BugZilla 438
[exim.git] / doc / doc-txt / ChangeLog
CommitLineData
a0540757 1$Cambridge: exim/doc/doc-txt/ChangeLog,v 1.447 2006/12/19 15:09:58 ph10 Exp $
495ae4b0
PH
2
3Change log file for Exim from version 4.21
4-------------------------------------------
5
944e9e9c
TF
6Exim version 4.64
7-----------------
aa41d2de 8
21d74bd9
TK
9TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
10 leftover -K file (the existence of which was triggered by #402).
11 While we were at it, introduced process PID as part of the -K
12 filename. This should rule out race conditions when creating
13 these files.
14
15TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
16 processing considerably. Previous code took too long for large mails,
17 triggering a timeout which in turn triggers #401.
18
19TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
20 in the DK code in transports.c. sendfile() is not really portable,
21 hence the _LINUX specificness.
944e9e9c
TF
22
23TF/01 In the add_headers option to the mail command in an Exim filter,
24 there was a bug that Exim would claim a syntax error in any
25 header after the first one which had an odd number of characters
26 in the field name.
27
2b1c6e3a
PH
28PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
29 callout verification, Exim cached a "reject" for the entire domain. This
30 is correct for most verifications, but it is not correct for a recipient
31 verification with use_sender or use_postmaster set, because in that case
32 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
33 case of MAIL FROM:<> rejection from other early rejections (e.g.
34 rejection of HELO). When verifying a recipient using a non-null MAIL
35 address, the cache is ignored if it shows MAIL FROM:<> rejection.
36 Whatever the result of the callout, the value of the domain cache is
37 left unchanged (for any other kind of callout, getting as far as trying
38 RCPT means that the domain itself is ok).
39
1f872c80
PH
40PH/02 Tidied a number of unused variable and signed/unsigned warnings that
41 gcc 4.1.1 threw up.
42
43PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
44 manifest itself as EPIPE rather than ECONNECT. When tidying away a
45 session, the daemon ignores ECONNECT errors and logs others; it now
46 ignores EPIPE as well.
47
d203e649
PH
48PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
49 (quoted-printable decoding).
50
cc2ed8f7 51PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
21a04aa3 52 later the small subsequent patch to fix an introduced bug.
f951fd57 53
ddfcd446
PH
54PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
55
d45b1de8
PH
56PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
57
58PH/08 An error is now given if message_size_limit is specified negative.
59
38a0a95f 60PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
641cb756 61 to be given (somewhat) arbitrary names.
38a0a95f 62
a2405d83
JJ
63JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
64 in 4.64-PH/09.
65
66JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
67 miscellaneous code fixes
68
6ea85e9a
PH
69PH/10 Added the log_reject_target ACL modifier to specify where to log
70 rejections.
71
26da7e20
PH
72PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
73 hostname. This is wrong, because it relates to the incoming message (and
74 probably the interface on which it is arriving) and not to the outgoing
75 callout (which could be using a different interface). This has been
76 changed to use the value of the helo_data option from the smtp transport
77 instead - this is what is used when a message is actually being sent. If
78 there is no remote transport (possible with a router that sets up host
79 addresses), $smtp_active_hostname is used.
6ea85e9a 80
14aa5a05 81PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
7befa435 82 tweaks were necessary in order to get it to work (see also 21 below):
14aa5a05
PH
83 (a) The code assumed that strncpy() returns a negative number on buffer
84 overflow, which isn't the case. Replaced with Exim's string_format()
85 function.
86 (b) There were several signed/unsigned issues. I just did the minimum
87 hacking in of casts. There is scope for a larger refactoring.
88 (c) The code used strcasecmp() which is not a standard C function.
89 Replaced with Exim's strcmpic() function.
90 (d) The code set only $1; it now sets $auth1 as well.
91 (e) A simple test gave the error "authentication client didn't specify
92 service in request". It would seem that Dovecot has changed its
93 interface. Fortunately there's a specification; I followed it and
94 changed what the client sends and it appears to be working now.
95
ff75a1f7
PH
96PH/13 Added $message_headers_raw to provide the headers without RFC 2047
97 decoding.
98
e6f6568e
PH
99PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
100 address A is aliased to B and C, where B exists and C does not. Without
101 -v the output is "A verified" because verification stops after a
102 successful redirection if more than one address is generated. However,
103 with -v the child addresses are also verified. Exim was outputting "A
104 failed to verify" and then showing the successful verification for C,
105 with its parentage. It now outputs "B failed to verify", showing B's
106 parentage before showing the successful verification of C.
107
d6f6e0dc
PH
108PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
109 look up a TXT record in a specific list after matching in a combined
110 list.
111
322050c2
PH
112PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
113 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
114 they consult the DNS. I had assumed they would set it the way they
115 wanted; and indeed my experiments on Linux seem to show that in some
116 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
117 To be on the safe side, however, I have now made the interface to
118 host_find_byname() similar to host_find_bydns(), with an argument
119 containing the DNS resolver options. The host_find_byname() function now
120 sets these options at its start, just as host_find_bydns() does. The smtp
121 transport options dns_qualify_single and dns_search_parents are passed to
122 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
123 of host_find_byname() use the default settings of RES_DEFNAMES
124 (qualify_single) but not RES_DNSRCH (search_parents).
125
08955dd3
PH
126PH/17 Applied (a modified version of) Nico Erfurth's patch to make
127 spool_read_header() do less string testing, by means of a preliminary
128 switch on the second character of optional "-foo" lines. (This is
129 overdue, caused by the large number of possibilities that now exist.
130 Originally there were few.) While I was there, I also converted the
131 str(n)cmp tests so they don't re-test the leading "-" and the first
132 character, in the hope this might squeeze out yet more improvement.
133
1eccaa59
PH
134PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
135 flag allowing group syntax was set by the header_syntax check but not
136 turned off, possible causing trouble later; (2) The flag was not being
137 set at all for the header_verify test, causing "group"-style headers to
138 be rejected. I have now set it in this case, and also caused header_
139 verify to ignore an empty address taken from a group. While doing this, I
140 came across some other cases where the code for allowing group syntax
141 while scanning a header line wasn't quite right (mostly, not resetting
142 the flag correctly in the right place). These bugs could have caused
143 trouble for malformed header lines. I hope it is now all correct.
144
602e59e5
PH
145PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
146 with the "reply" argument non-NULL. The code, however (which originally
147 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
148 but it didn't always do it. This confused somebody who was copying the
149 code for some other use. I have removed all the tests.
150
411ef850
PH
151PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
152 feature that was used to support insecure browsers during the U.S. crypto
153 embargo. It requires special client support, and Exim is probably the
154 only MTA that supported it -- and would never use it because real RSA is
155 always available. This code has been removed, because it had the bad
156 effect of slowing Exim down by computing (never used) parameters for the
157 RSA_EXPORT functionality.
158
7befa435
PH
159PH/21 On the advice of Timo Sirainen, added a check to the dovecot
160 authenticator to fail if there's a tab character in the incoming data
161 (there should never be unless someone is messing about, as it's supposed
162 to be base64-encoded). Also added, on Timo's advice, the "secured" option
163 if the connection is using TLS or if the remote IP is the same as the
164 local IP, and the "valid-client-cert option" if a client certificate has
165 been verified.
166
48da4259 167PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
16ff981e
PH
168 authenticators. This can be used for authorization after authentication
169 succeeds. (In the case of plaintext, it servers for both authentication
170 and authorization.)
171
48da4259
PH
172PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
173 if any retry times were supplied.
174
d1d5595c
PH
175PH/24 Exim crashed if verify=helo was activated during an incoming -bs
176 connection, where there is no client IP address to check. In this
177 situation, the verify now always succeeds.
178
0ef732d9
PH
179PH/25 Applied John Jetmore's -Mset patch.
180
328895cc
PH
181PH/26 Added -bem to be like -Mset, but loading a message from a file.
182
fd700877
PH
183PH/27 In a string expansion for a processed (not raw) header when multiple
184 headers of the same name were present, leading whitespace was being
185 removed from all of them, but trailing whitespace was being removed only
186 from the last one. Now trailing whitespace is removed from each header
f6c332bd
PH
187 before concatenation. Completely empty headers in a concatenation (as
188 before) are ignored.
fd700877 189
8dce1a6f
PH
190PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
191 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
192
17af4a17
PH
193PH/29 [Removed. This was a change that I later backed out, and forgot to
194 correct the ChangeLog entry (that I had efficiently created) before
195 committing the later change.]
f6c332bd
PH
196
197PH/30 Exim was sometimes attempting to deliver messages that had suffered
198 address errors (4xx response to RCPT) over the same connection as other
199 messages routed to the same hosts. Such deliveries are always "forced",
200 so retry times are not inspected. This resulted in far too many retries
201 for the affected addresses. The effect occurred only when there were more
202 hosts than the hosts_max_try setting in the smtp transport when it had
203 the 4xx errors. Those hosts that it had tried were not added to the list
204 of hosts for which the message was waiting, so if all were tried, there
205 was no problem. Two fixes have been applied:
206
207 (i) If there are any address or message errors in an SMTP delivery, none
208 of the hosts (tried or untried) are now added to the list of hosts
209 for which the message is waiting, so the message should not be a
210 candidate for sending over the same connection that was used for a
211 successful delivery of some other message. This seems entirely
212 reasonable: after all the message is NOT "waiting for some host".
213 This is so "obvious" that I'm not sure why it wasn't done
214 previously. Hope I haven't missed anything, but it can't do any
215 harm, as the worst effect is to miss an optimization.
216
217 (ii) If, despite (i), such a delivery is accidentally attempted, the
218 routing retry time is respected, so at least it doesn't keep
219 hammering the server.
220
c1114884
PH
221PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
222 in ${readsocket because some servers need this prod.
223
7a0743eb
PH
224PH/32 Added some extra debug output when updating a wait-xxx database.
225
0d85fa3f
PH
226PH/33 The hint "could be header name not terminated by colon", which has been
227 given for certain expansion errors for a long time, was not being given
228 for the ${if def:h_colon_omitted{... case.
229
1bf43b78
PH
230PH/34 The spec says: "With one important exception, whenever a domain list is
231 being scanned, $domain contains the subject domain." There was at least
232 one case where this was not true.
233
520de300
PH
234PH/35 The error "getsockname() failed: connection reset by peer" was being
235 written to the panic log as well as the main log, but it isn't really
236 panic-worthy as it just means the connection died rather early on. I have
237 removed the panic log writing for the ECONNRESET error when getsockname()
238 fails.
239
48c7f9e2
PH
240PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
241 runs only) independently of the message's sender address. This meant
242 that, if the 4xx error was in fact related to the sender, a different
243 message to the same recipient with a different sender could confuse
244 things. In particualar, this can happen when sending to a greylisting
245 server, but other circumstances could also provoke similar problems.
246 I have changed the default so that the retry time for these errors is now
247 based a combination of the sender and recipient addresses. This change
248 can be overridden by setting address_retry_include_sender=false in the
249 smtp transport.
250
99ea1c86
PH
251PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
252 remote server are returned as part of bounce messages. This was not
253 happening for LMTP over a pipe (the lmtp transport), but now it is the
254 same for both kinds of LMTP.
255
a2042e78
PH
256PH/38 Despite being documented as not happening, Exim was rewriting addresses
257 in header lines that were in fact CNAMEs. This is no longer the case.
258
4fbcfc2e
PH
259PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
260 and queue runs started by the daemon processed all messages. This has
261 been fixed so that -R and -S can now usefully be given with -q<time>.
262
aa41d2de
PH
263PH/40 Import PCRE release 6.7 (fixes some bugs).
264
af561417
PH
265PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
266
3cc66b45
PH
267PH/42 Give an error if -q is specified more than once.
268
194cc0e4
PH
269PH/43 Renamed the variables $interface_address and $interface_port as
270 $received_ip_address and $received_port, to make it clear that these
271 values apply to message reception, and not to the outgoing interface when
272 a message is delivered. (The old names remain recognized, of course.)
273
a401ddaa
PH
274PH/44 There was no timeout on the connect() call when using a Unix domain
275 socket in the ${readsocket expansion. There now is.
276
4e88a19f
PH
277PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
278 be meaningful with "accept".
279
d7d7b289
SC
280SC/01 Eximstats V1.43
281 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
282
283SC/02 Eximstats V1.44
284 Use a glob alias rather than an array ref in the generated
285 parser. This improves both readability and performance.
286
287SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
288 Collect SpamAssassin and rejection statistics.
289 Don't display local sender or destination tables unless
290 there is data to show.
291 Added average volumes into the top table text output.
292
293SC/04 Eximstats V1.46
294 Collect data on the number of addresses (recipients)
295 as well as the number of messages.
296
297SC/05 Eximstats V1.47
298 Added 'Message too big' to the list of mail rejection
299 reasons (thanks to Marco Gaiarin).
300
301SC/06 Eximstats V1.48
302 Mainlog lines which have GMT offsets and are too short to
303 have a flag are now skipped.
304
305SC/07 Eximstats V1.49 (Alain Williams)
306 Added the -emptyok flag.
307
308SC/08 Eximstats V1.50
309 Fixes for obtaining the IP address from reject messages.
310
0ea2a468
JJ
311JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
312 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
313 whitesspace changes from 4.64-PH/27
314
315JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
316 match 4.64-PH/13
317
318JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
319 are found, allow negative numbers in numeric criteria)
320
321JJ/06 exipick.20061117.2, added new $message_body_missing variable
322
323JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
324 to match changes made in 4.64-PH/43
325
8a10f5a4
PH
326PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
327
30e18802
PH
328PH/47 Put in an explicit test for a DNS lookup of an address record where the
329 "domain" is actually an IP address, and force a failure. This locks out
330 those revolvers/nameservers that support "A-for-A" lookups, in
331 contravention of the specifications.
332
55728a4f
PH
333PH/48 When a host name was looked up from an IP address, and the subsequent
334 forward lookup of the name timed out, the host name was left in
335 $sender_host_name, contrary to the specification.
d7d7b289 336
d7837193
PH
337PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
338 restricted to single-key lookups, Exim was not diagnosing an error if
339 * or *@ was used with a query-style lookup.
340
87054a31
PH
341PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
342
ea2c01d2
MH
343MH/01 local_scan ABI version incremented to 1.1. It should have been updated
344 long ago, but noone interested enough thought of it. Let's just say that
345 the "1.1" means that there are some new functions that weren't there at
346 some point in the past.
347
e4fa6968
PH
348PH/51 Error processing for expansion failure of helo_data from an smtp
349 transport during callout processing was broken.
350
56f5d9bd
PH
351PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
352 tested/used via the -bh/-bhc/-bs options.
353
922e1c28
PH
354PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
355 bug, fixed in subsequent PCRE releases).
356
21eb6e72
PH
357PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
358 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
359
a0540757
PH
360PH/55 Check for a ridiculously long file name in exim_dbmbuild.
361
944e9e9c 362
478be7b0
SC
363Exim version 4.63
364-----------------
365
366SC/01 Use a glob alias rather than an array ref in eximstats generated
367 parser. This improves both readability and performance.
368
369SC/02 Collect SpamAssassin and rejection statistics in eximstats.
370 Don't display local sender or destination tables in eximstats unless
371 there is data to show.
372 Added average volumes into the eximstats top table text output.
373
374SC/03 Collect data on the number of addresses (recipients) as well
375 as the number of messages in eximstats.
376
2b965a65
TF
377TF/01 Correct an error in the documentation for the redirect router. Exim
378 does (usually) call initgroups() when daemonizing.
478be7b0 379
45b91596
PH
380TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
381 with consistent privilege compared to when running as a daemon.
478be7b0 382
c59f5781
TF
383TF/03 Note in the spec that $authenticated_id is not set for local
384 submissions from trusted users.
385
90fc3069
TF
386TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
387 Thanks to Dean Brooks <dean@iglou.com> for the patch.
388
6083aca0
TF
389TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
390 by adding some example configuration directives to the default
391 configuration file. A little bit of work is required to uncomment the
392 directives and define how usernames and passwords are checked, but
393 there is now a framework to start from.
394
765b530f
PH
395PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
396 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
397 without this. I don't know how relevant this is to other LDAP libraries.
398
4e167a8c
PH
399PH/02 Add the verb name to the "unknown ACL verb" error.
400
4608d683
PH
401PH/03 Magnus Holmgren's patch for filter_prepend_home.
402
b8dc3e4a
PH
403PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
404
5418e93b
PH
405PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
406 directory not expanded when it should be if an expanded home directory
407 was set for the address (which is overridden by the transport).
408
b4a9bda2
PH
409PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
410 libradius.
411
45b91596
PH
412PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
413 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
414 because it is too late at that time, and has no effect.
415
5547e2c5
PH
416PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
417 security issue with \' (bugzilla #107). I could not use the
418 PQescapeStringConn() function, because it needs a PGconn value as one of
419 its arguments.
420
dbcef0ea
PH
421PH/08 When testing addresses using -bt, indicate those final addresses that
422 are duplicates that would not cause an additional delivery. At least one
423 person was confused, thinking that -bt output corresponded to deliveries.
424 (Suppressing duplicates isn't a good idea as you lose the information
425 about possibly different redirections that led to the duplicates.)
426
25257489
PH
427PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
428 systems where poll() doesn't work, in particular OS X.
429
c816d124
PH
430PH/10 Added more information to debugging output for retry time not reached.
431
a9ccd69a
PH
432PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
433 operations in malware.c.
434
75fa1910
PH
435PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
436 signatures.
437
a7d7aa58
PH
438PH/13 If write_rejectlog was set false when logging was sent to syslog with
439 syslog_duplication set false, log lines that would normally be written
440 both the the main log and to the reject log were not written to syslog at
441 all.
442
42119b09
PH
443PH/14 In the default configuration, change the use of "message" in ACL warn
444 statements to "add_header".
445
41609df5
PH
446PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
447 not followed by a command (e.g. "seen endif").
448
a5bd321b
PH
449PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
450 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
451 latter.
452
e85a7ad5 453PH/17 Added extra conditions to the default value of delay_warning_condition
5dff5817
PH
454 so that it is now:
455
e85a7ad5
PH
456 ${if or { \
457 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
458 { match{$h_precedence:}{(?i)bulk|list|junk} } \
459 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
5dff5817
PH
460 }{no}{yes}}
461
e85a7ad5
PH
462 The Auto-Submitted: and various List- headers are standardised, whereas I
463 don't think Precedence: ever was.
5dff5817 464
d8fe1c03
PH
465PH/18 Refactored debugging code in route_finduser() to show more information,
466 in particular, the error code if getpwnam() issues one.
467
16282d2b
PH
468PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
469 This is apparently needed in addition to the PH/07 change above to avoid
470 any possible encoding problems.
471
35d40a98
PH
472PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
473 but not after initializing Perl.
474
034d99ab
PH
475PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
476 output them only if debugging. By default they are written stderr,
477 apparently, which is not desirable.
478
6ec97b1b
PH
479PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
480 queries.
481
e22ca4ac
JJ
482JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
483 --not options
484
485JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
486
33d73e3b
PH
487PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
488 authenticated or an ident call has been made. Suppress the default
489 values for $authenticated_id and $authenticated_sender (but permit -oMai
490 and -oMas) when testing with -bh.
491
9ecb03f3
PH
492PH/24 Re-jigged the order of the tests in the default configuration so that the
493 tests for valid domains and recipients precede the DNS black list and CSA
494 tests, on the grounds that those ones are more expensive.
495
084efe8d
PH
496PH/25 Exim was not testing for a space following SMTP commands such as EHLO
497 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
498 This bug exists in every version of Exim that I still have, right back to
499 0.12.
500
366fc9f0
PH
501PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
502 However, an attempt to turn on case-sensitivity in a regex key by
503 including (?-i) didn't work because the subject string was already
504 lowercased, and the effects were non-intuitive. It turns out that a
505 one-line patch can be used to allow (?-i) to work as expected.
506
c59f5781 507
c887c79e
TF
508Exim version 4.62
509-----------------
510
511TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
512 other effects) broke the use of negated acl sub-conditions.
513
1cce3af8
PH
514PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
515 patch).
516
afb3eaaf
PH
517PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
518 "Deny" causes Exim to reject the incoming connection with a 554 error.
519 Unfortunately, if there is a major crisis, such as a disk failure,
520 tcp-wrappers gives "deny", whereas what one would like would be some
521 kind of temporary error. A kludge has been added to help with this.
522 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
523 554 error is used if errno is still zero or contains ENOENT (which occurs
524 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
525 451 error is used.
526
e173618b
PH
527PH/03 Add -lutil to the default FreeBSD LIBS setting.
528
dd16e114
PH
529PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
530 errors. Otherwise a message that provokes a temporary error (when other
531 messages do not) can cause a whole host to time out.
532
f7fd3850
PH
533PH/05 Batch deliveries by appendfile and pipe transports did not work when the
534 addresses were routed directly to files or pipes from a redirect router.
535 File deliveries just didn't batch; pipe deliveries might have suffered
536 odd errors.
537
d87df92c
PH
538PH/06 A failure to get a lock for a hints database would erroneously always say
539 "Failed to get write lock", even when it was really a read lock.
540
7e9f683d
PH
541PH/07 The appendfile transport was creating MBX lock files with a fixed mode
542 of 0600. This has been changed to use the value of the lockfile_mode
543 option (which defaults to 0600).
544
bfad5236
PH
545PH/08 Applied small patch from the Sieve maintainer.
546
01c490df
PH
547PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
548 folder from quota calculations, a direct delivery into this folder messed
549 up the contents of the maildirsize file. This was because the regex was
550 used only to exclude .Trash (or whatever) when the size of the mailbox
551 was calculated. There was no check that a delivery was happening into an
552 excluded directory. This bug has been fixed by ignoring all quota
553 processing for deliveries into excluded directories.
554
d6629cdc
PH
555PH/10 Added the maildirfolder_create_regex option to appendfile.
556
1cce3af8 557
214e2000
PH
558Exim version 4.61
559-----------------
560
561PH/01 The code for finding all the local interface addresses on a FreeBSD
562 system running IPv6 was broken. This may well have applied to all BSD
563 systems, as well as to others that have similar system calls. The broken
564 code found IPv4 interfaces correctly, but gave incorrect values for the
565 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
566 that it would not match correctly against @[] and not recognize the IPv6
567 addresses as local.
568
f9daeae0
PH
569PH/02 The ipliteral router was not recognizing addresses of the form user@
570 [ipv6:....] because it didn't know about the "ipv6:" prefix.
571
7e66e54d
PH
572PH/03 Added disable_ipv6.
573
c8ea1597
PH
574PH/04 Changed $reply_address to use the raw form of the headers instead of the
575 decoded form, because it is most often used to construct To: headers
576 lines in autoreplies, and the decoded form may well be syntactically
577 invalid. However, $reply_address has leading white space removed, and all
578 newlines turned into spaces so that the autoreply transport does not
579 grumble.
580
911f6fde
PH
581PH/05 If group was specified without a user on a router, and no group or user
582 was specified on a transport, the group from the router was ignored.
583
47ca6d6c
PH
584PH/06 Increased the number of ACL variables to 20 of each type, and arranged
585 for visible compile-time settings that can be used to change these
586 numbers, for those that want even more. Backwards compatibility with old
587 spool files has been maintained. However, going back to a previous Exim
588 release will lost any variables that are in spool files.
589
ed0e9820
PH
590PH/07 Two small changes when running in the test harness: increase delay when
591 passing a TCP/IP connection to a new process, in case the original
592 process has to generate a bounce, and remove special handling of
593 127.0.0.2 (sic), which is no longer necessary.
594
eff37e47
PH
595PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
596 be the same on different OS.
597
1921d2ea
PH
598PH/09 Moved a debug statement in filter processing to avoid a race problem when
599 testing.
600
b3f69ca8
JJ
601JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
602 whether --show-vars was specified or not
603
604JJ/02 exipick: Added support for new ACL variable spool format introduced
605 in 4.61-PH/06
606
424a1c63
PH
607PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
608 syntactically invalid From: or Reply-to: line, and a filter used this to
609 generate an autoreply, and therefore failed to obtain an address for the
610 autoreply, Exim could try to deliver to a non-existent relative file
611 name, causing unrelated and misleading errors. What now happens is that
612 it logs this as a hard delivery error, but does not attempt to create a
613 bounce message.
614
7a100415
PH
615PH/11 The exinext utility has a -C option for testing purposes, but although
616 the given file was scanned by exinext itself; it wasn't being passed on
617 when Exim was called.
618
19b9dc85
PH
619PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
620 an end-of-file indication when reading a command response.
621
309bd837
PH
622PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
623 compiled. In many other places in Exim, IPv6 addresses are always
624 recognized, so I have changed this. It also means that IPv4 domain
625 literals of the form [IPV4:n.n.n.n] are now always recognized.
626
59e82a2a
PH
627PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
628 used if the router is not running as root, for example, when verifying at
629 ACL time, or when using -bh. The debugging output from this situation was
630 non-existent - all you got was a failure to exec. I have made two
631 changes:
632
633 (a) Failures to set uid/gid, the current directory, or a process leader
634 in a subprocess such as that created by queryprogram now generate
635 suitable debugging ouput when -d is set.
636
637 (b) The queryprogram router detects when it is not running as root,
638 outputs suitable debugging information if -d is set, and then runs
639 the subprocess without attempting to change uid/gid.
640
9edc04ce
PH
641PH/15 Minor change to Makefile for building test_host (undocumented testing
642 feature).
643
1349e1e5
PH
644PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
645 additional section of a DNS packet that returns MX or SRV records.
646 Instead, it always explicitly searches for A/AAAA records. This avoids
647 major problems that occur when a DNS server includes only records of one
648 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
649 fixed another bug: if SRV records were looked up and the corresponding
650 address records were *not* found in the additional section, the port
651 values from the SRV records were lost.
652
ea49d0e1
PH
653PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
654 using the correct key (the original address) when searching the retry
655 rules in order to find which one to use for generating the retry hint.
656
064a94c9
PH
657PH/18 If quota_warn_message contains a From: header, Exim now refrains from
658 adding the default one. Similarly, if it contains a Reply-To: header, the
659 errors_reply_to option, if set, is not used.
660
727071f8
PH
661PH/19 When calculating a retry time, Exim used to measure the "time since
662 failure" by looking at the "first failed" field in the retry record. Now
663 it does not use this if it is later than than the arrival time of the
664 message. Instead it uses the arrival time. This makes for better
665 behaviour in cases where some deliveries succeed, thus re-setting the
666 "first failed" field. An example is a quota failure for a huge message
667 when small messages continue to be delivered. Without this change, the
668 "time since failure" will always be short, possible causing more frequent
669 delivery attempts for the huge message than are intended.
dd16e114 670 [Note: This change was subsequently modified - see PH/04 for 4.62.]
727071f8 671
f78eb7c6
PH
672PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
673 $1, $2, $3) because the numerical variables can be reset during some
674 expansion items (e.g. "match"), thereby losing the authentication data.
675
21c28500
PH
676PH/21 Make -bV show the size of off_t variables so that the test suite can
677 decide whether to run tests for quotas > 2G.
678
679PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
680 mailbox_size, and mailbox_filecount in the appendfile transport. If a
681 filecount value is greater than 2G or if a quota value is greater than 2G
682 on a system where the size of off_t is not greater than 4, a panic error
683 is given.
684
1688f43b
PH
685PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
686 never match. The debug and -bh output now contains an explicit error
687 message indicating a malformed IPv4 address or mask.
688
689PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
690 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
691 PH/23 above applies.
692
9675b384
PH
693PH/25 Do not write to syslog when running in the test harness. The only
694 occasion when this arises is a failure to open the main or panic logs
695 (for which there is an explicit test).
696
6a3f1455
PH
697PH/26 Added the /no_tell option to "control=freeze".
698
dac79d3e
PH
699PH/27 If a host name lookup failed very early in a connection, for example, if
700 the IP address matched host_lookup and the reverse lookup yielded a name
701 that did not have a forward lookup, an error message of the form "no IP
702 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
703 could be logged. Now it outputs the IP address instead of "NULL".
1349e1e5 704
5977a0b3
PH
705PH/28 An enabling patch from MH: add new function child_open_exim2() which
706 allows the sender and the authenticated sender to be set when
707 submitting a message from within Exim. Since child_open_exim() is
708 documented for local_scan(), the new function should be too.
709
c91535f3
PH
710PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
711 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
712 results in an empty string is now treated as unset.
713
0d46a8c8
PH
714PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
715
278c6e6c
PH
716PH/31 Added sender_verify_fail logging option.
717
2cbb4081
PH
718PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
719 needed by RFC 822 but not by RFC 2822 was commented out. I have now
720 tidied the source and removed it altogether.
721
3eef829e
PH
722PH/33 When a queue run was abandoned because the load average was too high, a
723 log line was always written; now it is written only if the queue_run log
724 selector is set. In addition, the log line for abandonment now contains
725 information about the queue run such as the pid. This is always present
726 in "start" and "stop" lines but was omitted from the "abandon" line.
727
1ab95fa6
PH
728PH/34 Omit spaces between a header name and the colon in the error message that
729 is given when verify = headers_syntax fails (if there are lots of them,
730 the message gets confusing).
731
230205fc
PH
732PH/35 Change the default for dns_check_names_pattern to allow slashes within
733 names, as there are now some PTR records that contain slashes. This check
734 is only to protect against broken name servers that fall over on strange
735 characters, so the fact that it applies to all lookups doesn't matter.
736
75e0e026
PH
737PH/36 Now that the new test suite is complete, we can remove some of the
738 special code in Exim that was needed for the old test suite. For example,
739 sorting DNS records because real resolvers return them in an arbitrary
740 order. The new test suite's fake resolver always returns records in the
741 same order.
742
743PH/37 When running in the test harness, use -odi for submitted messages (e.g.
744 bounces) except when queue_only is set, to avoid logging races between
745 the different processes.
746
145396a6
PH
747PH/38 Panic-die if .include specifies a non-absolute path.
748
3cd34f13
PH
749PH/39 A tweak to the "H" retry rule from its user.
750
11121d3d
JJ
751JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
752 a label. They prevented compilation on older perls.
753
754JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
755 a warning to be raised on newish perls.
756
757JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
758 on queue. Changes to match documented behaviour of showing count of
759 messages matching specified criteria.
760
8def5aaf
PH
761PH/40 Changed the default ident timeout from 30s to 5s.
762
929ba01c
PH
763PH/41 Added support for the use of login_cap features, on those BSD systems
764 that have them, for controlling the resources used by pipe deliveries.
765
2632889e
PH
766PH/42 The content-scanning code uses fopen() to create files in which to put
767 message data. Previously it was not paying any attention to the mode of
768 the files. Exim runs with umask(0) because the rest of the code creates
769 files with open(), and sets the required mode explicitly. Thus, these
770 files were ending up world-writeable. This was not a big issue, because,
771 being within the spool directory, they were not world-accessible. I have
772 created a function called modefopen, which takes an additional mode
773 argument. It sets umask(777), creates the file, chmods it to the required
774 mode, then resets the umask. All the relevant calls to fopen() in the
775 content scanning code have been changed to use this function.
776
944a9c55
PH
777PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
778 to 24 hours. This avoids potential overflow problems when processing G
779 and H retry rules. I suspect nobody ever tinkers with this value.
780
4a23603b
PH
781PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
782
4730f942
PH
783PH/45 When the plaintext authenticator is running as a client, the server's
784 challenges are checked to ensure they are valid base64 strings. By
785 default, the authentication attempt is cancelled if an invalid string is
786 received. Setting client_ignore_invalid_base64 true ignores these errors.
787 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
788 they are received. Thus, the responses can be made to depend on the
789 challenges. If an invalid string is ignored, an empty string is placed in
790 the variable.
791
30dba1e6
PH
792PH/46 Messages that are created by the autoreply transport now contains a
793 References: header, in accordance with RFCs 2822 and 3834.
794
382afc6b
PH
795PH/47 Added authenticated_sender_force to the smtp transport.
796
a86229cf
PH
797PH/48 The ${prvs expansion was broken on systems where time_t was long long.
798
50c99ba6
PH
799PH/49 Installed latest patch from the Sieve maintainer.
800
d35e429d
PH
801PH/50 When an Exim quota was set without a file count quota, and mailbox_size
802 was also set, the appendfile transport was unnecessarily scanning a
803 directory of message files (e.g. for maildir delivery) to find the count
804 of files (along with the size), even though it did not need this
805 information. It now does the scan only if it needs to find either the
806 size of the count of files.
807
f90d018c
PH
808PH/51 Added ${time_eval: to convert Exim time strings into seconds.
809
75def545
PH
810PH/52 Two bugs concerned with error handling when the smtp transport is
811 used in LMTP mode:
812
813 (i) Exim was not creating retry information for temporary errors given
814 for individual recipients after the DATA command when the smtp transport
815 was used in LMTP mode. This meant that they could be retried too
816 frequently, and not timed out correctly.
817
818 (ii) Exim was setting the flag that allows error details to be returned
819 for LMTP errors on RCPT commands, but not for LMTP errors for individual
820 recipients that were returned after the DATA command.
821
822PH/53 This is related to PH/52, but is more general: for any failing address,
823 when detailed error information was permitted to be returned to the
824 sender, but the error was temporary, then after the final timeout, only
825 "retry timeout exceeded" was returned. Now it returns the full error as
826 well as "retry timeout exceeded".
827
c46782ef
PH
828PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
829 do this, and (what is worse) MTAs that accept it.
830
71fafd95
PH
831PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
832 will now be deprecated.
833
2c5db4fd
PH
834PH/56 New os.c-cygwin from the Cygwin maintainer.
835
9cf6b11a
JJ
836JJ/06 exipick: added --unsorted option to allow unsorted output in all output
837 formats (previously only available in exim formats via -bpr, -bpru,
838 and -bpra. Now also available in native and exiqgrep formats)
839
840JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
841 with very large, slow to parse queues
842
843JJ/08 exipick: added ! as generic prefix to negate any criteria format
844
845JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
846
898d150f
PH
847PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
848 responses to authentication challenges, though it was showing the
849 challenges; (ii) I've removed the CR characters from the debug output for
850 SMTP output lines.
851
46218253
PH
852PH/58 Allow for the insertion of a newline as well as a space when a string
853 is turned into more than one encoded-word during RFC 2047 encoding. The
854 Sieve code now uses this.
855
e97957bc
PH
856PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
857 data_4xx, lost_connection, tls_required.
858
81e509d7
PH
859PH/60 When a VRFY deferred or FAILED, the log message rather than the user
860 message was being sent as an SMTP response.
861
3d240ff7
PH
862PH/61 Add -l and -k options to exicyclog.
863
b37c4101
PH
864PH/62 When verifying, if an address was redirected to one new address, so that
865 verification continued, and the new address failed or deferred after
866 having set something in $address_data, the value of $address_data was not
867 passed back to the ACL. This was different to the case when no
868 redirection occurred. The value is now passed back in both cases.
869
79378e0f
PH
870PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
871 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
872 use login_cap.h, so on its own it isn't the distinguishing feature. The
873 new name refers directly to the setclassresources() function.
874
e49c7bb4
PH
875PH/65 Added configuration files for NetBSD3.
876
d114ec46
PH
877PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
878
f3d7df6c
PH
879PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
880 is preferred over IPv4.
881
715ab376
PH
882PH/68 The bounce_return_message and bounce_return_body options were not being
883 honoured for bounces generated during the reception of non-SMTP messages.
884 In particular, this applied to messages rejected by the ACL. This bug has
885 been fixed. However, if bounce_return_message is true and bounce_return_
886 body is false, the headers that are returned for a non-SMTP message
887 include only those that have been read before the error was detected.
888 (In the case of an ACL rejection, they have all been read.)
889
6b31b150
PH
890PH/69 The HTML version of the specification is now built in a directory called
891 spec_html instead of spec.html, because the latter looks like a path with
892 a MIME-type, and this confuses some software.
893
894PH/70 Catch two compiler warnings in sieve.c.
895
d515a917
PH
896PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
897 function verify_get_ident() calls ip_connect() to connect a socket, but
898 if the "connect()" function timed out, ip_connect() used to close the
899 socket. However, verify_get_ident() also closes the socket later, and in
900 between Exim writes to the log, which may get opened at this point. When
901 the socket was closed in ip_connect(), the log could get the same file
902 descriptor number as the socket. This naturally causes chaos. The fix is
903 not to close the socket in ip_connect(); the socket should be closed by
904 the function that creates it. There was only one place in the code where
905 this was missing, in the iplookup router, which I don't think anybody now
906 uses, but I've fixed it anyway.
907
9b8fadde
PH
908PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
909 well as to direct DNS lookups. Otherwise the handling of names in host
910 lists is inconsistent and therefore confusing.
911
214e2000 912
5de37277
PH
913Exim version 4.60
914-----------------
915
cc38ddbf
PH
916PH/01 Two changes to the default runtime configuration:
917
918 (1) Move the checks for relay_from_hosts and authenticated clients from
919 after to before the (commented out) DNS black list checks.
920
921 (2) Add control=submission to the relay_from_hosts and authenticated
922 clients checks, on the grounds that messages accepted by these
923 statements are most likely to be submissions.
5de37277 924
72fdd6ae
PH
925PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
926
927 (1) Generate an error if the third argument for the ${prvs expansion is
928 not a single digit.
929
930 (2) Treat a missing third argument of ${prvscheck as if it were an empty
931 string.
932
933 (3) Reset the variables that are obtained from the first argument of
934 ${prvscheck and used in the second argument before leaving the code,
935 because their memory is reclaimed, so using them afterwards may do
936 silly things.
937
938 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
939 one (it's much easier than Tom thought :-).
940
941 (5) Because of (4), we can now allow for the use of $prvscheck_result
942 inside the third argument.
cb9328de 943
cb741023
PH
944PH/03 For some reason, the default setting of PATH when running a command from
945 a pipe transport was just "/usr/bin". I have changed it to
946 "/bin:/usr/bin".
947
f174f16e
PH
948PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
949 anything to be listed in the output from -bV.
b2f5a032 950
c25242d7
PH
951PH/05 When a filter generated an autoreply, the entire To: header line was
952 quoted in the delivery log line, like this:
953
954 => >A.N.Other <ano@some.domain> <original@ddress> ...
955
956 This has been changed so that it extracts the operative address. There
957 may be more than one such address. If so, they are comma-separated, like
958 this:
959
960 => >ano@some.domain,ona@other.domain <original@ddress> ...
961
82c19f95
PH
962PH/06 When a client host used a correct literal IP address in a HELO or EHLO
963 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
964 not being looked up in the rDNS to get a host name, Exim was showing the
965 IP address twice in Received: lines, even though the IP addresses were
966 identical. For example:
967
968 Received: from [1.2.3.4] (helo=[1.2.3.4])
969
970 However, if the real host name was known, it was omitting the HELO data
971 if it matched the actual IP address. This has been tidied up so that it
972 doesn't show the same IP address twice.
973
d7ffbc12
PH
974PH/07 When both +timestamp and +memory debugging was on, the value given by
975 $tod_xxx expansions could be wrong, because the tod_stamp() function was
976 called by the debug printing, thereby overwriting the timestamp buffer.
977 Debugging no longer uses the tod_stamp() function when +timestamp is set.
978
9f526266
PH
979PH/08 When the original message was included in an autoreply transport, it
980 always said "this is a copy of the message, including all the headers",
981 even if body_only or headers_only was set. It now gives an appropriate
982 message.
983
87fcc8b9
PH
984PH/09 Applied a patch from the Sieve maintainer which:
985
986 o fixes some comments
987 o adds the (disabled) notify extension core
988 o adds some debug output for the result of if/elsif tests
989 o points to the current vacation draft in the documentation
990 and documents the missing references header update
991
992 and most important:
993
994 o fixes a bug in processing the envelope test (when testing
995 multiple envelope elements, the last element determinted the
996 result)
997
456682f5
PH
998PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
999 Electronic Mail") by including:
1000
1001 Auto-submitted: auto-generated
1002
1003 in the messages that it generates (bounce messages and others, such as
1004 warnings). In the case of bounce messages for non-SMTP mesages, there was
1005 also a typo: it was using "Auto_submitted" (underscore instead of
1006 hyphen). Since every message generated by Exim is necessarily in response
1007 to another message, thes have all been changed to:
1008
1009 Auto-Submitted: auto-replied
1010
1011 in accordance with these statements in the RFC:
1012
1013 The auto-replied keyword:
1014
1015 - SHOULD be used on messages sent in direct response to another
1016 message by an automatic process,
1017
1018 - MUST NOT be used on manually-generated messages,
1019
1020 - MAY be used on Delivery Status Notifications (DSNs) and Message
1021 Disposition Notifications (MDNs),
1022
1023 - MUST NOT be used on messages generated by automatic or periodic
1024 processes, except for messages which are automatic responses to
1025 other messages.
1026
3e46c1aa
PH
1027PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1028 to the default Received: header definition.
456682f5 1029
49826d12
PH
1030PH/12 Added log selector acl_warn_skipped (default on).
1031
eba0c039
PH
1032PH/13 After a successful wildlsearch lookup, discard the values of numeric
1033 variables because (a) they are in the wrong storage pool and (b) even if
1034 they were copied, it wouldn't work properly because of the caching.
1035
a0d6ba8a
PH
1036PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1037 checking when decoding. Apparently there are clients that generate
1038 overlong encoded strings. Why am I not surprised?
1039
f0917727
PH
1040PH/15 If the first argument of "${if match_address" was not empty, but did not
1041 contain an "@" character, Exim crashed. Now it writes a panic log message
1042 and treats the condition as false.
1043
096fee00
PH
1044PH/16 In autoreply, treat an empty string for "once" the same as unset.
1045
024bd3c2
PH
1046PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1047 extension "envelope-auth". The code is finished and in agreement with
1048 other implementations, but there is no documentation so far and in fact,
1049 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1050 not changing the active code.
1051
1052 Print executed "if" and "elsif" statements when debugging is used. This
1053 helps a great deal to understand what a filter does.
1054
1055 Document more things not specified clearly in RFC3028. I had all this
1056 sorted out, when out of a sudden new issues came to my mind. Oops."
1057
df199fec
PH
1058PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1059 (Bugzilla #53).
1060
d27f1df3
PH
1061PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1062 canonical form (as documented). However, after a host name lookup from
1063 the IP address, check_host() was doing a simple string comparison with
1064 addresses acquired from the DNS when checking that the found name did
1065 have the original IP as one of its addresses. Since any found IPv6
1066 addresses are likely to be in abbreviated form, the comparison could
1067 fail. Luckily, there already exists a function for doing the comparison
1068 by converting both addresses to binary, so now that is used instead of
1069 the text comparison.
1070
96776534
PH
1071PH/20 There was another similar case to PH/19, when a complete host name was
1072 given in a host list; looking up its IP address could give an abbreviated
1073 form, whereas the current host's name might or might not be abbreviated.
1074 The same fix has been applied.
1075
5de37277 1076
9a799bc0
PH
1077Exim version 4.54
1078-----------------
1079
1080PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1081 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1082 It now does.
1083
99a4b039
PH
1084PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1085 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1086
1087PH/03 Typo: missing ".o" in src/pcre/Makefile.
1088
4b233853
PH
1089PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1090 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1091 Also, for "Auto-Submitted", treat anything other than "no" as
1092 non-personal, in accordance with RFC 3834. (Previously it treated
1093 anything starting "auto-" as non-personal.)
1094
8857ccfd
PH
1095TF/01 The control=submission/name=... option had a problem with syntax
1096 errors if the name included a slash character. The /name= option
1097 now slurps the rest of the string, so it can include any characters
1098 but it must come last in the list of options (after /sender_retain
1099 or /domain=).
1100
433a2980
PH
1101PH/05 Some modifications to the interface to the fake nameserver for the new
1102 testing suite.
1103
3e46c1aa 1104
9a799bc0 1105
e3a311ba
TK
1106Exim version 4.53
1107-----------------
1108
1109TK/01 Added the "success_on_redirect" address verification option. See
1110 NewStuff for rationale and an example.
1111
13b685f9
PH
1112PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1113
395ff96d
PH
1114PH/02 Patch to exigrep to allow it to work on syslog lines.
1115
5b68f6e4
PH
1116PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1117 fread() to skip over the body file's header line, because in Cygwin the
1118 header line is locked and is inaccessible.
1119
1ab52c69
PH
1120PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1121 co-exist for some time) to make it clear that it is the Exim ID that is
1122 referenced, not the Message-ID: header line.
1123
b07e6aa3
PH
1124PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1125 string_format() function, because snprintf() does not exist on all
1126 operating systems.
1127
254e032f
PH
1128PH/06 The use of forbid_filter_existstest now also locks out the use of the
1129 ${stat: expansion item.
1130
3af76a81
PH
1131PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1132 protocol synchronization error", to keep the pedants happy.
1133
2548ba04
PH
1134PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1135 well as for IRIX systems, when gcc is being used. See the host.c source
1136 file for comments.
1137
b6c6011d
PH
1138PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1139
cf39cf57
PH
1140PH/10 Named domain lists were not working if used in a queue_smtp_domains
1141 setting.
1142
f1513293
PH
1143PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1144 transport and to the smtp transport in LMTP mode.
1145
727549a4
PH
1146TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1147
af46795e
PH
1148PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1149 run a filter in a subprocess. This could lead to confusion in subsequent
1150 lookups in the parent process. There should also be a search_tidyup() at
1151 the end of the subprocess.
1152
d7b47fd0
PH
1153PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1154 only if the host matched helo_try_verify_hosts, which caused the
1155 verification to occur when the EHLO/HELO command was issued. The ACL just
1156 tested the remembered result. Now, if a previous verification attempt has
1157 not happened, "verify = helo" does it there and then.
1158
ee744174
JJ
1159JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1160
b582ab87
PH
1161TK/03 Fix log output including CR from clamd.
1162
41a13e0a
PH
1163PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1164 exist provoked a memory error which could cause a segfault.
1165
f625cc5a
PH
1166PH/15 Installed PCRE 6.2
1167
1168PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1169
21f7af35
PH
1170PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1171 of the problem. Specifically, suggested +O2 rather than +O1 for the
1172 HP-UX compiler.
1173
31480e42
PH
1174PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1175
2d280592
PH
1176PH/20 If a delivery was routed to a non-standard port by means of an SRV
1177 record, the port was not correctly logged when the outgoing_port log
1178 selector was set (it logged the transort's default port).
1179
7cd1141b
PH
1180PH/21 Added support for host-specific ports to manualroute, queryprogram,
1181 fallback_hosts, and "hosts" in the smtp transport.
1182
1183PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1184 host errors such as "Connection refused".
1185
750af86e
PH
1186PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1187 authentication with radiusclient 0.4.9:
1188
1189 - Error returned from rc_read_config was caught wrongly
1190 - Username/password not passed on to radius server due to wrong length.
1191
1192 The presumption is that some radiusclient API changes for 4.51/PH/17
1193 were not taken care of correctly. The code is still untested by me (my
1194 Linux distribution still has 0.3.2 of radiusclient), but it was
1195 contributed by a Radius user.
1196
1197PH/24 When doing a callout, the value of $domain wasn't set correctly when
1198 expanding the "port" option of the smtp transport.
1199
4304270b
TK
1200TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1201 while reading a MIME header. Thanks to Tom Hughes for a patch.
1202
750af86e
PH
1203PH/24 Include config.h inside local_scan.h so that configuration settings are
1204 available.
1205
64ffc24f
PH
1206PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1207 that in an ACL for RCPT (for example), you can examine exactly what was
1208 received.
1209
5dd9625b
PH
1210PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1211 commands, but it was not correctly comparing the address with the actual
1212 client host address. Thus, it would show the EHLO address in Received:
1213 header lines when this was not necessary.
1214
5591031b
PH
1215PH/27 Added the % operator to ${eval:}.
1216
ba18e66a
PH
1217PH/28 Exim tries to create and chdir to its spool directory when it starts;
1218 it should be ignoring failures (because with -C, for example, it has lost
1219 privilege). It wasn't ignoring creation failures other than "already
1220 exists".
1221
9cec981f
PH
1222PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1223 -bV or -d is used.
1224
aa2b5c79
PH
1225PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1226 because an input line was too long, either on its own, or by virtue of
1509d3a8
PH
1227 too many continuations, the temporary file was not being removed, and the
1228 return code was incorrect.
aa2b5c79 1229
48a53b7f
PH
1230PH/31 Missing "BOOL" in function definition in filtertest.c.
1231
1c59d63b
PH
1232PH/32 Applied Sieve patches from the maintainer.
1233
671012da
TK
1234TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1235
1509d3a8
PH
1236PH/33 Added "verify = not_blind".
1237
1238PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1239 Local/Makefile (with some defaults set). These are used in built scripts
1240 such as exicyclog, but they have never been used in the exim_install
1241 script (though there are many overriding facilities there). I have
1242 arranged that the exim_install script now takes note of these two
1243 settings.
1244
1245PH/35 Installed configuration files for Dragonfly.
1246
2fe1a124
PH
1247PH/36 When a locally submitted message by a trusted user did not contain a
1248 From: header, and the sender address was obtained from -f or from an SMTP
1249 MAIL command, and the trusted user did not use -F to supply a sender
1250 name, $originator_name was incorrectly used when constructing a From:
1251 header. Furthermore, $originator_name was used for submission mode
1252 messages from external hosts without From: headers in a similar way,
1253 which is clearly wrong.
1254
8800895a
PH
1255PH/37 Added control=suppress_local_fixups.
1256
ccfdb010
PH
1257PH/38 When log_selector = +received_sender was set, and the addition of the
1258 sender made the log line's construction buffer exactly full, or one byte
1259 less than full, an overflow happened when the terminating "\n" was
1260 subsequently added.
1261
1130bfb0
PH
1262PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1263 when the result of a list match is failure because a DNS lookup failed.
1264
ebcb507f
PH
1265PH/40 RM_COMMAND is now used in the building process.
1266
c35e155c
PH
1267PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1268 the "build-* directories that it finds.
1269
95d1f782
PH
1270PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1271 address in a domain literal was a prefix of an interface address.
1272
fd6de02e
PH
1273PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1274 when verifying a sender address, unless rewrite_headers is false.
1275
58de37c5
PH
1276PH/44 Wrote a long comment about why errors_to addresses are verified as
1277 recipients, not senders.
1278
261cf466
TF
1279TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1280 the ratelimit ACL was added.
1281
3ee512ff
PH
1282PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1283
e08c430f
PH
1284PH/46 Added extra information about PostgreSQL errors to the error string.
1285
bef5a11f
PH
1286PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1287 avoiding the need to install special zones in a real server. This is
1288 backwards compatible; if it can't find the fake resolver, it drops back.
1289 Thus, both old and new test suites can be run.
1290
7546de58
TF
1291TF/02 Added util/ratelimit.pl
1292
e5d5a95f
TF
1293TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1294 clock is set back in time.
1295
2e88a017
TF
1296TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1297 Candler <B.Candler@pobox.com>.
1298
a5f65aa4
TF
1299TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1300 OK for addresses that are the result of redirections.
1301
e7726cbf
PH
1302PH/48 A number of further additions for the benefit of the new test suite,
1303 including a fake gethostbyname() that interfaces to the fake DNS resolver
1304 (see PH/47 above).
1305
a7fdad5b
TF
1306TF/06 The fix for widen_domains has also been applied to qualify_single and
1307 search_parents which are the other dnslookup options that can cause
1308 header rewrites.
1309
6af56900
PH
1310PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1311 type ("H").
1312
0925ede6
PH
1313PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1314
66afa403
TF
1315TF/07 Exim produced the error message "an SRV record indicated no SMTP
1316 service" if it encountered an MX record with an empty target hostname.
1317 The message is now "an MX or SRV record indicated no SMTP service".
1318
0154e85a
TF
1319TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1320 if the DNS of the sending site is misconfigured. This is quite a
1321 common situation. This change restores the behaviour of treating a
1322 helo verification defer as a failure.
1323
16f12c76
PH
1324PH/51 If self=fail was set on a router, the bounce message did not include the
1325 actual error message.
1326
bbe902f0 1327
e5a9dba6
PH
1328Exim version 4.52
1329-----------------
1330
1331TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1332
22c3b60b
PH
1333PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1334 command itself ended in error, the underlying message about the transport
1335 filter timeout was being overwritten with the pipe command error. Now the
1336 underlying error message should be appended to the second error message.
1337
06a9b4b5
PH
1338TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1339
c1ac6996
PH
1340PH/02 Reduce the amount of output that "make" produces by default. Full output
1341 can still be requested.
1342
9c7a242c
PH
1343PH/03 The warning log line about a condition test deferring for a "warn" verb
1344 was being output only once per connection, rather than after each
1345 occurrence (because it was using the same function as for successful
1346 "warn" verbs). This seems wrong, so I have changed it.
1347
87ba3f5f
PH
1348TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1349 it should not have, which might have caused a crash in the right
1350 circumstances, but probably never did.
1351
1352PH/04 Installed a modified version of Tony Finch's patch to make submission
1353 mode fix the return path as well as the Sender: header line, and to
1354 add a /name= option so that you can make the user's friendly name appear
1355 in the header line.
1356
29aba418
TF
1357TF/03 Added the control = fakedefer ACL modifier.
1358
fe0dab11
TF
1359TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1360 Mark Lowes for thorough testing.
870f6ba8 1361
11d337a4
TK
1362TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1363
1364TK/03 Merged latest SRS patch from Miles Wilton.
1365
415c8f3b
PH
1366PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1367 with the definition in sysexits.h (which is #included earlier).
1368 Fortunately, Exim does not actually use EX_OK. The code used to try to
1369 preserve the sysexits.h value, by assumimg that macro definitions were
1370 scanned for macro replacements. I have been disabused of this notion,
1371 so now the code just undefines EX_OK before #including unistd.h.
11d337a4 1372
958541e9
PH
1373PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1374 in the smtp transport. When a block could not be written in a single
1375 write() function, the timeout was being re-applied to each part-write.
1376 This seems wrong - if the receiver was accepting one byte at a time it
1377 would take for ever. The timeout is now adjusted when this happens. It
1378 doesn't have to be particularly precise.
1379
c206415f
TK
1380TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1381 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1382
2a4be8f9
PH
1383PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1384 without a domain if the check to <postmaster@domain> fails.
1385
1cba11c5
SC
1386SC/01 Eximstats: added -xls and the ability to specify output files
1387 (patch written by Frank Heydlauf).
1388
1389SC/02 Eximstats: use FileHandles for outputing results.
1390
1391SC/03 Eximstats: allow any combination of xls, txt, and html output.
1392
1393SC/04 Eximstats: fixed display of large numbers with -nvr option
1394
1395SC/05 Eximstats: fixed merging of reports with empty tables.
1396
1397SC/06 Eximstats: added the -include_original_destination flag
1398
1399SC/07 Eximstats: removed tabs and trailing whitespace.
1400
1005d00e
TK
1401TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1402
1403TK/06 MBOX spool code: Add real "From " MBOX separator line
1404 so the .eml file is really in mbox format (even though
1405 most programs do not really care). Patch from Alex Miller.
1406
1407TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1408 The latter is generated from $received_to and is only set if the
1409 message has one envelope recipient. SA can use these headers,
1410 obviously out-of-the-box. Patch from Alex Miller.
1411
9b4768fa
PH
1412PH/08 The ${def test on a variable was returning false if the variable's
1413 value was "0", contrary to what the specification has always said!
1414 The result should be true unless the variable is empty.
1415
1416PH/09 The syntax error of a character other than { following "${if
1417 def:variable_name" (after optional whitespace) was not being diagnosed.
1418 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1419 accidental colon was present, for example, could give incorrect results.
1420
0d7eb84a
PH
1421PH/10 Tidied the code in a number of places where the st_size field of a stat()
1422 result is used (not including appendfile, where other changes are about
1423 to be made).
1424
1425PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1426 This involved changing a lot of size variables from int to off_t. It
1427 should work with maildirs and everything.
1428
40727bee
TK
1429TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1430 spamd dying while we are connected to it.
1431
554d2369
TF
1432TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1433 <jgh@wizmail.org>
1434
1f922db1
PH
1435PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1436 function for BDB 4.3.
1437
ef213c3b
PH
1438PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1439
8ac170f3
PH
1440PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1441 the sources for PCRE have been re-arranged and more files are now
1442 involved.
1443
b1c749bb
PH
1444PH/15 The code I had for printing potentially long long variables in PH/11
1445 above was not the best (it lost precision). The length of off_t variables
1446 is now inspected at build time, and an appropriate printing format (%ld
c6c2dc1d
PH
1447 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1448 to be "long long int" or "long int". This is needed for the internal
1449 formatting function string_vformat().
b1c749bb 1450
4aac9b49
PH
1451PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1452 the configuration file to be ":syslog", then the script "guesses" where
1453 the logs files are, rather than using the compiled in default. In our
1454 case the guess is not the same as the compiled default, so the script
1455 suddenly stopped working when I started to use syslog. The patch checks
1456 to see if log_file_path is "". If so, it attempts to read it from exim
1457 with no configuration file to get the compiled in version, before it
1458 falls back to the previous guessing code."
1459
294520c8
TK
1460TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1461 implementing BATV in an Exim configuration. See NewStuff for the gory
1462 details.
1463
5bd022fe
PH
1464PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1465 Makefile that are specific to HP-UX.
1466
90e9ce59
PH
1467PH/18 If the "use_postmaster" option was set for a recipient callout together
1468 with the "random" option, the postmaster address was used as the MAIL
1469 FROM address for the random test, but not for the subsequent recipient
1470 test. It is now used for both.
1471
5ea81592
PH
1472PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1473 patch removes a few documentation additions to RFC 3028, because the
1474 latest draft now contains them. It adds the new en;ascii-case comparator
1475 and a new error check for 8bit text in MIME parts. Comparator and
1476 require names are now matched exactly. I enabled the subaddress
1477 extension, but it is not well tested yet (read: it works for me)."
1478
c6c2dc1d
PH
1479PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1480 rework some of the code of TK/09 above to avoid the hardwired use of
1481 "%lld" and "long long". Replaced the call to snprintf() with a call to
1482 string_vformat().
1483
fffffe4c
PH
1484PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1485 records point to non-existent hosts", "retry timeout exceeded", and
1486 "retry time not reached for any host after a long failure period".
ca02eafb 1487
9a26b6b2
PH
1488PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1489 experimental DomainKeys support:
1490
1491 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1492 (2) On an error such as an illegally used "control", the wrong name for
1493 the control was given.
1494
1495 These problems did NOT occur unless DomainKeys support was compiled.
1496
4aee0225
PH
1497PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1498
32d668a5
PH
1499PH/24 Added ${if match_ip condition.
1500
8187c3f3
PH
1501PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1502 databases so that it will be absolutely obvious if a crash occurs in the
1503 DB library. This is a regular occurrence (often caused by mis-matched
1504 db.h files).
1505
ff790e47 1506PH/26 Insert a lot of missing (void) casts for functions such as chown(),
f1e894f3
PH
1507 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1508 picked up on a user's system that detects such things. There doesn't seem
1509 to be a gcc warning option for this - only an attribute that has to be
1510 put on the function's prototype. It seems that in Fedora Core 4 they have
1511 set this on a number of new functions. No doubt there will be more in due
1512 course.
ff790e47 1513
5417f6d1
PH
1514PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1515 specify a transport. However, if an address that was verified by such a
1516 router was the subject of a callout, Exim crashed because it tried to
1517 read the rcpt_include_affixes from the non-existent transport. Now it
1518 just assumes that the setting of that option is false. This bug was
1519 introduced by 4.51/PH/31.
1520
59cf8544
PH
1521PH/28 Changed -d+all to exclude +memory, because that information is very
1522 rarely of interest, but it makes the output a lot bigger. People tend to
1523 do -d+all out of habit.
1524
e7ad8a65
PH
1525PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1526 code in os-type was giving problems when libc.so lives in lib64, like on
1527 x86_64 Fedora Core.
1528
ade42478
PH
1529PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1530 aren't the modern standard, and it seems that some systems' include files
1531 don't always have them. Exim was already checking for some of the newer
1532 ones like T_AAAA, and defining it itself. I've added checks for all the
1533 record types that Exim uses.
1534
182ad5cf
PH
1535PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1536 not automatically generating a new one, as it is supposed to. This
1537 prevented TLS from working. If the file did exist, but contained invalid
1538 data, a new version was generated, as expected. It was only the case of a
1539 non-existent file that was broken.
1540
b0d9fc80
TK
1541TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1542 with a change in libdomainkeys > 0.64.
1543
1544TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1545 from DNS. If the selector record carries the flag, it now has
1546 precedence over the domain-wide flag.
1547
1548TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1549
47c7a64a
PH
1550PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1551 the use of a port name that isn't defined in /etc/services) occurred, the
1552 message was deferred as in a normal delivery, and thus remained on the
1553 spool, instead of being failed because of the mua_wrapper setting. This
1554 is now fixed, and I tidied up some of the mua_wrapper messages at the
1555 same time.
1556
a388bce4
SC
1557SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1558 the messages in a hash of arrays rather than using individual hashes.
1559 This is a bit cleaner and results in dramatic memory savings, albeit
1560 at a slight CPU cost.
1561
1562SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1563 as requested by Marc Sherman.
1564
1565SC/10 Eximstats: added histograms for user specified patterns as requested
1566 by Marc Sherman.
1567
0793e4ed
SC
1568SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1569
c58b88df
PH
1570PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1571 fopen() in the content-scanning modules that did not already have it.
1572
e7ad8a65 1573
7982096b
PH
1574Exim version 4.51
1575-----------------
1576
1a46a8c5
PH
1577TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1578 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1579
2f079f46 1580TK/02 Fix ACL "control" statement not being available in MIME ACL.
1a46a8c5
PH
1581
1582TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1583
1584PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1585 to test Sieve filters that use "vacation".
1586
1587PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1588 that changes the way the GnuTLS parameters are stored in the cache file.
1589 The new format can be generated externally. For backward compatibility,
1590 if the data in the cache doesn't make sense, Exim assumes it has read an
1591 old-format file, and it generates new data and writes a new file. This
1592 means that you can't go back to an older release without removing the
1593 file.
1594
1595PH/03 A redirect router that has both "unseen" and "one_time" set does not
1596 work if there are any delivery delays because "one_time" forces the
1597 parent to be marked "delivered", so its unseen clone is never tried
1598 again. For this reason, Exim now forbids the simultaneous setting of
1599 these two options.
1600
1601PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1602 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1603 entry if you want to know the details. The fix, however, neglected to
1604 consider the case when local delivery batching is involved. The test for
1605 "previously delivered" was not happening when checking to see if an
1606 address could be batched with a previous (undelivered) one; under
1607 certain circumstances this could lead to multiple deliveries to the same
c2c19e9d 1608 address.
1a46a8c5
PH
1609
1610PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1611 in its include files, and this causes problems building Exim.
1612
1613PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1614 header_syntax) but Exim was just ignoring anything given after a slash.
1615 In particular, this caused confusion with an attempt to use "verify =
1616 reverse_host_lookup/defer_ok". An error is now given when options are
1617 supplied for verify items that do not have them. (Maybe reverse_host_
1618 lookup should have a defer_ok option, but that's a different point.)
1619
1620PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1621 defined by RFC 821) to 2048, because there were problems with some AUTH
1622 commands, and RFC 1869 says the size should be increased for extended
1623 SMTP commands that take arguments.
1624
1625PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1626 Finch).
1627
1628PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1629 "unknown" error; now it says that the functionality isn't in the binary.
8d67ada3 1630
49c2d5ea
PH
1631PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1632 an address' error message when a string expansion fails (syntax or
f331f3b6
PH
1633 whatever). Otherwise the password may appear in the log. Following change
1634 PH/42 below, there is no longer a chance of it appearing in a bounce
1635 message.
49c2d5ea 1636
bf759a8b
PH
1637PH/11 Installed exipick version 20050225.0 from John Jetmore.
1638
83364d30
PH
1639PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1640 of its addresses was ever tried. (Bugzilla bug #2.)
1641
7999bbd7
PH
1642PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1643 the result incorrectly in the debug output. (It correctly added a newline
1644 to what was transported.)
1645
7dbf77c9
PH
1646TF/01 Added $received_time.
1647
74e0617f
PH
1648PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1649 commented out examples of how to interface to a virus scanner and to
1650 SpamAssassin. Also added commented examples of av_scanner and
1651 spamd_address settings.
1652
2f079f46
PH
1653PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1654 and controls are allowed in which ACLs. There were a couple of minor
1655 errors. Some of the entries in the conditions table (which is a table of
1656 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1657 negation of where the condition IS allowed.
1658
8c841523
PH
1659PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1660
7766a4f0
PH
1661PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1662 header file does not have a version number, so I've had to invent a new
1663 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1664 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1665 radiusclient), but it was contributed by a Radius user.
1666
8b417f2c
PH
1667PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1668 files or directories, for OpenSSL.
1669
901f42cb
PH
1670PH/19 When an Exim process that is running as root has to create an Exim log
1671 file, it does so in a subprocess that runs as exim:exim so as to get the
1672 ownership right at creation (otherwise, other Exim processes might see
1673 the file with the wrong ownership). There was no test for failure of this
1674 fork() call, which would lead to the process getting stuck as it waited
1675 for a non-existent subprocess. Forks do occasionally fail when resources
1676 run out. I reviewed all the other calls to fork(); they all seem to check
1677 for failure.
1678
f9b9210e
PH
1679PH/20 When checking for unexpected SMTP input at connect time (before writing
1680 the banner), Exim was not dealing correctly with a non-positive return
1681 from the read() function. If the client had disconnected by this time,
1682 the result was a log entry for a synchronization error with an empty
1683 string after "input=" when read() returned zero. If read() returned -1
1684 (an event I could not check), uninitialized data bytes were printed.
1685 There were reports of junk text (parts of files, etc) appearing after
1686 "input=".
1687
54cdb463
PH
1688PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1689
cf00dad6
PH
1690PH/22 Added support for macro redefinition, and (re)definition in between
1691 driver and ACL definitions.
1692
acb1b346
PH
1693PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1694 forgetting to use the resulting value; it was using the unexpanded value.
1695
c5ddb310
PH
1696PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1697 hadn't been configured. The fix is from Juergen Kreileder, who
1698 understands it better than I do:
1699
1700 "Here's what I see happening with three configured cyrus_sasl
1701 authenticators configured (plain, login, cram-md5):
1702
1703 On startup auth_cyrus_sasl_init() gets called for each of these.
1704 This means three calls to sasl_listmech() without a specified mech_list.
1705 => SASL tests which mechs of all available mechs actually work
1706 => three warnings about OTP not working
1707 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1708
1709 With the patch, sasl_listmech() also gets called three times. But now
1710 SASL's mech_list option is set to the server_mech specified in the the
1711 authenticator. Or in other words, the answer from sasl_listmech()
1712 gets limited to just the mech you're testing for (which is different
1713 for each call.)
1714 => the return list contains just 'plain' or 'login', 'cram-md5' or
1715 nothing depending on the value of ob->server_mech.
1716
1717 I've just tested the patch: Authentication still works fine,
1718 unavailable mechs specified in the exim configuration are still
1719 caught, and the auth.log warnings about OTP are gone."
1720
31619da6
PH
1721PH/25 When debugging is enabled, the contents of the command line are added
1722 to the debugging output, even when log_selector=+arguments is not
1723 specified.
1724
bebaf0fc
PH
1725PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1726 answer is "GNU", and only if the return is "GNU/something" is the answer
1727 "Linux".
1728
475fe28a
PH
1729PH/27 $acl_verify_message is now set immediately after the failure of a
1730 verification in an ACL, and so is available in subsequent modifiers. In
1731 particular, the message can be preserved by coding like this:
1732
1733 warn !verify = sender
1734 set acl_m0 = $acl_verify_message
1735
1736 Previously, $acl_verify_message was set only while expanding "message"
1737 and "log_message" when a very denied access.
1738
7e8bec7a
PH
1739PH/28 Modified OS/os.c-Linux with
1740
1741 -#ifndef OS_LOAD_AVERAGE
1742 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1743
1744 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1745 nomenclature these days.)
1746
e4a89c47
PH
1747PH/29 Installed patch from the Sieve maintainer that adds the options
1748 sieve_useraddress and sieve_subaddress to the redirect router.
1749
5ca2a9a1
PH
1750PH/30 In these circumstances:
1751 . Two addresses routed to the same list of hosts;
1752 . First host does not offer TLS;
1753 . First host accepts first address;
1754 . First host gives temporary error to second address;
1755 . Second host offers TLS and a TLS session is established;
1756 . Second host accepts second address.
1757 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1758 and peerdn, if requested) that were in fact used only for the second
1759 address.
7e8bec7a 1760
c688b954
PH
1761PH/31 When doing a callout as part of verifying an address, Exim was not paying
1762 attention to any local part prefix or suffix that was matched by the
1763 router that accepted the address. It now behaves in the same way as it
1764 does for delivery: the affixes are removed from the local part unless
1765 rcpt_include_affixes is set on the transport.
1766
fed77020
PH
1767PH/32 Add the sender address, as F=<...>, to the log line when logging a
1768 timeout during the DATA phase of an incoming message.
1769
7fe1560f
PH
1770PH/33 Sieve envelope tests were broken for match types other than :is. I have
1771 applied a patch sanctioned by the Sieve maintainer.
c688b954 1772
ebb6e6d5
PH
1773PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1774 the uid or gid is negative. A case of a negative gid caused this to be
1775 noticed. The fix allows for either to be negative.
1776
9c4e8f60
PH
1777PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1778 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1779 overlooked.
1780
1781PH/36 The change PH/12 above was broken. Fixed it.
1782
d7174846
PH
1783PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1784 the grounds that routing the same address twice would always produce the
1785 same answer. This might have been true once, but it is certainly no
1786 longer true now. Routing a child address may depend on the previous
1787 routing that produced that child. Some complicated redirection strategies
1788 went wrong when messages had multiple recipients, and made Exim's
1789 behaviour dependent on the order in which the addresses were given.
1790
1791 I have moved the duplicate checking until after the routing is complete.
1792 Exim scans the addresses that are assigned to local and remote
1793 transports, and removes any duplicates. This means that more work will be
1794 done, as duplicates will always all be routed, but duplicates are
1795 presumably rare, so I don't expect this is of any significance.
1796
1797 For deliveries to pipes, files, and autoreplies, the duplicate checking
1798 still happens during the routing process, since they are not going to be
1799 routed further.
1800
cfe75fc3
PH
1801PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1802 It corrects a timeout issue with spamd. This is Ian's comment: "The
1803 background is that sometimes spamd either never reads data from a
1804 connection it has accepted, or it never writes response data. The exiscan
1805 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1806 blindly assumes that writes won't block so it may never time out."
1807
be22d70e
PH
1808PH/39 Allow G after quota size as well as K and M.
1809
0612b098
PH
1810PH/40 The value set for $authenticated_id in an authenticator may not contain
1811 binary zeroes or newlines because the value is written to log lines and
1812 to spool files. There was no check on this. Now the value is run through
1813 the string_printing() function so that such characters are converted to
1814 printable escape sequences.
1815
2e0c1448
PH
1816PH/41 $message_linecount is a new variable that contains the total number of
1817 lines in the message. Compare $body_linecount, which is the count for the
1818 body only.
1819
447d236c
PH
1820PH/42 Exim no longer gives details of delivery errors for specific addresses in
1821 bounce and delay warning messages, except in certain special cases, which
1822 are as follows:
1823
1824 (a) An SMTP error message from a remote host;
1825 (b) A message specified in a :fail: redirection;
1826 (c) A message specified in a "fail" command in a system filter;
1827 (d) A message specified in a FAIL return from the queryprogram router;
1828 (e) A message specified by the cannot_route_message router option.
1829
1830 In these cases only, Exim does include the error details in bounce and
1831 warning messages. There are also a few cases where bland messages such
1832 as "unrouteable address" or "local delivery error" are given.
1833
d20976dc
PH
1834PH/43 $value is now also set for the "else" part of a ${run expansion.
1835
f656d135
PH
1836PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1837 being worked on, but at least Exim now implements the latest version to
1838 play with."
1839
2e2a30b4
PH
1840PH/45 In a pipe transport, although a timeout while waiting for the pipe
1841 process to complete was treated as a delivery failure, a timeout while
1842 writing the message to the pipe was logged, but erroneously treated as a
1843 successful delivery. Such timeouts include transport filter timeouts. For
1844 consistency with the overall process timeout, these timeouts are now
1845 treated as errors, giving rise to delivery failures by default. However,
1846 there is now a new Boolean option for the pipe transport called
1847 timeout_defer, which, if set TRUE, converts the failures into defers for
1848 both kinds of timeout. A transport filter timeout is now identified in
1849 the log output.
1850
9176e9f0
PH
1851PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1852 systems where "make" and "gmake" are different, calling "gmake" at top
1853 level broke things. I've arranged for the value of $(MAKE) to be passed
1854 from the Makefile to this script so that it can call the same version of
1855 "make".
1856
7982096b 1857
bbe902f0
PH
1858A note about Exim versions 4.44 and 4.50
1859----------------------------------------
1860
1861Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1862changes of various kinds. As a consequence, a big documentation update was
1863needed. This delayed the release for rather longer than seemed good, especially
1864in the light of a couple of (minor) security issues. Therefore, the changes
1865that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1866release. So 4.44 and 4.50 are in effect two different branches that both start
1867from 4.43.
1868
1869I have left the 4.50 change log unchanged; it contains all the changes since
18704.43. The change log for 4.44 is below; many of its items are identical to
1871those for 4.50. This seems to be the most sensible way to preserve the
1872historical information.
1873
1874
f7b63901 1875Exim version 4.50
495ae4b0
PH
1876-----------------
1877
5fe762f6
PH
1878 1. Minor wording change to the doc/README.SIEVE file.
1879
139059f6 1880 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
5fe762f6 1881 computation of the current number of files was incorrect.
495ae4b0 1882
7086e875
PH
1883 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1884 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1885 place.
1886
35af9f61
PH
1887 4. Give more explanation in the error message when the command for a transport
1888 filter fails to execute.
1889
b668c215
PH
1890 5. There are several places where Exim runs a non-Exim command in a
1891 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1892 was being done only for the command run by the queryprogram router. It is
1893 now done for all such subprocesses. The other cases are: ${run, transport
1894 filters, and the commands run by the lmtp and pipe transports.
1895
a494b1e1
PH
1896 6. Added CONFIGURE_GROUP build-time option.
1897
1898 7. Some older OS have a limit of 256 on the maximum number of file
1899 descriptors. Exim was using setrlimit() to set 1000 as a large value
1900 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1901 systems. I've change it so that if it can't get 1000, it tries for 256.
35edf2ff 1902
c5fcb476
PH
1903 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1904 was an oversight, and furthermore, ever since the addition of extra
1905 controls (e.g. 4.43/32), the checks on when to allow different forms of
1906 "control" were broken. There should now be diagnostics for all cases when a
1907 control that does not make sense is encountered.
1908
69358f02
PH
1909 9. Added the /retain_sender option to "control=submission".
1910
5be20824
PH
191110. $recipients is now available in the predata ACL (oversight).
1912
eb2c0248
PH
191311. Tidy the search cache before the fork to do a delivery from a message
1914 received from the command line. Otherwise the child will trigger a lookup
1915 failure and thereby defer the delivery if it tries to use (for example) a
1916 cached ldap connection that the parent has called unbind on.
1917
2a3eea10
PH
191812. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1919 of $address_data from the recipient verification was clobbered by the
1920 sender verification.
1921
192213. The value of address_data from a sender verification is now available in
1923 $sender_address_data in subsequent conditions in the ACL statement.
1924
23c7ff99
PH
192514. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1926
4deaf07d
PH
192715. Added a new option "connect=<time>" to callout options, to set a different
1928 connection timeout.
1929
926e1192
PH
193016. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1931 was its contents. (It was OK if the option was not defined at all.)
1932
650edc6f
PH
193317. A "Completed" log line is now written for messages that are removed from
1934 the spool by the -Mrm option.
1935
2c7db3f5
PH
193618. New variables $sender_verify_failure and $recipient_verify_failure contain
1937 information about exactly what failed.
1938
3d235903
PH
193919. Added -dd to debug only the daemon process.
1940
7c7ad977
PH
194120. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1942 handles timeouts, both on the server side and network timeouts. Renamed the
1943 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1944
981756db
PH
194521. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1946 It is now set to "smtps".
1947
d4eb88df
PH
194822. $host_address is now set to the target address during the checking of
1949 ignore_target_hosts.
1950
195123. When checking ignore_target_hosts for an ipliteral router, no host name was
1952 being passed; this would have caused $sender_host_name to have been used if
1953 matching the list had actually called for a host name (not very likely,
1954 since this list is usually IP addresses). A host name is now passed as
1955 "[x.x.x.x]".
1956
7d468ab8
PH
195724. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1958 code that specifies a non-restarting handler (typically sigaction() in
1959 modern systems) in an attempt to fix a rare and obscure crash bug.
1960
196125. Narrowed the window for a race in the daemon that could cause it to ignore
1962 SIGCHLD signals. This is not a major problem, because they are used only to
1963 wake it up if nothing else does.
1964
62c0818f
PH
196526. A malformed maildirsize file could cause Exim to calculate negative values
1966 for the mailbox size or file count. Odd effects could occur as a result.
1967 The maildirsize information is now recalculated if the size or filecount
1968 end up negative.
1969
26034054
PH
197027. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1971 support for a long time. Removed HAVE_SYS_VFS_H.
1972
af66f652
PH
197328. Installed the latest version of exipick from John Jetmore.
1974
90af77f4
PH
197529. In an address list, if the pattern was not a regular expression, an empty
1976 subject address (from a bounce message) matched only if the pattern was an
1977 empty string. Non-empty patterns were not even tested. This was the wrong
1978 because it is perfectly reasonable to use an empty address as part of a
1979 database query. An empty address is now tested by patterns that are
1980 lookups. However, all the other forms of pattern expect the subject to
1981 contain a local part and a domain, and therefore, for them, an empty
1982 address still always fails if the pattern is not itself empty.
1983
d8ef3577
PH
198430. Exim went into a mad DNS loop when attempting to do a callout where the
1985 host was specified on an smtp transport, and looking it up yielded more
1986 than one IP address.
1987
5cb8cbc6
PH
198831. Re-factored the code for checking spool and log partition space into a
1989 function that finds that data and another that does the check. The former
1990 is then used to implement four new variables: $spool_space, $log_space,
1991 $spool_inodes, and $log_inodes.
1992
14702f5b
PH
199332. The RFC2047 encoding function was originally intended for short strings
1994 such as real names; it was not keeping to the 75-character limit for
1995 encoded words that the RFC imposes. It now respects the limit, and
1996 generates multiple encoded words if necessary. To be on the safe side, I
1997 have increased the buffer size for the ${rfc2047: expansion operator from
1998 1024 to 2048 bytes.
1999
063b1e99
PH
200033. It is now permitted to omit both strings after an "if" condition; if the
2001 condition is true, the result is "true". As before, when the second string
2002 is omitted, a false condition yields an empty string. This makes it less
2003 cumbersome to write custom ACL and router conditions.
2004
652e1b65
PH
200534. Failure to deliver a bounce message always caused it to be frozen, even if
2006 there was an errors_to setting on the router. The errors_to setting is now
2007 respected.
2008
6f0c9a4f
PH
200935. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2010 canonical form (fully expanded) before being placed in
2011 $sender_host_address.
2012
33397d19
PH
201336. The table in the code that translates DNS record types into text (T_A to
2014 "A" for instance) was missing entries for NS and CNAME. It is just possible
2015 that this could have caused confusion if both these types were looked up
2016 for the same domain, because the text type is used as part of Exim's
2017 per-process caching. But the chance of anyone hitting this buglet seems
2018 very small.
2019
7bb56e1f
PH
202037. The dnsdb lookup has been extended in a number of ways.
2021
2022 (1) There is a new type, "zns", which walks up the domain tree until it
2023 finds some nameserver records. It should be used with care.
2024
ea3bc19b
PH
2025 (2) There is a new type, "mxh", which is like "mx" except that it returns
2026 just the host names, not the priorities.
2027
2028 (3) It is now possible to give a list of domains (or IP addresses) to be
ff4dbb19
PH
2029 looked up. The behaviour when one of the lookups defers can be
2030 controlled by a keyword.
7bb56e1f 2031
ea3bc19b 2032 (4) It is now possible to specify the separator character for use when
7bb56e1f 2033 multiple records are returned.
33397d19 2034
0bcb2a0e
PH
203538. The dnslists ACL condition has been extended: it is now possible to supply
2036 a list of IP addresses and/or domains to be looked up in a particular DNS
2037 domain.
2038
2ac0e484
PH
203939. Added log_selector=+queue_time_overall.
2040
4e1fde53
PH
204140. When running the queue in the test harness, wait just a tad after forking a
2042 delivery process, to get repeatability of debugging output.
2043
de365ded
PH
204441. Include certificate and key file names in error message when GnuTLS fails
2045 to set them up, because the GnuTLS error message doesn't include the name
2046 of the failing file when there is a problem reading it.
2047
f05da2e8
PH
204842. Allow both -bf and -bF in the same test run.
2049
d6453af2
PH
205043. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2051
f7b63901
PH
205244. The "Exiscan patch" is now merged into the mainline Exim source.
2053
205445. Sometimes the final signoff response after QUIT could fail to get
2055 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2056 < 0 before doing a fflush(). This bug looks as though it goes back to the
2057 introduction of TLS in release 3.20, but "sometimes" must have been rare
2058 because the tests only now provoked it.
2059
a444213a
PH
206046. Reset the locale to "C" after calling embedded Perl, in case it was changed
2061 (this can affect the format of dates).
2062
0ec020ea
PH
206347. exim_tidydb, when checking for the continued existence of a message for
2064 which it has found a message-specific retry record, was not finding
2065 messages that were in split spool directories. Consequently, it was
2066 deleting retry records that should have stayed in existence.
2067
b1206957
PH
206848. Steve fixed some bugs in eximstats.
2069
207049. The SPA authentication driver was not abandoning authentication and moving
2071 on to the next authenticator when an expansion was forced to fail,
2072 contradicting the general specification for all authenticators. Instead it
2073 was generating a temporary error. It now behaves as specified.
2074
26dd5a95
PH
207550. The default ordering of permitted cipher suites for GnuTLS was pessimal
2076 (the order specifies the preference for clients). The order is now AES256,
2077 AES128, 3DES, ARCFOUR128.
2078
343b2385
PH
207951. Small patch to Sieve code - explicitly set From: when generating an
2080 autoreply.
2081
1c5466b9
PH
208252. Exim crashed if a remote delivery caused a very long error message to be
2083 recorded - for instance if somebody sent an entire SpamAssassin report back
2084 as a large number of 550 error lines. This bug was coincidentally fixed by
2085 increasing the size of one of Exim's internal buffers (big_buffer) that
2086 happened as part of the Exiscan merge. However, to be on the safe side, I
2087 have made the code more robust (and fixed the comments that describe what
2088 is going on).
2089
55ee9ee3
PH
209053. Now that there can be additional text after "Completed" in log lines (if
2091 the queue_time_overall log selector is set), a one-byte patch to exigrep
2092 was needed to allow it to recognize "Completed" as not the last thing in
2093 the line.
2094
d38f8232
PH
209554. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2096 patch that reportedly fixes this has been added. I am not expert enough to
2097 create a test for it. This is what the patch creator wrote:
2098
2099 "I found a little strange behaviour of ldap code when working with
2100 Windows 2003 AD Domain, where users was placed in more than one
2101 Organization Units. When I tried to give exim partial DN, the exit code
2102 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2103 But simultaneously result of request was absolutely normal ldap result,
2104 so I produce this patch..."
2105
3295e65b
PH
2106 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2107 REFERENCE, so I have modified the code to exclude the patch when that macro
2108 is not defined.
2109
7102e136
PH
211055. Some experimental protocols are using DNS PTR records for new purposes. The
2111 keys for these records are domain names, not reversed IP addresses. The
b975ba52
PH
2112 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2113 leaves it alone. Component reversal etc. now happens only for IP addresses.
ea3a6f44 2114 CAN-2005-0021
7102e136 2115
3ca0ba97
PH
211656. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2117
c2bcbe20
PH
211857. Double the size of the debug message buffer (to 2048) so that more of very
2119 long debug lines gets shown.
2120
18ce445d
PH
212158. The exicyclog utility now does better if the number of log files to keep
2122 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2123
1f5b4c3d
PH
212459. Two changes related to the smtp_active_hostname option:
2125
2126 (1) $smtp_active_hostname is now available as a variable.
2127 (2) The default for smtp_banner uses $smtp_active_hostname instead
2128 of $primary_hostname.
2129
b975ba52
PH
213060. The host_aton() function is supposed to be passed a string that is known
2131 to be a valid IP address. However, in the case of IPv6 addresses, it was
2132 not checking this. This is a hostage to fortune. Exim now panics and dies
2133 if the condition is not met. A case was found where this could be provoked
85b87bc2
PH
2134 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2135 components; fortuitously, this particular loophole had already been fixed
2136 by change 4.50/55 above.
2137
2138 If there are any other similar loopholes, the new check in host_aton()
2139 itself should stop them being exploited. The report I received stated that
2140 data on the command line could provoke the exploit when Exim was running as
2141 exim, but did not say which command line option was involved. All I could
2142 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2143 running as the user.
ea3a6f44 2144 CAN-2005-0021
85b87bc2
PH
2145
214661. There was a buffer overflow vulnerability in the SPA authentication code
2147 (which came originally from the Samba project). I have added a test to the
2148 spa_base64_to_bits() function which I hope fixes it.
ea3a6f44 2149 CAN-2005-0022
b975ba52 2150
17ffcae7
PH
215162. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2152 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2153
d95f9fdb
PH
215463. The daemon start-up calls getloadavg() while still root for those OS that
2155 need the first call to be done as root, but it missed one case: when
2156 deliver_queue_load_max is set with deliver_drop_privilege. This is
2157 necessary for the benefit of the queue runner, because there is no re-exec
2158 when deliver_drop_privilege is set.
2159
86b8287f
PH
216064. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2161 This has been fixed.
2162
60dc5e56
PH
216365. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2164 was in use, was not putting the data itself into the right store pool;
2165 consequently, it could be overwritten for a subsequent message in the same
2166 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2167 the caching.)
2168
533244af
PH
216966. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2170
a5a28604
PH
217167. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2172 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2173 the function were treating the return as a boolean value, which happened to
2174 work because 0=false and not-0=true, but is not correct code.
2175
7e634d24
PH
217668. The host_aton() function was not handling scoped IPv6 addresses (those
2177 with, for example, "%eth0" on the end) correctly.
2178
3e11c26b
PH
217969. Fixed some compiler warnings in acl.c for the bitmaps specified with
2180 negated items (that is, ~something) in unsigned ints. Some compilers
2181 apparently mutter when there is no cast.
2182
6729cf78
PH
218370. If an address verification called from an ACL failed, and did not produce a
2184 user-specific message (i.e. there was only a "system" message), nothing was
2185 put in $acl_verify_message. In this situation, it now puts the system
2186 message there.
2187
00f00ca5
PH
218871. Change 4.23/11 added synchronization checking at the start of an SMTP
2189 session; change 4.31/43 added the unwanted input to the log line - except
2190 that it did not do this in the start of session case. It now does.
2191
c9bdd01c
PH
219272. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2193 This is wrong and can cause the other end to generate a synchronization
2194 error if it is another Exim or anything else that does the synchronization
2195 check. A QUIT command is no longer sent after a timeout.
2196
d43194df
PH
219773. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2198 during host lookups.
2199
fe5b5d0b
PH
220074. The defer_ok option of callout verification was not working if it was used
2201 when verifying addresses in header lines, that is, for this case:
2202
2203 verify = header_sender/callout=defer_ok
2204
76a2d7ba
PH
220575. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2206 those file descriptors could be used for SMTP connections. If anything
2207 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2208 could be sent to the SMTP client, causing chaos. The daemon now opens
2209 stdin, stdout, and stderr to /dev/null when it puts itself into the
2210 background.
2211
221276. Arrange for output from Perl's "warn" command to be written to Exim's main
2213 log by default. The user can override this with suitable Perl magic.
2214
04f7d5b9
PH
221577. The use of log_message on a "discard" ACL verb, which is supposed to add to
2216 the log message when discard triggers, was not working for the DATA ACL or
2217 for the non-SMTP ACL.
2218
bc60667e
PH
221978. Error message wording change in sieve.c.
2220
bb6e88ff
PH
222179. If smtp_accept_max_per_host was set, the number of connections could be
2222 restricted to fewer than expected, because the daemon was trying to set up
2223 a new connection before checking whether the processes handling previous
2224 connections had finished. The check for completed processes is now done
2225 earlier. On busy systems, this bug wouldn't be noticed because something
2226 else would have woken the daemon, and it would have reaped the completed
2227 process earlier.
2228
1e70f85b
PH
222980. If a message was submitted locally by a user whose login name contained one
2230 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2231 It caused a spool format error. I have fixed the spool reading code. A
2232 related problem was that the "from" clause in the Received: line became
2233 illegal because of the space(s). It is now covered by ${quote_local_part.
2234
223581. Included the latest eximstats from Steve (adds average sizes to HTML Top
2236 tables).
2237
4e01f9d6
PH
223882. Updated OS/Makefile-AIX as per message from Mike Meredith.
2239
1ee1cef2
PH
224083. Patch from Sieve maintainer to fix unterminated string problem in
2241 "vacation" handling.
2242
6e2b4ccc
PH
224384. Some minor changes to the Linux configuration files to help with other
2244 OS variants using glibc.
2245
8e669ac1
PH
224685. One more patch for Sieve to update vacation handling to latest spec.
2247
495ae4b0 2248
bbe902f0
PH
2249----------------------------------------------------
2250See the note above about the 4.44 and 4.50 releases.
2251----------------------------------------------------
2252
2253
2254Exim version 4.44
2255-----------------
2256
2257 1. Change 4.43/35 introduced a bug that caused file counts to be
2258 incorrectly computed when quota_filecount was set in an appendfile
2259 transport
2260
2261 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2262 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2263 place.
2264
2265 3. Give more explanation in the error message when the command for a transport
2266 filter fails to execute.
2267
2268 4. There are several places where Exim runs a non-Exim command in a
2269 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2270 was being done only for the command run by the queryprogram router. It is
2271 now done for all such subprocesses. The other cases are: ${run, transport
2272 filters, and the commands run by the lmtp and pipe transports.
2273
2274 5. Some older OS have a limit of 256 on the maximum number of file
2275 descriptors. Exim was using setrlimit() to set 1000 as a large value
2276 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2277 systems. I've change it so that if it can't get 1000, it tries for 256.
2278
2279 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2280 was an oversight, and furthermore, ever since the addition of extra
2281 controls (e.g. 4.43/32), the checks on when to allow different forms of
2282 "control" were broken. There should now be diagnostics for all cases when a
2283 control that does not make sense is encountered.
2284
2285 7. $recipients is now available in the predata ACL (oversight).
2286
2287 8. Tidy the search cache before the fork to do a delivery from a message
2288 received from the command line. Otherwise the child will trigger a lookup
2289 failure and thereby defer the delivery if it tries to use (for example) a
2290 cached ldap connection that the parent has called unbind on.
2291
2292 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2293 of $address_data from the recipient verification was clobbered by the
2294 sender verification.
2295
229610. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2297 was its contents. (It was OK if the option was not defined at all.)
2298
229911. A "Completed" log line is now written for messages that are removed from
2300 the spool by the -Mrm option.
2301
230212. $host_address is now set to the target address during the checking of
2303 ignore_target_hosts.
2304
230513. When checking ignore_target_hosts for an ipliteral router, no host name was
2306 being passed; this would have caused $sender_host_name to have been used if
2307 matching the list had actually called for a host name (not very likely,
2308 since this list is usually IP addresses). A host name is now passed as
2309 "[x.x.x.x]".
2310
231114. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2312 code that specifies a non-restarting handler (typically sigaction() in
2313 modern systems) in an attempt to fix a rare and obscure crash bug.
2314
231515. Narrowed the window for a race in the daemon that could cause it to ignore
2316 SIGCHLD signals. This is not a major problem, because they are used only to
2317 wake it up if nothing else does.
2318
231916. A malformed maildirsize file could cause Exim to calculate negative values
2320 for the mailbox size or file count. Odd effects could occur as a result.
2321 The maildirsize information is now recalculated if the size or filecount
2322 end up negative.
2323
232417. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2325 support for a long time. Removed HAVE_SYS_VFS_H.
2326
ea3a6f44 232718. Updated exipick to current release from John Jetmore.
bbe902f0
PH
2328
232919. Allow an empty sender to be matched against a lookup in an address list.
2330 Previously the only cases considered were a regular expression, or an
2331 empty pattern.
2332
233320. Exim went into a mad DNS lookup loop when doing a callout where the
2334 host was specified on the transport, if the DNS lookup yielded more than
2335 one IP address.
2336
ea3a6f44
NM
233721. The RFC2047 encoding function was originally intended for short strings
2338 such as real names; it was not keeping to the 75-character limit for
2339 encoded words that the RFC imposes. It now respects the limit, and
2340 generates multiple encoded words if necessary. To be on the safe side, I
2341 have increased the buffer size for the ${rfc2047: expansion operator from
2342 1024 to 2048 bytes.
bbe902f0 2343
ea3a6f44
NM
234422. Failure to deliver a bounce message always caused it to be frozen, even if
2345 there was an errors_to setting on the router. The errors_to setting is now
2346 respected.
bbe902f0
PH
2347
234823. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2349 canonical form (fully expanded) before being placed in
2350 $sender_host_address.
2351
235224. Updated eximstats to version 1.33
2353
ea3a6f44
NM
235425. Include certificate and key file names in error message when GnuTLS fails
2355 to set them up, because the GnuTLS error message doesn't include the name
2356 of the failing file when there is a problem reading it.
bbe902f0
PH
2357
235826. Expand error message when OpenSSL has problems setting up cert/key files.
ea3a6f44 2359 As per change 25.
bbe902f0 2360
ea3a6f44
NM
236127. Reset the locale to "C" after calling embedded Perl, in case it was changed
2362 (this can affect the format of dates).
bbe902f0 2363
ea3a6f44
NM
236428. exim_tidydb, when checking for the continued existence of a message for
2365 which it has found a message-specific retry record, was not finding
2366 messages that were in split spool directories. Consequently, it was
2367 deleting retry records that should have stayed in existence.
bbe902f0
PH
2368
236929. eximstats updated to version 1.35
2370 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2371 1.35 - bugfix such that pie charts by volume are generated correctly
2372
ea3a6f44
NM
237330. The SPA authentication driver was not abandoning authentication and moving
2374 on to the next authenticator when an expansion was forced to fail,
2375 contradicting the general specification for all authenticators. Instead it
2376 was generating a temporary error. It now behaves as specified.
bbe902f0 2377
ea3a6f44
NM
237831. The default ordering of permitted cipher suites for GnuTLS was pessimal
2379 (the order specifies the preference for clients). The order is now AES256,
2380 AES128, 3DES, ARCFOUR128.
bbe902f0 2381
ea3a6f44
NM
238231. Small patch to Sieve code - explicitly set From: when generating an
2383 autoreply.
bbe902f0 2384
ea3a6f44
NM
238532. Exim crashed if a remote delivery caused a very long error message to be
2386 recorded - for instance if somebody sent an entire SpamAssassin report back
2387 as a large number of 550 error lines. This bug was coincidentally fixed by
2388 increasing the size of one of Exim's internal buffers (big_buffer) that
2389 happened as part of the Exiscan merge. However, to be on the safe side, I
2390 have made the code more robust (and fixed the comments that describe what
2391 is going on).
bbe902f0 2392
ea3a6f44
NM
239333. Some experimental protocols are using DNS PTR records for new purposes. The
2394 keys for these records are domain names, not reversed IP addresses. The
2395 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2396 leaves it alone. Component reversal etc. now happens only for IP addresses.
bbe902f0
PH
2397 CAN-2005-0021
2398
ea3a6f44
NM
239934. The host_aton() function is supposed to be passed a string that is known
2400 to be a valid IP address. However, in the case of IPv6 addresses, it was
2401 not checking this. This is a hostage to fortune. Exim now panics and dies
2402 if the condition is not met. A case was found where this could be provoked
2403 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2404 components; fortuitously, this particular loophole had already been fixed
2405 by change 4.50/55 or 4.44/33 above.
2406
2407 If there are any other similar loopholes, the new check in host_aton()
2408 itself should stop them being exploited. The report I received stated that
2409 data on the command line could provoke the exploit when Exim was running as
2410 exim, but did not say which command line option was involved. All I could
2411 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2412 running as the user.
bbe902f0
PH
2413 CAN-2005-0021
2414
ea3a6f44
NM
241535. There was a buffer overflow vulnerability in the SPA authentication code
2416 (which came originally from the Samba project). I have added a test to the
2417 spa_base64_to_bits() function which I hope fixes it.
bbe902f0
PH
2418 CAN-2005-0022
2419
ea3a6f44
NM
242036. The daemon start-up calls getloadavg() while still root for those OS that
2421 need the first call to be done as root, but it missed one case: when
2422 deliver_queue_load_max is set with deliver_drop_privilege. This is
2423 necessary for the benefit of the queue runner, because there is no re-exec
2424 when deliver_drop_privilege is set.
bbe902f0 2425
ea3a6f44
NM
242637. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2427 was in use, was not putting the data itself into the right store pool;
2428 consequently, it could be overwritten for a subsequent message in the same
2429 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2430 the caching.)
bbe902f0 2431
ea3a6f44
NM
243238. Sometimes the final signoff response after QUIT could fail to get
2433 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2434 < 0 before doing a fflush(). This bug looks as though it goes back to the
2435 introduction of TLS in release 3.20, but "sometimes" must have been rare
2436 because the tests only now provoked it.
bbe902f0
PH
2437
2438
495ae4b0
PH
2439Exim version 4.43
2440-----------------
2441
2442 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2443 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2444 Now it returns an integer. A number of calls were still expecting a T/F
2445 return. Fortuitously, in all cases, the tests worked in OK situations,
2446 which is the norm. However, things would have gone wrong on any write
2447 failures on the smtp file descriptor. This function is used when sending
2448 messages over SMTP and also when doing verify callouts.
2449
2450 2. When Exim is called to do synchronous delivery of a locally submitted
2451 message (the -odf or -odi options), it no longer closes stderr before doing
2452 the delivery.
2453
2454 3. Implemented the mua_wrapper option.
2455
2456 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2457
2458 5. Implemented the functions header_remove(), header_testname(),
2459 header_add_at_position(), and receive_remove_recipient(), and exported them
2460 to local_scan().
2461
2462 6. If an ACL "warn" statement specified the addition of headers, Exim already
2463 inserted X-ACL-Warn: at the start if there was no header name. However, it
2464 was not making this test for the second and subsequent header lines if
2465 there were newlines in the string. This meant that an invalid header could
2466 be inserted if Exim was badly configured.
2467
2468 7. Allow an ACL "warn" statement to add header lines at the start or after all
2469 the Received: headers, as well as at the end.
2470
2471 8. Added the rcpt_4xx retry error code.
2472
2473 9. Added postmaster_mailfrom=xxx to callout verification option.
2474
247510. Added mailfrom=xxxx to the callout verification option, for verify=
2476 header_sender only.
2477
247811. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2479 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2480
248112. Inserted some casts to stop certain compilers warning when using pointer
2482 differences as field lengths or precisions in printf-type calls (mostly
2483 affecting debugging statements).
2484
248513. Added optional readline() support for -be (dynamically loaded).
2486
248714. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2488 same clock tick as a message's arrival, so that its received time was the
2489 same as the "first fail" time on the retry record, and that message
2490 remained on the queue past the ultimate address timeout, every queue runner
2491 would try a delivery (because it was past the ultimate address timeout) but
2492 after another failure, the ultimate address timeout, which should have then
2493 bounced the address, did not kick in. This was a "< instead of <=" error;
2494 in most cases the first failure would have been in the next clock tick
2495 after the received time, and all would be well.
2496
249715. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2498 being recognized when the domain list was tested by the match_domain
2499 condition in an expansion string.
2500
250116. Added the ${str2b64: operator.
2502
250317. Exim was always calling setrlimit() to set a large limit for the number of
2504 processes, without checking whether the existing limit was already
2505 adequate. (It did check for the limit on file descriptors.) Furthermore,
2506 errors from getrlimit() and setrlimit() were being ignored. Now they are
2507 logged to the main and panic logs, but Exim does carry on, to try to do its
2508 job under whatever limits there are.
2509
251018. Imported PCRE 5.0.
2511
251219. Trivial typo in log message " temporarily refused connection" (the leading
2513 space).
2514
251520. If the log selector return_path_on_delivery was set and an address was
2516 redirected to /dev/null, the delivery process crashed because it assumed
2517 that a return path would always be set for a "successful" delivery. In this
2518 case, the whole delivery is bypassed as an optimization, and therefore no
2519 return path is set.
2520
252121. Internal re-arrangement: the function for sending a challenge and reading
2522 a response while authentication was assuming a zero-terminated challenge
2523 string. It's now changed to take a pointer and a length, to allow for
2524 binary data in such strings.
2525
252622. Added the cyrus_sasl authenticator (code supplied by MBM).
2527
252823. Exim was not respecting finduser_retries when seeking the login of the
2529 uid under which it was called; it was always trying 10 times. (The default
2530 setting of finduser_retries is zero.) Also, it was sleeping after the final
2531 failure, which is pointless.
2532
253324. Implemented tls_on_connect_ports.
2534
253525. Implemented acl_smtp_predata.
2536
253726. If the domain in control=submission is set empty, Exim assumes that the
2538 authenticated id is a complete email address when it generates From: or
2539 Sender: header lines.
2540
254127. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2542 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2543 chown and chgrp in /bin and hostname in /usr/bin.
2544
254528. Exim was keeping the "process log" file open after each use, just as it
2546 does for the main log. This opens the possibility of it remaining open for
2547 long periods when the USR1 signal hits a daemon. Occasional processlog
2548 errors were reported, that could have been caused by this. Anyway, it seems
2549 much more sensible not to leave this file open at all, so that is what now
2550 happens.
2551
255229. The long-running daemon process does not normally write to the log once it
2553 has entered its main loop, and it closes the log before doing so. This is
2554 so that log files can straightforwardly be renamed and moved. However,
2555 there are a couple of unusual error situations where the daemon does write
2556 log entries, and I had neglected to close the log afterwards.
2557
255830. The text of an SMTP error response that was received during a remote
2559 delivery was being truncated at 512 bytes. This is too short for some of
2560 the long messages that one sometimes sees. I've increased the limit to
2561 1024.
2562
256331. It is now possible to make retry rules that apply only when a message has a
2564 specific sender, in particular, an empty sender.
2565
256632. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2567 it possible to be selective about when SMTP synchronization is enforced.
2568
256933. Added "control = caseful_local_part" and "control = "caselower_local_part".
2570
257132. Implemented hosts_connection_nolog.
2572
257333. Added an ACL for QUIT.
2574
257534. Setting "delay_warning=" to disable warnings was not working; it gave a
2576 syntax error.
2577
257835. Added mailbox_size and mailbox_filecount to appendfile.
2579
258036. Added control = no_multiline_responses to ACLs.
2581
258237. There was a bug in the logic of the code that waits for the clock to tick
2583 in the case where the clock went backwards by a substantial amount such
2584 that the microsecond fraction of "now" was more than the microsecond
2585 fraction of "then" (but the whole seconds number was less).
2586
258738. Added support for the libradius Radius client library this is found on
2588 FreeBSD (previously only the radiusclient library was supported).
2589
2590
2591Exim version 4.42
2592-----------------
2593
2594 1. When certain lookups returned multiple values in the form name=value, the
2595 quoting of the values was not always being done properly. Specifically:
2596 (a) If the value started with a double quote, but contained no whitespace,
2597 it was not quoted.
2598 (b) If the value contained whitespace other than a space character (i.e.
2599 tabs or newlines or carriage returns) it was not quoted.
2600 This fix has been applied to the mysql and pgsql lookups by writing a
2601 separate quoting function and calling it from the lookup code. The fix
2602 should probably also be applied to nisplus, ibase and oracle lookups, but
2603 since I cannot test any of those, I have not disturbed their existing code.
2604
2605 2. A hit in the callout cache for a specific address caused a log line with no
2606 reason for rejecting RCPT. Now it says "Previous (cached) callout
2607 verification failure".
2608
2609 3. There was an off-by-one bug in the queryprogram router. An over-long
2610 return line was truncated at 256 instead of 255 characters, thereby
2611 overflowing its buffer with the terminating zero. As well as fixing this, I
2612 have increased the buffer size to 1024 (and made a note to document this).
2613
2614 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2615 when Exim is waiting for an SMTP response from a remote server, Exim
2616 restarts its select() call on the socket, thereby resetting its timeout.
2617 This is not a problem when such interrupts are rare. Somebody set up a cron
2618 job to run exiwhat every 2 minutes, which is less than the normal select()
2619 timeout (5 or 10 minutes). This meant that the select() timeout never
2620 kicked in because it was always reset. I have fixed this by comparing the
2621 time when an interrupt arrives with the time at the start of the first call
2622 to select(). If more time than the timeout has elapsed, the interrupt is
2623 treated as a timeout.
2624
2625 5. Some internal re-factoring in preparation for the addition of Sieve
2626 extensions (by MH). In particular, the "personal" test is moved to a
2627 separate function, and given an option for scanning Cc: and Bcc: (which is
2628 not set for Exim filters).
2629
2630 6. When Exim created an email address using the login of the caller as the
2631 local part (e.g. when creating a From: or Sender: header line), it was not
2632 quoting the local part when it contained special characters such as @.
2633
2634 7. Installed new OpenBSD configuration files.
2635
2636 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2637 try to make them clearer.
2638
2639 9. Callout options, other than the timeout value, were being ignored when
2640 verifying sender addresses in header lines. For example, when using
2641
2642 verify = header_sender/callout=no_cache
2643
2644 the cache was (incorrectly) being used.
2645
264610. Added a missing instance of ${EXE} to the exim_install script; this affects
2647 only the Cygwin environment.
2648
264911. When return_path_on_delivery was set as a log selector, if different remote
2650 addresses in the same message used different return paths and parallel
2651 remote delivery occurred, the wrong values would sometimes be logged.
2652 (Whenever a remote delivery process finished, the return path value from
2653 the most recently started remote delivery process was logged.)
2654
265512. RFC 3848 specifies standard names for the "with" phrase in Received: header
2656 lines when AUTH and/or TLS are in use. This is the "received protocol"
2657 field. Exim used to use "asmtp" for authenticated SMTP, without any
2658 indication (in the protocol name) for TLS use. Now it follows the RFC and
2659 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2660 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2661 names appear in log lines as well as in Received: header lines.
2662
266313. Installed MH's patches for Sieve to add the "copy" and "vacation"
2664 extensions, and comparison tests, and to fix some bugs.
2665
266614. Changes to the "personal" filter test:
2667
2668 (1) The test was buggy in that it was just doing the equivalent of
2669 "contains" tests on header lines. For example, if a user's address was
2670 anne@some.where, the "personal" test would incorrectly be true for
2671
2672 To: susanne@some.where
2673
2674 This test is now done by extracting each address from the header in turn,
2675 and checking the entire address. Other tests that are part of "personal"
2676 are now done using regular expressions (for example, to check local parts
2677 of addresses in From: header lines).
2678
2679 (2) The list of non-personal local parts in From: addresses has been
2680 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2681 taken from the Sieve specification recommendations.
2682
2683 (3) If the message contains any header line starting with "List-" it is
2684 treated as non-personal.
2685
2686 (4) The test for "circular" in the Subject: header line has been removed
2687 because it now seems ill-conceived.
2688
268915. Minor typos in src/EDITME comments corrected.
2690
269116. Installed latest exipick from John Jetmore.
2692
269317. If headers_add on a router specified a text string that was too long for
2694 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2695 of string_sprintf() is now avoided.
2696
269718. $message_body_size was not set (it was always zero) when running the DATA
2698 ACL and the local_scan() function.
2699
270019. For the "mail" command in an Exim filter, no default was being set for
2701 the once_repeat time, causing a random time value to be used if "once" was
2702 specified. (If the value happened to be <= 0, no repeat happened.) The
2703 default is now 0s, meaning "never repeat". The "vacation" command was OK
2704 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2705 (I found it when inspecting the code).
2706
270720. There is now an overall timeout for performing a callout verification. It
2708 defaults to 4 times the callout timeout, which applies to individual SMTP
2709 commands during the callout. The overall timeout applies when there is more
2710 than one host that can be tried. The timeout is checked before trying the
2711 next host. This prevents very long delays if there are a large number of
2712 hosts and all are timing out (e.g. when the network connections are timing
2713 out). The value of the overall timeout can be changed by specifying an
2714 additional sub-option for "callout", called "maxwait". For example:
2715
2716 verify = sender/callout=5s,maxwait=20s
2717
271821. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2719 to the end before writing, but this should make it even safer).
2720
272122. Exim was forgetting that it had advertised PIPELINING for the second and
2722 subsequent messages on an SMTP connection. It was also not resetting its
2723 memory on STARTTLS and an internal HELO.
2724
272523. When Exim logs an SMTP synchronization error within a session, it now
2726 records whether PIPELINING has been advertised or not.
2727
272824. Added 3 instances of "(long int)" casts to time_t variables that were being
2729 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2730 rather than long int.
2731
273225. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2733
273426. Added the never_mail option to autoreply.
2735
2736
2737Exim version 4.41
2738-----------------
2739
2740 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2741 crash if the getsockname() call failed; this can happen if a connection is
2742 closed very soon after it is established. The problem was simply in the
2743 order in which certain operations were done, causing Exim to try to write
2744 to the SMTP stream before it had set up the file descriptor. The bug has
2745 been fixed by making things happen in the correct order.
2746
2747
2748Exim version 4.40
2749-----------------
2750
2751 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2752 before the connection was closed, thus losing the rejection response.
2753
2754 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2755 some early Solaris releases, but causes trouble in current releases where
2756 socklen_t is defined.
2757
2758 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2759 always exist.
2760
2761 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2762 configured.
2763
2764 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2765 that releasing the top of it at the end releases what was used for sub-
2766 expansions (unless the block got too big). However, discard this block if
2767 the first thing is a variable or header, so that we can use its block when
2768 it is dynamic (useful for very large $message_headers, for example).
2769
2770 6. Lookups now cache *every* query, not just the most recent. A new, separate
2771 store pool is used for this. It can be recovered when all lookup caches are
2772 flushed. Lookups now release memory at the end of their result strings.
2773 This has involved some general refactoring of the lookup sources.
2774
2775 7. Some code has been added to the store_xxx() functions to reduce the amount
2776 of flapping under certain conditions.
2777
2778 8. log_incoming_interface used to affect only the <= reception log lines. Now
2779 it causes the local interface and port to be added to several more SMTP log
2780 lines, for example "SMTP connection from", and rejection lines.
2781
2782 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2783
278410. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2785
278611. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2787 could be overwritten at the end of the current message (or the start of a
2788 new message if it was set in a HELO ACL). The value is now preserved for
2789 the duration of the SMTP connection.
2790
279112. If a transport had a headers_rewrite setting, and a matching header line
2792 contained an unqualified address, that address was qualified, even if it
2793 did not match any rewriting rules. The underlying bug was that the values
2794 of the flags that permit the existence of unqualified sender and recipient
2795 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2796 non-local messages, and by -bnq for local messages) were not being
2797 preserved with the message after it was received.
2798
279913. When Exim was logging an SMTP synchronization error, it could sometimes log
2800 "next input=" as part of the text comprising the host identity instead of
2801 the correct text. The code was using the same buffer for two different
2802 strings. However, depending on which order the printing function evaluated
2803 its arguments, the bug did not always show up. Under Linux, for example, my
2804 test suite worked just fine.
2805
280614. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2807 doesn't work with some older versions of Perl. It has been changed to "my",
2808 which in any case is probably the better facility to use.
2809
281015. A really picky compiler found some instances of statements for creating
2811 error messages that either had too many or two few arguments for the format
2812 string.
2813
281416. The size of the buffer for calls to the DNS resolver has been increased
2815 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2816 for addresses that have a lot of PTR records. This alleviates a problem; it
2817 does not fully solve it.
2818
281917. A dnsdb lookup for PTR records that receives more data than will fit in the
2820 buffer now truncates the list and logs the incident, which is the same
2821 action as happens when Exim is looking up a host name and its aliases.
2822 Previously in this situation something unpredictable would happen;
2823 sometimes it was "internal error: store_reset failed".
2824
282518. If a server dropped the connection unexpectedly when an Exim client was
2826 using GnuTLS and trying to read a response, the client delivery process
2827 crashed while trying to generate an error log message.
2828
282919. If a "warn" verb in an ACL added multiple headers to a message in a single
2830 string, for example:
2831
2832 warn message = H1: something\nH2: something
2833
2834 the text was added as a single header line from Exim's point of view
2835 though it ended up OK in the delivered message. However, searching for the
2836 second and subsequent header lines using $h_h2: did not work. This has been
2837 fixed. Similarly, if a system filter added multiple headers in this way,
2838 the routers could not see them.
2839
284020. Expanded the error message when iplsearch is called with an invalid key to
2841 suggest using net-iplsearch in a host list.
2842
284321. When running tests using -bh, any delays imposed by "delay" modifiers in
2844 ACLs are no longer actually imposed (and a message to that effect is
2845 output).
2846
284722. If a "gecos" field in a passwd entry contained escaped characters, in
2848 particular, if it contained a \" sequence, Exim got it wrong when building
2849 a From: or a Sender: header from that name. A second bug also caused
2850 incorrect handling when an unquoted " was present following a character
2851 that needed quoting.
2852
285323. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2854 was not being matched caselessly.
2855
285624. Arranged for all hyphens in the exim.8 source to be escaped with
2857 backslashes.
2858
285925. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2860 properly. Recipient callout cache records were still being keyed to include
2861 the sender, even when use_sender was set false. This led to far more
2862 callouts that were necessary. The sender is no longer included in the key
2863 when use_sender is false.
2864
286526. Added "control = submission" modifier to ACLs.
2866
286727. Added the ${base62d: operator to decode base 62 numbers.
2868
286928. dnsdb lookups can now access SRV records.
2870
287129. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2872 the configuration file.
2873
287430. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2875 (-v) mode. This makes the output for a verbose queue run more intelligible.
2876
287731. Added a use_postmaster feature to recipient callouts.
2878
287932. Added the $body_zerocount variable, containing the number of binary zero
2880 bytes in the message body.
2881
288233. The time of last modification of the "new" subdirectory is now used as the
2883 "mailbox time last read" when there is a quota error for a maildir
2884 delivery.
2885
288634. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2887
288835. Added +ignore_unknown as a special item in host lists.
2889
289036. Code for decoding IPv6 addresses in host lists is now included, even if
2891 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2892 address was recognized as an IP address, but was then not correctly decoded
2893 into binary, causing unexpected and incorrect effects when compared with
2894 another IP address.
2895
2896
2897Exim version 4.34
2898-----------------
2899
2900 1. Very minor rewording of debugging text in manualroute to say "list of
2901 hosts" instead of "hostlist".
2902
2903 2. If verify=header_syntax was set, and a header line with an unqualified
2904 address (no domain) and a large number of spaces between the end of the
2905 name and the colon was received, the reception process suffered a buffer
2906 overflow, and (when I tested it) crashed. This was caused by some obsolete
2907 code that should have been removed. The fix is to remove it!
2908
2909 3. When running in the test harness, delay a bit after writing a bounce
2910 message to get a bit more predictability in the log output.
2911
2912 4. Added a call to search_tidyup() just before forking a reception process. In
2913 theory, someone could use a lookup in the expansion of smtp_accept_max_
2914 per_host which, without the tidyup, could leave open a database connection.
2915
2916 5. Added the variables $recipient_data and $sender_data which get set from a
2917 lookup success in an ACL "recipients" or "senders" condition, or a router
2918 "senders" option, similar to $domain_data and $local_part_data.
2919
2920 6. Moved the writing of debug_print from before to after the "senders" test
2921 for routers.
2922
2923 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2924 problems for message scanning, either using a data ACL, or using
2925 local_scan() because the Received: header was not generated till after they
2926 were called (in order to set the time as the time of reception completion).
2927 I have revised the way this works. The header is now generated after the
2928 body is received, but before the ACL or local_scan() are called. After they
2929 are run, the timestamp in the header is updated.
2930
2931
2932Exim version 4.33
2933-----------------
2934
2935 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2936 before starting a queue runner without re-exec. This happened only when
2937 deliver_drop_privilege was set or when the Exim user was set to root. The
2938 effect of the bug was that timeouts during subsequent deliveries caused
2939 crashes instead of being properly handled. The handler is now left at its
2940 default (and expected) setting.
2941
2942 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2943 message, again when deliver_drop_privilege is set or Exim is run as root.
2944 The bug described in (1) was not present in this case, but the tidying up
2945 of the other signals was missing. I have made the two cases consistent.
2946
2947 3. The ignore_target_hosts setting on a manualroute router was being ignored
2948 for hosts that were looked up using the /MX notation.
2949
2950 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2951 in domain lists.
2952
2953 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2954 operated on the sender address. After changing the $sender_address to <>
2955 for the sender address verify, Exim was re-instated it as the original
2956 (before rewriting) address, but remembering that it had rewritten it, so it
2957 wasn't rewriting it again. This bug also had the effect of breaking the
2958 sender address verification caching when the sender address was rewritten.
2959
2960 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2961 This has been changed so that if the ip literal address matches
2962 ignore_target_hosts, the router declines.
2963
2964 7. Added expansion conditions match_domain, match_address, and match_local_
2965 part (NOT match_host).
2966
2967 8. The placeholder for the Received: header didn't have a length field set.
2968
2969 9. Added code to Exim itself and to exim_lock to test for a specific race
2970 condition that could lead to file corruption when using MBX delivery. The
2971 issue is with the lockfile that is created in /tmp. If this file is removed
2972 after a process has opened it but before that process has acquired a lock,
2973 there is the potential for a second process to recreate the file and also
2974 acquire a lock. This could lead to two Exim processes writing to the file
2975 at the same time. The added code performs the same test as UW imapd; it
2976 checks after acquiring the lock that its file descriptor still refers to
2977 the same named file.
2978
297910. The buffer for building added header lines was of fixed size, 8192 bytes.
2980 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2981 when Exim is built.
2982
298311. Added the smtp_active_hostname option. If used, this will typically be made
2984 to depend on the incoming interface address. Because $interface_address is
2985 not set up until the daemon has forked a reception process, error responses
2986 that can happen earlier (such as "too many connections") no longer contain
2987 a host name.
2988
298912. If an expansion in a condition on a "warn" statement fails because a lookup
2990 defers, the "warn" statement is abandoned, and the next ACL statement is
2991 processed. Previously this caused the whole ACL to be aborted.
2992
299313. Added the iplsearch lookup type.
2994
299514. Added ident_timeout as a log selector.
2996
299715. Added tls_certificate_verified as a log selector.
2998
299916. Added a global option tls_require_ciphers (compare the smtp transport
3000 option of the same name). This controls incoming TLS connections.
3001
300217. I finally figured out how to make tls_require_ciphers do a similar thing
3003 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3004 before starting the TLS session.
3005
300618. Tabs are now shown as \t in -bP output.
3007
300819. If the log selector return_path_on_delivery was set, Exim crashed when
3009 bouncing a message because it had too many Received: header lines.
3010
301120. If two routers both had headers_remove settings, and the first one included
3012 a superfluous trailing colon, the final name in the first list and the
3013 first name in the second list were incorrectly joined into one item (with a
3014 colon in the middle).
3015
3016
3017Exim version 4.32
3018-----------------
3019
3020 1. Added -C and -D options to the exinext utility, mainly to make it easier
3021 to include in the automated testing, but these could be helpful when
3022 multiple configurations are in use.
3023
3024 2. The exinext utility was not formatting the output nicely when there was
3025 an alternate port involved in the retry record key, nor when there was a
3026 message id as well (for retries that were specific to a specific message
3027 and a specific host). It was also confused by IPv6 addresses, because of
3028 the additional colons they contain. I have fixed the IPv4 problem, and
3029 patched it up to do a reasonable job for IPv6.
3030
3031 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3032 delivery, the log line now contains "pipelined" if PIPELINING was used.
3033
3034 4. An SMTP transport process used to panic and die if the bind() call to set
3035 an explicit outgoing interface failed. This has been changed; it is now
3036 treated in the same way as a connect() failure.
3037
3038 5. A reference to $sender_host_name in the part of a conditional expansion
3039 that was being skipped was still causing a DNS lookup. This no longer
3040 occurs.
3041
3042 6. The def: expansion condition was not recognizing references to header lines
3043 that used bh_ and bheader_.
3044
3045 7. Added the _cache feature to named lists.
3046
3047 8. The code for checking quota_filecount in the appendfile transport was
3048 allowing one more file than it should have been.
3049
3050 9. For compatibility with Sendmail, the command line option
3051
3052 -prval:sval
3053
3054 is equivalent to
3055
3056 -oMr rval -oMs sval
3057
3058 and sets the incoming protocol and host name (for trusted callers). The
3059 host name and its colon can be omitted when only the protocol is to be set.
3060 Note the Exim already has two private options, -pd and -ps, that refer to
3061 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3062 "s", but I don't think that's a major issue.
3063
306410. A number of refactoring changes to the code, none of which should affect
3065 Exim's behaviour:
3066
3067 (a) The number of logging options was getting close to filling up the
3068 32-bit word that was used as a bit map. I have split them into two classes:
3069 those that are passed in the argument to log_write(), and those that are
3070 only ever tested independently outside of that function. These are now in
3071 separate 32-bit words, so there is plenty of room for expansion again.
3072 There is no change in the user interface or the logging behaviour.
3073
3074 (b) When building, for example, log lines, the code previously used a
3075 macro that called string_cat() twice, in order to add two strings. This is
3076 not really sufficiently general. Furthermore, there was one instance where
3077 it was actually wrong because one of the argument was used twice, and in
3078 one call a function was used. (As it happened, calling the function twice
3079 did not affect the overall behaviour.) The macro has been replaced by a
3080 function that can join an arbitrary number of extra strings onto a growing
3081 string.
3082
3083 (c) The code for expansion conditions now uses a table and a binary chop
3084 instead of a serial search (which was left over from when there were very
3085 few conditions). Also, it now recognizes conditions like "pam" even when
3086 the relevant support is not compiled in: a suitably worded error message is
3087 given if an attempt is made to use such a condition.
3088
308911. Added ${time_interval:xxxxx}.
3090
309112. A bug was causing one of the ddress fields not to be passed back correctly
3092 from remote delivery subprocesses. The field in question was not being
3093 subsequently used, so this caused to problems in practice.
3094
309513. Added new log selectors queue_time and deliver_time.
3096
309714. Might have fixed a bug in maildirsizefile handling that threw up
3098 "unexpected character" debug warnings, and recalculated the data
3099 unnecessarily. In any case, I expanded the warning message to give more
3100 information.
3101
310215. Added the message "Restricted characters in address" to the statements in
3103 the default ACL that block characters like @ and % in local parts.
3104
310516. Change 71 for release 4.31 proved to be much less benign that I imagined.
3106 Three changes have been made:
3107
3108 (a) There was a serious bug; a negative response to MAIL caused the whole
3109 recipient domain to be cached as invalid, thereby blocking all messages
3110 to all local parts at the same domain, from all senders. This bug has
3111 been fixed. The domain is no longer cached after a negative response to
3112 MAIL if the sender used is not empty.
3113
3114 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3115 been restored.
3116
3117 (c) A new callout option, "use_sender" has been added for people who want
3118 the modified behaviour.
3119
3120
3121Exim version 4.31
3122-----------------
3123
3124 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3125 Larry Rosenman.
3126
3127 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3128 indeed breaks things for older releases.
3129
3130 3. Added additional logging to the case where there is a problem reading data
3131 from a filter that is running in a subprocess using a pipe, in order to
3132 try to track down a specific problem.
3133
3134 4. Testing facility fudge: when running in the test harness and attempting
3135 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3136 getting "No route to host". Convert this to a timeout.
3137
3138 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3139 warning.
3140
3141 6. Some OS don't have socklen_t but use size_t instead. This affects the
3142 fifth argument of getsockopt() amongst other things. This is now
3143 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3144 can be set for individual OS. I have set it for SunOS5, OSF1, and
3145 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3146 some earlier ones do not.
3147
3148 7. Change 4.30/15 was not doing the test caselessly.
3149
3150 8. The standard form for an IPv6 address literal was being rejected by address
3151 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3152 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3153 this, as well as the form without the "IPv6" on the front (but only when
3154 address literals are enabled, of course).
3155
3156 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3157
315810. Exim crashed if a message with an empty sender address specified by -f
3159 encountered a router with an errors_to setting. This could be provoked only
3160 by a command such as
3161
3162 exim -f "" ...
3163
3164 where an empty string was supplied; "<>" did not hit this bug.
3165
316611. Installed PCRE release 4.5.
3167
316812. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3169 remained set. It is now erased.
3170
317113. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3172 times from message ids (which are base 36 rather than the normal 62).
3173
317414. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3175 were being counted as actual protocol errors, and logged if the log
3176 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3177 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3178 and DATA following a set of rejected RCPTs do not count as protocol errors.
3179 In other words, Exim assumes they were pipelined, though this may not
3180 actually be the case. Of course, in all cases the client gets an
3181 appropriate error code.
3182
318315. If a lookup fails in an ACL condition, a message about the failure may
3184 be available; it is used if testing the ACL cannot continue, because most
3185 such messages specify what the cause of the deferral is. However, some
3186 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3187 that caused an old message to be retained and used if a later statement
3188 caused a defer, replacing the real cause of the deferral.
3189
319016. If an IP address had so many PTR records that the DNS lookup buffer
3191 was not large enough to hold them, Exim could crash while trying to process
3192 the truncated data. It now detects and logs this case.
3193
319417. Further to 4.21/58, another change has been made: if (and only if) the
3195 first line of a message (the first header line) ends with CRLF, a bare LF
3196 in a subsequent header line has a space inserted after it, so as not to
3197 terminate the header.
3198
319918. Refactoring: tidied an ugly bit of code in appendfile that copied data
3200 unnecessarily, used atoi() instead of strtol(), and didn't check the
3201 termination when getting file sizes from file names by regex.
3202
320319. Completely re-implemented the support for maildirsize files, in the light
3204 of a number of problems with the previous contributed implementation
3205 (4.30/29). In particular:
3206
3207 . If the quota is zero, the maildirsize file is maintained, but no quota is
3208 imposed.
3209
3210 . If the maildir directory does not exist, it is created before any attempt
3211 to write a maildirsize file.
3212
3213 . The quota value in the file is just a cache; if the quota is changed in
3214 the transport, the new value overrides.
3215
3216 . A regular expression is available for excluding directories from the
3217 count.
3218
321920. The autoreply transport checks the characters in options that define the
3220 message's headers; it allows continued headers, but it was checking with
3221 isspace() after an embedded newline instead of explicitly looking for a
3222 space or a tab.
3223
322421. If all the "regular" hosts to which an address was routed had passed their
3225 expiry times, and had not reached their retry times, the address was
3226 bounced, even if fallback hosts were defined. Now Exim should go on to try
3227 the fallback hosts.
3228
322922. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3230 equivalent code in the SMTP transport. Some hosts send humungous responses
3231 to HELO/EHLO, more than 1024 it seems.
3232
323323. Refactoring: code in filter.c used (void *) for "any old type" but this
3234 gives compiler warnings in some environments. I've now done it "properly",
3235 using a union.
3236
323724. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3238 (because of problems with the built-in one) was declared to return uschar *
3239 instead of char *, causing compiler failure.
3240
324125. Fixed a file descriptor leak when processing alias/forward files.
3242
324326. Fixed a minor format string issue in dbfn.c.
3244
324527. Typo in exim.c: ("dmbnz" for "dbmnz").
3246
324728. If a filter file refered to $h_xxx or $message_headers, and the headers
3248 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3249 become corrupted.
3250
325129. When a sender address is verified, it is cached, to save repeating the test
3252 when there is more than one recipient in a message. However, when the
3253 verification involves a callout, it is possible for different callout
3254 options to be set for different recipients. It is too complicated to keep
3255 track of this in the cache, so now Exim always runs a verification when a
3256 callout is required, relying on the callout cache for the optimization.
3257 The overhead is duplication of the address routing, but this should not be
3258 too great.
3259
326030. Fixed a bug in callout caching. If a RCPT command caused the sender address
3261 to be verified with callout=postmaster, and the main callout worked but the
3262 postmaster check failed, the verification correctly failed. However, if a
3263 subsequent RCPT command asked for sender verification *without* the
3264 postmaster check, incorrect caching caused this verification also to fail,
3265 incorrectly.
3266
326731. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3268 it was not caching the DNS options (qualify_single, search_parents) that
3269 were used when the lookup failed. A subsequent lookup with different
3270 options therefore always gave the same answer, though there were cases
3271 where it should not have. (Example: a "domains = !$mx_any" option on a
3272 dnslookup router: the "domains" option is always processed without any
3273 widening, but the router might have qualify_single set.) Now Exim uses the
3274 cached value only when the same options are set.
3275
327632. Added John Jetmore's "exipick" utility to the distribution.
3277
327833. GnuTLS: When an attempt to start a TLS session fails for any reason other
3279 than a timeout (e.g. a certificate is required, and is not provided), an
3280 Exim server now closes the connection immediately. Previously it waited for
3281 the client to close - but if the client is SSL, it seems that they each
3282 wait for each other, leading to a delay before one of them times out.
3283
328434: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3285 maintained 0.8.x compatibility because I don't think many are using it, and
3286 it is clearly obsolete.
3287
328835. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3289 transport.
3290
329136. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3292 client certificate was expired. A simple patch fixes this, though I don't
3293 understand the full logic of why the verify callback is called multiple
3294 times.
3295
329637. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3297 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3298 which causes problems with some clients (such as the Certicom SSL Plus
3299 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3300 disables the coutermeasure allowing Eudora to connect."
3301
330238. Exim was not checking that a write() to a log file succeeded. This could
3303 lead to Bad Things if a log got too big, in particular if it hit a file
3304 size limit. Exim now panics and dies if it cannot write to a log file, just
3305 as it does if it cannot open a log file.
3306
330739. Modified OS/Makefile-Linux so that it now contains
3308
3309 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3310
3311 The two -D definitions ensure that Exim is compiled with large file
3312 support, which makes it possible to handle log files that are bigger than
3313 2^31.
3314
331540. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3316 instance) a domain was checked against a named list that involved a lookup,
3317 causing $domain_data to be set, then another domain was checked against the
3318 same list, then the first domain was re-checked, the value of $domain_data
3319 after the final check could be wrong. In particular, if the second check
3320 failed, it could be set empty. This bug probably also applied to
3321 $localpart_data.
3322
332341. The strip_trailing_dot option was not being applied to the address given
3324 with the -f command-line option.
3325
332642. The code for reading a message's header from the spool was incrementing
3327 $received_count, but never initializing it. This meant that the value was
3328 incorrect (doubled) while delivering a message in the same process in which
3329 it was received. In the most common configuration of Exim, this never
3330 happens - a fresh exec is done - but it can happen when
3331 deliver_drop_privilege is set.
3332
333343. When Exim logs an SMTP synchronization error - client data sent too soon -
3334 it now includes up to 150 characters of the unexpected data in the log
3335 line.
3336
333744. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3338 and building data strings. The size of both of these buffers was 10 000
3339 bytes - far larger than anybody would *ever* want, thought I. Needless to
3340 say, somebody hit the limit. I have increased the maximum line length to
3341 20 000 and the maximum data length of concatenated lines to 100 000. I have
3342 also fixed two bugs, because there was no checking on these buffers. Tsk,
3343 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3344 buffer is too small.
3345
334645. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3347 lsearch lookups. Now it does.
3348
334946. When parsing a route_list item in a manualroute router, a fixed-length
3350 buffer was used for the list of hosts. I made this 1024 bytes long,
3351 thinking that nobody would ever have a list of hosts that long. Wrong.
3352 Somebody had a whole pile of complicated expansion conditions, and the
3353 string was silently truncated, leading to an expansion error. It turns out
3354 that it is easier to change to an unlimited length (owing to other changes
3355 that have happened since this code was originally written) than to build
3356 structure for giving a limitation error. The length of the item that
3357 expands into the list of hosts is now unlimited.
3358
335947. The lsearch lookup could not handle data where the length of text line was
3360 more than 4095 characters. Such lines were truncated, leading to shortened
3361 data being returned. It should now handle lines of any length.
3362
336348. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3364 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3365 ACL").
3366
336749. Cosmetic tidy to scripts like exicyclog that are generated by globally
3368 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3369 no longer happens in comment lines. A list of replacements is now placed
3370 at the head of all of the source files, except those whose only change is
3371 to replace PERL_COMMAND in the very first #! line.
3372
337350. Replaced the slow insertion sort in queue.c, for sorting the list of
3374 messages on the queue, with a bottom-up merge sort, using code contributed
3375 by Michael Haardt. This should make operations like -bp somewhat faster on
3376 large queues. It won't affect queue runners, except when queue_run_in_order
3377 is set.
3378
337951. Installed eximstats 1.31 in the distribution.
3380
338152. Added support for SRV lookups to the dnslookup router.
3382
338353. If an ACL referred to $message_body or $message_body_end, the value was not
3384 reset for any messages that followed in the same SMTP session.
3385
338654. The store-handling optimization for building very long strings was not
3387 differentiating between the different store pools. I don't think this
3388 actually made any difference in practice, but I've tidied it.
3389
339055. While running the routers to verify a sender address, $sender_address
3391 was still set to the sender address. This is wrong, because when routing to
3392 send a bounce to the sender, it would be empty. Therefore, I have changed
3393 it so that, while verifying a sender address, $sender_address is set to <>.
3394 (There is no change to what happens when verifying a recipient address.)
3395
339656. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3397 target A or AAAA records (if not already returned) without resetting the
3398 qualify_single or search_parents options of the DNS resolver. These are
3399 inappropriate in this case because the targets of MX and SRV records must
3400 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3401 target that, when qualified, matched something in the local domain. These
3402 two options are now turned off when doing these lookups.
3403
340457. It seems that at least some releases of Reiserfs (which does not have the
3405 concept of a fixed number of inodes) returns zero and not -1 for the
3406 number of available inodes. This interacted badly with check_spool_inodes,
3407 which assumed that -1 was the "no such thing" setting. What I have done is
3408 to check that the total number of inodes is greater than zero before doing
3409 the test of how many are available.
3410
341158. When a "warn" ACL statement has a log_message modifier, the message is
3412 remembered, and not repeated. This is to avoid a lot of repetition when a
3413 message has many recipients that cause the same warning to be written.
3414 Howewer, Exim was preserving the list of already written lines for an
3415 entire SMTP session, which doesn't seem right. The memory is now reset if a
3416 new message is started.
3417
341859. The "rewrite" debugging flag was not showing the result of rewriting in the
3419 debugging output unless log_rewrite was also set.
3420
342160. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3422 of (int)(handle) when we know that handle contains (void *)(-1).
3423
342461. The Exim daemon panic-logs an error return when it closes the incoming
3425 connection. However "connection reset by peer" seems to be common, and
3426 isn't really an error worthy of noting specially, so that particular error
3427 is no long logged.
3428
342962. When Exim is trying to find all the local interfaces, it used to panic and
3430 die if the ioctl to get the interface flags failed. However, it seems that
3431 on at least one OS (Solaris 9) it is possible to have an interface that is
3432 included in the list of interfaces, but for which you get a failure error
3433 for this call. This happens when the interface is not "plumbed" into a
3434 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3435 failure of the "get flags" call assumes that the interface is down.
3436
343763. Added a ${eval10: operator, which assumes all numbers are decimal. This
3438 makes life easier for people who are doing arithmetic on fields extracted
3439 from dates, where you often get leading zeros that should not be
3440 interpreted as octal.
3441
344264. Added qualify_domain to the redirect router, to override the global
3443 setting.
3444
344565. If a pathologically long header line contained very many addresses (the
3446 report of this problem mentioned 10 000) and each of them was rewritten,
3447 Exim could use up a very large amount of memory. (It kept on making new
3448 copies of the header line as it rewrote, and never released the old ones.)
3449 At the expense of a bit more processing, the header rewriting function has
3450 been changed so that it no longer eats memory in this way.
3451
345266. The generation of the Received: header has been moved from the time that a
3453 message starts to be received, to the time that it finishes. The timestamp
3454 in the Received: header should now be very close to that of the <= log
3455 line. There are two side-effects of this change:
3456
3457 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3458 logged header lines no longer include the local Received: line, because
3459 it has not yet been created. The same applies to a copy of the message
3460 that is returned to a non-SMTP sender when a message is rejected.
3461
3462 (b) When a filter file is tested using -bf, no additional Received: header
3463 is added to the test message. After some thought, I decided that this
3464 is a bug fix.
3465
3466 This change does not affect the value of $received_for. It is still set
3467 after address rewriting, but before local_scan() is called.
3468
346967. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3470
347168. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3472 gave an unhelpful panic error message, and a defer error. I have managed to
3473 change this behaviour so that it now rejects any supplied certificate,
3474 which seems right, as the list of acceptable certificates is empty.
3475
347669. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3477 gave an unhelpful defer error. I have not managed to make this reject any
3478 supplied certificates, but the error message it gives is "no certificate
3479 supplied", which is not helpful.
3480
348170. exigrep's output now also includes lines that are not associated with any
3482 message, but which match the given pattern. Implemented by a patch from
3483 Martin Sluka, which also tidied up the Perl a bit.
3484
348571. Recipient callout verification, like sender verification, was using <> in
3486 the MAIL FROM command. This isn't really the right thing, since the actual
3487 sender may affect whether the remote host accepts the recipient or not. I
3488 have changed it to use the actual sender in the callout; this means that
3489 the cache record is now keyed on a recipient/sender pair, not just the
3490 recipient address. There doesn't seem to be a real danger of callout loops,
3491 since a callout by the remote host to check the sender would use <>.
3492 [SEE ABOVE: changed after hitting problems.]
3493
349472. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3495 temporary errors. However, in the case of such a code being given after
3496 the end of a data transmission (i.e. after ".") Exim was failing to write
3497 a retry record for the message. (Yes, there was some broken host that was
3498 actually sending 8xx at this point.)
3499
350073. An unknown lookup type in a host list could cause Exim to panic-die when
3501 the list was checked. (An example that provoked this was putting <; in the
3502 middle of a list instead of at the start.) If this happened during a DATA
3503 ACL check, a -D file could be left lying around. This kind of configuration
3504 error no longer causes Exim to die; instead it causes a defer errror. The
3505 incident is still logged to the main and panic logs.
3506
350774. Buglet left over from Exim 3 conversion. The message "too many messages
3508 in one connection" was written to the rejectlog but not the mainlog, except
3509 when address rewriting (yes!) was being logged.
3510
351175. Added write_rejectlog option.
3512
351376. When a system filter was run not as root (that is, when system_filter_user
3514 was set), the values of the $n variables were not being returned to the
3515 main process; thus, they were not subsequently available in the $sn
3516 variables.
3517
351877. Added +return_path_on_delivery log selector.
3519
352078. A connection timeout was being treated differently from recipients deferred
3521 when testing hosts_max_try with a message that was older than the host's
3522 retry timeout. (The host should not be counted, thus allowing all hosts to
3523 be tried at least once before bouncing.) This may have been the cause of an
3524 occasionally reported bug whereby a message would remain on the queue
3525 longer than the retry timeout, but would be bounced if a delivery was
3526 forced. I say "may" because I never totally pinned down the problem;
3527 setting up timeout/retry tests is difficult. See also the next item.
3528
352979. The ultimate address timeout was not being applied to errors that involved
3530 a combination of host plus message (for example, a timeout on a MAIL
3531 command). When an address resolved to a number of possible hosts, and they
3532 were not all tried for each delivery (e.g. because of hosts_max_try), a
3533 message could remain on the queue longer than the retry timeout.
3534
353580. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3536 Haardt.
3537
353881. Fixed an obscure SMTP outgoing bug which required at least the following
3539 conditions: (a) there was another message waiting for the same server;
3540 (b) the server returned 5xx to all RCPT commands in the first message so
3541 that the message was not completed; (c) the server dropped the connection
3542 or gave a negative response to the RSET that Exim sends to abort the
3543 transaction. The observed case was a dropped connection after DATA that had
3544 been sent in pipelining mode. That is, the server had advertised PIPELINING
3545 but was not implementing it correctly. The effect of the bug was incorrect
3546 behaviour, such as trying another host, and this could lead to a crash.
3547
3548
3549Exim version 4.30
3550-----------------
3551
3552 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3553 and daemon.c were passed as pointers to ints; they should have been
3554 pointers to socklen_t variables (which are typically unsigned ints).
3555
3556 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3557 fixed.
3558
3559 3. Fixed a really odd bug that affected only the testing scheme; patching a
3560 certain fixed string in the binary changed the value of another string that
3561 happened to be identical to the end of the original first string.
3562
3563 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3564 name", it returns that address as the IP address. On some operating
3565 systems (e.g. Solaris), it also passes back the IP address string as the
3566 "host name". However, on others (e.g. Linux), it passes back an empty
3567 string. Exim wasn't checking for this, and was changing the host name to an
3568 empty string, assuming it had been canonicized.
3569
3570 5. Although rare, it is permitted to have more than one PTR record for a given
3571 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3572 all the names associated with an address, because they do in Solaris.
3573 However, it seems that they do not in Linux for data that comes from the
3574 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3575 I found this out when I moved to a new Linux workstation and tried to run
3576 the Exim test suite.
3577
3578 To get round this problem I have changed the code so that it now does its
3579 own call to the DNS to look up PTR records when searching for a host name.
3580 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3581 addresses that are only in /etc/hosts are still found.
3582
3583 This behaviour is, however, controlled by an option called host_lookup_
3584 order, which defaults to "bydns:byaddr". If people want to use the other
3585 order, or indeed, just use one or the other means of lookup, they can
3586 specify it in this variable.
3587
3588 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3589 some operating systems, this comes back from gethostbyaddr() as an empty
3590 string, and this is what Exim used to test for. However, it seems that in
3591 other systems, "." is yielded. Exim now tests for this case too.
3592
3593 7. The values of check_spool_space and check_log_space are now held internally
3594 as a number of kilobytes instead of an absolute number of bytes. If a
3595 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3596 kilobyte. This means that much larger values can be stored.
3597