
2623
2624 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2625 configured.
2626
2627 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2628 that releasing the top of it at the end releases what was used for sub-
2629 expansions (unless the block got too big). However, discard this block if
2630 the first thing is a variable or header, so that we can use its block when
2631 it is dynamic (useful for very large $message_headers, for example).
2632
2633 6. Lookups now cache *every* query, not just the most recent. A new, separate
2634 store pool is used for this. It can be recovered when all lookup caches are
2635 flushed. Lookups now release memory at the end of their result strings.
2636 This has involved some general refactoring of the lookup sources.
2637
2638 7. Some code has been added to the store_xxx() functions to reduce the amount
2639 of flapping under certain conditions.
2640
2641 8. log_incoming_interface used to affect only the <= reception log lines. Now
2642 it causes the local interface and port to be added to several more SMTP log
2643 lines, for example "SMTP connection from", and rejection lines.
2644
2645 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2646
264710. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2648
264911. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2650 could be overwritten at the end of the current message (or the start of a
2651 new message if it was set in a HELO ACL). The value is now preserved for
2652 the duration of the SMTP connection.
2653
265412. If a transport had a headers_rewrite setting, and a matching header line
2655 contained an unqualified address, that address was qualified, even if it
2656 did not match any rewriting rules. The underlying bug was that the values
2657 of the flags that permit the existence of unqualified sender and recipient
2658 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2659 non-local messages, and by -bnq for local messages) were not being
2660 preserved with the message after it was received.
2661
266213. When Exim was logging an SMTP synchronization error, it could sometimes log
2663 "next input=" as part of the text comprising the host identity instead of
2664 the correct text. The code was using the same buffer for two different
2665 strings. However, depending on which order the printing function evaluated
2666 its arguments, the bug did not always show up. Under Linux, for example, my
2667 test suite worked just fine.
2668
266914. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2670 doesn't work with some older versions of Perl. It has been changed to "my",
2671 which in any case is probably the better facility to use.
2672
267315. A really picky compiler found some instances of statements for creating
2674 error messages that either had too many or two few arguments for the format
2675 string.
2676
267716. The size of the buffer for calls to the DNS resolver has been increased
2678 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2679 for addresses that have a lot of PTR records. This alleviates a problem; it
2680 does not fully solve it.
2681
268217. A dnsdb lookup for PTR records that receives more data than will fit in the
2683 buffer now truncates the list and logs the incident, which is the same
2684 action as happens when Exim is looking up a host name and its aliases.
2685 Previously in this situation something unpredictable would happen;
2686 sometimes it was "internal error: store_reset failed".
2687
268818. If a server dropped the connection unexpectedly when an Exim client was
2689 using GnuTLS and trying to read a response, the client delivery process
2690 crashed while trying to generate an error log message.
2691
269219. If a "warn" verb in an ACL added multiple headers to a message in a single
2693 string, for example:
2694
2695 warn message = H1: something\nH2: something
2696
2697 the text was added as a single header line from Exim's point of view
2698 though it ended up OK in the delivered message. However, searching for the
2699 second and subsequent header lines using $h_h2: did not work. This has been
2700 fixed. Similarly, if a system filter added multiple headers in this way,
2701 the routers could not see them.
2702
270320. Expanded the error message when iplsearch is called with an invalid key to
2704 suggest using net-iplsearch in a host list.
2705
270621. When running tests using -bh, any delays imposed by "delay" modifiers in
2707 ACLs are no longer actually imposed (and a message to that effect is
2708 output).
2709
271022. If a "gecos" field in a passwd entry contained escaped characters, in
2711 particular, if it contained a \" sequence, Exim got it wrong when building
2712 a From: or a Sender: header from that name. A second bug also caused
2713 incorrect handling when an unquoted " was present following a character
2714 that needed quoting.
2715
271623. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2717 was not being matched caselessly.
2718
271924. Arranged for all hyphens in the exim.8 source to be escaped with
2720 backslashes.
2721
272225. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2723 properly. Recipient callout cache records were still being keyed to include
2724 the sender, even when use_sender was set false. This led to far more
2725 callouts that were necessary. The sender is no longer included in the key
2726 when use_sender is false.
2727
272826. Added "control = submission" modifier to ACLs.
2729
273027. Added the ${base62d: operator to decode base 62 numbers.
2731
273228. dnsdb lookups can now access SRV records.
2733
273429. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2735 the configuration file.
2736
273730. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2738 (-v) mode. This makes the output for a verbose queue run more intelligible.
2739
274031. Added a use_postmaster feature to recipient callouts.
2741
274232. Added the $body_zerocount variable, containing the number of binary zero
2743 bytes in the message body.
2744
274533. The time of last modification of the "new" subdirectory is now used as the
2746 "mailbox time last read" when there is a quota error for a maildir
2747 delivery.
2748
274934. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2750
275135. Added +ignore_unknown as a special item in host lists.
2752
275336. Code for decoding IPv6 addresses in host lists is now included, even if
2754 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2755 address was recognized as an IP address, but was then not correctly decoded
2756 into binary, causing unexpected and incorrect effects when compared with
2757 another IP address.
2758
2759
2760Exim version 4.34
2761-----------------
2762
2763 1. Very minor rewording of debugging text in manualroute to say "list of
2764 hosts" instead of "hostlist".
2765
2766 2. If verify=header_syntax was set, and a header line with an unqualified
2767 address (no domain) and a large number of spaces between the end of the
2768 name and the colon was received, the reception process suffered a buffer
2769 overflow, and (when I tested it) crashed. This was caused by some obsolete
2770 code that should have been removed. The fix is to remove it!
2771
2772 3. When running in the test harness, delay a bit after writing a bounce
2773 message to get a bit more predictability in the log output.
2774
2775 4. Added a call to search_tidyup() just before forking a reception process. In
2776 theory, someone could use a lookup in the expansion of smtp_accept_max_
2777 per_host which, without the tidyup, could leave open a database connection.
2778
2779 5. Added the variables $recipient_data and $sender_data which get set from a
2780 lookup success in an ACL "recipients" or "senders" condition, or a router
2781 "senders" option, similar to $domain_data and $local_part_data.
2782
2783 6. Moved the writing of debug_print from before to after the "senders" test
2784 for routers.
2785
2786 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2787 problems for message scanning, either using a data ACL, or using
2788 local_scan() because the Received: header was not generated till after they
2789 were called (in order to set the time as the time of reception completion).
2790 I have revised the way this works. The header is now generated after the
2791 body is received, but before the ACL or local_scan() are called. After they
2792 are run, the timestamp in the header is updated.
2793
2794
2795Exim version 4.33
2796-----------------
2797
2798 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2799 before starting a queue runner without re-exec. This happened only when
2800 deliver_drop_privilege was set or when the Exim user was set to root. The
2801 effect of the bug was that timeouts during subsequent deliveries caused
2802 crashes instead of being properly handled. The handler is now left at its
2803 default (and expected) setting.
2804
2805 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2806 message, again when deliver_drop_privilege is set or Exim is run as root.
2807 The bug described in (1) was not present in this case, but the tidying up
2808 of the other signals was missing. I have made the two cases consistent.
2809
2810 3. The ignore_target_hosts setting on a manualroute router was being ignored
2811 for hosts that were looked up using the /MX notation.
2812
2813 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2814 in domain lists.
2815
2816 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2817 operated on the sender address. After changing the $sender_address to <>
2818 for the sender address verify, Exim was re-instated it as the original
2819 (before rewriting) address, but remembering that it had rewritten it, so it
2820 wasn't rewriting it again. This bug also had the effect of breaking the
2821 sender address verification caching when the sender address was rewritten.
2822
2823 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2824 This has been changed so that if the ip literal address matches
2825 ignore_target_hosts, the router declines.
2826
2827 7. Added expansion conditions match_domain, match_address, and match_local_
2828 part (NOT match_host).
2829
2830 8. The placeholder for the Received: header didn't have a length field set.
2831
2832 9. Added code to Exim itself and to exim_lock to test for a specific race
2833 condition that could lead to file corruption when using MBX delivery. The
2834 issue is with the lockfile that is created in /tmp. If this file is removed
2835 after a process has opened it but before that process has acquired a lock,
2836 there is the potential for a second process to recreate the file and also
2837 acquire a lock. This could lead to two Exim processes writing to the file
2838 at the same time. The added code performs the same test as UW imapd; it
2839 checks after acquiring the lock that its file descriptor still refers to
2840 the same named file.
2841
284210. The buffer for building added header lines was of fixed size, 8192 bytes.
2843 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2844 when Exim is built.
2845
284611. Added the smtp_active_hostname option. If used, this will typically be made
2847 to depend on the incoming interface address. Because $interface_address is
2848 not set up until the daemon has forked a reception process, error responses
2849 that can happen earlier (such as "too many connections") no longer contain
2850 a host name.
2851
285212. If an expansion in a condition on a "warn" statement fails because a lookup
2853 defers, the "warn" statement is abandoned, and the next ACL statement is
2854 processed. Previously this caused the whole ACL to be aborted.
2855
285613. Added the iplsearch lookup type.
2857
285814. Added ident_timeout as a log selector.
2859
286015. Added tls_certificate_verified as a log selector.
2861
286216. Added a global option tls_require_ciphers (compare the smtp transport
2863 option of the same name). This controls incoming TLS connections.
2864
286517. I finally figured out how to make tls_require_ciphers do a similar thing
2866 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2867 before starting the TLS session.
2868
286918. Tabs are now shown as \t in -bP output.
2870
287119. If the log selector return_path_on_delivery was set, Exim crashed when
2872 bouncing a message because it had too many Received: header lines.
2873
287420. If two routers both had headers_remove settings, and the first one included
2875 a superfluous trailing colon, the final name in the first list and the
2876 first name in the second list were incorrectly joined into one item (with a
2877 colon in the middle).
2878
2879
2880Exim version 4.32
2881-----------------
2882
2883 1. Added -C and -D options to the exinext utility, mainly to make it easier
2884 to include in the automated testing, but these could be helpful when
2885 multiple configurations are in use.
2886
2887 2. The exinext utility was not formatting the output nicely when there was
2888 an alternate port involved in the retry record key, nor when there was a
2889 message id as well (for retries that were specific to a specific message
2890 and a specific host). It was also confused by IPv6 addresses, because of
2891 the additional colons they contain. I have fixed the IPv4 problem, and
2892 patched it up to do a reasonable job for IPv6.
2893
2894 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2895 delivery, the log line now contains "pipelined" if PIPELINING was used.
2896
2897 4. An SMTP transport process used to panic and die if the bind() call to set
2898 an explicit outgoing interface failed. This has been changed; it is now
2899 treated in the same way as a connect() failure.
2900
2901 5. A reference to $sender_host_name in the part of a conditional expansion
2902 that was being skipped was still causing a DNS lookup. This no longer
2903 occurs.
2904
2905 6. The def: expansion condition was not recognizing references to header lines
2906 that used bh_ and bheader_.
2907
2908 7. Added the _cache feature to named lists.
2909
2910 8. The code for checking quota_filecount in the appendfile transport was
2911 allowing one more file than it should have been.
2912
2913 9. For compatibility with Sendmail, the command line option
2914
2915 -prval:sval
2916
2917 is equivalent to
2918
2919 -oMr rval -oMs sval
2920
2921 and sets the incoming protocol and host name (for trusted callers). The
2922 host name and its colon can be omitted when only the protocol is to be set.
2923 Note the Exim already has two private options, -pd and -ps, that refer to
2924 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2925 "s", but I don't think that's a major issue.
2926
292710. A number of refactoring changes to the code, none of which should affect
2928 Exim's behaviour:
2929
2930 (a) The number of logging options was getting close to filling up the
2931 32-bit word that was used as a bit map. I have split them into two classes:
2932 those that are passed in the argument to log_write(), and those that are
2933 only ever tested independently outside of that function. These are now in
2934 separate 32-bit words, so there is plenty of room for expansion again.
2935 There is no change in the user interface or the logging behaviour.
2936
2937 (b) When building, for example, log lines, the code previously used a
2938 macro that called string_cat() twice, in order to add two strings. This is
2939 not really sufficiently general. Furthermore, there was one instance where
2940 it was actually wrong because one of the argument was used twice, and in
2941 one call a function was used. (As it happened, calling the function twice
2942 did not affect the overall behaviour.) The macro has been replaced by a
2943 function that can join an arbitrary number of extra strings onto a growing
2944 string.
2945
2946 (c) The code for expansion conditions now uses a table and a binary chop
2947 instead of a serial search (which was left over from when there were very
2948 few conditions). Also, it now recognizes conditions like "pam" even when
2949 the relevant support is not compiled in: a suitably worded error message is
2950 given if an attempt is made to use such a condition.
2951
295211. Added ${time_interval:xxxxx}.
2953
295412. A bug was causing one of the ddress fields not to be passed back correctly
2955 from remote delivery subprocesses. The field in question was not being
2956 subsequently used, so this caused to problems in practice.
2957
295813. Added new log selectors queue_time and deliver_time.
2959
296014. Might have fixed a bug in maildirsizefile handling that threw up
2961 "unexpected character" debug warnings, and recalculated the data
2962 unnecessarily. In any case, I expanded the warning message to give more
2963 information.
2964
296515. Added the message "Restricted characters in address" to the statements in
2966 the default ACL that block characters like @ and % in local parts.
2967
296816. Change 71 for release 4.31 proved to be much less benign that I imagined.
2969 Three changes have been made:
2970
2971 (a) There was a serious bug; a negative response to MAIL caused the whole
2972 recipient domain to be cached as invalid, thereby blocking all messages
2973 to all local parts at the same domain, from all senders. This bug has
2974 been fixed. The domain is no longer cached after a negative response to
2975 MAIL if the sender used is not empty.
2976
2977 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2978 been restored.
2979
2980 (c) A new callout option, "use_sender" has been added for people who want
2981 the modified behaviour.
2982
2983
2984Exim version 4.31
2985-----------------
2986
2987 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2988 Larry Rosenman.
2989
2990 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2991 indeed breaks things for older releases.
2992
2993 3. Added additional logging to the case where there is a problem reading data
2994 from a filter that is running in a subprocess using a pipe, in order to
2995 try to track down a specific problem.
2996
2997 4. Testing facility fudge: when running in the test harness and attempting
2998 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2999 getting "No route to host". Convert this to a timeout.
3000
3001 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3002 warning.
3003
3004 6. Some OS don't have socklen_t but use size_t instead. This affects the
3005 fifth argument of getsockopt() amongst other things. This is now
3006 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3007 can be set for individual OS. I have set it for SunOS5, OSF1, and
3008 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3009 some earlier ones do not.
3010
3011 7. Change 4.30/15 was not doing the test caselessly.
3012
3013 8. The standard form for an IPv6 address literal was being rejected by address
3014 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3015 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3016 this, as well as the form without the "IPv6" on the front (but only when
3017 address literals are enabled, of course).
3018
3019 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3020
302110. Exim crashed if a message with an empty sender address specified by -f
3022 encountered a router with an errors_to setting. This could be provoked only
3023 by a command such as
3024
3025 exim -f "" ...
3026
3027 where an empty string was supplied; "<>" did not hit this bug.
3028
302911. Installed PCRE release 4.5.
3030
303112. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3032 remained set. It is now erased.
3033
303413. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3035 times from message ids (which are base 36 rather than the normal 62).
3036
303714. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3038 were being counted as actual protocol errors, and logged if the log
3039 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3040 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3041 and DATA following a set of rejected RCPTs do not count as protocol errors.
3042 In other words, Exim assumes they were pipelined, though this may not
3043 actually be the case. Of course, in all cases the client gets an
3044 appropriate error code.
3045
304615. If a lookup fails in an ACL condition, a message about the failure may
3047 be available; it is used if testing the ACL cannot continue, because most
3048 such messages specify what the cause of the deferral is. However, some
3049 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3050 that caused an old message to be retained and used if a later statement
3051 caused a defer, replacing the real cause of the deferral.
3052
305316. If an IP address had so many PTR records that the DNS lookup buffer
3054 was not large enough to hold them, Exim could crash while trying to process
3055 the truncated data. It now detects and logs this case.
3056
305717. Further to 4.21/58, another change has been made: if (and only if) the
3058 first line of a message (the first header line) ends with CRLF, a bare LF
3059 in a subsequent header line has a space inserted after it, so as not to
3060 terminate the header.
3061
306218. Refactoring: tidied an ugly bit of code in appendfile that copied data
3063 unnecessarily, used atoi() instead of strtol(), and didn't check the
3064 termination when getting file sizes from file names by regex.
3065
306619. Completely re-implemented the support for maildirsize files, in the light
3067 of a number of problems with the previous contributed implementation
3068 (4.30/29). In particular:
3069
3070 . If the quota is zero, the maildirsize file is maintained, but no quota is
3071 imposed.
3072
3073 . If the maildir directory does not exist, it is created before any attempt
3074 to write a maildirsize file.
3075
3076 . The quota value in the file is just a cache; if the quota is changed in
3077 the transport, the new value overrides.
3078
3079 . A regular expression is available for excluding directories from the
3080 count.
3081
308220. The autoreply transport checks the characters in options that define the
3083 message's headers; it allows continued headers, but it was checking with
3084 isspace() after an embedded newline instead of explicitly looking for a
3085 space or a tab.
3086
308721. If all the "regular" hosts to which an address was routed had passed their
3088 expiry times, and had not reached their retry times, the address was
3089 bounced, even if fallback hosts were defined. Now Exim should go on to try
3090 the fallback hosts.
3091
309222. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3093 equivalent code in the SMTP transport. Some hosts send humungous responses
3094 to HELO/EHLO, more than 1024 it seems.
3095
309623. Refactoring: code in filter.c used (void *) for "any old type" but this
3097 gives compiler warnings in some environments. I've now done it "properly",
3098 using a union.
3099
310024. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3101 (because of problems with the built-in one) was declared to return uschar *
3102 instead of char *, causing compiler failure.
3103
310425. Fixed a file descriptor leak when processing alias/forward files.
3105
310626. Fixed a minor format string issue in dbfn.c.
3107
310827. Typo in exim.c: ("dmbnz" for "dbmnz").
3109
311028. If a filter file refered to $h_xxx or $message_headers, and the headers
3111 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3112 become corrupted.
3113
311429. When a sender address is verified, it is cached, to save repeating the test
3115 when there is more than one recipient in a message. However, when the
3116 verification involves a callout, it is possible for different callout
3117 options to be set for different recipients. It is too complicated to keep
3118 track of this in the cache, so now Exim always runs a verification when a
3119 callout is required, relying on the callout cache for the optimization.
3120 The overhead is duplication of the address routing, but this should not be
3121 too great.
3122
312330. Fixed a bug in callout caching. If a RCPT command caused the sender address
3124 to be verified with callout=postmaster, and the main callout worked but the
3125 postmaster check failed, the verification correctly failed. However, if a
3126 subsequent RCPT command asked for sender verification *without* the
3127 postmaster check, incorrect caching caused this verification also to fail,
3128 incorrectly.
3129
313031. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3131 it was not caching the DNS options (qualify_single, search_parents) that
3132 were used when the lookup failed. A subsequent lookup with different
3133 options therefore always gave the same answer, though there were cases
3134 where it should not have. (Example: a "domains = !$mx_any" option on a
3135 dnslookup router: the "domains" option is always processed without any
3136 widening, but the router might have qualify_single set.) Now Exim uses the
3137 cached value only when the same options are set.
3138
313932. Added John Jetmore's "exipick" utility to the distribution.
3140
314133. GnuTLS: When an attempt to start a TLS session fails for any reason other
3142 than a timeout (e.g. a certificate is required, and is not provided), an
3143 Exim server now closes the connection immediately. Previously it waited for
3144 the client to close - but if the client is SSL, it seems that they each
3145 wait for each other, leading to a delay before one of them times out.
3146
314734: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3148 maintained 0.8.x compatibility because I don't think many are using it, and
3149 it is clearly obsolete.
3150
315135. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3152 transport.
3153
315436. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3155 client certificate was expired. A simple patch fixes this, though I don't
3156 understand the full logic of why the verify callback is called multiple
3157 times.
3158
315937. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3160 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3161 which causes problems with some clients (such as the Certicom SSL Plus
3162 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3163 disables the coutermeasure allowing Eudora to connect."
3164
316538. Exim was not checking that a write() to a log file succeeded. This could
3166 lead to Bad Things if a log got too big, in particular if it hit a file
3167 size limit. Exim now panics and dies if it cannot write to a log file, just
3168 as it does if it cannot open a log file.
3169
317039. Modified OS/Makefile-Linux so that it now contains
3171
3172 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3173
3174 The two -D definitions ensure that Exim is compiled with large file
3175 support, which makes it possible to handle log files that are bigger than
3176 2^31.
3177
317840. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3179 instance) a domain was checked against a named list that involved a lookup,
3180 causing $domain_data to be set, then another domain was checked against the
3181 same list, then the first domain was re-checked, the value of $domain_data
3182 after the final check could be wrong. In particular, if the second check
3183 failed, it could be set empty. This bug probably also applied to
3184 $localpart_data.
3185
318641. The strip_trailing_dot option was not being applied to the address given
3187 with the -f command-line option.
3188
318942. The code for reading a message's header from the spool was incrementing
3190 $received_count, but never initializing it. This meant that the value was
3191 incorrect (doubled) while delivering a message in the same process in which
3192 it was received. In the most common configuration of Exim, this never
3193 happens - a fresh exec is done - but it can happen when
3194 deliver_drop_privilege is set.
3195
319643. When Exim logs an SMTP synchronization error - client data sent too soon -
3197 it now includes up to 150 characters of the unexpected data in the log
3198 line.
3199
320044. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3201 and building data strings. The size of both of these buffers was 10 000
3202 bytes - far larger than anybody would *ever* want, thought I. Needless to
3203 say, somebody hit the limit. I have increased the maximum line length to
3204 20 000 and the maximum data length of concatenated lines to 100 000. I have
3205 also fixed two bugs, because there was no checking on these buffers. Tsk,
3206 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3207 buffer is too small.
3208
320945. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3210 lsearch lookups. Now it does.
3211
321246. When parsing a route_list item in a manualroute router, a fixed-length
3213 buffer was used for the list of hosts. I made this 1024 bytes long,
3214 thinking that nobody would ever have a list of hosts that long. Wrong.
3215 Somebody had a whole pile of complicated expansion conditions, and the
3216 string was silently truncated, leading to an expansion error. It turns out
3217 that it is easier to change to an unlimited length (owing to other changes
3218 that have happened since this code was originally written) than to build
3219 structure for giving a limitation error. The length of the item that
3220 expands into the list of hosts is now unlimited.
3221
322247. The lsearch lookup could not handle data where the length of text line was
3223 more than 4095 characters. Such lines were truncated, leading to shortened
3224 data being returned. It should now handle lines of any length.
3225
322648. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3227 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3228 ACL").
3229
323049. Cosmetic tidy to scripts like exicyclog that are generated by globally
3231 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3232 no longer happens in comment lines. A list of replacements is now placed
3233 at the head of all of the source files, except those whose only change is
3234 to replace PERL_COMMAND in the very first #! line.
3235
323650. Replaced the slow insertion sort in queue.c, for sorting the list of
3237 messages on the queue, with a bottom-up merge sort, using code contributed
3238 by Michael Haardt. This should make operations like -bp somewhat faster on
3239 large queues. It won't affect queue runners, except when queue_run_in_order
3240 is set.
3241
324251. Installed eximstats 1.31 in the distribution.
3243
324452. Added support for SRV lookups to the dnslookup router.
3245
324653. If an ACL referred to $message_body or $message_body_end, the value was not
3247 reset for any messages that followed in the same SMTP session.
3248
324954. The store-handling optimization for building very long strings was not
3250 differentiating between the different store pools. I don't think this
3251 actually made any difference in practice, but I've tidied it.
3252
325355. While running the routers to verify a sender address, $sender_address
3254 was still set to the sender address. This is wrong, because when routing to
3255 send a bounce to the sender, it would be empty. Therefore, I have changed
3256 it so that, while verifying a sender address, $sender_address is set to <>.
3257 (There is no change to what happens when verifying a recipient address.)
3258
325956. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3260 target A or AAAA records (if not already returned) without resetting the
3261 qualify_single or search_parents options of the DNS resolver. These are
3262 inappropriate in this case because the targets of MX and SRV records must
3263 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3264 target that, when qualified, matched something in the local domain. These
3265 two options are now turned off when doing these lookups.
3266
326757. It seems that at least some releases of Reiserfs (which does not have the
3268 concept of a fixed number of inodes) returns zero and not -1 for the
3269 number of available inodes. This interacted badly with check_spool_inodes,
3270 which assumed that -1 was the "no such thing" setting. What I have done is
3271 to check that the total number of inodes is greater than zero before doing
3272 the test of how many are available.
3273
327458. When a "warn" ACL statement has a log_message modifier, the message is
3275 remembered, and not repeated. This is to avoid a lot of repetition when a
3276 message has many recipients that cause the same warning to be written.
3277 Howewer, Exim was preserving the list of already written lines for an
3278 entire SMTP session, which doesn't seem right. The memory is now reset if a
3279 new message is started.
3280
328159. The "rewrite" debugging flag was not showing the result of rewriting in the
3282 debugging output unless log_rewrite was also set.
3283
328460. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3285 of (int)(handle) when we know that handle contains (void *)(-1).
3286
328761. The Exim daemon panic-logs an error return when it closes the incoming
3288 connection. However "connection reset by peer" seems to be common, and
3289 isn't really an error worthy of noting specially, so that particular error
3290 is no long logged.
3291
329262. When Exim is trying to find all the local interfaces, it used to panic and
3293 die if the ioctl to get the interface flags failed. However, it seems that
3294 on at least one OS (Solaris 9) it is possible to have an interface that is
3295 included in the list of interfaces, but for which you get a failure error
3296 for this call. This happens when the interface is not "plumbed" into a
3297 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3298 failure of the "get flags" call assumes that the interface is down.
3299
330063. Added a ${eval10: operator, which assumes all numbers are decimal. This
3301 makes life easier for people who are doing arithmetic on fields extracted
3302 from dates, where you often get leading zeros that should not be
3303 interpreted as octal.
3304
330564. Added qualify_domain to the redirect router, to override the global
3306 setting.
3307
330865. If a pathologically long header line contained very many addresses (the
3309 report of this problem mentioned 10 000) and each of them was rewritten,
3310 Exim could use up a very large amount of memory. (It kept on making new
3311 copies of the header line as it rewrote, and never released the old ones.)
3312 At the expense of a bit more processing, the header rewriting function has
3313 been changed so that it no longer eats memory in this way.
3314
331566. The generation of the Received: header has been moved from the time that a
3316 message starts to be received, to the time that it finishes. The timestamp
3317 in the Received: header should now be very close to that of the <= log
3318 line. There are two side-effects of this change:
3319
3320 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3321 logged header lines no longer include the local Received: line, because
3322 it has not yet been created. The same applies to a copy of the message
3323 that is returned to a non-SMTP sender when a message is rejected.
3324
3325 (b) When a filter file is tested using -bf, no additional Received: header
3326 is added to the test message. After some thought, I decided that this
3327 is a bug fix.
3328
3329 This change does not affect the value of $received_for. It is still set
3330 after address rewriting, but before local_scan() is called.
3331
333267. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3333
333468. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3335 gave an unhelpful panic error message, and a defer error. I have managed to
3336 change this behaviour so that it now rejects any supplied certificate,
3337 which seems right, as the list of acceptable certificates is empty.
3338
333969. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3340 gave an unhelpful defer error. I have not managed to make this reject any
3341 supplied certificates, but the error message it gives is "no certificate
3342 supplied", which is not helpful.
3343
334470. exigrep's output now also includes lines that are not associated with any
3345 message, but which match the given pattern. Implemented by a patch from
3346 Martin Sluka, which also tidied up the Perl a bit.
3347
334871. Recipient callout verification, like sender verification, was using <> in
3349 the MAIL FROM command. This isn't really the right thing, since the actual
3350 sender may affect whether the remote host accepts the recipient or not. I
3351 have changed it to use the actual sender in the callout; this means that
3352 the cache record is now keyed on a recipient/sender pair, not just the
3353 recipient address. There doesn't seem to be a real danger of callout loops,
3354 since a callout by the remote host to check the sender would use <>.
3355 [SEE ABOVE: changed after hitting problems.]
3356
335772. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3358 temporary errors. However, in the case of such a code being given after
3359 the end of a data transmission (i.e. after ".") Exim was failing to write
3360 a retry record for the message. (Yes, there was some broken host that was
3361 actually sending 8xx at this point.)
3362
336373. An unknown lookup type in a host list could cause Exim to panic-die when
3364 the list was checked. (An example that provoked this was putting <; in the
3365 middle of a list instead of at the start.) If this happened during a DATA
3366 ACL check, a -D file could be left lying around. This kind of configuration
3367 error no longer causes Exim to die; instead it causes a defer errror. The
3368 incident is still logged to the main and panic logs.
3369
337074. Buglet left over from Exim 3 conversion. The message "too many messages
3371 in one connection" was written to the rejectlog but not the mainlog, except
3372 when address rewriting (yes!) was being logged.
3373
337475. Added write_rejectlog option.
3375
337676. When a system filter was run not as root (that is, when system_filter_user
3377 was set), the values of the $n variables were not being returned to the
3378 main process; thus, they were not subsequently available in the $sn
3379 variables.
3380
338177. Added +return_path_on_delivery log selector.
3382
338378. A connection timeout was being treated differently from recipients deferred
3384 when testing hosts_max_try with a message that was older than the host's
3385 retry timeout. (The host should not be counted, thus allowing all hosts to
3386 be tried at least once before bouncing.) This may have been the cause of an
3387 occasionally reported bug whereby a message would remain on the queue
3388 longer than the retry timeout, but would be bounced if a delivery was
3389 forced. I say "may" because I never totally pinned down the problem;
3390 setting up timeout/retry tests is difficult. See also the next item.
3391
339279. The ultimate address timeout was not being applied to errors that involved
3393 a combination of host plus message (for example, a timeout on a MAIL
3394 command). When an address resolved to a number of possible hosts, and they
3395 were not all tried for each delivery (e.g. because of hosts_max_try), a
3396 message could remain on the queue longer than the retry timeout.
3397
339880. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3399 Haardt.
3400
340181. Fixed an obscure SMTP outgoing bug which required at least the following
3402 conditions: (a) there was another message waiting for the same server;
3403 (b) the server returned 5xx to all RCPT commands in the first message so
3404 that the message was not completed; (c) the server dropped the connection
3405 or gave a negative response to the RSET that Exim sends to abort the
3406 transaction. The observed case was a dropped connection after DATA that had
3407 been sent in pipelining mode. That is, the server had advertised PIPELINING
3408 but was not implementing it correctly. The effect of the bug was incorrect
3409 behaviour, such as trying another host, and this could lead to a crash.
3410
3411
3412Exim version 4.30
3413-----------------
3414
3415 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3416 and daemon.c were passed as pointers to ints; they should have been
3417 pointers to socklen_t variables (which are typically unsigned ints).
3418
3419 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3420 fixed.
3421
3422 3. Fixed a really odd bug that affected only the testing scheme; patching a
3423 certain fixed string in the binary changed the value of another string that
3424 happened to be identical to the end of the original first string.
3425
3426 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3427 name", it returns that address as the IP address. On some operating
3428 systems (e.g. Solaris), it also passes back the IP address string as the
3429 "host name". However, on others (e.g. Linux), it passes back an empty
3430 string. Exim wasn't checking for this, and was changing the host name to an
3431 empty string, assuming it had been canonicized.
3432
3433 5. Although rare, it is permitted to have more than one PTR record for a given
3434 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3435 all the names associated with an address, because they do in Solaris.
3436 However, it seems that they do not in Linux for data that comes from the
3437 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3438 I found this out when I moved to a new Linux workstation and tried to run
3439 the Exim test suite.
3440
3441 To get round this problem I have changed the code so that it now does its
3442 own call to the DNS to look up PTR records when searching for a host name.
3443 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3444 addresses that are only in /etc/hosts are still found.
3445
3446 This behaviour is, however, controlled by an option called host_lookup_
3447 order, which defaults to "bydns:byaddr". If people want to use the other
3448 order, or indeed, just use one or the other means of lookup, they can
3449 specify it in this variable.
3450
3451 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3452 some operating systems, this comes back from gethostbyaddr() as an empty
3453 string, and this is what Exim used to test for. However, it seems that in
3454 other systems, "." is yielded. Exim now tests for this case too.
3455
3456 7. The values of check_spool_space and check_log_space are now held internally
3457 as a number of kilobytes instead of an absolute number of bytes. If a
3458 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3459 kilobyte. This means that much larger values can be stored.
3460
3461 8. Exim monitor: an attempt to get the action menu when not actually pointing
3462 at a message produces an empty menu entitled "No message selected". This
3463 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3464 no entries in it ("Shell widget menu has zero width and/or height"). So I
3465 have added a single, blank menu entry in this case.
3466
3467 9. Added ${quote_local_part.
3468
346910. MIME decoding is now applied to the contents of Subject: header lines when
3470 they are logged.
3471
347211. Now that a reference to $sender_host_address automatically causes a reverse
3473 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3474 host lookup before query-style lookups in lists that might use this
3475 variable. This has therefore been abolished, and the "net-" prefix is no
3476 longer necessary for query-style lookups.
3477
347812. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3479 have been a typo for LFLAGS, so it has been changed.
3480
348113. The install script calls Exim with "-C /dev/null" in order to find the
3482 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3483 to be output. Howeve, since Exim outputs its version number before the
3484 error, it didn't break the script. It just looked ugly. I fixed this by
3485 always allowing "-C /dev/null" if the caller is root.
3486
348714. Ignore overlarge ACL variable number when reading spool file - insurance
3488 against a later release with more variables having written the file.
3489
349015. The standard form for an IPv6 address literal was being rejected by EHLO.
3491 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3492 this, as well as the form without the "IPv6" on the front.
3493
349416. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3495 OS/Makefile-Darwin file.
3496
349717. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3498 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3499
350018. After change 4.21/52, "%ld" was used to format the contents of the $inode
3501 variable. However, some OS use ints for inodes. I've added cast to long int
3502 to get rid of the compiler warning.
3503
350419. I had forgotten to lock out "/../" in configuration file names when
3505 ALT_CONFIG_PREFIX was set.
3506
350720. Routers used for verification do not need to specify transports. However,
3508 if such a router generated a host list, and callout was configured, Exim
3509 crashed, because it could not find a port number from the (non-existent)
3510 transport. It now assumes port 25 in this circumstance.
3511
351221. Added the -t option to exigrep.
3513
351422. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3515 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3516 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3517 them is set, LOOKUP_LSEARCH is forced.
3518
351923. "exim -bV" now outputs a list of lookups that are included in the binary.
3520
352124. Added sender and host information to the "rejected by local_scan()" log
3522 line; previously there was no indication of these.
3523
352425. Added .include_if_exists.
3525
352626. Change 3.952/11 added an explicit directory sync on top of a file sync for
3527 Linux. It turns out that not all file systems support this. Apparently some
3528 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3529 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3530 not supported on the file descriptor, is now ignored when Exim is trying to
3531 sync a directory. This applies only to Linux.
3532
353327. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3534
353528. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3536 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3537 this.
3538
353929. Added support for maildirsize files from supplied patch (modified a bit).
3540
354130. The use of :fail: followed by an empty string could lead Exim to respond to
3542 sender verification failures with (e.g.):
3543
3544 550 Verification failed for <xxx>
3545 550 Sender verify failed
3546
3547 where the first response line was missing the '-' that indicates it is not
3548 the final line of the response.
3549
355031. The loop for finding the name of the user that called Exim had a hardwired
3551 limit of 10; it now uses the value of finduser_retries, which is used for
3552 all other user lookups.
3553
355432. Added $received_count variable, available in data and not_smtp ACLs, and at
3555 delivery time.
3556
355733. Exim was neglecting to zero errno before one call of strtol() when
3558 expanding a string and expecting an integer value. On some systems this
3559 resulted in spurious "integer overflow" errors. Also, it was casting the
3560 result into an int without checking.
3561
356234. Testing for a connection timeout using "timeout_connect" in the retry rules
3563 did not work. The code looks as if it has *never* worked, though it appears
3564 to have been documented since at least releast 1.62. I have made it work.
3565
356635. The "timeout_DNS" error in retry rules, also documented since at least
3567 1.62, also never worked. As it isn't clear exactly what this means, and
3568 clearly it isn't a major issue, I have abolished the feature by treating it
3569 as "timeout", and writing a warning to the main and panic logs.
3570
357136. The display of retry rules for -brt wasn't always showing the error code
3572 correctly.
3573
357437. Added new error conditions to retry rules: timeout_A, timeout_MX,
3575 timeout_connect_A, timeout_connect_MX.
3576
357738. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3578 to the empty sender.
3579
358039. The daemon was not analysing the content of -oX till after it had closed
3581 stderr and disconnected from the controlling terminal. This meant that any
3582 syntax errors were only noted on the panic log, and the return code from
3583 the command was 0. By re-arranging the code a little, I've made the
3584 decoding happen first, so such errors now appear on stderr, and the return
3585 code is 1. However, the actual setting up of the sockets still happens in
3586 the disconnected process, so errors there are still only recorded on the
3587 panic log.
3588
358940. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3590 connections (as happens on some IP stacks) was logged at start up time as
3591 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3592 it from "IPv6 and IPv4", which means that two separate sockets are being
3593 used.
3594
359541. The debug output for gethostbyname2() or getipnodebyname() failures now
3596 says whether AF_INET or AF_INET6 was passed as an argument.
3597
359842. Exiwhat output was messed up when time zones were included in log
3599 timestamps.
3600
360143. Exiwhat now gives more information about the daemon's listening ports,
3602 and whether -tls-on-connect was used.
3603
360444. The "port" option of the smtp transport is now expanded.
3605
360645. A "message" modifier in a "warn" statement in a non-message ACL was being
3607 silently ignored. Now an error message is written to the main and panic
3608 logs.
3609