DMARC documentation and license
[exim.git] / doc / doc-txt / ChangeLog
CommitLineData
495ae4b0
PH
1Change log file for Exim from version 4.21
2-------------------------------------------
3
2c422e6f 4Exim version 4.82
98a90c36
PP
5-----------------
6
7PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
8
12f69989
PP
9PP/02 Make -n do something, by making it not do something.
10 When combined with -bP, the name of an option is not output.
11
54c90be1
PP
12PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
13 by GnuTLS.
14
1f4a55da
PP
15PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
16 $sender_host_name and config options to manage this, and basic check
17 routines.
18
13363eba 19PP/05 DSCP support for outbound connections and control modifier for inbound.
36a3ae5f 20
66645890 21PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
e402235f
PP
22 (Only plugin which currently uses this is kerberos4, which nobody should
23 be using, but we should make it available and other future plugins might
24 conceivably use it, even though it would break NAT; stuff *should* be
25 using channel bindings instead).
66645890 26
a3fb9793 27PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
f4ee74ac
PP
28 name; added for Sendmail compatibility; requires admin caller.
29 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
30 just ignore it); requires trusted caller.
a3fb9793 31 Also parse but ignore: -Ac -Am -X<logfile>
f4ee74ac 32 Bugzilla 1117.
a3fb9793 33
d27f98fe 34TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
98a90c36 35
6822b909
TL
36TL/02 Add +smtp_confirmation as a default logging option.
37
e7568d51
TL
38TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
39 Patch by Magnus Holmgren from 2007-02-20.
40
ae0e32ee 41TL/04 Bugzilla 1281 - Spec typo.
ca0ff207 42 Bugzilla 1283 - Spec typo.
97f42f10 43 Bugzilla 1290 - Spec grammar fixes.
ca0ff207
TL
44
45TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
ae0e32ee 46
976b7e9f
JH
47JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
48
be4a1376
JH
49JH/02 Support "G" suffix to numbers in ${if comparisons.
50
ec4b68e5
PP
51PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
52
d7148a07
NM
53NM/01 Bugzilla 1197 - Spec typo
54 Bugzilla 1196 - Spec examples corrections
ec4b68e5 55
585121e2 56JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
ec4b68e5 57
a5f239e4
PP
58PP/09 Add gnutls_enable_pkcs11 option.
59
13d08c90
PP
60PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
61 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
62
bef3ea7f
JH
63JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
64 "acl {{name}{arg}...}", and optional args on acl condition
65 "acl = name arg..."
a5f239e4 66
846726c5
JH
67JH/05 Permit multiple router/transport headers_add/remove lines.
68
3a796370
JH
69JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
70
ea722490 71JH/07 Avoid using a waiting database for a single-message-only transport.
8b260705
PP
72 Performance patch from Paul Fisher. Bugzilla 1262.
73
b1b05573
JH
74JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
75 Bugzilla 884.
76
362145b5
JH
77JH/09 Add $headers_added variable, with content from use of ACL modifier
78 add_header (but not yet added to the message). Bugzilla 199.
79
3c0a92dc
JH
80JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
81 Pulled from Bugzilla 817 by Wolfgang Breyha.
82
6d7c6175
PP
83PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
84 CVE-2012-5671
85
6f123593
JH
86JH/11 Add A= logging on delivery lines, and a client_set_id option on
87 authenticators.
88
c8e2fc1e
JH
89JH/12 Add optional authenticated_sender logging to A= and a log_selector
90 for control.
91
005ac57f
PP
92PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
93
3f1df0e3
PP
94PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
95 advertise SMTP AUTH mechanism to us, instead of a generic
96 protocol violation error. Also, make Exim more robust to bad
97 data from the Dovecot auth socket.
98
67bd1ab3
TF
99TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
100
101 When a queue runner is handling a message, Exim first routes the
102 recipient addresses, during which it prunes them based on the retry
103 hints database. After that it attempts to deliver the message to
104 any remaining recipients. It then updates the hints database using
105 the retry rules.
106
107 So if a recipient address works intermittently, it can get repeatedly
108 deferred at routing time. The retry hints record remains fresh so the
109 address never reaches the final cutoff time.
110
111 This is a fairly common occurrence when a user is bumping up against
112 their storage quota. Exim had some logic in its local delivery code
113 to deal with this. However it did not apply to per-recipient defers
114 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
115
1ddeb334
TF
116 This change adds a proper retry rule check during routing so that the
117 final cutoff time is checked against the message's age. We only do
118 this check if there is an address retry record and there is not a
119 domain retry record; this implies that previous attempts to handle
120 the address had the retry_use_local_parts option turned on. We use
121 this as an approximation for the destination being like a local
122 delivery, as in LMTP.
67bd1ab3
TF
123
124 I suspect this new check makes the old local delivery cutoff check
125 redundant, but I have not verified this so I left the code in place.
126
326cdc37
TF
127TF/02 Correct gecos expansion when From: is a prefix of the username.
128
129 Test 0254 submits a message to Exim with the header
130
131 Resent-From: f
132
133 When I ran the test suite under the user fanf2, Exim expanded
134 the header to contain my full name, whereas it should have added
135 a Resent-Sender: header. It erroneously treats any prefix of the
136 username as equal to the username.
137
138 This change corrects that bug.
139
f62514b3
GF
140GF/01 DCC debug and logging tidyup
141 Error conditions log to paniclog rather than rejectlog.
142 Debug lines prefixed by "DCC: " to remove any ambiguity.
143
eb505532
TF
144TF/03 Avoid unnecessary rebuilds of lookup-related code.
145
14c7b357
PP
146PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
147 Bug spotted by Jeremy Harris; was flawed since initial commit.
148 Would have resulted in OCSP responses post-SNI triggering an Exim
149 NULL dereference and crash.
150
94eaf700
PP
151JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
152
6f5a440a
PP
153PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
154 Bug detection, analysis and fix by Samuel Thibault.
155 Bugzilla 1331, Debian bug #698092.
156
514ee161
SC
157SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
158
fd98a5c6
JH
159JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
160 Server implementation by Todd Lyons, client by JH.
161 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
162 config variable "prdr_enable" controls whether the server
163 advertises the facility. If the client requests PRDR a new
164 acl_data_smtp_prdr ACL is called once for each recipient, after
165 the body content is received and before the acl_smtp_data ACL.
166 The client is controlled by bolth of: a hosts_try_prdr option
167 on the smtp transport, and the server advertisement.
168 Default client logging of deliveries and rejections involving
169 PRDR are flagged with the string "PRDR".
170
035c7f1e
PP
171PP/16 Fix problems caused by timeouts during quit ACLs trying to double
172 fclose(). Diagnosis by Todd Lyons.
173
ff284120
PP
174PP/17 Update configure.default to handle IPv6 localhost better.
175 Patch by Alain Williams (plus minor tweaks).
176 Bugzilla 880.
177
26e72755
PP
178PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
179 This is now consistent with GnuTLS, and is now documented: the
180 previous undocumented portable approach to treating the option as
181 unset was to force an expansion failure. That still works, and
182 an empty string is now equivalent.
183
0fbd9bff
PP
184PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
185 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
186 not performing validation itself.
187
700d22f3
PP
188PP/20 Added force_command boolean option to pipe transport.
189 Patch from Nick Koston, of cPanel Inc.
190
6d7c6175 191
4263f395
PP
192Exim version 4.80.1
193-------------------
194
195PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
196 CVE-2012-5671
2c422e6f 197 This, or similar/improved, will also be change PP/11 of 4.82.
3c0a92dc 198
ea722490 199
b1770b6e 200Exim version 4.80
0599f9cf
PP
201-----------------
202
203PP/01 Handle short writes when writing local log-files.
204 In practice, only affects FreeBSD (8 onwards).
205 Bugzilla 1053, with thanks to Dmitry Isaikin.
206
23c7e742
NM
207NM/01 Bugzilla 949 - Documentation tweak
208
b322aac8
NM
209NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
210 improved.
211
4a891427
NM
212NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
213
c1e794ba 214PP/02 Implemented gsasl authenticator.
b322aac8 215
97753960
PP
216PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
217
218PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
219 `pkg-config foo` for cflags/libs.
220
df6303fa
PP
221PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
222 with rest of GSASL and with heimdal_gssapi.
223
7e6a8985
PP
224PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
225 `pkg-config foo` for cflags/libs for the TLS implementation.
226
f1e05cc7 227PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
20aa9dbd
PP
228 properties get this fed in as external SSF. A number of robustness
229 and debugging improvements to the cyrus_sasl authenticator.
b322aac8 230
4c287009
PP
231PP/08 cyrus_sasl server now expands the server_realm option.
232
b98bb9ac
PP
233PP/09 Bugzilla 1214 - Log authentication information in reject log.
234 Patch by Jeremy Harris.
235
4a6a987a
PP
236PP/10 Added dbmjz lookup type.
237
c45dd180 238PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
c7955b11 239
7db8d074
PP
240PP/12 MAIL args handles TAB as well as SP, for better interop with
241 non-compliant senders.
242 Analysis and variant patch by Todd Lyons.
243
eae0036b 244NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
e0df1c83
DM
245 Bug report from Lars Müller <lars@samba.org> (via SUSE),
246 Patch from Dirk Mueller <dmueller@suse.com>
247
dec5017e
PP
248PP/13 tls_peerdn now print-escaped for spool files.
249 Observed some $tls_peerdn in wild which contained \n, which resulted
250 in spool file corruption.
251
c80c5570
PP
252PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
253 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
254 or write after TLS renegotiation, which otherwise led to messages
255 "Got SSL error 2".
256
076b11e2
PP
257TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
258 as a tracking header (ie: a signed header comes before the signature).
259 Patch from Wolfgang Breyha.
260
5407bfff
JH
261JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
262 comma-sep list; embedded commas doubled.
263
9e45c72b
PP
264JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
265
e74376d8
PP
266PP/15 LDAP: Check for errors of TLS initialisation, to give correct
267 diagnostics.
268 Report and patch from Dmitry Banschikov.
269
da3ad30d
PP
270PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
271 Removed SSL_clear() after SSL_new() which led to protocol negotiation
272 failures. We appear to now support TLS1.1+ with Exim.
273
7be682ca
PP
274PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
275 lets Exim select keys and certificates based upon TLS SNI from client.
3f0945ff
PP
276 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
277 before an outbound SMTP session. New log_selector, +tls_sni.
7be682ca 278
ef840681
PP
279PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
280 NULL dereference. Report and patch from Alun Jones.
281
5bfb4cdf
PP
282PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
283 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
284 Not seeing resolver debug output on NetBSD, but suspect this is a
285 resolver implementation change.
286
c6e95d22
PP
287PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
288 Left warnings. Added "eximon gdb" invocation mode.
289
9cbad13b
PP
290PP/21 Defaulting "accept_8bitmime" to true, not false.
291
9ee44efb
PP
292PP/22 Added -bw for inetd wait mode support.
293
6a6084f8
PP
294PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
295 locate the relevant includes and libraries. Made this the default.
296
12dd53c7
PP
297PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
298 Bugzilla 1246, report and most of solution from Tomasz Kusy.
299
9e45c72b 300JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
97d17305
JH
301 This may cause build issues on older platforms.
302
17c76198
PP
303PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
304 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
305 gnutls_require_mac & gnutls_require_protocols (no longer supported).
306 Added SNI support via GnuTLS too.
af3498d6 307 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
17c76198 308
53947857 309PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
3f7eeb86 310
eae0036b 311PP/27 Applied dnsdb SPF support patch from Janne Snabb.
8ee4b30e
PP
312 Applied second patch from Janne, implementing suggestion to default
313 multiple-strings-in-record handling to match SPF spec.
eae0036b 314
9e45c72b 315JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
2605c55b 316
7390e768
PP
317PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
318 read-only, out of scope).
319 Patch from Wolfgang Breyha, report from Stuart Northfield.
320
08488c86
PP
321PP/29 Fix three issues highlighted by clang analyser static analysis.
322 Only crash-plausible issue would require the Cambridge-specific
323 iplookup router and a misconfiguration.
324 Report from Marcin Mirosław.
325
6475bd82
PP
326PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
327
81f91683
PP
328PP/31 %D in printf continues to cause issues (-Wformat=security), so for
329 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
330 As part of this, removing so much warning spew let me fix some minor
331 real issues in debug logging.
332
5779e6aa
PP
333PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
334 assignment on my part. Fixed.
335
3375e053
PP
336PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
337 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
338 Janne Snabb (who went above and beyond: thank you).
339
340PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
341 string otherwise requires a connection and a bunch more work and it's
78e0c7a3
PP
342 relatively easy to get wrong. Should also expose TLS library linkage
343 problems.
3375e053 344
9d26b8c0
PP
345PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
346 64-bit ${eval} (JH/03).
347
57eb9e91 348PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
b87a6e0e
PP
349 GNU libc to support some of the 64-bit stuff, should not lead to
350 conflicts. Defined before os.h is pulled in, so if a given platform
351 needs to override this, it can.
352
16880d1a
PP
353PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
354 protection layer was required, which is not implemented.
355 Bugzilla 1254, patch from Wolfgang Breyha.
356
a799883d
PP
357PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
358 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
359 tls_dhparam take prime identifiers. Also unbreak combination of
360 OpenSSL+DH_params+TLSSNI.
361
3ecab157 362PP/39 Disable SSLv2 by default in OpenSSL support.
f0f5a555 363
0599f9cf 364
867fcbf5
PP
365Exim version 4.77
366-----------------
367
368PP/01 Solaris build fix for Oracle's LDAP libraries.
369 Bugzilla 1109, patch from Stephen Usher.
370
f1a29782
TF
371TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
372
ab42bd23
TK
373TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
374 whitespace trailer
867fcbf5 375
0ca0cf52
TF
376TF/02 Fix a couple more cases where we did not log the error message
377 when unlink() failed. See also change 4.74-TF/03.
378
921b12ca
TF
379TF/03 Make the exiwhat support code safe for signals. Previously Exim might
380 lock up or crash if it happened to be inside a call to libc when it
381 got a SIGUSR1 from exiwhat.
382
383 The SIGUSR1 handler appends the current process status to the process
384 log which is later printed by exiwhat. It used to use the general
385 purpose logging code to do this, but several functions it calls are
386 not safe for signals.
387
388 The new output code in the SIGUSR1 handler is specific to the process
389 log, and simple enough that it's easy to inspect for signal safety.
390 Removing some special cases also simplifies the general logging code.
391 Removing the spurious timestamps from the process log simplifies
392 exiwhat.
393
c99ce5c9
TF
394TF/04 Improved ratelimit ACL condition.
395
396 The /noupdate option has been deprecated in favour of /readonly which
397 has clearer semantics. The /leaky, /strict, and /readonly update modes
398 are mutually exclusive. The update mode is no longer included in the
399 database key; it just determines when the database is updated. (This
400 means that when you upgrde Exim will forget old rate measurements.)
401
402 Exim now checks that the per_* options are used with an update mode that
403 makes sense for the current ACL. For example, when Exim is processing a
404 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
405 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
406 must specify per_mail/readonly. If you omit the update mode it defaults to
407 /leaky where that makes sense (as before) or /readonly where required.
408
409 The /noupdate option is now undocumented but still supported for
410 backwards compatibility. It is equivalent to /readonly except that in
411 ACLs where /readonly is required you may specify /leaky/noupdate or
412 /strict/noupdate which are treated the same as /readonly.
413
414 A useful new feature is the /count= option. This is a generalization
415 of the per_byte option, so that you can measure the throughput of other
416 aggregate values. For example, the per_byte option is now equivalent
417 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
418
419 The per_rcpt option has been generalized using the /count= mechanism
420 (though it's more complicated than the per_byte equivalence). When it is
421 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
422 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
423 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
424 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
425 Note that using per_rcpt with a non-readonly update mode in more than
426 one ACL will cause the recipients to be double-counted. (The per_mail
427 and per_byte options don't have this problem.)
428
429 The handling of very low rates has changed slightly. If the computed rate
430 is less than the event's count (usually one) then this event is the first
431 after a long gap. In this case the rate is set to the same as this event's
432 count, so that the first message of a spam run is counted properly.
433
434 The major new feature is a mechanism for counting the rate of unique
435 events. The new per_addr option counts the number of different
436 recipients that someone has sent messages to in the last time period. It
437 behaves like per_rcpt if all the recipient addresses are different, but
438 duplicate recipient addresses do not increase the measured rate. Like
439 the /count= option this is a general mechanism, so the per_addr option
440 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
441 example, measure the rate that a client uses different sender addresses
442 with the options per_mail/unique=$sender_address. There are further
443 details in the main documentation.
444
3634fc25
TF
445TF/05 Removed obsolete $Cambridge$ CVS revision strings.
446
792e8a19
TF
447TF/06 Removed a few PCRE remnants.
448
5901f0ab
TF
449TF/07 Automatically extract Exim's version number from tags in the git
450 repository when doing development or release builds.
451
7f2a2a43
PP
452PP/02 Raise smtp_cmd_buffer_size to 16kB.
453 Bugzilla 879. Patch from Paul Fisher.
e2ca7082 454
061b7ebd
PP
455PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
456 Heavily based on revision 40f9a89a from Simon Arlott's tree.
457 Bugzilla 97.
458
e12f8c32
PP
459PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
460
9e949f00 461PP/05 Variable $av_failed, true if the AV scanner deferred.
7f2a2a43
PP
462 Bugzilla 1078. Patch from John Horne.
463
464PP/06 Stop make process more reliably on build failure.
465 Bugzilla 1087. Patch from Heiko Schlittermann.
9e949f00 466
555ae6af 467PP/07 Make maildir_use_size_file an _expandable_ boolean.
ac53fcda
PP
468 Bugzilla 1089. Patch from Heiko Schlittermann.
469
470PP/08 Handle ${run} returning more data than OS pipe buffer size.
471 Bugzilla 1131. Patch from Holger Weiß.
555ae6af 472
6f7fe114
PP
473PP/09 Handle IPv6 addresses with SPF.
474 Bugzilla 860. Patch from Wolfgang Breyha.
475
c566dd90
PP
476PP/10 GnuTLS: support TLS 1.2 & 1.1.
477 Bugzilla 1156.
89f897c3
PP
478 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
479 Bugzilla 1095.
c566dd90 480
d6cc7c78 481PP/11 match_* no longer expand right-hand-side by default.
39257585
PP
482 New compile-time build option, EXPAND_LISTMATCH_RHS.
483 New expansion conditions, "inlist", "inlisti".
484
0d0e4455
PP
485PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
486
3399bb60 487PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
d690cbdc
PP
488
489PP/14 fix log_write() format string regression from TF/03.
490 Bugzilla 1152. Patch from Dmitry Isaikin.
491
0ca0cf52 492
10906672
PP
493Exim version 4.76
494-----------------
495
496PP/01 The new ldap_require_cert option would segfault if used. Fixed.
497
754a0503
PP
498PP/02 Harmonised TLS library version reporting; only show if debugging.
499 Layout now matches that introduced for other libraries in 4.74 PP/03.
500
c0c7b2da
PP
501PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
502
e97d1f08
PP
503PP/04 New "dns_use_edns0" global option.
504
084c1d8c
PP
505PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
506 Bugzilla 1098.
507
4e7ee012
PP
508PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
509 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
da80c2a8 510
c8d52a00
PP
511TK/01 Updated PolarSSL code to 0.14.2.
512 Bugzilla 1097. Patch from Andreas Metzler.
513
54e7ce4a
PP
514PP/07 Catch divide-by-zero in ${eval:...}.
515 Fixes bugzilla 1102.
516
5ee6f336
PP
517PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
518 Bugzilla 1104.
519
c8d52a00 520TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
6ea4a851
PP
521 format-string attack -- SECURITY: remote arbitrary code execution.
522
523TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
524 time unintentionally subject to list matching rules, letting the header
525 cause arbitrary Exim lookups (of items which can occur in lists, *not*
526 arbitrary string expansion). This allowed for information disclosure.
527
528PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
529 INT_MIN/-1 -- value coerced to INT_MAX.
c8d52a00 530
10906672 531
aa097c4c
NM
532Exim version 4.75
533-----------------
534
4c9ef03a 535NM/01 Workround for PCRE version dependency in version reporting
aa097c4c
NM
536 Bugzilla 1073
537
7f3d9eff
TF
538TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
539 This fixes portability to compilers other than gcc, notably
540 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
541
159f52d2
TF
542TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
543 makefiles for portability to HP-UX and POSIX correctness.
544
0cc9542a
PP
545PP/01 Permit LOOKUP_foo enabling on the make command-line.
546 Also via indented variable definition in the Makefile.
547 (Debugging by Oliver Heesakkers).
548
f7274286
PP
549PP/02 Restore caching of spamd results with expanded spamd_address.
550 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
551
7b797365
PP
552PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
553 Improves build reliability. Fix from: Frank Elsner
554
caacae52
NM
555NM/02 Fix wide character breakage in the rfc2047 coding
556 Fixes bug 1064. Patch from Andrey N. Oktyabrski
557
09dcaba9
NM
558NM/03 Allow underscore in dnslist lookups
559 Fixes bug 1026. Patch from Graeme Fowler
560
bc19a55b
PP
561PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
562 Code patches from Adam Ciarcinski of NetBSD.
caacae52 563
bd4c9759
NM
564NM/04 Fixed exiqgrep to cope with mailq missing size issue
565 Fixes bug 943.
566
b72aab72
PP
567PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
568 is logged, to avoid truncation. Patch from John Horne.
569
2fe76745
PP
570PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
571 Patch from Jakob Hirsch.
572
76aa570c
PP
573PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
574 SQL string expansion failure details.
575 Patch from Andrey Oktyabrski.
576
f1e5fef5
PP
577PP/08 Bugzilla 486: implement %M datestamping in log filenames.
578 Patch from Simon Arlott.
579
4d805ee9
PP
580PP/09 New lookups functionality failed to compile on old gcc which rejects
581 extern declarations in function scope.
582 Patch from Oliver Fleischmann
583
cd59ab18
PP
584PP/10 Use sig_atomic_t for flags set from signal handlers.
585 Check getgroups() return and improve debugging.
586 Fixed developed for diagnosis in bug 927 (which turned out to be
587 a kernel bug).
588
332f5cf3
PP
589PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
590 Patch from Mark Zealey.
591
29cfeb94
PP
592PP/12 Bugzilla 1056: Improved spamd server selection.
593 Patch from Mark Zealey.
594
660242ad
PP
595PP/13 Bugzilla 1086: Deal with maildir quota file races.
596 Based on patch from Heiko Schlittermann.
597
bc4bc4c5
PP
598PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
599 Patch from Uwe Doering, sign-off by Michael Haardt.
600
2e64baa9
NM
601NM/05 Fix to spam.c to accommodate older gcc versions which dislike
602 variable declaration deep within a block. Bug and patch from
603 Dennis Davis.
604
bddd7526
PP
605PP/15 lookups-Makefile IRIX compatibilty coercion.
606
6bac1a9a
PP
607PP/16 Make DISABLE_DKIM build knob functional.
608
552193f0
NM
609NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
610 Patch by Simon Arlott
baeee2c1 611
1b587e48
TF
612TF/03 Fix valgrind.h portability to C89 compilers that do not support
613 variable argument macros. Our copy now differs from upstream.
614
aa097c4c 615
8c07b69f
TF
616Exim version 4.74
617-----------------
618
619TF/01 Failure to get a lock on a hints database can have serious
620 consequences so log it to the panic log.
621
c0ea85ab
TF
622TF/02 Log LMTP confirmation messages in the same way as SMTP,
623 controlled using the smtp_confirmation log selector.
624
0761d44e
TF
625TF/03 Include the error message when we fail to unlink a spool file.
626
0a349494
PP
627DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
628 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
629 for maintaining out-of-tree patches for some time.
630
631PP/01 Bugzilla 139: Documentation and portability issues.
632 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
633 Handle per-OS dynamic-module compilation flags.
634
fea24b2e
PP
635PP/02 Let /dev/null have normal permissions.
636 The 4.73 fixes were a little too stringent and complained about the
637 permissions on /dev/null. Exempt it from some checks.
638 Reported by Andreas M. Kirchwitz.
639
6545de78
PP
640PP/03 Report version information for many libraries, including
641 Exim version information for dynamically loaded libraries. Created
642 version.h, now support a version extension string for distributors
643 who patch heavily. Dynamic module ABI change.
644
1670ef10
PP
645PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
646 privilege escalation vulnerability whereby the Exim run-time user
647 can cause root to append content of the attacker's choosing to
648 arbitrary files.
649
c0886197
PP
650PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
651 (Wolfgang Breyha)
652
b7487bce
PP
653PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
654 If dropping privileges for untrusted macros, we disabled normal logging
655 on the basis that it would fail; for the Exim run-time user, this is not
656 the case, and it resulted in successful deliveries going unlogged.
657 Fixed. Reported by Andreas Metzler.
658
8c07b69f 659
97fd1e48 660Exim version 4.73
ed7f7860 661-----------------
97fd1e48
PP
662
663PP/01 Date: & Message-Id: revert to normally being appended to a message,
664 only prepend for the Resent-* case. Fixes regression introduced in
665 Exim 4.70 by NM/22 for Bugzilla 607.
666
6901c596
PP
667PP/02 Include check_rfc2047_length in configure.default because we're seeing
668 increasing numbers of administrators be bitten by this.
669
a8c8d6b5
JJ
670JJ/01 Added DISABLE_DKIM and comment to src/EDITME
671
77bb000f
PP
672PP/03 Bugzilla 994: added openssl_options main configuration option.
673
a29e5231
PP
674PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
675
ec5a0394 676PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
a29e5231 677
55c75993
PP
678PP/06 Adjust NTLM authentication to handle SASL Initial Response.
679
453a6645 680PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
ec5a0394
PP
681 without a peer certificate, leading to a segfault because of an
682 assumption that peers always have certificates. Be a little more
453a6645
PP
683 paranoid. Problem reported by Martin Tscholak.
684
8544e77a
PP
685PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
686 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
687 NB: ClamAV planning to remove STREAM in "middle of 2010".
3346ab01
PP
688 CL also introduces -bmalware, various -d+acl logging additions and
689 more caution in buffer sizes.
8544e77a 690
83e029d5
PP
691PP/09 Implemented reverse_ip expansion operator.
692
ed7f7860
PP
693PP/10 Bugzilla 937: provide a "debug" ACL control.
694
7d9f747b
PP
695PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
696
4b2241d2
PP
697PP/12 Bugzilla 973: Implement --version.
698
10385c15
PP
699PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
700
dbc4b90d
PP
701PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
702
532be449
PP
703PP/15 Bugzilla 816: support multiple condition rules on Routers.
704
6a8de854 705PP/16 Add bool_lax{} expansion operator and use that for combining multiple
71265ae9
PP
706 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
707 ignore trailing whitespace.
6a8de854 708
5dc43717
JJ
709JJ/02 prevent non-panic DKIM error from being sent to paniclog
710
711JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
712 "exim" to be used
55c75993 713
3346ab01
PP
714PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
715 Notification from Dr Andrew Aitchison.
716
491fab4c
PP
717PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
718 ExtendedDetectionInfo response format.
719 Notification from John Horne.
720
13eb9497
PP
721PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
722 compatible.
723
724PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
725 XSL and documented dependency on system catalogs, with examples of how
726 it normally works.
727
7f36d675
DW
728DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
729 access.
730
c1d94452
DW
731DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
732 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
733 configuration file which is writeable by the Exim user or group.
734
e2f5dc15
DW
735DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
736 of configuration files to cover files specified with the -C option if
737 they are going to be used with root privileges, not just the default
738 configuration file.
739
cd25e41d
DW
740DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
741 option (effectively making it always true).
742
261dc43e
DW
743DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
744 files to be used while preserving root privileges.
745
fa32850b
DW
746DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
747 that rogue child processes cannot use them.
748
79d4bc3d
PP
749PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
750 run-time user, instead of root.
751
43236f35 752PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2cfd3221
PP
753 Exim run-time user without dropping privileges.
754
fb08281f
DW
755DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
756 result string, instead of calling string_vformat() twice with the same
757 arguments.
3346ab01 758
74935b98
DW
759DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
760 for other users. Others should always drop root privileges if they use
761 -C on the command line, even for a whitelisted configure file.
762
90b6341f
DW
763DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
764
57730b52
ML
765NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
766
66581d1e 767
465e92cf
JJ
768Exim version 4.72
769-----------------
770
453a6645
PP
771JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
772 $data_path, and $header_path variables; fixed documentation bugs and
773 typos
465e92cf 774
453a6645
PP
775JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
776 exipick to access non-standard spools, including the "frozen" queue
777 (Finput)
edae0343 778
9bd3e22c
NM
779NM/01 Bugzilla 965: Support mysql stored procedures.
780 Patch from Alain Williams
781
bb576ff7
NM
782NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
783
5a1a5845
NM
784NM/03 Bugzilla 955: Documentation fix for max_rcpts.
785 Patch from Andreas Metzler
786
981a9fad
NM
787NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
788 Patch from Kirill Miazine
789
7fc497ee
NM
790NM/05 Bugzilla 671: Added umask to procmail example.
791
1a41defa
JJ
792JJ/03 installed exipick 20100323.0, fixing doc bug
793
a466095c 794NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
b26eacf1 795 directory. Notification and patch from Dan Rosenberg.
a466095c 796
94a6bd0b
NM
797TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
798
799TK/02 Improve log output when DKIM signing operation fails.
800
801MH/01 Treat the transport option dkim_domain as a colon separated
802 list, not as a single string, and sign the message with each element,
803 omitting multiple occurences of the same signer.
804
c1b141a8
NM
805NM/07 Null terminate DKIM strings, Null initialise DKIM variable
806 Bugzilla 985, 986. Patch by Simon Arlott
94a6bd0b 807
b26eacf1 808NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
0d0c6357
NM
809 Patch by Simon Arlott
810
179c5980 811PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
b26eacf1 812 MBX locking. Notification from Dan Rosenberg.
179c5980 813
9bd3e22c 814
7c6d71af
NM
815Exim version 4.71
816-----------------
817
7d9f747b 818TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
7c6d71af 819
f013fb92
NM
820NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
821
0eb8eedd
NM
822NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
823
663ee6d9
NM
824NM/03 Bugzilla 847: Enable DNSDB lookup by default.
825
177ebd9b
NM
826NM/04 Bugzilla 915: Flag broken perl installation during build.
827
7c6d71af 828
210f147e
NM
829Exim version 4.70
830-----------------
831
cdd3bb85 832TK/01 Added patch by Johannes Berg that expands the main option
e739e3d9 833 "spamd_address" if it starts with a dollar sign.
cdd3bb85
TK
834
835TK/02 Write list of recipients to X-Envelope-Sender header when building
836 the mbox-format spool file for content scanning (suggested by Jakob
7d9f747b 837 Hirsch).
cdd3bb85
TK
838
839TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
840 (http://www.dcc-servers.net/) support via dccifd. Activated by
e739e3d9 841 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
cdd3bb85
TK
842
843TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
844 by Mark Daniel Reidel <mr@df.eu>.
845
210f147e
NM
846NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
847 When building exim an external PCRE library is now needed -
848 PCRE is a system library on the majority of modern systems.
849 See entry on PCRE_LIBS in EDITME file.
850
deafd5b3
NM
851NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
852 conversation. Added nologin parameter to request.
7d9f747b 853 Patch contributed by Kirill Miazine.
deafd5b3 854
089793a4
TF
855TF/01 Do not log submission mode rewrites if they do not change the address.
856
5f16ca82
TF
857TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
858
dae9d94e 859NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
7d9f747b 860 log files in place. Contributed by Roberto Lima.
dae9d94e 861
7d9f747b 862NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
3f0da4d0 863
06864c44
TF
864TF/03 Bugzilla 615: When checking the local_parts router precondition
865 after a local_part_suffix or local_part_prefix option, Exim now
866 does not use the address's named list lookup cache, since this
867 contains cached lookups for the whole local part.
868
65a7d8c3 869NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
7d9f747b 870 Robert Millan. Documentation is in experimental-spec.txt.
65a7d8c3 871
23510047 872TF/04 Bugzilla 668: Fix parallel build (make -j).
65a7d8c3 873
7d9f747b 874NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
5f28a6e8 875
7d8eec3a 876NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
7d9f747b 877 Patch provided by Jan Srzednicki.
6c588e74 878
89dec7b6
TF
879TF/05 Leading white space used to be stripped from $spam_report which
880 wrecked the formatting. Now it is preserved.
5f28a6e8 881
a99de90c
TF
882TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
883 that they are available at delivery time.
884
e2803e40
TF
885TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
886
7199e1ee
TF
887TF/08 TLS error reporting now respects the incoming_interface and
888 incoming_port log selectors.
889
e276e04b
TF
890TF/09 Produce a more useful error message if an SMTP transport's hosts
891 setting expands to an empty string.
892
ce552449 893NM/06 Bugzilla 744: EXPN did not work under TLS.
7d9f747b 894 Patch provided by Phil Pennock.
ce552449 895
e765a0f1 896NM/07 Bugzilla 769: Extraneous comma in usage fprintf
7d9f747b 897 Patch provided by Richard Godbee.
e765a0f1 898
4f054c63 899NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
447de4b0 900 acl_smtp_notquit, added index entry.
4f054c63 901
7d9f747b
PP
902NM/09 Bugzilla 787: Potential buffer overflow in string_format.
903 Patch provided by Eugene Bujak.
24c929a2 904
7d9f747b
PP
905NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
906 accept(). Patch provided by Maxim Dounin.
cf73943b 907
b52bc06e 908NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
7d9f747b 909 Patch provided by Phil Pennock.
b52bc06e 910
447de4b0
NM
911NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
912
4c69d561 913NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
7d9f747b 914 Patch provided by Brad "anomie" Jorsch.
4c69d561 915
d5c39246 916NM/14 Bugzilla 622: Added timeout setting to transport filter.
7d9f747b 917 Patch provided by Dean Brooks.
9b989985 918
0b23848a
TK
919TK/05 Add native DKIM support (does not depend on external libraries).
920
8f3414a1 921NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
7d9f747b 922 Patch provided by Graeme Fowler.
e2aacdfd 923
fb6f955d
NM
924NM/16 Bugzilla 851: Documentation example syntax fix.
925
926NM/17 Changed NOTICE file to remove references to embedded PCRE.
8f3414a1 927
7d9f747b
PP
928NM/18 Bugzilla 894: Fix issue with very long lines including comments in
929 lsearch.
dbb0bf41 930
7d9f747b
PP
931NM/19 Bugzilla 745: TLS version reporting.
932 Patch provided by Phil Pennock.
f3766eb5 933
7d9f747b
PP
934NM/20 Bugzilla 167: bool: condition support.
935 Patch provided by Phil Pennock.
36f12725 936
7d9f747b
PP
937NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
938 clients. Patch provided by Phil Pennock.
e6060e2c 939
7d9f747b
PP
940NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
941 Patch provided by Brad "anomie" Jorsch.
5eb690a1 942
7d9f747b
PP
943NM/23 Bugzilla 687: Fix misparses in eximstats.
944 Patch provided by Heiko Schlittermann.
d5c13d66 945
7d9f747b
PP
946NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
947 Patch provided by Heiko Schlittermann.
b2335c0b 948
7d9f747b 949NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1da77999 950 plus update to original patch.
f4cd9433 951
7d9f747b 952NM/26 Bugzilla 799: Documentation correction for ratelimit.
dc988b7e 953
7d9f747b
PP
954NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
955 Patch provided by David Brownlee.
8dc71ab3 956
7d9f747b 957NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
400eda43 958
7d9f747b 959NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
ec5a421b 960
7d9f747b 961NM/30 Bugzilla 888: TLS documentation bugfixes.
07af267e 962
7d9f747b 963NM/31 Bugzilla 896: Dovecot buffer overrun fix.
51473862 964
17792b53 965NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
7d9f747b 966 Unlike the original bugzilla I have changed all shell scripts in src tree.
17792b53 967
7d9f747b
PP
968NM/33 Bugzilla 898: Transport filter timeout fix.
969 Patch by Todd Rinaldo.
52383f8f 970
91576cec 971NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
7d9f747b 972 Patch by Serge Demonchaux.
5ca6d115 973
7d9f747b
PP
974NM/35 Bugzilla 39: Base64 decode bug fixes.
975 Patch by Jakob Hirsch.
baee9eee 976
7d9f747b 977NM/36 Bugzilla 909: Correct connect() call in dcc code.
e93a964c 978
7d9f747b 979NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
9bf3d68f 980
7d9f747b 981NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
96535b98 982
7d9f747b 983NM/39 Bugzilla 911: Fixed MakeLinks build script.
30339e0f 984
deafd5b3 985
47db1125
NM
986Exim version 4.69
987-----------------
988
4b3504d0
TK
989TK/01 Add preliminary DKIM support. Currently requires a forked version of
990 ALT-N's libdkim that I have put here:
991 http://duncanthrax.net/exim-experimental/
992
993 Note to Michael Haardt: I had to rename some vars in sieve.c. They
994 were called 'true' and it seems that C99 defines that as a reserved
995 keyword to be used with 'bool' variable types. That means you could
996 not include C99-style headers which use bools without triggering
997 build errors in sieve.c.
998
81ea09ca
NM
999NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1000 as mailq or other aliases. Changed the --help handling significantly
1001 to do whats expected. exim_usage() emits usage/help information.
1002
f13cddcb
SC
1003SC/01 Added the -bylocaldomain option to eximstats.
1004
7d9f747b 1005NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
8ad076b2 1006
7d9f747b 1007NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
a843aaa6 1008
7d9f747b 1009NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
47db1125
NM
1010
1011
eb4c0de6
PH
1012Exim version 4.68
1013-----------------
1014
1015PH/01 Another patch from the Sieve maintainer.
1016
6a3bceb1
PH
1017PH/02 When an IPv6 address is converted to a string for single-key lookup
1018 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1019 dots are used instead of colons so that keys in lsearch files need not
1020 contain colons. This was done some time before quoting was made available
1021 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1022 (notated using the quote facility) so as to distinguish them from IPv4
1023 keys. This meant that lookups for IP addresses in host lists did not work
1024 for iplsearch lookups.
1025
1026 This has been fixed by arranging for IPv6 addresses to be expressed with
1027 colons if the lookup type is iplsearch. This is not incompatible, because
1028 previously such lookups could never work.
1029
1030 The situation is now rather anomolous, since one *can* have colons in
1031 ordinary lsearch keys. However, making the change in all cases is
1032 incompatible and would probably break a number of configurations.
1033
2e30fa9d
TK
1034TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1035 version.
1036
0806a9c5
MH
1037MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1038 conversion specification without a maximum field width, thereby enabling
1039 a rogue spamd server to cause a buffer overflow. While nobody in their
1040 right mind would setup Exim to query an untrusted spamd server, an
1041 attacker that gains access to a server running spamd could potentially
1042 exploit this vulnerability to run arbitrary code as the Exim user.
1043
ae276964
TK
1044TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1045 $primary_hostname instead of what libspf2 thinks the hosts name is.
1046
0f2cbd1b
MH
1047MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1048 a directory entry by the name of the lookup key. Previously, if a
1049 symlink pointed to a non-existing file or a file in a directory that
1050 Exim lacked permissions to read, a lookup for a key matching that
1051 symlink would fail. Now it is enough that a matching directory entry
1052 exists, symlink or not. (Bugzilla 503.)
1053
2b85bce7
PH
1054PH/03 The body_linecount and body_zerocount variables are now exported in the
1055 local_scan API.
1056
93655c46
PH
1057PH/04 Added the $dnslist_matched variable.
1058
6c512171
PH
1059PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1060 This means they are set thereafter only if the connection becomes
1061 encrypted.
1062
1063PH/06 Added the client_condition to authenticators so that some can be skipped
1064 by clients under certain conditions.
1065
aa6dc513
PH
1066PH/07 The error message for a badly-placed control=no_multiline_responses left
1067 "_responses" off the end of the name.
1068
a96603a0
PH
1069PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1070
8f240103
PH
1071PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1072 (without spaces) instead of just copying the configuration text.
1073
1074PH/10 Added the /noupdate option to the ratelimit ACL condition.
1075
d677b2f2
PH
1076PH/11 Added $max_received_linelength.
1077
d52120f2
PH
1078PH/12 Added +ignore_defer and +include_defer to host lists.
1079
64f2600a
PH
1080PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1081 way in which PCRE > 7.0 is built.
1082
8669f003
PH
1083PH/14 Implemented queue_only_load_latch.
1084
a4dc33a8
PH
1085PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1086 MAIL command. The effect was to mangle the value on 64-bit systems.
1087
d6a60c0f
PH
1088PH/16 Another patch from the Sieve maintainer.
1089
8f128379
PH
1090PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1091
8932dffe
PH
1092PH/18 If a system quota error occurred while trying to create the file for
1093 a maildir delivery, the message "Mailbox is full" was not appended to the
1094 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1095 only to a quota excession during the actual writing of the file.
d6a60c0f 1096
ddea74fa 1097PH/19 It seems that peer DN values may contain newlines (and other non-printing
48ed62d9
PH
1098 characters?) which causes problems in log lines. The DN values are now
1099 passed through string_printing() before being added to log lines.
1100
ddea74fa 1101PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
b7670459
PH
1102 and InterBase are left for another time.)
1103
ddea74fa
PH
1104PH/21 Added message_body_newlines option.
1105
ce9f225c
PH
1106PH/22 Guard against possible overflow in moan_check_errorcopy().
1107
19897d52
PH
1108PH/23 POSIX allows open() to be a macro; guard against that.
1109
bc64a74d
PH
1110PH/24 If the recipient of an error message contained an @ in the local part
1111 (suitably quoted, of course), incorrect values were put in $domain and
1112 $local_part during the evaluation of errors_copy.
1113
eb4c0de6 1114
b4ed4da0
PH
1115Exim version 4.67
1116-----------------
1117
22ad45c9
MH
1118MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1119 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1120 Jan Srzednicki.
1121
b4ed4da0
PH
1122PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1123 issue a MAIL command.
1124
431b7361
PH
1125PH/02 In an ACL statement such as
1126
1127 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1128
1129 if a client was not listed at all, or was listed with a value other than
1130 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1131 the condition was not true (as it should be), so access was not denied.
1132 The bug was that the ! inversion was incorrectly passed on to the second
1133 item. This has been fixed.
1134
1135PH/03 Added additional dnslists conditions == and =& which are different from
1136 = and & when the dns lookup returns more than one IP address.
1137
83da1223
PH
1138PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1139 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1140
54fc8428
PH
1141PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1142 FSYNC, which compiles an option called disable_fsync that allows for
1143 bypassing fsync(). The documentation is heavily laced with warnings.
1144
34c5e8dd
SC
1145SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1146
bbe15da8
PH
1147PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1148 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1149 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1150 including adding "make clean"; (3) Added -fPIC when compiling the test
1151 dynamically loaded module, to get rid of a warning.
1152
0e8a9471
MH
1153MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1154 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1155 = 0s. The bug is otherwise harmless.
1156
f0872424
PH
1157PH/07 There was a bug in the dovecot authenticator such that the value of
1158 $auth1 could be overwritten, and so not correctly preserved, after a
1159 successful authentication. This usually meant that the value preserved by
1160 the server_setid option was incorrect.
1161
b01dd148
PH
1162PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1163
6bf342e1
PH
1164PH/09 Installed PCRE release 7.0.
1165
273f34d0
PH
1166PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1167 run for batched SMTP input. It is now run at the start of every message
1168 in the batch. While fixing this I discovered that the process information
1169 (output by running exiwhat) was not always getting set for -bs and -bS
1170 input. This is fixed, and it now also says "batched" for BSMTP.
1171
cf8b11a5
PH
1172PH/11 Added control=no_pipelining.
1173
41c7c167
PH
1174PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1175 patch, slightly modified), and move the expansion of helo_data till after
1176 the connection is made in the smtp transport (so it can use these
1177 values).
1178
9c57cbc0
PH
1179PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1180
f3f065bb
PH
1181PH/14 Added log_selector = +pid.
1182
047bdd8c
PH
1183PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1184
0ce9abe6
PH
1185PH/16 Add ${if forany and ${if forall.
1186
0e22dfd1
PH
1187PH/17 Added dsn_from option to vary the From: line in DSNs.
1188
4c590bd1
PH
1189PH/18 Flush SMTP output before performing a callout, unless control =
1190 no_callout_flush is set.
1191
09945f1e
PH
1192PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1193 was true (the default) a successful delivery failed to delete the retry
1194 item, thus causing premature timeout of the address. The bug is now
1195 fixed.
1196
c51b8e75
PH
1197PH/20 Added hosts_avoid_pipelining to the smtp transport.
1198
e28326d8
PH
1199PH/21 Long custom messages for fakedefer and fakereject are now split up
1200 into multiline reponses in the same way that messages for "deny" and
1201 other ACL rejections are.
1202
75b1493f
PH
1203PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1204 with slight modification.
1205
7c5214ec
PH
1206PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1207 draft, changing the syntax and factoring some duplicate code".
1208
4311097e
PH
1209PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1210 for deliveries of the second and subsequent messages over the same SMTP
1211 connection.
1212
29f89cad
PH
1213PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1214 ${reduce, with only minor "tidies".
1215
5e687460
SC
1216SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1217
c3611384
PH
1218PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1219 expansion side effects.
1220
5a11a7b4
PH
1221PH/27 When a message times out after an over-quota error from an Exim-imposed
1222 quota, the bounce message says "mailbox is full". This message was not
1223 being given when it was a system quota that was exceeded. It now should
1224 be the same.
1225
0e20aff9
MH
1226MH/03 Made $recipients available in local_scan(). local_scan() already has
1227 better access to the recipient list through recipients_list[], but
1228 $recipients can be useful in postmaster-provided expansion strings.
1229
ca86f471
PH
1230PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1231 in the case of a MAIL command with additional options following the
1232 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1233 were accidentally chopped off.
1234
a14e5636
PH
1235PH/29 SMTP synchronization checks are implemented when a command is read -
1236 there is a check that no more input is waiting when there shouldn't be
1237 any. However, for some commands, a delay in an ACL can mean that it is
1238 some time before the response is written. In this time, more input might
1239 arrive, invalidly. So now there are extra checks after an ACL has run for
1240 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1241 pipelining has not been advertised.
1242
ec95d1a6
PH
1243PH/30 MH's patch to allow iscntrl() characters to be list separators.
1244
42855d71
PH
1245PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1246 returned in the SMTP response when smtp_return_error_details was false.
1247 This has been fixed.
1248
57c2c631
PH
1249PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1250 instead of the C I/O that was originally supplied, because problems were
1251 reported on Solaris.
1252
58c01c94
PH
1253PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1254 Exim which did not show up earlier: it was assuming that a call to
1255 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1256 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1257 was a macro that became an assignment, so it seemed to work. This has
1258 changed to a proper function call with a void return, hence the compile
1259 error. Exim's code has been fixed.
1260
dee5a20a
PH
1261PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1262 cpus.
1263
d2ee6114
PH
1264PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1265
b2d5182b
PH
1266PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1267
79749a79
PH
1268PH/37 If a message is not accepted after it has had an id assigned (e.g.
1269 because it turns out to be too big or there is a timeout) there is no
3ce62588
PH
1270 "Completed" line in the log. When some messages of this type were
1271 selected by exigrep, they were listed as "not completed". Others were
1272 picked up by some special patterns. I have improved the selection
1273 criteria to be more general.
79749a79 1274
c456d9bb
PH
1275PH/38 The host_find_failed option in the manualroute router can now be set
1276 to "ignore", to completely ignore a host whose IP address cannot be
1277 found. If all hosts are ignored, the behaviour is controlled by the new
1278 host_all_ignored option.
1279
cd9868ec
PH
1280PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1281 homing or because of multiple MX records with /mx) generated more than
1282 one IP address, and the following item turned out to be the local host,
1283 all the secondary addresses of the first item were incorrectly removed
1284 from the list, along with the local host and any following hosts (which
1285 is what is supposed to happen).
1286
ebeaf996
PH
1287PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1288 whoever called Exim into the -H file. In the case of the daemon it was
1289 behaving confusingly. When first started, it used values for whoever
1290 started the daemon, but after a SIGHUP it used the Exim user (because it
1291 calls itself on a restart). I have changed the code so that it now always
1292 uses the Exim user.
1293
2679d413
PH
1294PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1295 message are rejected with the same error (e.g. no authentication or bad
1296 sender address), and a DATA command is nevertheless sent (as can happen
1297 with PIPELINING or a stupid MUA), the error message that was given to the
1298 RCPT commands is included in the rejection of the DATA command. This is
1299 intended to be helpful for MUAs that show only the final error to their
1300 users.
1301
84024b72
PH
1302PH/42 Another patch from the Sieve maintainer.
1303
8005d38e
SC
1304SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1305 Eximstats - Fixed some broken HTML links and added missing column headers
1306 (Jez Hancock).
1307 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1308 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1309
3298c6c6
SC
1310SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1311
a43a27c5
PH
1312PH/43 Yet another patch from the Sieve maintainer.
1313
58eb016e 1314PH/44 I found a way to check for a TCP/IP connection going away before sending
563b63fa
PH
1315 the response to the final '.' that terminates a message, but only in the
1316 case where the client has not sent further data following the '.'
1317 (unfortunately, this is allowed). However, in many cases there won't be
1318 any further data because there won't be any more messages to send. A call
1319 to select() can be used: if it shows that the input is "ready", there is
1320 either input waiting, or the socket has been closed. An attempt to read
1321 the next input character can distinguish the two cases. Previously, Exim
58eb016e 1322 would have sent an OK response which the client would never have see.
563b63fa
PH
1323 This could lead to message repetition. This fix should cure that, at
1324 least in a lot of common cases.
58eb016e 1325
b43a74ea
PH
1326PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1327 advertised in response to EHLO.
1328
b4ed4da0 1329
5dd1517f
PH
1330Exim version 4.66
1331-----------------
1332
1333PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1334 fixed by 4.65/MH/01 (is this a record?) are fixed:
1335
1336 (i) An empty string was always treated as zero by the numeric comparison
1337 operators. This behaviour has been restored.
1338
1339 (ii) It is documented that the numeric comparison operators always treat
1340 their arguments as decimal numbers. This was broken in that numbers
1341 starting with 0 were being interpreted as octal.
1342
1343 While fixing these problems I realized that there was another issue that
1344 hadn't been noticed. Values of message_size_limit (both the global option
1345 and the transport option) were treated as octal if they started with 0.
1346 The documentation was vague. These values are now always treated as
1347 decimal, and I will make that clear in the documentation.
1348
1349
93cfa765
TK
1350Exim version 4.65
1351-----------------
1352
1353TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1354 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1355 versions. (#438)
1356
d6066548
MH
1357MH/01 Don't check that the operands of numeric comparison operators are
1358 integers when their expansion is in "skipping" mode (fixes bug
1359 introduced by 4.64-PH/07).
1360
4362ff0d
PH
1361PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1362 child addresses, Exim now panics and dies. Previously, because the count
1363 is held in a short int, deliveries were likely to be lost. As such a
1364 large number of recipients for a single message is ridiculous
1365 (performance will be very, very poor), I have chosen to impose a limit
1366 rather than extend the field.
1367
93cfa765 1368
944e9e9c
TF
1369Exim version 4.64
1370-----------------
aa41d2de 1371
21d74bd9
TK
1372TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1373 leftover -K file (the existence of which was triggered by #402).
1374 While we were at it, introduced process PID as part of the -K
1375 filename. This should rule out race conditions when creating
1376 these files.
1377
1378TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1379 processing considerably. Previous code took too long for large mails,
1380 triggering a timeout which in turn triggers #401.
1381
1382TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1383 in the DK code in transports.c. sendfile() is not really portable,
1384 hence the _LINUX specificness.
944e9e9c
TF
1385
1386TF/01 In the add_headers option to the mail command in an Exim filter,
1387 there was a bug that Exim would claim a syntax error in any
1388 header after the first one which had an odd number of characters
1389 in the field name.
1390
2b1c6e3a
PH
1391PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1392 callout verification, Exim cached a "reject" for the entire domain. This
1393 is correct for most verifications, but it is not correct for a recipient
1394 verification with use_sender or use_postmaster set, because in that case
1395 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1396 case of MAIL FROM:<> rejection from other early rejections (e.g.
1397 rejection of HELO). When verifying a recipient using a non-null MAIL
1398 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1399 Whatever the result of the callout, the value of the domain cache is
1400 left unchanged (for any other kind of callout, getting as far as trying
1401 RCPT means that the domain itself is ok).
1402
1f872c80
PH
1403PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1404 gcc 4.1.1 threw up.
1405
1406PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1407 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1408 session, the daemon ignores ECONNECT errors and logs others; it now
1409 ignores EPIPE as well.
1410
d203e649
PH
1411PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1412 (quoted-printable decoding).
1413
cc2ed8f7 1414PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
21a04aa3 1415 later the small subsequent patch to fix an introduced bug.
f951fd57 1416
ddfcd446
PH
1417PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1418
d45b1de8
PH
1419PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1420
1421PH/08 An error is now given if message_size_limit is specified negative.
1422
38a0a95f 1423PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
641cb756 1424 to be given (somewhat) arbitrary names.
38a0a95f 1425
a2405d83
JJ
1426JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1427 in 4.64-PH/09.
1428
1429JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1430 miscellaneous code fixes
1431
6ea85e9a
PH
1432PH/10 Added the log_reject_target ACL modifier to specify where to log
1433 rejections.
1434
26da7e20
PH
1435PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1436 hostname. This is wrong, because it relates to the incoming message (and
1437 probably the interface on which it is arriving) and not to the outgoing
1438 callout (which could be using a different interface). This has been
1439 changed to use the value of the helo_data option from the smtp transport
1440 instead - this is what is used when a message is actually being sent. If
1441 there is no remote transport (possible with a router that sets up host
1442 addresses), $smtp_active_hostname is used.
6ea85e9a 1443
14aa5a05 1444PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
7befa435 1445 tweaks were necessary in order to get it to work (see also 21 below):
14aa5a05
PH
1446 (a) The code assumed that strncpy() returns a negative number on buffer
1447 overflow, which isn't the case. Replaced with Exim's string_format()
1448 function.
1449 (b) There were several signed/unsigned issues. I just did the minimum
1450 hacking in of casts. There is scope for a larger refactoring.
1451 (c) The code used strcasecmp() which is not a standard C function.
1452 Replaced with Exim's strcmpic() function.
1453 (d) The code set only $1; it now sets $auth1 as well.
1454 (e) A simple test gave the error "authentication client didn't specify
1455 service in request". It would seem that Dovecot has changed its
1456 interface. Fortunately there's a specification; I followed it and
1457 changed what the client sends and it appears to be working now.
1458
ff75a1f7
PH
1459PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1460 decoding.
1461
e6f6568e
PH
1462PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1463 address A is aliased to B and C, where B exists and C does not. Without
1464 -v the output is "A verified" because verification stops after a
1465 successful redirection if more than one address is generated. However,
1466 with -v the child addresses are also verified. Exim was outputting "A
1467 failed to verify" and then showing the successful verification for C,
1468 with its parentage. It now outputs "B failed to verify", showing B's
1469 parentage before showing the successful verification of C.
1470
d6f6e0dc
PH
1471PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1472 look up a TXT record in a specific list after matching in a combined
1473 list.
1474
322050c2
PH
1475PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1476 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1477 they consult the DNS. I had assumed they would set it the way they
1478 wanted; and indeed my experiments on Linux seem to show that in some
1479 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1480 To be on the safe side, however, I have now made the interface to
1481 host_find_byname() similar to host_find_bydns(), with an argument
1482 containing the DNS resolver options. The host_find_byname() function now
1483 sets these options at its start, just as host_find_bydns() does. The smtp
1484 transport options dns_qualify_single and dns_search_parents are passed to
1485 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
1486 of host_find_byname() use the default settings of RES_DEFNAMES
1487 (qualify_single) but not RES_DNSRCH (search_parents).
1488
08955dd3
PH
1489PH/17 Applied (a modified version of) Nico Erfurth's patch to make
1490 spool_read_header() do less string testing, by means of a preliminary
1491 switch on the second character of optional "-foo" lines. (This is
1492 overdue, caused by the large number of possibilities that now exist.
1493 Originally there were few.) While I was there, I also converted the
1494 str(n)cmp tests so they don't re-test the leading "-" and the first
1495 character, in the hope this might squeeze out yet more improvement.
1496
1eccaa59
PH
1497PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
1498 flag allowing group syntax was set by the header_syntax check but not
1499 turned off, possible causing trouble later; (2) The flag was not being
1500 set at all for the header_verify test, causing "group"-style headers to
1501 be rejected. I have now set it in this case, and also caused header_
1502 verify to ignore an empty address taken from a group. While doing this, I
1503 came across some other cases where the code for allowing group syntax
1504 while scanning a header line wasn't quite right (mostly, not resetting
1505 the flag correctly in the right place). These bugs could have caused
1506 trouble for malformed header lines. I hope it is now all correct.
1507
602e59e5
PH
1508PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
1509 with the "reply" argument non-NULL. The code, however (which originally
1510 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
1511 but it didn't always do it. This confused somebody who was copying the
1512 code for some other use. I have removed all the tests.
1513
411ef850
PH
1514PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
1515 feature that was used to support insecure browsers during the U.S. crypto
1516 embargo. It requires special client support, and Exim is probably the
1517 only MTA that supported it -- and would never use it because real RSA is
1518 always available. This code has been removed, because it had the bad
1519 effect of slowing Exim down by computing (never used) parameters for the
1520 RSA_EXPORT functionality.
1521
7befa435
PH
1522PH/21 On the advice of Timo Sirainen, added a check to the dovecot
1523 authenticator to fail if there's a tab character in the incoming data
1524 (there should never be unless someone is messing about, as it's supposed
1525 to be base64-encoded). Also added, on Timo's advice, the "secured" option
1526 if the connection is using TLS or if the remote IP is the same as the
1527 local IP, and the "valid-client-cert option" if a client certificate has
1528 been verified.
1529
48da4259 1530PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
16ff981e
PH
1531 authenticators. This can be used for authorization after authentication
1532 succeeds. (In the case of plaintext, it servers for both authentication
1533 and authorization.)
1534
48da4259
PH
1535PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1536 if any retry times were supplied.
1537
d1d5595c
PH
1538PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1539 connection, where there is no client IP address to check. In this
1540 situation, the verify now always succeeds.
1541
0ef732d9
PH
1542PH/25 Applied John Jetmore's -Mset patch.
1543
328895cc
PH
1544PH/26 Added -bem to be like -Mset, but loading a message from a file.
1545
fd700877
PH
1546PH/27 In a string expansion for a processed (not raw) header when multiple
1547 headers of the same name were present, leading whitespace was being
1548 removed from all of them, but trailing whitespace was being removed only
1549 from the last one. Now trailing whitespace is removed from each header
f6c332bd
PH
1550 before concatenation. Completely empty headers in a concatenation (as
1551 before) are ignored.
fd700877 1552
8dce1a6f
PH
1553PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1554 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1555
17af4a17
PH
1556PH/29 [Removed. This was a change that I later backed out, and forgot to
1557 correct the ChangeLog entry (that I had efficiently created) before
1558 committing the later change.]
f6c332bd
PH
1559
1560PH/30 Exim was sometimes attempting to deliver messages that had suffered
1561 address errors (4xx response to RCPT) over the same connection as other
1562 messages routed to the same hosts. Such deliveries are always "forced",
1563 so retry times are not inspected. This resulted in far too many retries
1564 for the affected addresses. The effect occurred only when there were more
1565 hosts than the hosts_max_try setting in the smtp transport when it had
1566 the 4xx errors. Those hosts that it had tried were not added to the list
1567 of hosts for which the message was waiting, so if all were tried, there
1568 was no problem. Two fixes have been applied:
1569
1570 (i) If there are any address or message errors in an SMTP delivery, none
1571 of the hosts (tried or untried) are now added to the list of hosts
1572 for which the message is waiting, so the message should not be a
1573 candidate for sending over the same connection that was used for a
1574 successful delivery of some other message. This seems entirely
1575 reasonable: after all the message is NOT "waiting for some host".
1576 This is so "obvious" that I'm not sure why it wasn't done
1577 previously. Hope I haven't missed anything, but it can't do any
1578 harm, as the worst effect is to miss an optimization.
1579
1580 (ii) If, despite (i), such a delivery is accidentally attempted, the
1581 routing retry time is respected, so at least it doesn't keep
1582 hammering the server.
1583
c1114884
PH
1584PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1585 in ${readsocket because some servers need this prod.
1586
7a0743eb
PH
1587PH/32 Added some extra debug output when updating a wait-xxx database.
1588
0d85fa3f
PH
1589PH/33 The hint "could be header name not terminated by colon", which has been
1590 given for certain expansion errors for a long time, was not being given
1591 for the ${if def:h_colon_omitted{... case.
1592
1bf43b78
PH
1593PH/34 The spec says: "With one important exception, whenever a domain list is
1594 being scanned, $domain contains the subject domain." There was at least
1595 one case where this was not true.
1596
520de300
PH
1597PH/35 The error "getsockname() failed: connection reset by peer" was being
1598 written to the panic log as well as the main log, but it isn't really
1599 panic-worthy as it just means the connection died rather early on. I have
1600 removed the panic log writing for the ECONNRESET error when getsockname()
1601 fails.
1602
48c7f9e2
PH
1603PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1604 runs only) independently of the message's sender address. This meant
1605 that, if the 4xx error was in fact related to the sender, a different
1606 message to the same recipient with a different sender could confuse
1607 things. In particualar, this can happen when sending to a greylisting
1608 server, but other circumstances could also provoke similar problems.
1609 I have changed the default so that the retry time for these errors is now
1610 based a combination of the sender and recipient addresses. This change
1611 can be overridden by setting address_retry_include_sender=false in the
1612 smtp transport.
1613
99ea1c86
PH
1614PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1615 remote server are returned as part of bounce messages. This was not
1616 happening for LMTP over a pipe (the lmtp transport), but now it is the
1617 same for both kinds of LMTP.
1618
a2042e78
PH
1619PH/38 Despite being documented as not happening, Exim was rewriting addresses
1620 in header lines that were in fact CNAMEs. This is no longer the case.
1621
4fbcfc2e
PH
1622PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1623 and queue runs started by the daemon processed all messages. This has
1624 been fixed so that -R and -S can now usefully be given with -q<time>.
1625
aa41d2de
PH
1626PH/40 Import PCRE release 6.7 (fixes some bugs).
1627
af561417
PH
1628PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1629
3cc66b45
PH
1630PH/42 Give an error if -q is specified more than once.
1631
194cc0e4
PH
1632PH/43 Renamed the variables $interface_address and $interface_port as
1633 $received_ip_address and $received_port, to make it clear that these
1634 values apply to message reception, and not to the outgoing interface when
1635 a message is delivered. (The old names remain recognized, of course.)
1636
a401ddaa
PH
1637PH/44 There was no timeout on the connect() call when using a Unix domain
1638 socket in the ${readsocket expansion. There now is.
1639
4e88a19f
PH
1640PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1641 be meaningful with "accept".
1642
d7d7b289
SC
1643SC/01 Eximstats V1.43
1644 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1645
1646SC/02 Eximstats V1.44
1647 Use a glob alias rather than an array ref in the generated
1648 parser. This improves both readability and performance.
1649
1650SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1651 Collect SpamAssassin and rejection statistics.
1652 Don't display local sender or destination tables unless
1653 there is data to show.
1654 Added average volumes into the top table text output.
1655
1656SC/04 Eximstats V1.46
1657 Collect data on the number of addresses (recipients)
1658 as well as the number of messages.
1659
1660SC/05 Eximstats V1.47
1661 Added 'Message too big' to the list of mail rejection
1662 reasons (thanks to Marco Gaiarin).
1663
1664SC/06 Eximstats V1.48
1665 Mainlog lines which have GMT offsets and are too short to
1666 have a flag are now skipped.
1667
1668SC/07 Eximstats V1.49 (Alain Williams)
1669 Added the -emptyok flag.
1670
1671SC/08 Eximstats V1.50
1672 Fixes for obtaining the IP address from reject messages.
1673
0ea2a468
JJ
1674JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1675 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1676 whitesspace changes from 4.64-PH/27
1677
1678JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1679 match 4.64-PH/13
1680
1681JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1682 are found, allow negative numbers in numeric criteria)
1683
1684JJ/06 exipick.20061117.2, added new $message_body_missing variable
1685
1686JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1687 to match changes made in 4.64-PH/43
1688
8a10f5a4
PH
1689PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1690
30e18802
PH
1691PH/47 Put in an explicit test for a DNS lookup of an address record where the
1692 "domain" is actually an IP address, and force a failure. This locks out
1693 those revolvers/nameservers that support "A-for-A" lookups, in
1694 contravention of the specifications.
1695
55728a4f
PH
1696PH/48 When a host name was looked up from an IP address, and the subsequent
1697 forward lookup of the name timed out, the host name was left in
1698 $sender_host_name, contrary to the specification.
d7d7b289 1699
d7837193
PH
1700PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1701 restricted to single-key lookups, Exim was not diagnosing an error if
1702 * or *@ was used with a query-style lookup.
1703
87054a31
PH
1704PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1705
ea2c01d2
MH
1706MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1707 long ago, but noone interested enough thought of it. Let's just say that
1708 the "1.1" means that there are some new functions that weren't there at
1709 some point in the past.
1710
e4fa6968
PH
1711PH/51 Error processing for expansion failure of helo_data from an smtp
1712 transport during callout processing was broken.
1713
56f5d9bd
PH
1714PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1715 tested/used via the -bh/-bhc/-bs options.
1716
922e1c28
PH
1717PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1718 bug, fixed in subsequent PCRE releases).
1719
21eb6e72
PH
1720PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1721 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1722
a0540757
PH
1723PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1724
944e9e9c 1725
478be7b0
SC
1726Exim version 4.63
1727-----------------
1728
1729SC/01 Use a glob alias rather than an array ref in eximstats generated
1730 parser. This improves both readability and performance.
1731
1732SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1733 Don't display local sender or destination tables in eximstats unless
1734 there is data to show.
1735 Added average volumes into the eximstats top table text output.
1736
1737SC/03 Collect data on the number of addresses (recipients) as well
1738 as the number of messages in eximstats.
1739
2b965a65
TF
1740TF/01 Correct an error in the documentation for the redirect router. Exim
1741 does (usually) call initgroups() when daemonizing.
478be7b0 1742
45b91596
PH
1743TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1744 with consistent privilege compared to when running as a daemon.
478be7b0 1745
c59f5781
TF
1746TF/03 Note in the spec that $authenticated_id is not set for local
1747 submissions from trusted users.
1748
90fc3069
TF
1749TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1750 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1751
6083aca0
TF
1752TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1753 by adding some example configuration directives to the default
1754 configuration file. A little bit of work is required to uncomment the
1755 directives and define how usernames and passwords are checked, but
1756 there is now a framework to start from.
1757
765b530f
PH
1758PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1759 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1760 without this. I don't know how relevant this is to other LDAP libraries.
1761
4e167a8c
PH
1762PH/02 Add the verb name to the "unknown ACL verb" error.
1763
4608d683
PH
1764PH/03 Magnus Holmgren's patch for filter_prepend_home.
1765
b8dc3e4a
PH
1766PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1767
5418e93b
PH
1768PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1769 directory not expanded when it should be if an expanded home directory
1770 was set for the address (which is overridden by the transport).
1771
b4a9bda2
PH
1772PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1773 libradius.
1774
45b91596
PH
1775PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1776 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1777 because it is too late at that time, and has no effect.
1778
5547e2c5
PH
1779PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1780 security issue with \' (bugzilla #107). I could not use the
1781 PQescapeStringConn() function, because it needs a PGconn value as one of
1782 its arguments.
1783
dbcef0ea
PH
1784PH/08 When testing addresses using -bt, indicate those final addresses that
1785 are duplicates that would not cause an additional delivery. At least one
1786 person was confused, thinking that -bt output corresponded to deliveries.
1787 (Suppressing duplicates isn't a good idea as you lose the information
1788 about possibly different redirections that led to the duplicates.)
1789
25257489
PH
1790PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1791 systems where poll() doesn't work, in particular OS X.
1792
c816d124
PH
1793PH/10 Added more information to debugging output for retry time not reached.
1794
a9ccd69a
PH
1795PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1796 operations in malware.c.
1797
75fa1910
PH
1798PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1799 signatures.
1800
a7d7aa58
PH
1801PH/13 If write_rejectlog was set false when logging was sent to syslog with
1802 syslog_duplication set false, log lines that would normally be written
1803 both the the main log and to the reject log were not written to syslog at
1804 all.
1805
42119b09
PH
1806PH/14 In the default configuration, change the use of "message" in ACL warn
1807 statements to "add_header".
1808
41609df5
PH
1809PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1810 not followed by a command (e.g. "seen endif").
1811
a5bd321b
PH
1812PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1813 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1814 latter.
1815
e85a7ad5 1816PH/17 Added extra conditions to the default value of delay_warning_condition
5dff5817
PH
1817 so that it is now:
1818
e85a7ad5
PH
1819 ${if or { \
1820 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1821 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1822 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
5dff5817
PH
1823 }{no}{yes}}
1824
e85a7ad5
PH
1825 The Auto-Submitted: and various List- headers are standardised, whereas I
1826 don't think Precedence: ever was.
5dff5817 1827
d8fe1c03
PH
1828PH/18 Refactored debugging code in route_finduser() to show more information,
1829 in particular, the error code if getpwnam() issues one.
1830
16282d2b
PH
1831PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1832 This is apparently needed in addition to the PH/07 change above to avoid
1833 any possible encoding problems.
1834
35d40a98
PH
1835PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1836 but not after initializing Perl.
1837
034d99ab
PH
1838PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1839 output them only if debugging. By default they are written stderr,
1840 apparently, which is not desirable.
1841
6ec97b1b
PH
1842PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1843 queries.
1844
e22ca4ac
JJ
1845JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1846 --not options
1847
1848JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1849
33d73e3b
PH
1850PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1851 authenticated or an ident call has been made. Suppress the default
1852 values for $authenticated_id and $authenticated_sender (but permit -oMai
1853 and -oMas) when testing with -bh.
1854
9ecb03f3
PH
1855PH/24 Re-jigged the order of the tests in the default configuration so that the
1856 tests for valid domains and recipients precede the DNS black list and CSA
1857 tests, on the grounds that those ones are more expensive.
1858
084efe8d
PH
1859PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1860 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1861 This bug exists in every version of Exim that I still have, right back to
1862 0.12.
1863
366fc9f0
PH
1864PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1865 However, an attempt to turn on case-sensitivity in a regex key by
1866 including (?-i) didn't work because the subject string was already
1867 lowercased, and the effects were non-intuitive. It turns out that a
1868 one-line patch can be used to allow (?-i) to work as expected.
1869
c59f5781 1870
c887c79e
TF
1871Exim version 4.62
1872-----------------
1873
1874TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1875 other effects) broke the use of negated acl sub-conditions.
1876
1cce3af8
PH
1877PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1878 patch).
1879
afb3eaaf
PH
1880PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1881 "Deny" causes Exim to reject the incoming connection with a 554 error.
1882 Unfortunately, if there is a major crisis, such as a disk failure,
1883 tcp-wrappers gives "deny", whereas what one would like would be some
1884 kind of temporary error. A kludge has been added to help with this.
1885 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1886 554 error is used if errno is still zero or contains ENOENT (which occurs
1887 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1888 451 error is used.
1889
e173618b
PH
1890PH/03 Add -lutil to the default FreeBSD LIBS setting.
1891
dd16e114
PH
1892PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1893 errors. Otherwise a message that provokes a temporary error (when other
1894 messages do not) can cause a whole host to time out.
1895
f7fd3850
PH
1896PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1897 addresses were routed directly to files or pipes from a redirect router.
1898 File deliveries just didn't batch; pipe deliveries might have suffered
1899 odd errors.
1900
d87df92c
PH
1901PH/06 A failure to get a lock for a hints database would erroneously always say
1902 "Failed to get write lock", even when it was really a read lock.
1903
7e9f683d
PH
1904PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1905 of 0600. This has been changed to use the value of the lockfile_mode
1906 option (which defaults to 0600).
1907
bfad5236
PH
1908PH/08 Applied small patch from the Sieve maintainer.
1909
01c490df
PH
1910PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1911 folder from quota calculations, a direct delivery into this folder messed
1912 up the contents of the maildirsize file. This was because the regex was
1913 used only to exclude .Trash (or whatever) when the size of the mailbox
1914 was calculated. There was no check that a delivery was happening into an
1915 excluded directory. This bug has been fixed by ignoring all quota
1916 processing for deliveries into excluded directories.
1917
d6629cdc
PH
1918PH/10 Added the maildirfolder_create_regex option to appendfile.
1919
1cce3af8 1920
214e2000
PH
1921Exim version 4.61
1922-----------------
1923
1924PH/01 The code for finding all the local interface addresses on a FreeBSD
1925 system running IPv6 was broken. This may well have applied to all BSD
1926 systems, as well as to others that have similar system calls. The broken
1927 code found IPv4 interfaces correctly, but gave incorrect values for the
1928 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1929 that it would not match correctly against @[] and not recognize the IPv6
1930 addresses as local.
1931
f9daeae0
PH
1932PH/02 The ipliteral router was not recognizing addresses of the form user@
1933 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1934
7e66e54d
PH
1935PH/03 Added disable_ipv6.
1936
c8ea1597
PH
1937PH/04 Changed $reply_address to use the raw form of the headers instead of the
1938 decoded form, because it is most often used to construct To: headers
1939 lines in autoreplies, and the decoded form may well be syntactically
1940 invalid. However, $reply_address has leading white space removed, and all
1941 newlines turned into spaces so that the autoreply transport does not
1942 grumble.
1943
911f6fde
PH
1944PH/05 If group was specified without a user on a router, and no group or user
1945 was specified on a transport, the group from the router was ignored.
1946
47ca6d6c
PH
1947PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1948 for visible compile-time settings that can be used to change these
1949 numbers, for those that want even more. Backwards compatibility with old
1950 spool files has been maintained. However, going back to a previous Exim
1951 release will lost any variables that are in spool files.
1952
ed0e9820
PH
1953PH/07 Two small changes when running in the test harness: increase delay when
1954 passing a TCP/IP connection to a new process, in case the original
1955 process has to generate a bounce, and remove special handling of
1956 127.0.0.2 (sic), which is no longer necessary.
1957
eff37e47
PH
1958PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1959 be the same on different OS.
1960
1921d2ea
PH
1961PH/09 Moved a debug statement in filter processing to avoid a race problem when
1962 testing.
1963
b3f69ca8
JJ
1964JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1965 whether --show-vars was specified or not
1966
1967JJ/02 exipick: Added support for new ACL variable spool format introduced
1968 in 4.61-PH/06
1969
424a1c63
PH
1970PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1971 syntactically invalid From: or Reply-to: line, and a filter used this to
1972 generate an autoreply, and therefore failed to obtain an address for the
1973 autoreply, Exim could try to deliver to a non-existent relative file
1974 name, causing unrelated and misleading errors. What now happens is that
1975 it logs this as a hard delivery error, but does not attempt to create a
1976 bounce message.
1977
7a100415
PH
1978PH/11 The exinext utility has a -C option for testing purposes, but although
1979 the given file was scanned by exinext itself; it wasn't being passed on
1980 when Exim was called.
1981
19b9dc85
PH
1982PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1983 an end-of-file indication when reading a command response.
1984
309bd837
PH
1985PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1986 compiled. In many other places in Exim, IPv6 addresses are always
1987 recognized, so I have changed this. It also means that IPv4 domain
1988 literals of the form [IPV4:n.n.n.n] are now always recognized.
1989
59e82a2a
PH
1990PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1991 used if the router is not running as root, for example, when verifying at
1992 ACL time, or when using -bh. The debugging output from this situation was
1993 non-existent - all you got was a failure to exec. I have made two
1994 changes:
1995
1996 (a) Failures to set uid/gid, the current directory, or a process leader
1997 in a subprocess such as that created by queryprogram now generate
1998 suitable debugging ouput when -d is set.
1999
2000 (b) The queryprogram router detects when it is not running as root,
2001 outputs suitable debugging information if -d is set, and then runs
2002 the subprocess without attempting to change uid/gid.
2003
9edc04ce
PH
2004PH/15 Minor change to Makefile for building test_host (undocumented testing
2005 feature).
2006
1349e1e5
PH
2007PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2008 additional section of a DNS packet that returns MX or SRV records.
2009 Instead, it always explicitly searches for A/AAAA records. This avoids
2010 major problems that occur when a DNS server includes only records of one
2011 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2012 fixed another bug: if SRV records were looked up and the corresponding
2013 address records were *not* found in the additional section, the port
2014 values from the SRV records were lost.
2015
ea49d0e1
PH
2016PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2017 using the correct key (the original address) when searching the retry
2018 rules in order to find which one to use for generating the retry hint.
2019
064a94c9
PH
2020PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2021 adding the default one. Similarly, if it contains a Reply-To: header, the
2022 errors_reply_to option, if set, is not used.
2023
727071f8
PH
2024PH/19 When calculating a retry time, Exim used to measure the "time since
2025 failure" by looking at the "first failed" field in the retry record. Now
2026 it does not use this if it is later than than the arrival time of the
2027 message. Instead it uses the arrival time. This makes for better
2028 behaviour in cases where some deliveries succeed, thus re-setting the
2029 "first failed" field. An example is a quota failure for a huge message
2030 when small messages continue to be delivered. Without this change, the
2031 "time since failure" will always be short, possible causing more frequent
2032 delivery attempts for the huge message than are intended.
dd16e114 2033 [Note: This change was subsequently modified - see PH/04 for 4.62.]
727071f8 2034
f78eb7c6
PH
2035PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2036 $1, $2, $3) because the numerical variables can be reset during some
2037 expansion items (e.g. "match"), thereby losing the authentication data.
2038
21c28500
PH
2039PH/21 Make -bV show the size of off_t variables so that the test suite can
2040 decide whether to run tests for quotas > 2G.
2041
2042PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2043 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2044 filecount value is greater than 2G or if a quota value is greater than 2G
2045 on a system where the size of off_t is not greater than 4, a panic error
2046 is given.
2047
1688f43b
PH
2048PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2049 never match. The debug and -bh output now contains an explicit error
2050 message indicating a malformed IPv4 address or mask.
2051
2052PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2053 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2054 PH/23 above applies.
2055
9675b384
PH
2056PH/25 Do not write to syslog when running in the test harness. The only
2057 occasion when this arises is a failure to open the main or panic logs
2058 (for which there is an explicit test).
2059
6a3f1455
PH
2060PH/26 Added the /no_tell option to "control=freeze".
2061
dac79d3e
PH
2062PH/27 If a host name lookup failed very early in a connection, for example, if
2063 the IP address matched host_lookup and the reverse lookup yielded a name
2064 that did not have a forward lookup, an error message of the form "no IP
2065 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2066 could be logged. Now it outputs the IP address instead of "NULL".
1349e1e5 2067
5977a0b3
PH
2068PH/28 An enabling patch from MH: add new function child_open_exim2() which
2069 allows the sender and the authenticated sender to be set when
2070 submitting a message from within Exim. Since child_open_exim() is
2071 documented for local_scan(), the new function should be too.
2072
c91535f3
PH
2073PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2074 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2075 results in an empty string is now treated as unset.
2076
0d46a8c8
PH
2077PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2078
278c6e6c
PH
2079PH/31 Added sender_verify_fail logging option.
2080
2cbb4081
PH
2081PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2082 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2083 tidied the source and removed it altogether.
2084
3eef829e
PH
2085PH/33 When a queue run was abandoned because the load average was too high, a
2086 log line was always written; now it is written only if the queue_run log
2087 selector is set. In addition, the log line for abandonment now contains
2088 information about the queue run such as the pid. This is always present
2089 in "start" and "stop" lines but was omitted from the "abandon" line.
2090
1ab95fa6
PH
2091PH/34 Omit spaces between a header name and the colon in the error message that
2092 is given when verify = headers_syntax fails (if there are lots of them,
2093 the message gets confusing).
2094
230205fc
PH
2095PH/35 Change the default for dns_check_names_pattern to allow slashes within
2096 names, as there are now some PTR records that contain slashes. This check
2097 is only to protect against broken name servers that fall over on strange
2098 characters, so the fact that it applies to all lookups doesn't matter.
2099
75e0e026
PH
2100PH/36 Now that the new test suite is complete, we can remove some of the
2101 special code in Exim that was needed for the old test suite. For example,
2102 sorting DNS records because real resolvers return them in an arbitrary
2103 order. The new test suite's fake resolver always returns records in the
2104 same order.
2105
2106PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2107 bounces) except when queue_only is set, to avoid logging races between
2108 the different processes.
2109
145396a6
PH
2110PH/38 Panic-die if .include specifies a non-absolute path.
2111
3cd34f13
PH
2112PH/39 A tweak to the "H" retry rule from its user.
2113
11121d3d
JJ
2114JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2115 a label. They prevented compilation on older perls.
2116
2117JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2118 a warning to be raised on newish perls.
2119
2120JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2121 on queue. Changes to match documented behaviour of showing count of
2122 messages matching specified criteria.
2123
8def5aaf
PH
2124PH/40 Changed the default ident timeout from 30s to 5s.
2125
929ba01c
PH
2126PH/41 Added support for the use of login_cap features, on those BSD systems
2127 that have them, for controlling the resources used by pipe deliveries.
2128
2632889e
PH
2129PH/42 The content-scanning code uses fopen() to create files in which to put
2130 message data. Previously it was not paying any attention to the mode of
2131 the files. Exim runs with umask(0) because the rest of the code creates
2132 files with open(), and sets the required mode explicitly. Thus, these
2133 files were ending up world-writeable. This was not a big issue, because,
2134 being within the spool directory, they were not world-accessible. I have
2135 created a function called modefopen, which takes an additional mode
2136 argument. It sets umask(777), creates the file, chmods it to the required
2137 mode, then resets the umask. All the relevant calls to fopen() in the
2138 content scanning code have been changed to use this function.
2139
944a9c55
PH
2140PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2141 to 24 hours. This avoids potential overflow problems when processing G
2142 and H retry rules. I suspect nobody ever tinkers with this value.
2143
4a23603b
PH
2144PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2145
4730f942
PH
2146PH/45 When the plaintext authenticator is running as a client, the server's
2147 challenges are checked to ensure they are valid base64 strings. By
2148 default, the authentication attempt is cancelled if an invalid string is
2149 received. Setting client_ignore_invalid_base64 true ignores these errors.
2150 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2151 they are received. Thus, the responses can be made to depend on the
2152 challenges. If an invalid string is ignored, an empty string is placed in
2153 the variable.
2154
30dba1e6
PH
2155PH/46 Messages that are created by the autoreply transport now contains a
2156 References: header, in accordance with RFCs 2822 and 3834.
2157
382afc6b
PH
2158PH/47 Added authenticated_sender_force to the smtp transport.
2159
a86229cf
PH
2160PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2161
50c99ba6
PH
2162PH/49 Installed latest patch from the Sieve maintainer.
2163
d35e429d
PH
2164PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2165 was also set, the appendfile transport was unnecessarily scanning a
2166 directory of message files (e.g. for maildir delivery) to find the count
2167 of files (along with the size), even though it did not need this
2168 information. It now does the scan only if it needs to find either the
2169 size of the count of files.
2170
f90d018c
PH
2171PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2172
75def545
PH
2173PH/52 Two bugs concerned with error handling when the smtp transport is
2174 used in LMTP mode:
2175
2176 (i) Exim was not creating retry information for temporary errors given
2177 for individual recipients after the DATA command when the smtp transport
2178 was used in LMTP mode. This meant that they could be retried too
2179 frequently, and not timed out correctly.
2180
2181 (ii) Exim was setting the flag that allows error details to be returned
2182 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2183 recipients that were returned after the DATA command.
2184
2185PH/53 This is related to PH/52, but is more general: for any failing address,
2186 when detailed error information was permitted to be returned to the
2187 sender, but the error was temporary, then after the final timeout, only
2188 "retry timeout exceeded" was returned. Now it returns the full error as
2189 well as "retry timeout exceeded".
2190
c46782ef
PH
2191PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2192 do this, and (what is worse) MTAs that accept it.
2193
71fafd95
PH
2194PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2195 will now be deprecated.
2196
2c5db4fd
PH
2197PH/56 New os.c-cygwin from the Cygwin maintainer.
2198
9cf6b11a
JJ
2199JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2200 formats (previously only available in exim formats via -bpr, -bpru,
2201 and -bpra. Now also available in native and exiqgrep formats)
2202
2203JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2204 with very large, slow to parse queues
2205
2206JJ/08 exipick: added ! as generic prefix to negate any criteria format
2207
2208JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2209
898d150f
PH
2210PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2211 responses to authentication challenges, though it was showing the
2212 challenges; (ii) I've removed the CR characters from the debug output for
2213 SMTP output lines.
2214
46218253
PH
2215PH/58 Allow for the insertion of a newline as well as a space when a string
2216 is turned into more than one encoded-word during RFC 2047 encoding. The
2217 Sieve code now uses this.
2218
e97957bc
PH
2219PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2220 data_4xx, lost_connection, tls_required.
2221
81e509d7
PH
2222PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2223 message was being sent as an SMTP response.
2224
3d240ff7
PH
2225PH/61 Add -l and -k options to exicyclog.
2226
b37c4101
PH
2227PH/62 When verifying, if an address was redirected to one new address, so that
2228 verification continued, and the new address failed or deferred after
2229 having set something in $address_data, the value of $address_data was not
2230 passed back to the ACL. This was different to the case when no
2231 redirection occurred. The value is now passed back in both cases.
2232
79378e0f
PH
2233PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2234 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2235 use login_cap.h, so on its own it isn't the distinguishing feature. The
2236 new name refers directly to the setclassresources() function.
2237
e49c7bb4
PH
2238PH/65 Added configuration files for NetBSD3.
2239
d114ec46
PH
2240PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2241
f3d7df6c
PH
2242PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2243 is preferred over IPv4.
2244
715ab376
PH
2245PH/68 The bounce_return_message and bounce_return_body options were not being
2246 honoured for bounces generated during the reception of non-SMTP messages.
2247 In particular, this applied to messages rejected by the ACL. This bug has
2248 been fixed. However, if bounce_return_message is true and bounce_return_
2249 body is false, the headers that are returned for a non-SMTP message
2250 include only those that have been read before the error was detected.
2251 (In the case of an ACL rejection, they have all been read.)
2252
6b31b150
PH
2253PH/69 The HTML version of the specification is now built in a directory called
2254 spec_html instead of spec.html, because the latter looks like a path with
2255 a MIME-type, and this confuses some software.
2256
2257PH/70 Catch two compiler warnings in sieve.c.
2258
d515a917
PH
2259PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2260 function verify_get_ident() calls ip_connect() to connect a socket, but
2261 if the "connect()" function timed out, ip_connect() used to close the
2262 socket. However, verify_get_ident() also closes the socket later, and in
2263 between Exim writes to the log, which may get opened at this point. When
2264 the socket was closed in ip_connect(), the log could get the same file
2265 descriptor number as the socket. This naturally causes chaos. The fix is
2266 not to close the socket in ip_connect(); the socket should be closed by
2267 the function that creates it. There was only one place in the code where
2268 this was missing, in the iplookup router, which I don't think anybody now
2269 uses, but I've fixed it anyway.
2270
9b8fadde
PH
2271PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2272 well as to direct DNS lookups. Otherwise the handling of names in host
2273 lists is inconsistent and therefore confusing.
2274
214e2000 2275
5de37277
PH
2276Exim version 4.60
2277-----------------
2278
cc38ddbf
PH
2279PH/01 Two changes to the default runtime configuration:
2280
2281 (1) Move the checks for relay_from_hosts and authenticated clients from
2282 after to before the (commented out) DNS black list checks.
2283
2284 (2) Add control=submission to the relay_from_hosts and authenticated
2285 clients checks, on the grounds that messages accepted by these
2286 statements are most likely to be submissions.
5de37277 2287
72fdd6ae
PH
2288PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2289
2290 (1) Generate an error if the third argument for the ${prvs expansion is
2291 not a single digit.
2292
2293 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2294 string.
2295
2296 (3) Reset the variables that are obtained from the first argument of
2297 ${prvscheck and used in the second argument before leaving the code,
2298 because their memory is reclaimed, so using them afterwards may do
2299 silly things.
2300
2301 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2302 one (it's much easier than Tom thought :-).
2303
2304 (5) Because of (4), we can now allow for the use of $prvscheck_result
2305 inside the third argument.
cb9328de 2306
cb741023
PH
2307PH/03 For some reason, the default setting of PATH when running a command from
2308 a pipe transport was just "/usr/bin". I have changed it to
2309 "/bin:/usr/bin".
2310
f174f16e
PH
2311PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2312 anything to be listed in the output from -bV.
b2f5a032 2313
c25242d7
PH
2314PH/05 When a filter generated an autoreply, the entire To: header line was
2315 quoted in the delivery log line, like this:
2316
2317 => >A.N.Other <ano@some.domain> <original@ddress> ...
2318
2319 This has been changed so that it extracts the operative address. There
2320 may be more than one such address. If so, they are comma-separated, like
2321 this:
2322
2323 => >ano@some.domain,ona@other.domain <original@ddress> ...
2324
82c19f95
PH
2325PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2326 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2327 not being looked up in the rDNS to get a host name, Exim was showing the
2328 IP address twice in Received: lines, even though the IP addresses were
2329 identical. For example:
2330
2331 Received: from [1.2.3.4] (helo=[1.2.3.4])
2332
2333 However, if the real host name was known, it was omitting the HELO data
2334 if it matched the actual IP address. This has been tidied up so that it
2335 doesn't show the same IP address twice.
2336
d7ffbc12
PH
2337PH/07 When both +timestamp and +memory debugging was on, the value given by
2338 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2339 called by the debug printing, thereby overwriting the timestamp buffer.
2340 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2341
9f526266
PH
2342PH/08 When the original message was included in an autoreply transport, it
2343 always said "this is a copy of the message, including all the headers",
2344 even if body_only or headers_only was set. It now gives an appropriate
2345 message.
2346
87fcc8b9
PH
2347PH/09 Applied a patch from the Sieve maintainer which:
2348
2349 o fixes some comments
2350 o adds the (disabled) notify extension core
2351 o adds some debug output for the result of if/elsif tests
2352 o points to the current vacation draft in the documentation
2353 and documents the missing references header update
2354
2355 and most important:
2356
2357 o fixes a bug in processing the envelope test (when testing
2358 multiple envelope elements, the last element determinted the
2359 result)
2360
456682f5
PH
2361PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2362 Electronic Mail") by including:
2363
2364 Auto-submitted: auto-generated
2365
2366 in the messages that it generates (bounce messages and others, such as
2367 warnings). In the case of bounce messages for non-SMTP mesages, there was
2368 also a typo: it was using "Auto_submitted" (underscore instead of
2369 hyphen). Since every message generated by Exim is necessarily in response
2370 to another message, thes have all been changed to:
2371
2372 Auto-Submitted: auto-replied
2373
2374 in accordance with these statements in the RFC:
2375
2376 The auto-replied keyword:
2377
2378 - SHOULD be used on messages sent in direct response to another
2379 message by an automatic process,
2380
2381 - MUST NOT be used on manually-generated messages,
2382
2383 - MAY be used on Delivery Status Notifications (DSNs) and Message
2384 Disposition Notifications (MDNs),
2385
2386 - MUST NOT be used on messages generated by automatic or periodic
2387 processes, except for messages which are automatic responses to
2388 other messages.
2389
3e46c1aa
PH
2390PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2391 to the default Received: header definition.
456682f5 2392
49826d12
PH
2393PH/12 Added log selector acl_warn_skipped (default on).
2394
eba0c039
PH
2395PH/13 After a successful wildlsearch lookup, discard the values of numeric
2396 variables because (a) they are in the wrong storage pool and (b) even if
2397 they were copied, it wouldn't work properly because of the caching.
2398
a0d6ba8a
PH
2399PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2400 checking when decoding. Apparently there are clients that generate
2401 overlong encoded strings. Why am I not surprised?
2402
f0917727
PH
2403PH/15 If the first argument of "${if match_address" was not empty, but did not
2404 contain an "@" character, Exim crashed. Now it writes a panic log message
2405 and treats the condition as false.
2406
096fee00
PH
2407PH/16 In autoreply, treat an empty string for "once" the same as unset.
2408
024bd3c2
PH
2409PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2410 extension "envelope-auth". The code is finished and in agreement with
2411 other implementations, but there is no documentation so far and in fact,
2412 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2413 not changing the active code.
2414
2415 Print executed "if" and "elsif" statements when debugging is used. This
2416 helps a great deal to understand what a filter does.
2417
2418 Document more things not specified clearly in RFC3028. I had all this
2419 sorted out, when out of a sudden new issues came to my mind. Oops."
2420
df199fec
PH
2421PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2422 (Bugzilla #53).
2423
d27f1df3
PH
2424PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2425 canonical form (as documented). However, after a host name lookup from
2426 the IP address, check_host() was doing a simple string comparison with
2427 addresses acquired from the DNS when checking that the found name did
2428 have the original IP as one of its addresses. Since any found IPv6
2429 addresses are likely to be in abbreviated form, the comparison could
2430 fail. Luckily, there already exists a function for doing the comparison
2431 by converting both addresses to binary, so now that is used instead of
2432 the text comparison.
2433
96776534
PH
2434PH/20 There was another similar case to PH/19, when a complete host name was
2435 given in a host list; looking up its IP address could give an abbreviated
2436 form, whereas the current host's name might or might not be abbreviated.
2437 The same fix has been applied.
2438
5de37277 2439
9a799bc0
PH
2440Exim version 4.54
2441-----------------
2442
2443PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2444 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2445 It now does.
2446
99a4b039
PH
2447PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2448 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2449
2450PH/03 Typo: missing ".o" in src/pcre/Makefile.
2451
4b233853
PH
2452PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2453 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2454 Also, for "Auto-Submitted", treat anything other than "no" as
2455 non-personal, in accordance with RFC 3834. (Previously it treated
2456 anything starting "auto-" as non-personal.)
2457
8857ccfd
PH
2458TF/01 The control=submission/name=... option had a problem with syntax
2459 errors if the name included a slash character. The /name= option
2460 now slurps the rest of the string, so it can include any characters
2461 but it must come last in the list of options (after /sender_retain
2462 or /domain=).
2463
433a2980
PH
2464PH/05 Some modifications to the interface to the fake nameserver for the new
2465 testing suite.
2466
3e46c1aa 2467
9a799bc0 2468
e3a311ba
TK
2469Exim version 4.53
2470-----------------
2471
2472TK/01 Added the "success_on_redirect" address verification option. See
2473 NewStuff for rationale and an example.
2474
13b685f9
PH
2475PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2476
395ff96d
PH
2477PH/02 Patch to exigrep to allow it to work on syslog lines.
2478
5b68f6e4
PH
2479PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2480 fread() to skip over the body file's header line, because in Cygwin the
2481 header line is locked and is inaccessible.
2482
1ab52c69
PH
2483PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
2484 co-exist for some time) to make it clear that it is the Exim ID that is
2485 referenced, not the Message-ID: header line.
2486
b07e6aa3
PH
2487PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
2488 string_format() function, because snprintf() does not exist on all
2489 operating systems.
2490
254e032f
PH
2491PH/06 The use of forbid_filter_existstest now also locks out the use of the
2492 ${stat: expansion item.
2493
3af76a81
PH
2494PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
2495 protocol synchronization error", to keep the pedants happy.
2496
2548ba04
PH
2497PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
2498 well as for IRIX systems, when gcc is being used. See the host.c source
2499 file for comments.
2500
b6c6011d
PH
2501PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
2502
cf39cf57
PH
2503PH/10 Named domain lists were not working if used in a queue_smtp_domains
2504 setting.
2505
f1513293
PH
2506PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
2507 transport and to the smtp transport in LMTP mode.
2508
727549a4
PH
2509TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
2510
af46795e
PH
2511PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
2512 run a filter in a subprocess. This could lead to confusion in subsequent
2513 lookups in the parent process. There should also be a search_tidyup() at
2514 the end of the subprocess.
2515
d7b47fd0
PH
2516PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
2517 only if the host matched helo_try_verify_hosts, which caused the
2518 verification to occur when the EHLO/HELO command was issued. The ACL just
2519 tested the remembered result. Now, if a previous verification attempt has
2520 not happened, "verify = helo" does it there and then.
2521
ee744174
JJ
2522JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
2523
b582ab87
PH
2524TK/03 Fix log output including CR from clamd.
2525
41a13e0a
PH
2526PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
2527 exist provoked a memory error which could cause a segfault.
2528
f625cc5a
PH
2529PH/15 Installed PCRE 6.2
2530
2531PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2532
21f7af35
PH
2533PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2534 of the problem. Specifically, suggested +O2 rather than +O1 for the
2535 HP-UX compiler.
2536
31480e42
PH
2537PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2538
2d280592
PH
2539PH/20 If a delivery was routed to a non-standard port by means of an SRV
2540 record, the port was not correctly logged when the outgoing_port log
2541 selector was set (it logged the transort's default port).
2542
7cd1141b
PH
2543PH/21 Added support for host-specific ports to manualroute, queryprogram,
2544 fallback_hosts, and "hosts" in the smtp transport.
2545
2546PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2547 host errors such as "Connection refused".
2548
750af86e
PH
2549PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2550 authentication with radiusclient 0.4.9:
2551
2552 - Error returned from rc_read_config was caught wrongly
2553 - Username/password not passed on to radius server due to wrong length.
2554
2555 The presumption is that some radiusclient API changes for 4.51/PH/17
2556 were not taken care of correctly. The code is still untested by me (my
2557 Linux distribution still has 0.3.2 of radiusclient), but it was
2558 contributed by a Radius user.
2559
2560PH/24 When doing a callout, the value of $domain wasn't set correctly when
2561 expanding the "port" option of the smtp transport.
2562
4304270b
TK
2563TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2564 while reading a MIME header. Thanks to Tom Hughes for a patch.
2565
750af86e
PH
2566PH/24 Include config.h inside local_scan.h so that configuration settings are
2567 available.
2568
64ffc24f
PH
2569PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2570 that in an ACL for RCPT (for example), you can examine exactly what was
2571 received.
2572
5dd9625b
PH
2573PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2574 commands, but it was not correctly comparing the address with the actual
2575 client host address. Thus, it would show the EHLO address in Received:
2576 header lines when this was not necessary.
2577
5591031b
PH
2578PH/27 Added the % operator to ${eval:}.
2579
ba18e66a
PH
2580PH/28 Exim tries to create and chdir to its spool directory when it starts;
2581 it should be ignoring failures (because with -C, for example, it has lost
2582 privilege). It wasn't ignoring creation failures other than "already
2583 exists".
2584
9cec981f
PH
2585PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2586 -bV or -d is used.
2587
aa2b5c79
PH
2588PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2589 because an input line was too long, either on its own, or by virtue of
1509d3a8
PH
2590 too many continuations, the temporary file was not being removed, and the
2591 return code was incorrect.
aa2b5c79 2592
48a53b7f
PH
2593PH/31 Missing "BOOL" in function definition in filtertest.c.
2594
1c59d63b
PH
2595PH/32 Applied Sieve patches from the maintainer.
2596
671012da
TK
2597TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2598
1509d3a8
PH
2599PH/33 Added "verify = not_blind".
2600
2601PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2602 Local/Makefile (with some defaults set). These are used in built scripts
2603 such as exicyclog, but they have never been used in the exim_install
2604 script (though there are many overriding facilities there). I have
2605 arranged that the exim_install script now takes note of these two
2606 settings.
2607
2608PH/35 Installed configuration files for Dragonfly.
2609
2fe1a124
PH
2610PH/36 When a locally submitted message by a trusted user did not contain a
2611 From: header, and the sender address was obtained from -f or from an SMTP
2612 MAIL command, and the trusted user did not use -F to supply a sender
2613 name, $originator_name was incorrectly used when constructing a From:
2614 header. Furthermore, $originator_name was used for submission mode
2615 messages from external hosts without From: headers in a similar way,
2616 which is clearly wrong.
2617
8800895a
PH
2618PH/37 Added control=suppress_local_fixups.
2619
ccfdb010
PH
2620PH/38 When log_selector = +received_sender was set, and the addition of the
2621 sender made the log line's construction buffer exactly full, or one byte
2622 less than full, an overflow happened when the terminating "\n" was
2623 subsequently added.
2624
1130bfb0
PH
2625PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2626 when the result of a list match is failure because a DNS lookup failed.
2627
ebcb507f
PH
2628PH/40 RM_COMMAND is now used in the building process.
2629
c35e155c
PH
2630PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2631 the "build-* directories that it finds.
2632
95d1f782
PH
2633PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2634 address in a domain literal was a prefix of an interface address.
2635
fd6de02e
PH
2636PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2637 when verifying a sender address, unless rewrite_headers is false.
2638
58de37c5
PH
2639PH/44 Wrote a long comment about why errors_to addresses are verified as
2640 recipients, not senders.
2641
261cf466
TF
2642TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2643 the ratelimit ACL was added.
2644
3ee512ff
PH
2645PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2646
e08c430f
PH
2647PH/46 Added extra information about PostgreSQL errors to the error string.
2648
bef5a11f
PH
2649PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2650 avoiding the need to install special zones in a real server. This is
2651 backwards compatible; if it can't find the fake resolver, it drops back.
2652 Thus, both old and new test suites can be run.
2653
7546de58
TF
2654TF/02 Added util/ratelimit.pl
2655
e5d5a95f
TF
2656TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2657 clock is set back in time.
2658
2e88a017
TF
2659TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2660 Candler <B.Candler@pobox.com>.
2661
a5f65aa4
TF
2662TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2663 OK for addresses that are the result of redirections.
2664
e7726cbf
PH
2665PH/48 A number of further additions for the benefit of the new test suite,
2666 including a fake gethostbyname() that interfaces to the fake DNS resolver
2667 (see PH/47 above).
2668
a7fdad5b
TF
2669TF/06 The fix for widen_domains has also been applied to qualify_single and
2670 search_parents which are the other dnslookup options that can cause
2671 header rewrites.
2672
6af56900
PH
2673PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2674 type ("H").
2675
0925ede6
PH
2676PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2677
66afa403
TF
2678TF/07 Exim produced the error message "an SRV record indicated no SMTP
2679 service" if it encountered an MX record with an empty target hostname.
2680 The message is now "an MX or SRV record indicated no SMTP service".
2681
0154e85a
TF
2682TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2683 if the DNS of the sending site is misconfigured. This is quite a
2684 common situation. This change restores the behaviour of treating a
2685 helo verification defer as a failure.
2686
16f12c76
PH
2687PH/51 If self=fail was set on a router, the bounce message did not include the
2688 actual error message.
2689
bbe902f0 2690
e5a9dba6
PH
2691Exim version 4.52
2692-----------------
2693
2694TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2695
22c3b60b
PH
2696PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2697 command itself ended in error, the underlying message about the transport
2698 filter timeout was being overwritten with the pipe command error. Now the
2699 underlying error message should be appended to the second error message.
2700
06a9b4b5
PH
2701TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2702
c1ac6996
PH
2703PH/02 Reduce the amount of output that "make" produces by default. Full output
2704 can still be requested.
2705
9c7a242c
PH
2706PH/03 The warning log line about a condition test deferring for a "warn" verb
2707 was being output only once per connection, rather than after each
2708 occurrence (because it was using the same function as for successful
2709 "warn" verbs). This seems wrong, so I have changed it.
2710
87ba3f5f
PH
2711TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2712 it should not have, which might have caused a crash in the right
2713 circumstances, but probably never did.
2714
2715PH/04 Installed a modified version of Tony Finch's patch to make submission
2716 mode fix the return path as well as the Sender: header line, and to
2717 add a /name= option so that you can make the user's friendly name appear
2718 in the header line.
2719
29aba418
TF
2720TF/03 Added the control = fakedefer ACL modifier.
2721
fe0dab11
TF
2722TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2723 Mark Lowes for thorough testing.
870f6ba8 2724
11d337a4
TK
2725TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2726
2727TK/03 Merged latest SRS patch from Miles Wilton.
2728
415c8f3b
PH
2729PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2730 with the definition in sysexits.h (which is #included earlier).
2731 Fortunately, Exim does not actually use EX_OK. The code used to try to
2732 preserve the sysexits.h value, by assumimg that macro definitions were
2733 scanned for macro replacements. I have been disabused of this notion,
2734 so now the code just undefines EX_OK before #including unistd.h.
11d337a4 2735
958541e9
PH
2736PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2737 in the smtp transport. When a block could not be written in a single
2738 write() function, the timeout was being re-applied to each part-write.
2739 This seems wrong - if the receiver was accepting one byte at a time it
2740 would take for ever. The timeout is now adjusted when this happens. It
2741 doesn't have to be particularly precise.
2742
c206415f
TK
2743TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2744 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2745
2a4be8f9
PH
2746PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2747 without a domain if the check to <postmaster@domain> fails.
2748
1cba11c5
SC
2749SC/01 Eximstats: added -xls and the ability to specify output files
2750 (patch written by Frank Heydlauf).
2751
2752SC/02 Eximstats: use FileHandles for outputing results.
2753
2754SC/03 Eximstats: allow any combination of xls, txt, and html output.
2755
2756SC/04 Eximstats: fixed display of large numbers with -nvr option
2757
2758SC/05 Eximstats: fixed merging of reports with empty tables.
2759
2760SC/06 Eximstats: added the -include_original_destination flag
2761
2762SC/07 Eximstats: removed tabs and trailing whitespace.
2763
1005d00e
TK
2764TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2765
2766TK/06 MBOX spool code: Add real "From " MBOX separator line
2767 so the .eml file is really in mbox format (even though
2768 most programs do not really care). Patch from Alex Miller.
2769
2770TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2771 The latter is generated from $received_to and is only set if the
2772 message has one envelope recipient. SA can use these headers,
2773 obviously out-of-the-box. Patch from Alex Miller.
2774
9b4768fa
PH
2775PH/08 The ${def test on a variable was returning false if the variable's
2776 value was "0", contrary to what the specification has always said!
2777 The result should be true unless the variable is empty.
2778
2779PH/09 The syntax error of a character other than { following "${if
2780 def:variable_name" (after optional whitespace) was not being diagnosed.
2781 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2782 accidental colon was present, for example, could give incorrect results.
2783
0d7eb84a
PH
2784PH/10 Tidied the code in a number of places where the st_size field of a stat()
2785 result is used (not including appendfile, where other changes are about
2786 to be made).
2787
2788PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2789 This involved changing a lot of size variables from int to off_t. It
2790 should work with maildirs and everything.
2791
40727bee
TK
2792TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2793 spamd dying while we are connected to it.
2794
554d2369
TF
2795TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2796 <jgh@wizmail.org>
2797
1f922db1
PH
2798PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2799 function for BDB 4.3.
2800
ef213c3b
PH
2801PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2802
8ac170f3
PH
2803PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2804 the sources for PCRE have been re-arranged and more files are now
2805 involved.
2806
b1c749bb
PH
2807PH/15 The code I had for printing potentially long long variables in PH/11
2808 above was not the best (it lost precision). The length of off_t variables
2809 is now inspected at build time, and an appropriate printing format (%ld
c6c2dc1d
PH
2810 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2811 to be "long long int" or "long int". This is needed for the internal
2812 formatting function string_vformat().
b1c749bb 2813
4aac9b49
PH
2814PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2815 the configuration file to be ":syslog", then the script "guesses" where
2816 the logs files are, rather than using the compiled in default. In our
2817 case the guess is not the same as the compiled default, so the script
2818 suddenly stopped working when I started to use syslog. The patch checks
2819 to see if log_file_path is "". If so, it attempts to read it from exim
2820 with no configuration file to get the compiled in version, before it
2821 falls back to the previous guessing code."
2822
294520c8
TK
2823TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2824 implementing BATV in an Exim configuration. See NewStuff for the gory
2825 details.
2826
5bd022fe
PH
2827PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2828 Makefile that are specific to HP-UX.
2829
90e9ce59
PH
2830PH/18 If the "use_postmaster" option was set for a recipient callout together
2831 with the "random" option, the postmaster address was used as the MAIL
2832 FROM address for the random test, but not for the subsequent recipient
2833 test. It is now used for both.
2834
5ea81592
PH
2835PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2836 patch removes a few documentation additions to RFC 3028, because the
2837 latest draft now contains them. It adds the new en;ascii-case comparator
2838 and a new error check for 8bit text in MIME parts. Comparator and
2839 require names are now matched exactly. I enabled the subaddress
2840 extension, but it is not well tested yet (read: it works for me)."
2841
c6c2dc1d
PH
2842PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2843 rework some of the code of TK/09 above to avoid the hardwired use of
2844 "%lld" and "long long". Replaced the call to snprintf() with a call to
2845 string_vformat().
2846
fffffe4c
PH
2847PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2848 records point to non-existent hosts", "retry timeout exceeded", and
2849 "retry time not reached for any host after a long failure period".
ca02eafb 2850
9a26b6b2
PH
2851PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2852 experimental DomainKeys support:
2853
2854 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2855 (2) On an error such as an illegally used "control", the wrong name for
2856 the control was given.
2857
2858 These problems did NOT occur unless DomainKeys support was compiled.
2859
4aee0225
PH
2860PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2861
32d668a5
PH
2862PH/24 Added ${if match_ip condition.
2863
8187c3f3
PH
2864PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2865 databases so that it will be absolutely obvious if a crash occurs in the
2866 DB library. This is a regular occurrence (often caused by mis-matched
2867 db.h files).
2868
ff790e47 2869PH/26 Insert a lot of missing (void) casts for functions such as chown(),
f1e894f3
PH
2870 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2871 picked up on a user's system that detects such things. There doesn't seem
2872 to be a gcc warning option for this - only an attribute that has to be
2873 put on the function's prototype. It seems that in Fedora Core 4 they have
2874 set this on a number of new functions. No doubt there will be more in due
2875 course.
ff790e47 2876
5417f6d1
PH
2877PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2878 specify a transport. However, if an address that was verified by such a
2879 router was the subject of a callout, Exim crashed because it tried to
2880 read the rcpt_include_affixes from the non-existent transport. Now it
2881 just assumes that the setting of that option is false. This bug was
2882 introduced by 4.51/PH/31.
2883
59cf8544
PH
2884PH/28 Changed -d+all to exclude +memory, because that information is very
2885 rarely of interest, but it makes the output a lot bigger. People tend to
2886 do -d+all out of habit.
2887
e7ad8a65
PH
2888PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2889 code in os-type was giving problems when libc.so lives in lib64, like on
2890 x86_64 Fedora Core.
2891
ade42478
PH
2892PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2893 aren't the modern standard, and it seems that some systems' include files
2894 don't always have them. Exim was already checking for some of the newer
2895 ones like T_AAAA, and defining it itself. I've added checks for all the
2896 record types that Exim uses.
2897
182ad5cf
PH
2898PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2899 not automatically generating a new one, as it is supposed to. This
2900 prevented TLS from working. If the file did exist, but contained invalid
2901 data, a new version was generated, as expected. It was only the case of a
2902 non-existent file that was broken.
2903
b0d9fc80
TK
2904TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2905 with a change in libdomainkeys > 0.64.
2906
2907TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2908 from DNS. If the selector record carries the flag, it now has
2909 precedence over the domain-wide flag.
2910
2911TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2912
47c7a64a
PH
2913PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2914 the use of a port name that isn't defined in /etc/services) occurred, the
2915 message was deferred as in a normal delivery, and thus remained on the
2916 spool, instead of being failed because of the mua_wrapper setting. This
2917 is now fixed, and I tidied up some of the mua_wrapper messages at the
2918 same time.
2919
a388bce4
SC
2920SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2921 the messages in a hash of arrays rather than using individual hashes.
2922 This is a bit cleaner and results in dramatic memory savings, albeit
2923 at a slight CPU cost.
2924
2925SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2926 as requested by Marc Sherman.
2927
2928SC/10 Eximstats: added histograms for user specified patterns as requested
2929 by Marc Sherman.
2930
0793e4ed
SC
2931SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2932
c58b88df
PH
2933PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2934 fopen() in the content-scanning modules that did not already have it.
2935
e7ad8a65 2936
7982096b
PH
2937Exim version 4.51
2938-----------------
2939
1a46a8c5
PH
2940TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2941 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2942
2f079f46 2943TK/02 Fix ACL "control" statement not being available in MIME ACL.
1a46a8c5
PH
2944
2945TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2946
2947PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2948 to test Sieve filters that use "vacation".
2949
2950PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2951 that changes the way the GnuTLS parameters are stored in the cache file.
2952 The new format can be generated externally. For backward compatibility,
2953 if the data in the cache doesn't make sense, Exim assumes it has read an
2954 old-format file, and it generates new data and writes a new file. This
2955 means that you can't go back to an older release without removing the
2956 file.
2957
2958PH/03 A redirect router that has both "unseen" and "one_time" set does not
2959 work if there are any delivery delays because "one_time" forces the
2960 parent to be marked "delivered", so its unseen clone is never tried
2961 again. For this reason, Exim now forbids the simultaneous setting of
2962 these two options.
2963
2964PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2965 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2966 entry if you want to know the details. The fix, however, neglected to
2967 consider the case when local delivery batching is involved. The test for
2968 "previously delivered" was not happening when checking to see if an
2969 address could be batched with a previous (undelivered) one; under
2970 certain circumstances this could lead to multiple deliveries to the same
c2c19e9d 2971 address.
1a46a8c5
PH
2972
2973PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2974 in its include files, and this causes problems building Exim.
2975
2976PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2977 header_syntax) but Exim was just ignoring anything given after a slash.
2978 In particular, this caused confusion with an attempt to use "verify =
2979 reverse_host_lookup/defer_ok". An error is now given when options are
2980 supplied for verify items that do not have them. (Maybe reverse_host_
2981 lookup should have a defer_ok option, but that's a different point.)
2982
2983PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2984 defined by RFC 821) to 2048, because there were problems with some AUTH
2985 commands, and RFC 1869 says the size should be increased for extended
2986 SMTP commands that take arguments.
2987
2988PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2989 Finch).
2990
2991PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2992 "unknown" error; now it says that the functionality isn't in the binary.
8d67ada3 2993
49c2d5ea
PH
2994PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2995 an address' error message when a string expansion fails (syntax or
f331f3b6
PH
2996 whatever). Otherwise the password may appear in the log. Following change
2997 PH/42 below, there is no longer a chance of it appearing in a bounce
2998 message.
49c2d5ea 2999
bf759a8b
PH
3000PH/11 Installed exipick version 20050225.0 from John Jetmore.
3001
83364d30
PH
3002PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3003 of its addresses was ever tried. (Bugzilla bug #2.)
3004
7999bbd7
PH
3005PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3006 the result incorrectly in the debug output. (It correctly added a newline
3007 to what was transported.)
3008
7dbf77c9
PH
3009TF/01 Added $received_time.
3010
74e0617f
PH
3011PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3012 commented out examples of how to interface to a virus scanner and to
3013 SpamAssassin. Also added commented examples of av_scanner and
3014 spamd_address settings.
3015
2f079f46
PH
3016PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3017 and controls are allowed in which ACLs. There were a couple of minor
3018 errors. Some of the entries in the conditions table (which is a table of
3019 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3020 negation of where the condition IS allowed.
3021
8c841523
PH
3022PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3023
7766a4f0
PH
3024PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3025 header file does not have a version number, so I've had to invent a new
3026 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3027 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3028 radiusclient), but it was contributed by a Radius user.
3029
8b417f2c
PH
3030PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3031 files or directories, for OpenSSL.
3032
901f42cb
PH
3033PH/19 When an Exim process that is running as root has to create an Exim log
3034 file, it does so in a subprocess that runs as exim:exim so as to get the
3035 ownership right at creation (otherwise, other Exim processes might see
3036 the file with the wrong ownership). There was no test for failure of this
3037 fork() call, which would lead to the process getting stuck as it waited
3038 for a non-existent subprocess. Forks do occasionally fail when resources
3039 run out. I reviewed all the other calls to fork(); they all seem to check
3040 for failure.
3041
f9b9210e
PH
3042PH/20 When checking for unexpected SMTP input at connect time (before writing
3043 the banner), Exim was not dealing correctly with a non-positive return
3044 from the read() function. If the client had disconnected by this time,
3045 the result was a log entry for a synchronization error with an empty
3046 string after "input=" when read() returned zero. If read() returned -1
3047 (an event I could not check), uninitialized data bytes were printed.
3048 There were reports of junk text (parts of files, etc) appearing after
3049 "input=".
3050
54cdb463
PH
3051PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3052
cf00dad6
PH
3053PH/22 Added support for macro redefinition, and (re)definition in between
3054 driver and ACL definitions.
3055
acb1b346
PH
3056PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3057 forgetting to use the resulting value; it was using the unexpanded value.
3058
c5ddb310
PH
3059PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3060 hadn't been configured. The fix is from Juergen Kreileder, who
3061 understands it better than I do:
3062
3063 "Here's what I see happening with three configured cyrus_sasl
3064 authenticators configured (plain, login, cram-md5):
3065
3066 On startup auth_cyrus_sasl_init() gets called for each of these.
3067 This means three calls to sasl_listmech() without a specified mech_list.
3068 => SASL tests which mechs of all available mechs actually work
3069 => three warnings about OTP not working
3070 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3071
3072 With the patch, sasl_listmech() also gets called three times. But now
3073 SASL's mech_list option is set to the server_mech specified in the the
3074 authenticator. Or in other words, the answer from sasl_listmech()
3075 gets limited to just the mech you're testing for (which is different
3076 for each call.)
3077 => the return list contains just 'plain' or 'login', 'cram-md5' or
3078 nothing depending on the value of ob->server_mech.
3079
3080 I've just tested the patch: Authentication still works fine,
3081 unavailable mechs specified in the exim configuration are still
3082 caught, and the auth.log warnings about OTP are gone."
3083
31619da6
PH
3084PH/25 When debugging is enabled, the contents of the command line are added
3085 to the debugging output, even when log_selector=+arguments is not
3086 specified.
3087
bebaf0fc
PH
3088PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3089 answer is "GNU", and only if the return is "GNU/something" is the answer
3090 "Linux".
3091
475fe28a
PH
3092PH/27 $acl_verify_message is now set immediately after the failure of a
3093 verification in an ACL, and so is available in subsequent modifiers. In
3094 particular, the message can be preserved by coding like this:
3095
3096 warn !verify = sender
3097 set acl_m0 = $acl_verify_message
3098
3099 Previously, $acl_verify_message was set only while expanding "message"
3100 and "log_message" when a very denied access.
3101
7e8bec7a
PH
3102PH/28 Modified OS/os.c-Linux with
3103
3104 -#ifndef OS_LOAD_AVERAGE
3105 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3106
3107 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3108 nomenclature these days.)
3109
e4a89c47
PH
3110PH/29 Installed patch from the Sieve maintainer that adds the options
3111 sieve_useraddress and sieve_subaddress to the redirect router.
3112
5ca2a9a1
PH
3113PH/30 In these circumstances:
3114 . Two addresses routed to the same list of hosts;
3115 . First host does not offer TLS;
3116 . First host accepts first address;
3117 . First host gives temporary error to second address;
3118 . Second host offers TLS and a TLS session is established;
3119 . Second host accepts second address.
3120 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3121 and peerdn, if requested) that were in fact used only for the second
3122 address.
7e8bec7a 3123
c688b954
PH
3124PH/31 When doing a callout as part of verifying an address, Exim was not paying
3125 attention to any local part prefix or suffix that was matched by the
3126 router that accepted the address. It now behaves in the same way as it
3127 does for delivery: the affixes are removed from the local part unless
3128 rcpt_include_affixes is set on the transport.
3129
fed77020
PH
3130PH/32 Add the sender address, as F=<...>, to the log line when logging a
3131 timeout during the DATA phase of an incoming message.
3132
7fe1560f
PH
3133PH/33 Sieve envelope tests were broken for match types other than :is. I have
3134 applied a patch sanctioned by the Sieve maintainer.
c688b954 3135
ebb6e6d5
PH
3136PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3137 the uid or gid is negative. A case of a negative gid caused this to be
3138 noticed. The fix allows for either to be negative.
3139
9c4e8f60
PH
3140PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3141 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3142 overlooked.
3143
3144PH/36 The change PH/12 above was broken. Fixed it.
3145
d7174846
PH
3146PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3147 the grounds that routing the same address twice would always produce the
3148 same answer. This might have been true once, but it is certainly no
3149 longer true now. Routing a child address may depend on the previous
3150 routing that produced that child. Some complicated redirection strategies
3151 went wrong when messages had multiple recipients, and made Exim's
3152 behaviour dependent on the order in which the addresses were given.
3153
3154 I have moved the duplicate checking until after the routing is complete.
3155 Exim scans the addresses that are assigned to local and remote
3156 transports, and removes any duplicates. This means that more work will be
3157 done, as duplicates will always all be routed, but duplicates are
3158 presumably rare, so I don't expect this is of any significance.
3159
3160 For deliveries to pipes, files, and autoreplies, the duplicate checking
3161 still happens during the routing process, since they are not going to be
3162 routed further.
3163
cfe75fc3
PH
3164PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3165 It corrects a timeout issue with spamd. This is Ian's comment: "The
3166 background is that sometimes spamd either never reads data from a
3167 connection it has accepted, or it never writes response data. The exiscan
3168 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3169 blindly assumes that writes won't block so it may never time out."
3170
be22d70e
PH
3171PH/39 Allow G after quota size as well as K and M.
3172
0612b098
PH
3173PH/40 The value set for $authenticated_id in an authenticator may not contain
3174 binary zeroes or newlines because the value is written to log lines and
3175 to spool files. There was no check on this. Now the value is run through
3176 the string_printing() function so that such characters are converted to
3177 printable escape sequences.
3178
2e0c1448
PH
3179PH/41 $message_linecount is a new variable that contains the total number of
3180 lines in the message. Compare $body_linecount, which is the count for the
3181 body only.
3182
447d236c
PH
3183PH/42 Exim no longer gives details of delivery errors for specific addresses in
3184 bounce and delay warning messages, except in certain special cases, which
3185 are as follows:
3186
3187 (a) An SMTP error message from a remote host;
3188 (b) A message specified in a :fail: redirection;
3189 (c) A message specified in a "fail" command in a system filter;
3190 (d) A message specified in a FAIL return from the queryprogram router;
3191 (e) A message specified by the cannot_route_message router option.
3192
3193 In these cases only, Exim does include the error details in bounce and
3194 warning messages. There are also a few cases where bland messages such
3195 as "unrouteable address" or "local delivery error" are given.
3196
d20976dc
PH
3197PH/43 $value is now also set for the "else" part of a ${run expansion.
3198
f656d135
PH
3199PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3200 being worked on, but at least Exim now implements the latest version to
3201 play with."
3202
2e2a30b4
PH
3203PH/45 In a pipe transport, although a timeout while waiting for the pipe
3204 process to complete was treated as a delivery failure, a timeout while
3205 writing the message to the pipe was logged, but erroneously treated as a
3206 successful delivery. Such timeouts include transport filter timeouts. For
3207 consistency with the overall process timeout, these timeouts are now
3208 treated as errors, giving rise to delivery failures by default. However,
3209 there is now a new Boolean option for the pipe transport called
3210 timeout_defer, which, if set TRUE, converts the failures into defers for
3211 both kinds of timeout. A transport filter timeout is now identified in
3212 the log output.
3213
9176e9f0
PH
3214PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3215 systems where "make" and "gmake" are different, calling "gmake" at top
3216 level broke things. I've arranged for the value of $(MAKE) to be passed
3217 from the Makefile to this script so that it can call the same version of
3218 "make".
3219
7982096b 3220
bbe902f0
PH
3221A note about Exim versions 4.44 and 4.50
3222----------------------------------------
3223
3224Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3225changes of various kinds. As a consequence, a big documentation update was
3226needed. This delayed the release for rather longer than seemed good, especially
3227in the light of a couple of (minor) security issues. Therefore, the changes
3228that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3229release. So 4.44 and 4.50 are in effect two different branches that both start
3230from 4.43.
3231
3232I have left the 4.50 change log unchanged; it contains all the changes since
32334.43. The change log for 4.44 is below; many of its items are identical to
3234those for 4.50. This seems to be the most sensible way to preserve the
3235historical information.
3236
3237
f7b63901 3238Exim version 4.50
495ae4b0
PH
3239-----------------
3240
5fe762f6
PH
3241 1. Minor wording change to the doc/README.SIEVE file.
3242
139059f6 3243 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
5fe762f6 3244 computation of the current number of files was incorrect.
495ae4b0 3245
7086e875
PH
3246 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3247 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3248 place.
3249
35af9f61
PH
3250 4. Give more explanation in the error message when the command for a transport
3251 filter fails to execute.
3252
b668c215
PH
3253 5. There are several places where Exim runs a non-Exim command in a
3254 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3255 was being done only for the command run by the queryprogram router. It is
3256 now done for all such subprocesses. The other cases are: ${run, transport
3257 filters, and the commands run by the lmtp and pipe transports.
3258
a494b1e1
PH
3259 6. Added CONFIGURE_GROUP build-time option.
3260
3261 7. Some older OS have a limit of 256 on the maximum number of file
3262 descriptors. Exim was using setrlimit() to set 1000 as a large value
3263 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3264 systems. I've change it so that if it can't get 1000, it tries for 256.
35edf2ff 3265
c5fcb476
PH
3266 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3267 was an oversight, and furthermore, ever since the addition of extra
3268 controls (e.g. 4.43/32), the checks on when to allow different forms of
3269 "control" were broken. There should now be diagnostics for all cases when a
3270 control that does not make sense is encountered.
3271
69358f02
PH
3272 9. Added the /retain_sender option to "control=submission".
3273
5be20824
PH
327410. $recipients is now available in the predata ACL (oversight).
3275
eb2c0248
PH
327611. Tidy the search cache before the fork to do a delivery from a message
3277 received from the command line. Otherwise the child will trigger a lookup
3278 failure and thereby defer the delivery if it tries to use (for example) a
3279 cached ldap connection that the parent has called unbind on.
3280
2a3eea10
PH
328112. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3282 of $address_data from the recipient verification was clobbered by the
3283 sender verification.
3284
328513. The value of address_data from a sender verification is now available in
3286 $sender_address_data in subsequent conditions in the ACL statement.
3287
23c7ff99
PH
328814. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3289
4deaf07d
PH
329015. Added a new option "connect=<time>" to callout options, to set a different
3291 connection timeout.
3292
926e1192
PH
329316. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3294 was its contents. (It was OK if the option was not defined at all.)
3295
650edc6f
PH
329617. A "Completed" log line is now written for messages that are removed from
3297 the spool by the -Mrm option.
3298
2c7db3f5
PH
329918. New variables $sender_verify_failure and $recipient_verify_failure contain
3300 information about exactly what failed.
3301
3d235903
PH
330219. Added -dd to debug only the daemon process.
3303
7c7ad977
PH
330420. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3305 handles timeouts, both on the server side and network timeouts. Renamed the
3306 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3307
981756db
PH
330821. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3309 It is now set to "smtps".
3310
d4eb88df
PH
331122. $host_address is now set to the target address during the checking of
3312 ignore_target_hosts.
3313
331423. When checking ignore_target_hosts for an ipliteral router, no host name was
3315 being passed; this would have caused $sender_host_name to have been used if
3316 matching the list had actually called for a host name (not very likely,
3317 since this list is usually IP addresses). A host name is now passed as
3318 "[x.x.x.x]".
3319
7d468ab8
PH
332024. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3321 code that specifies a non-restarting handler (typically sigaction() in
3322 modern systems) in an attempt to fix a rare and obscure crash bug.
3323
332425. Narrowed the window for a race in the daemon that could cause it to ignore
3325 SIGCHLD signals. This is not a major problem, because they are used only to
3326 wake it up if nothing else does.
3327
62c0818f
PH
332826. A malformed maildirsize file could cause Exim to calculate negative values
3329 for the mailbox size or file count. Odd effects could occur as a result.
3330 The maildirsize information is now recalculated if the size or filecount
3331 end up negative.
3332
26034054
PH
333327. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3334 support for a long time. Removed HAVE_SYS_VFS_H.
3335
af66f652
PH
333628. Installed the latest version of exipick from John Jetmore.
3337
90af77f4
PH
333829. In an address list, if the pattern was not a regular expression, an empty
3339 subject address (from a bounce message) matched only if the pattern was an
3340 empty string. Non-empty patterns were not even tested. This was the wrong
3341 because it is perfectly reasonable to use an empty address as part of a
3342 database query. An empty address is now tested by patterns that are
3343 lookups. However, all the other forms of pattern expect the subject to
3344 contain a local part and a domain, and therefore, for them, an empty
3345 address still always fails if the pattern is not itself empty.
3346
d8ef3577
PH
334730. Exim went into a mad DNS loop when attempting to do a callout where the
3348 host was specified on an smtp transport, and looking it up yielded more
3349 than one IP address.
3350
5cb8cbc6
PH
335131. Re-factored the code for checking spool and log partition space into a
3352 function that finds that data and another that does the check. The former
3353 is then used to implement four new variables: $spool_space, $log_space,
3354 $spool_inodes, and $log_inodes.
3355
14702f5b
PH
335632. The RFC2047 encoding function was originally intended for short strings
3357 such as real names; it was not keeping to the 75-character limit for
3358 encoded words that the RFC imposes. It now respects the limit, and
3359 generates multiple encoded words if necessary. To be on the safe side, I
3360 have increased the buffer size for the ${rfc2047: expansion operator from
3361 1024 to 2048 bytes.
3362
063b1e99
PH
336333. It is now permitted to omit both strings after an "if" condition; if the
3364 condition is true, the result is "true". As before, when the second string
3365 is omitted, a false condition yields an empty string. This makes it less
3366 cumbersome to write custom ACL and router conditions.
3367
652e1b65
PH
336834. Failure to deliver a bounce message always caused it to be frozen, even if
3369 there was an errors_to setting on the router. The errors_to setting is now
3370 respected.
3371
6f0c9a4f
PH
337235. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3373 canonical form (fully expanded) before being placed in
3374 $sender_host_address.
3375
33397d19
PH
337636. The table in the code that translates DNS record types into text (T_A to
3377 "A" for instance) was missing entries for NS and CNAME. It is just possible
3378 that this could have caused confusion if both these types were looked up
3379 for the same domain, because the text type is used as part of Exim's
3380 per-process caching. But the chance of anyone hitting this buglet seems
3381 very small.
3382
7bb56e1f
PH
338337. The dnsdb lookup has been extended in a number of ways.
3384
3385 (1) There is a new type, "zns", which walks up the domain tree until it
3386 finds some nameserver records. It should be used with care.