OCSP observability: variables $tls_{in,out}_ocsp
[exim.git] / doc / doc-txt / ChangeLog
CommitLineData
495ae4b0
PH
1Change log file for Exim from version 4.21
2-------------------------------------------
3
c0e56233
TF
4
5Exim version 4.83
6-----------------
7
8TF/01 Correctly close the server side of TLS when forking for delivery.
9
10 When a message was received over SMTP with TLS, Exim failed to clear up
11 the incoming connection properly after forking off the child process to
12 deliver the message. In some situations the subsequent outgoing
13 delivery connection happened to have the same fd number as the incoming
14 connection previously had. Exim would try to use TLS and fail, logging
15 a "Bad file descriptor" error.
16
7245734e
TF
17TF/02 Portability fix for building lookup modules on Solaris when the xpg4
18 utilities have not been installed.
19
fd5dad68
JH
20JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
21 temporary space as the ACL may create new global variables.
22
5428a946
TL
23TL/01 LDAP support uses per connection or global context settings, depending
24 upon the detected version of the libraries at build time.
25
a3c86431
TL
26TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
27 to extract and use the src ip:port in logging and expansions as if it
28 were a direct connection from the outside internet.
29
aa26e137
JH
30JH/02 Add ${listextract {number}{list}{success}{fail}}.
31
5a1b8443
WB
32TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
33 Properly escape header and check for NULL return.
34
72c9e342
PP
35PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
36 not dns_use_dnssec.
37
76f44207
WB
38JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
39
770747fd
MFM
40TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
41 characters in header names, implemented as a verify condition.
42 Contributed by Michael Fischer v. Mollard.
43
8ddef691 44TL/05 Rename SPF condition results err_perm and err_temp to standardized
982650ec
TL
45 results permerror and temperror. Previous values are deprecated but
46 still accepted. In a future release, err_perm and err_temp will be
47 completely removed, which will be a backward incompatibility if the
48 ACL tests for either of these two old results. Patch contributed by
8ddef691 49 user bes-internal on the mailing list.
c0e56233 50
b9c2e32f
AR
51JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
52
e45a1c37
JH
53JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
54 selectors, in both main and reject logs.
55
67d81c10
JH
56JH/06 Log outbound-TLS and port details, subject to log selectors, for a
57 failed delivery.
58
b1f8e4f8
JH
59JH/07 Add malware type "sock" for talking to simple daemon.
60
511a6c14 61JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
511a6c14
JH
62
63JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
64 routers/transports under cutthrough routing.
214042d2 65
51c7471d
JH
66JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
67 numbers. Touch up "bool" conditional to keep the same definition.
68
3695be34
TL
69TL/06 Remove duplicated language in spec file from 4.82 TL/16.
70
1e06383a
TL
71JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
72
76146973
JH
73JH/12 Expand items in router/transport headers_add or headers_remove lists
74 individually rather than the list as a whole. Bug 1452.
75
76 Required for reasonable handling of multiple headers_ options when
77 they may be empty; requires that headers_remove items with embedded
78 colons must have them doubled (or the list-separator changed).
79
8c8b8274
TL
80TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
81 view the policy declared in the DMARC record. Currently, $dmarc_status
82 is a combined value of both the record presence and the result of the
83 analysis.
b1f8e4f8 84
35aba663
JH
85JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
86
8c51eead 87JH/14 New options dnssec_request_domains, dnssec_require_domains on the
578897ea
JH
88 dnslookup router and the smtp transport (applying to the forward
89 lookup).
8c51eead 90
deae092e
HS
91TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
92 of ldap servers used for a specific lookup. Patch provided by Heiko
93 Schlichting.
35aba663 94
fd3b6a4a 95JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
4e0983dc 96 New variable $lookup_dnssec_authenticated for observability.
fd3b6a4a 97
8d91c6dc
LT
98TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
99 Patch submitted by Lars Timman.
100
2b4a568d
JH
101JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
102
d2af03f4
HS
103TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
104 Requires trusted mode and valid format message id, aborts otherwise.
105 Patch contributed by Heiko Schlichting.
106
9d1c15ef
JH
107JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
108 certextract with support for various fields. Bug 1358.
109
44662487
JH
110JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
111 is requested by default, modifiable by smtp transport option
112 hosts_request_ocsp;
113
fd3b6a4a 114
2c422e6f 115Exim version 4.82
98a90c36
PP
116-----------------
117
118PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
119
12f69989
PP
120PP/02 Make -n do something, by making it not do something.
121 When combined with -bP, the name of an option is not output.
122
54c90be1
PP
123PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
124 by GnuTLS.
125
1f4a55da
PP
126PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
127 $sender_host_name and config options to manage this, and basic check
128 routines.
129
13363eba 130PP/05 DSCP support for outbound connections and control modifier for inbound.
36a3ae5f 131
66645890 132PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
e402235f
PP
133 (Only plugin which currently uses this is kerberos4, which nobody should
134 be using, but we should make it available and other future plugins might
135 conceivably use it, even though it would break NAT; stuff *should* be
136 using channel bindings instead).
66645890 137
a3fb9793 138PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
f4ee74ac
PP
139 name; added for Sendmail compatibility; requires admin caller.
140 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
141 just ignore it); requires trusted caller.
a3fb9793 142 Also parse but ignore: -Ac -Am -X<logfile>
f4ee74ac 143 Bugzilla 1117.
a3fb9793 144
d27f98fe 145TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
98a90c36 146
6822b909
TL
147TL/02 Add +smtp_confirmation as a default logging option.
148
e7568d51
TL
149TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
150 Patch by Magnus Holmgren from 2007-02-20.
151
ae0e32ee 152TL/04 Bugzilla 1281 - Spec typo.
ca0ff207 153 Bugzilla 1283 - Spec typo.
97f42f10 154 Bugzilla 1290 - Spec grammar fixes.
ca0ff207
TL
155
156TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
ae0e32ee 157
e2658fff
TL
158TL/06 Add Experimental DMARC support using libopendmarc libraries.
159
83712b39
TL
160TL/07 Fix an out of order global option causing a segfault. Reported to dev
161 mailing list by by Dmitry Isaikin.
162
976b7e9f
JH
163JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
164
be4a1376
JH
165JH/02 Support "G" suffix to numbers in ${if comparisons.
166
ec4b68e5
PP
167PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
168
d7148a07
NM
169NM/01 Bugzilla 1197 - Spec typo
170 Bugzilla 1196 - Spec examples corrections
ec4b68e5 171
585121e2 172JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
ec4b68e5 173
2519e60d
TL
174PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
175 gnutls_enable_pkcs11, but renamed to more accurately indicate its
176 function.
a5f239e4 177
13d08c90
PP
178PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
179 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
180
bef3ea7f
JH
181JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
182 "acl {{name}{arg}...}", and optional args on acl condition
183 "acl = name arg..."
a5f239e4 184
846726c5
JH
185JH/05 Permit multiple router/transport headers_add/remove lines.
186
3a796370
JH
187JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
188
ea722490 189JH/07 Avoid using a waiting database for a single-message-only transport.
8b260705
PP
190 Performance patch from Paul Fisher. Bugzilla 1262.
191
b1b05573
JH
192JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
193 Bugzilla 884.
194
362145b5
JH
195JH/09 Add $headers_added variable, with content from use of ACL modifier
196 add_header (but not yet added to the message). Bugzilla 199.
197
3c0a92dc
JH
198JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
199 Pulled from Bugzilla 817 by Wolfgang Breyha.
200
6d7c6175
PP
201PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
202 CVE-2012-5671
e78e6ecf 203 (nb: this is the same fix as in Exim 4.80.1)
6d7c6175 204
6f123593
JH
205JH/11 Add A= logging on delivery lines, and a client_set_id option on
206 authenticators.
207
c8e2fc1e
JH
208JH/12 Add optional authenticated_sender logging to A= and a log_selector
209 for control.
210
005ac57f
PP
211PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
212
3f1df0e3
PP
213PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
214 advertise SMTP AUTH mechanism to us, instead of a generic
215 protocol violation error. Also, make Exim more robust to bad
216 data from the Dovecot auth socket.
217
67bd1ab3
TF
218TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
219
220 When a queue runner is handling a message, Exim first routes the
221 recipient addresses, during which it prunes them based on the retry
222 hints database. After that it attempts to deliver the message to
223 any remaining recipients. It then updates the hints database using
224 the retry rules.
225
226 So if a recipient address works intermittently, it can get repeatedly
227 deferred at routing time. The retry hints record remains fresh so the
228 address never reaches the final cutoff time.
229
230 This is a fairly common occurrence when a user is bumping up against
231 their storage quota. Exim had some logic in its local delivery code
232 to deal with this. However it did not apply to per-recipient defers
233 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
234
1ddeb334
TF
235 This change adds a proper retry rule check during routing so that the
236 final cutoff time is checked against the message's age. We only do
237 this check if there is an address retry record and there is not a
238 domain retry record; this implies that previous attempts to handle
239 the address had the retry_use_local_parts option turned on. We use
240 this as an approximation for the destination being like a local
241 delivery, as in LMTP.
67bd1ab3
TF
242
243 I suspect this new check makes the old local delivery cutoff check
244 redundant, but I have not verified this so I left the code in place.
245
326cdc37
TF
246TF/02 Correct gecos expansion when From: is a prefix of the username.
247
248 Test 0254 submits a message to Exim with the header
249
250 Resent-From: f
251
252 When I ran the test suite under the user fanf2, Exim expanded
253 the header to contain my full name, whereas it should have added
254 a Resent-Sender: header. It erroneously treats any prefix of the
255 username as equal to the username.
256
257 This change corrects that bug.
258
f62514b3
GF
259GF/01 DCC debug and logging tidyup
260 Error conditions log to paniclog rather than rejectlog.
261 Debug lines prefixed by "DCC: " to remove any ambiguity.
262
eb505532
TF
263TF/03 Avoid unnecessary rebuilds of lookup-related code.
264
14c7b357
PP
265PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
266 Bug spotted by Jeremy Harris; was flawed since initial commit.
267 Would have resulted in OCSP responses post-SNI triggering an Exim
268 NULL dereference and crash.
269
94eaf700
PP
270JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
271
6f5a440a
PP
272PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
273 Bug detection, analysis and fix by Samuel Thibault.
274 Bugzilla 1331, Debian bug #698092.
275
514ee161
SC
276SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
277
fd98a5c6
JH
278JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
279 Server implementation by Todd Lyons, client by JH.
280 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
281 config variable "prdr_enable" controls whether the server
282 advertises the facility. If the client requests PRDR a new
283 acl_data_smtp_prdr ACL is called once for each recipient, after
284 the body content is received and before the acl_smtp_data ACL.
285 The client is controlled by bolth of: a hosts_try_prdr option
286 on the smtp transport, and the server advertisement.
287 Default client logging of deliveries and rejections involving
288 PRDR are flagged with the string "PRDR".
289
035c7f1e
PP
290PP/16 Fix problems caused by timeouts during quit ACLs trying to double
291 fclose(). Diagnosis by Todd Lyons.
292
ff284120
PP
293PP/17 Update configure.default to handle IPv6 localhost better.
294 Patch by Alain Williams (plus minor tweaks).
295 Bugzilla 880.
296
26e72755
PP
297PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
298 This is now consistent with GnuTLS, and is now documented: the
299 previous undocumented portable approach to treating the option as
300 unset was to force an expansion failure. That still works, and
301 an empty string is now equivalent.
302
0fbd9bff
PP
303PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
304 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
305 not performing validation itself.
306
700d22f3
PP
307PP/20 Added force_command boolean option to pipe transport.
308 Patch from Nick Koston, of cPanel Inc.
309
fcc8e047
JH
310JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
311 Bugzilla 321, 823.
312
7142daca
PP
313TF/04 Added udpsend ACL modifer and hexquote expansion operator
314
8c020188
PP
315PP/21 Fix eximon continuous updating with timestamped log-files.
316 Broken in a format-string cleanup in 4.80, missed when I repaired the
317 other false fix of the same issue.
318 Report and fix from Heiko Schlichting.
319 Bugzilla 1363.
320
d13cdd30
PP
321PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
322 Report from Prashanth Katuri.
323
e2fbf4a2
PP
324PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
325 It's SecureTransport, so affects any MacOS clients which use the
326 system-integrated TLS libraries, including email clients.
327
f4c1088b
PP
328PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
329 using a MIME ACL for non-SMTP local injection.
330 Report and assistance in diagnosis by Warren Baker.
331
c5c2182f
PP
332TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
333
73431ca9
JH
334JH/16 Fix comparisons for 64b. Bugzilla 1385.
335
2d07a215
TL
336TL/09 Add expansion variable $authenticated_fail_id to keep track of
337 last id that failed so it may be referenced in subsequent ACL's.
338
a30a8861
TL
339TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
340 Alexander Miroch.
341
33382dd9
TL
342TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
343 ldap library initialization, allowing self-signed CA's to be
344 used. Also properly sets require_cert option later in code by
345 using NULL (global ldap config) instead of ldap handle (per
346 session). Bug diagnosis and testing by alxgomz.
6d7c6175 347
046172e6
TL
348TL/12 Enhanced documentation in the ratelimit.pl script provided in
349 the src/util/ subdirectory.
350
1a7b746d
TL
351TL/13 Bug 1301 - Imported transport SQL logging patch from Axel Rau
352 renamed to Transport Post Delivery Action by Jeremy Harris, as
9bdd29ad
TL
353 EXPERIMENTAL_TPDA.
354
355TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
356 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
357 redis_servers = needs to be configured which will be used by the redis
358 lookup. Patch from Warren Baker, of The Packet Hub.
359
237b2cf2
TL
360TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
361
9fc5a352
TL
362TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
363 hostname or reverse DNS when processing a host list. Used suggestions
364 from multiple comments on this bug.
1a7b746d 365
b10e4ec2
TL
366TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
367
e2cebd74
TL
368TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
369 Missed a few lines, added it to make the runtest require no keyboard
370 interaction.
371
372TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
373 contains upper case chars. Make router use caseful_local_part.
374
2519e60d
TL
375TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
376 support when GnuTLS has been built with p11-kit.
377
e78e6ecf 378
4263f395
PP
379Exim version 4.80.1
380-------------------
381
382PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
383 CVE-2012-5671
2c422e6f 384 This, or similar/improved, will also be change PP/11 of 4.82.
3c0a92dc 385
ea722490 386
b1770b6e 387Exim version 4.80
0599f9cf
PP
388-----------------
389
390PP/01 Handle short writes when writing local log-files.
391 In practice, only affects FreeBSD (8 onwards).
392 Bugzilla 1053, with thanks to Dmitry Isaikin.
393
23c7e742
NM
394NM/01 Bugzilla 949 - Documentation tweak
395
b322aac8
NM
396NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
397 improved.
398
4a891427
NM
399NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
400
c1e794ba 401PP/02 Implemented gsasl authenticator.
b322aac8 402
97753960
PP
403PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
404
405PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
406 `pkg-config foo` for cflags/libs.
407
df6303fa
PP
408PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
409 with rest of GSASL and with heimdal_gssapi.
410
7e6a8985
PP
411PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
412 `pkg-config foo` for cflags/libs for the TLS implementation.
413
f1e05cc7 414PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
20aa9dbd
PP
415 properties get this fed in as external SSF. A number of robustness
416 and debugging improvements to the cyrus_sasl authenticator.
b322aac8 417
4c287009
PP
418PP/08 cyrus_sasl server now expands the server_realm option.
419
b98bb9ac
PP
420PP/09 Bugzilla 1214 - Log authentication information in reject log.
421 Patch by Jeremy Harris.
422
4a6a987a
PP
423PP/10 Added dbmjz lookup type.
424
c45dd180 425PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
c7955b11 426
7db8d074
PP
427PP/12 MAIL args handles TAB as well as SP, for better interop with
428 non-compliant senders.
429 Analysis and variant patch by Todd Lyons.
430
eae0036b 431NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
e0df1c83
DM
432 Bug report from Lars Müller <lars@samba.org> (via SUSE),
433 Patch from Dirk Mueller <dmueller@suse.com>
434
dec5017e
PP
435PP/13 tls_peerdn now print-escaped for spool files.
436 Observed some $tls_peerdn in wild which contained \n, which resulted
437 in spool file corruption.
438
c80c5570
PP
439PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
440 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
441 or write after TLS renegotiation, which otherwise led to messages
442 "Got SSL error 2".
443
076b11e2
PP
444TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
445 as a tracking header (ie: a signed header comes before the signature).
446 Patch from Wolfgang Breyha.
447
5407bfff
JH
448JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
449 comma-sep list; embedded commas doubled.
450
9e45c72b
PP
451JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
452
e74376d8
PP
453PP/15 LDAP: Check for errors of TLS initialisation, to give correct
454 diagnostics.
455 Report and patch from Dmitry Banschikov.
456
da3ad30d
PP
457PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
458 Removed SSL_clear() after SSL_new() which led to protocol negotiation
459 failures. We appear to now support TLS1.1+ with Exim.
460
7be682ca
PP
461PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
462 lets Exim select keys and certificates based upon TLS SNI from client.
3f0945ff
PP
463 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
464 before an outbound SMTP session. New log_selector, +tls_sni.
7be682ca 465
ef840681
PP
466PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
467 NULL dereference. Report and patch from Alun Jones.
468
5bfb4cdf
PP
469PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
470 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
471 Not seeing resolver debug output on NetBSD, but suspect this is a
472 resolver implementation change.
473
c6e95d22
PP
474PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
475 Left warnings. Added "eximon gdb" invocation mode.
476
9cbad13b
PP
477PP/21 Defaulting "accept_8bitmime" to true, not false.
478
9ee44efb
PP
479PP/22 Added -bw for inetd wait mode support.
480
6a6084f8
PP
481PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
482 locate the relevant includes and libraries. Made this the default.
483
12dd53c7
PP
484PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
485 Bugzilla 1246, report and most of solution from Tomasz Kusy.
486
9e45c72b 487JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
97d17305
JH
488 This may cause build issues on older platforms.
489
17c76198
PP
490PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
491 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
492 gnutls_require_mac & gnutls_require_protocols (no longer supported).
493 Added SNI support via GnuTLS too.
af3498d6 494 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
17c76198 495
53947857 496PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
3f7eeb86 497
eae0036b 498PP/27 Applied dnsdb SPF support patch from Janne Snabb.
8ee4b30e
PP
499 Applied second patch from Janne, implementing suggestion to default
500 multiple-strings-in-record handling to match SPF spec.
eae0036b 501
9e45c72b 502JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
2605c55b 503
7390e768
PP
504PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
505 read-only, out of scope).
506 Patch from Wolfgang Breyha, report from Stuart Northfield.
507
08488c86
PP
508PP/29 Fix three issues highlighted by clang analyser static analysis.
509 Only crash-plausible issue would require the Cambridge-specific
510 iplookup router and a misconfiguration.
511 Report from Marcin Mirosław.
512
6475bd82
PP
513PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
514
81f91683
PP
515PP/31 %D in printf continues to cause issues (-Wformat=security), so for
516 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
517 As part of this, removing so much warning spew let me fix some minor
518 real issues in debug logging.
519
5779e6aa
PP
520PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
521 assignment on my part. Fixed.
522
3375e053
PP
523PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
524 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
525 Janne Snabb (who went above and beyond: thank you).
526
527PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
528 string otherwise requires a connection and a bunch more work and it's
78e0c7a3
PP
529 relatively easy to get wrong. Should also expose TLS library linkage
530 problems.
3375e053 531
9d26b8c0
PP
532PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
533 64-bit ${eval} (JH/03).
534
57eb9e91 535PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
b87a6e0e
PP
536 GNU libc to support some of the 64-bit stuff, should not lead to
537 conflicts. Defined before os.h is pulled in, so if a given platform
538 needs to override this, it can.
539
16880d1a
PP
540PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
541 protection layer was required, which is not implemented.
542 Bugzilla 1254, patch from Wolfgang Breyha.
543
a799883d
PP
544PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
545 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
546 tls_dhparam take prime identifiers. Also unbreak combination of
547 OpenSSL+DH_params+TLSSNI.
548
3ecab157 549PP/39 Disable SSLv2 by default in OpenSSL support.
f0f5a555 550
0599f9cf 551
867fcbf5
PP
552Exim version 4.77
553-----------------
554
555PP/01 Solaris build fix for Oracle's LDAP libraries.
556 Bugzilla 1109, patch from Stephen Usher.
557
f1a29782
TF
558TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
559
ab42bd23
TK
560TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
561 whitespace trailer
867fcbf5 562
0ca0cf52
TF
563TF/02 Fix a couple more cases where we did not log the error message
564 when unlink() failed. See also change 4.74-TF/03.
565
921b12ca
TF
566TF/03 Make the exiwhat support code safe for signals. Previously Exim might
567 lock up or crash if it happened to be inside a call to libc when it
568 got a SIGUSR1 from exiwhat.
569
570 The SIGUSR1 handler appends the current process status to the process
571 log which is later printed by exiwhat. It used to use the general
572 purpose logging code to do this, but several functions it calls are
573 not safe for signals.
574
575 The new output code in the SIGUSR1 handler is specific to the process
576 log, and simple enough that it's easy to inspect for signal safety.
577 Removing some special cases also simplifies the general logging code.
578 Removing the spurious timestamps from the process log simplifies
579 exiwhat.
580
c99ce5c9
TF
581TF/04 Improved ratelimit ACL condition.
582
583 The /noupdate option has been deprecated in favour of /readonly which
584 has clearer semantics. The /leaky, /strict, and /readonly update modes
585 are mutually exclusive. The update mode is no longer included in the
586 database key; it just determines when the database is updated. (This
587 means that when you upgrde Exim will forget old rate measurements.)
588
589 Exim now checks that the per_* options are used with an update mode that
590 makes sense for the current ACL. For example, when Exim is processing a
591 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
592 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
593 must specify per_mail/readonly. If you omit the update mode it defaults to
594 /leaky where that makes sense (as before) or /readonly where required.
595
596 The /noupdate option is now undocumented but still supported for
597 backwards compatibility. It is equivalent to /readonly except that in
598 ACLs where /readonly is required you may specify /leaky/noupdate or
599 /strict/noupdate which are treated the same as /readonly.
600
601 A useful new feature is the /count= option. This is a generalization
602 of the per_byte option, so that you can measure the throughput of other
603 aggregate values. For example, the per_byte option is now equivalent
604 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
605
606 The per_rcpt option has been generalized using the /count= mechanism
607 (though it's more complicated than the per_byte equivalence). When it is
608 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
609 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
610 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
611 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
612 Note that using per_rcpt with a non-readonly update mode in more than
613 one ACL will cause the recipients to be double-counted. (The per_mail
614 and per_byte options don't have this problem.)
615
616 The handling of very low rates has changed slightly. If the computed rate
617 is less than the event's count (usually one) then this event is the first
618 after a long gap. In this case the rate is set to the same as this event's
619 count, so that the first message of a spam run is counted properly.
620
621 The major new feature is a mechanism for counting the rate of unique
622 events. The new per_addr option counts the number of different
623 recipients that someone has sent messages to in the last time period. It
624 behaves like per_rcpt if all the recipient addresses are different, but
625 duplicate recipient addresses do not increase the measured rate. Like
626 the /count= option this is a general mechanism, so the per_addr option
627 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
628 example, measure the rate that a client uses different sender addresses
629 with the options per_mail/unique=$sender_address. There are further
630 details in the main documentation.
631
3634fc25
TF
632TF/05 Removed obsolete $Cambridge$ CVS revision strings.
633
792e8a19
TF
634TF/06 Removed a few PCRE remnants.
635
5901f0ab
TF
636TF/07 Automatically extract Exim's version number from tags in the git
637 repository when doing development or release builds.
638
7f2a2a43
PP
639PP/02 Raise smtp_cmd_buffer_size to 16kB.
640 Bugzilla 879. Patch from Paul Fisher.
e2ca7082 641
061b7ebd
PP
642PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
643 Heavily based on revision 40f9a89a from Simon Arlott's tree.
644 Bugzilla 97.
645
e12f8c32
PP
646PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
647
9e949f00 648PP/05 Variable $av_failed, true if the AV scanner deferred.
7f2a2a43
PP
649 Bugzilla 1078. Patch from John Horne.
650
651PP/06 Stop make process more reliably on build failure.
652 Bugzilla 1087. Patch from Heiko Schlittermann.
9e949f00 653
555ae6af 654PP/07 Make maildir_use_size_file an _expandable_ boolean.
ac53fcda
PP
655 Bugzilla 1089. Patch from Heiko Schlittermann.
656
657PP/08 Handle ${run} returning more data than OS pipe buffer size.
658 Bugzilla 1131. Patch from Holger Weiß.
555ae6af 659
6f7fe114
PP
660PP/09 Handle IPv6 addresses with SPF.
661 Bugzilla 860. Patch from Wolfgang Breyha.
662
c566dd90
PP
663PP/10 GnuTLS: support TLS 1.2 & 1.1.
664 Bugzilla 1156.
89f897c3
PP
665 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
666 Bugzilla 1095.
c566dd90 667
d6cc7c78 668PP/11 match_* no longer expand right-hand-side by default.
39257585
PP
669 New compile-time build option, EXPAND_LISTMATCH_RHS.
670 New expansion conditions, "inlist", "inlisti".
671
0d0e4455
PP
672PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
673
3399bb60 674PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
d690cbdc
PP
675
676PP/14 fix log_write() format string regression from TF/03.
677 Bugzilla 1152. Patch from Dmitry Isaikin.
678
0ca0cf52 679
10906672
PP
680Exim version 4.76
681-----------------
682
683PP/01 The new ldap_require_cert option would segfault if used. Fixed.
684
754a0503
PP
685PP/02 Harmonised TLS library version reporting; only show if debugging.
686 Layout now matches that introduced for other libraries in 4.74 PP/03.
687
c0c7b2da
PP
688PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
689
e97d1f08
PP
690PP/04 New "dns_use_edns0" global option.
691
084c1d8c
PP
692PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
693 Bugzilla 1098.
694
4e7ee012
PP
695PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
696 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
da80c2a8 697
c8d52a00
PP
698TK/01 Updated PolarSSL code to 0.14.2.
699 Bugzilla 1097. Patch from Andreas Metzler.
700
54e7ce4a
PP
701PP/07 Catch divide-by-zero in ${eval:...}.
702 Fixes bugzilla 1102.
703
5ee6f336
PP
704PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
705 Bugzilla 1104.
706
c8d52a00 707TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
6ea4a851
PP
708 format-string attack -- SECURITY: remote arbitrary code execution.
709
710TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
711 time unintentionally subject to list matching rules, letting the header
712 cause arbitrary Exim lookups (of items which can occur in lists, *not*
713 arbitrary string expansion). This allowed for information disclosure.
714
715PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
716 INT_MIN/-1 -- value coerced to INT_MAX.
c8d52a00 717
10906672 718
aa097c4c
NM
719Exim version 4.75
720-----------------
721
4c9ef03a 722NM/01 Workround for PCRE version dependency in version reporting
aa097c4c
NM
723 Bugzilla 1073
724
7f3d9eff
TF
725TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
726 This fixes portability to compilers other than gcc, notably
727 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
728
159f52d2
TF
729TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
730 makefiles for portability to HP-UX and POSIX correctness.
731
0cc9542a
PP
732PP/01 Permit LOOKUP_foo enabling on the make command-line.
733 Also via indented variable definition in the Makefile.
734 (Debugging by Oliver Heesakkers).
735
f7274286
PP
736PP/02 Restore caching of spamd results with expanded spamd_address.
737 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
738
7b797365
PP
739PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
740 Improves build reliability. Fix from: Frank Elsner
741
caacae52
NM
742NM/02 Fix wide character breakage in the rfc2047 coding
743 Fixes bug 1064. Patch from Andrey N. Oktyabrski
744
09dcaba9
NM
745NM/03 Allow underscore in dnslist lookups
746 Fixes bug 1026. Patch from Graeme Fowler
747
bc19a55b
PP
748PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
749 Code patches from Adam Ciarcinski of NetBSD.
caacae52 750
bd4c9759
NM
751NM/04 Fixed exiqgrep to cope with mailq missing size issue
752 Fixes bug 943.
753
b72aab72
PP
754PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
755 is logged, to avoid truncation. Patch from John Horne.
756
2fe76745
PP
757PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
758 Patch from Jakob Hirsch.
759
76aa570c
PP
760PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
761 SQL string expansion failure details.
762 Patch from Andrey Oktyabrski.
763
f1e5fef5
PP
764PP/08 Bugzilla 486: implement %M datestamping in log filenames.
765 Patch from Simon Arlott.
766
4d805ee9
PP
767PP/09 New lookups functionality failed to compile on old gcc which rejects
768 extern declarations in function scope.
769 Patch from Oliver Fleischmann
770
cd59ab18
PP
771PP/10 Use sig_atomic_t for flags set from signal handlers.
772 Check getgroups() return and improve debugging.
773 Fixed developed for diagnosis in bug 927 (which turned out to be
774 a kernel bug).
775
332f5cf3
PP
776PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
777 Patch from Mark Zealey.
778
29cfeb94
PP
779PP/12 Bugzilla 1056: Improved spamd server selection.
780 Patch from Mark Zealey.
781
660242ad
PP
782PP/13 Bugzilla 1086: Deal with maildir quota file races.
783 Based on patch from Heiko Schlittermann.
784
bc4bc4c5
PP
785PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
786 Patch from Uwe Doering, sign-off by Michael Haardt.
787
2e64baa9
NM
788NM/05 Fix to spam.c to accommodate older gcc versions which dislike
789 variable declaration deep within a block. Bug and patch from
790 Dennis Davis.
791
bddd7526
PP
792PP/15 lookups-Makefile IRIX compatibilty coercion.
793
6bac1a9a
PP
794PP/16 Make DISABLE_DKIM build knob functional.
795
552193f0
NM
796NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
797 Patch by Simon Arlott
baeee2c1 798
1b587e48
TF
799TF/03 Fix valgrind.h portability to C89 compilers that do not support
800 variable argument macros. Our copy now differs from upstream.
801
aa097c4c 802
8c07b69f
TF
803Exim version 4.74
804-----------------
805
806TF/01 Failure to get a lock on a hints database can have serious
807 consequences so log it to the panic log.
808
c0ea85ab
TF
809TF/02 Log LMTP confirmation messages in the same way as SMTP,
810 controlled using the smtp_confirmation log selector.
811
0761d44e
TF
812TF/03 Include the error message when we fail to unlink a spool file.
813
0a349494
PP
814DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
815 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
816 for maintaining out-of-tree patches for some time.
817
818PP/01 Bugzilla 139: Documentation and portability issues.
819 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
820 Handle per-OS dynamic-module compilation flags.
821
fea24b2e
PP
822PP/02 Let /dev/null have normal permissions.
823 The 4.73 fixes were a little too stringent and complained about the
824 permissions on /dev/null. Exempt it from some checks.
825 Reported by Andreas M. Kirchwitz.
826
6545de78
PP
827PP/03 Report version information for many libraries, including
828 Exim version information for dynamically loaded libraries. Created
829 version.h, now support a version extension string for distributors
830 who patch heavily. Dynamic module ABI change.
831
1670ef10
PP
832PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
833 privilege escalation vulnerability whereby the Exim run-time user
834 can cause root to append content of the attacker's choosing to
835 arbitrary files.
836
c0886197
PP
837PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
838 (Wolfgang Breyha)
839
b7487bce
PP
840PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
841 If dropping privileges for untrusted macros, we disabled normal logging
842 on the basis that it would fail; for the Exim run-time user, this is not
843 the case, and it resulted in successful deliveries going unlogged.
844 Fixed. Reported by Andreas Metzler.
845
8c07b69f 846
97fd1e48 847Exim version 4.73
ed7f7860 848-----------------
97fd1e48
PP
849
850PP/01 Date: & Message-Id: revert to normally being appended to a message,
851 only prepend for the Resent-* case. Fixes regression introduced in
852 Exim 4.70 by NM/22 for Bugzilla 607.
853
6901c596
PP
854PP/02 Include check_rfc2047_length in configure.default because we're seeing
855 increasing numbers of administrators be bitten by this.
856
a8c8d6b5
JJ
857JJ/01 Added DISABLE_DKIM and comment to src/EDITME
858
77bb000f
PP
859PP/03 Bugzilla 994: added openssl_options main configuration option.
860
a29e5231
PP
861PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
862
ec5a0394 863PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
a29e5231 864
55c75993
PP
865PP/06 Adjust NTLM authentication to handle SASL Initial Response.
866
453a6645 867PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
ec5a0394
PP
868 without a peer certificate, leading to a segfault because of an
869 assumption that peers always have certificates. Be a little more
453a6645
PP
870 paranoid. Problem reported by Martin Tscholak.
871
8544e77a
PP
872PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
873 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
874 NB: ClamAV planning to remove STREAM in "middle of 2010".
3346ab01
PP
875 CL also introduces -bmalware, various -d+acl logging additions and
876 more caution in buffer sizes.
8544e77a 877
83e029d5
PP
878PP/09 Implemented reverse_ip expansion operator.
879
ed7f7860
PP
880PP/10 Bugzilla 937: provide a "debug" ACL control.
881
7d9f747b
PP
882PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
883
4b2241d2
PP
884PP/12 Bugzilla 973: Implement --version.
885
10385c15
PP
886PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
887
dbc4b90d
PP
888PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
889
532be449
PP
890PP/15 Bugzilla 816: support multiple condition rules on Routers.
891
6a8de854 892PP/16 Add bool_lax{} expansion operator and use that for combining multiple
71265ae9
PP
893 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
894 ignore trailing whitespace.
6a8de854 895
5dc43717
JJ
896JJ/02 prevent non-panic DKIM error from being sent to paniclog
897
898JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
899 "exim" to be used
55c75993 900
3346ab01
PP
901PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
902 Notification from Dr Andrew Aitchison.
903
491fab4c
PP
904PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
905 ExtendedDetectionInfo response format.
906 Notification from John Horne.
907
13eb9497
PP
908PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
909 compatible.
910
911PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
912 XSL and documented dependency on system catalogs, with examples of how
913 it normally works.
914
7f36d675
DW
915DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
916 access.
917
c1d94452
DW
918DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
919 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
920 configuration file which is writeable by the Exim user or group.
921
e2f5dc15
DW
922DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
923 of configuration files to cover files specified with the -C option if
924 they are going to be used with root privileges, not just the default
925 configuration file.
926
cd25e41d
DW
927DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
928 option (effectively making it always true).
929
261dc43e
DW
930DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
931 files to be used while preserving root privileges.
932
fa32850b
DW
933DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
934 that rogue child processes cannot use them.
935
79d4bc3d
PP
936PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
937 run-time user, instead of root.
938
43236f35 939PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2cfd3221
PP
940 Exim run-time user without dropping privileges.
941
fb08281f
DW
942DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
943 result string, instead of calling string_vformat() twice with the same
944 arguments.
3346ab01 945
74935b98
DW
946DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
947 for other users. Others should always drop root privileges if they use
948 -C on the command line, even for a whitelisted configure file.
949
90b6341f
DW
950DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
951
57730b52
ML
952NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
953
66581d1e 954
465e92cf
JJ
955Exim version 4.72
956-----------------
957
453a6645
PP
958JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
959 $data_path, and $header_path variables; fixed documentation bugs and
960 typos
465e92cf 961
453a6645
PP
962JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
963 exipick to access non-standard spools, including the "frozen" queue
964 (Finput)
edae0343 965
9bd3e22c
NM
966NM/01 Bugzilla 965: Support mysql stored procedures.
967 Patch from Alain Williams
968
bb576ff7
NM
969NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
970
5a1a5845
NM
971NM/03 Bugzilla 955: Documentation fix for max_rcpts.
972 Patch from Andreas Metzler
973
981a9fad
NM
974NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
975 Patch from Kirill Miazine
976
7fc497ee
NM
977NM/05 Bugzilla 671: Added umask to procmail example.
978
1a41defa
JJ
979JJ/03 installed exipick 20100323.0, fixing doc bug
980
a466095c 981NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
b26eacf1 982 directory. Notification and patch from Dan Rosenberg.
a466095c 983
94a6bd0b
NM
984TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
985
986TK/02 Improve log output when DKIM signing operation fails.
987
988MH/01 Treat the transport option dkim_domain as a colon separated
989 list, not as a single string, and sign the message with each element,
990 omitting multiple occurences of the same signer.
991
c1b141a8
NM
992NM/07 Null terminate DKIM strings, Null initialise DKIM variable
993 Bugzilla 985, 986. Patch by Simon Arlott
94a6bd0b 994
b26eacf1 995NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
0d0c6357
NM
996 Patch by Simon Arlott
997
179c5980 998PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
b26eacf1 999 MBX locking. Notification from Dan Rosenberg.
179c5980 1000
9bd3e22c 1001
7c6d71af
NM
1002Exim version 4.71
1003-----------------
1004
7d9f747b 1005TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
7c6d71af 1006
f013fb92
NM
1007NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1008
0eb8eedd
NM
1009NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1010
663ee6d9
NM
1011NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1012
177ebd9b
NM
1013NM/04 Bugzilla 915: Flag broken perl installation during build.
1014
7c6d71af 1015
210f147e
NM
1016Exim version 4.70
1017-----------------
1018
cdd3bb85 1019TK/01 Added patch by Johannes Berg that expands the main option
e739e3d9 1020 "spamd_address" if it starts with a dollar sign.
cdd3bb85
TK
1021
1022TK/02 Write list of recipients to X-Envelope-Sender header when building
1023 the mbox-format spool file for content scanning (suggested by Jakob
7d9f747b 1024 Hirsch).
cdd3bb85
TK
1025
1026TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1027 (http://www.dcc-servers.net/) support via dccifd. Activated by
e739e3d9 1028 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
cdd3bb85
TK
1029
1030TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1031 by Mark Daniel Reidel <mr@df.eu>.
1032
210f147e
NM
1033NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1034 When building exim an external PCRE library is now needed -
1035 PCRE is a system library on the majority of modern systems.
1036 See entry on PCRE_LIBS in EDITME file.
1037
deafd5b3
NM
1038NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1039 conversation. Added nologin parameter to request.
7d9f747b 1040 Patch contributed by Kirill Miazine.
deafd5b3 1041
089793a4
TF
1042TF/01 Do not log submission mode rewrites if they do not change the address.
1043
5f16ca82
TF
1044TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1045
dae9d94e 1046NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
7d9f747b 1047 log files in place. Contributed by Roberto Lima.
dae9d94e 1048
7d9f747b 1049NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
3f0da4d0 1050
06864c44
TF
1051TF/03 Bugzilla 615: When checking the local_parts router precondition
1052 after a local_part_suffix or local_part_prefix option, Exim now
1053 does not use the address's named list lookup cache, since this
1054 contains cached lookups for the whole local part.
1055
65a7d8c3 1056NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
7d9f747b 1057 Robert Millan. Documentation is in experimental-spec.txt.
65a7d8c3 1058
23510047 1059TF/04 Bugzilla 668: Fix parallel build (make -j).
65a7d8c3 1060
7d9f747b 1061NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
5f28a6e8 1062
7d8eec3a 1063NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
7d9f747b 1064 Patch provided by Jan Srzednicki.
6c588e74 1065
89dec7b6
TF
1066TF/05 Leading white space used to be stripped from $spam_report which
1067 wrecked the formatting. Now it is preserved.
5f28a6e8 1068
a99de90c
TF
1069TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1070 that they are available at delivery time.
1071
e2803e40
TF
1072TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1073
7199e1ee
TF
1074TF/08 TLS error reporting now respects the incoming_interface and
1075 incoming_port log selectors.
1076
e276e04b
TF
1077TF/09 Produce a more useful error message if an SMTP transport's hosts
1078 setting expands to an empty string.
1079
ce552449 1080NM/06 Bugzilla 744: EXPN did not work under TLS.
7d9f747b 1081 Patch provided by Phil Pennock.
ce552449 1082
e765a0f1 1083NM/07 Bugzilla 769: Extraneous comma in usage fprintf
7d9f747b 1084 Patch provided by Richard Godbee.
e765a0f1 1085
4f054c63 1086NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
447de4b0 1087 acl_smtp_notquit, added index entry.
4f054c63 1088
7d9f747b
PP
1089NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1090 Patch provided by Eugene Bujak.
24c929a2 1091
7d9f747b
PP
1092NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1093 accept(). Patch provided by Maxim Dounin.
cf73943b 1094
b52bc06e 1095NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
7d9f747b 1096 Patch provided by Phil Pennock.
b52bc06e 1097
447de4b0
NM
1098NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1099
4c69d561 1100NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
7d9f747b 1101 Patch provided by Brad "anomie" Jorsch.
4c69d561 1102
d5c39246 1103NM/14 Bugzilla 622: Added timeout setting to transport filter.
7d9f747b 1104 Patch provided by Dean Brooks.
9b989985 1105
0b23848a
TK
1106TK/05 Add native DKIM support (does not depend on external libraries).
1107
8f3414a1 1108NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
7d9f747b 1109 Patch provided by Graeme Fowler.
e2aacdfd 1110
fb6f955d
NM
1111NM/16 Bugzilla 851: Documentation example syntax fix.
1112
1113NM/17 Changed NOTICE file to remove references to embedded PCRE.
8f3414a1 1114
7d9f747b
PP
1115NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1116 lsearch.
dbb0bf41 1117
7d9f747b
PP
1118NM/19 Bugzilla 745: TLS version reporting.
1119 Patch provided by Phil Pennock.
f3766eb5 1120
7d9f747b
PP
1121NM/20 Bugzilla 167: bool: condition support.
1122 Patch provided by Phil Pennock.
36f12725 1123
7d9f747b
PP
1124NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1125 clients. Patch provided by Phil Pennock.
e6060e2c 1126
7d9f747b
PP
1127NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1128 Patch provided by Brad "anomie" Jorsch.
5eb690a1 1129
7d9f747b
PP
1130NM/23 Bugzilla 687: Fix misparses in eximstats.
1131 Patch provided by Heiko Schlittermann.
d5c13d66 1132
7d9f747b
PP
1133NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1134 Patch provided by Heiko Schlittermann.
b2335c0b 1135
7d9f747b 1136NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1da77999 1137 plus update to original patch.
f4cd9433 1138
7d9f747b 1139NM/26 Bugzilla 799: Documentation correction for ratelimit.
dc988b7e 1140
7d9f747b
PP
1141NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1142 Patch provided by David Brownlee.
8dc71ab3 1143
7d9f747b 1144NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
400eda43 1145
7d9f747b 1146NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
ec5a421b 1147
7d9f747b 1148NM/30 Bugzilla 888: TLS documentation bugfixes.
07af267e 1149
7d9f747b 1150NM/31 Bugzilla 896: Dovecot buffer overrun fix.
51473862 1151
17792b53 1152NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
7d9f747b 1153 Unlike the original bugzilla I have changed all shell scripts in src tree.
17792b53 1154
7d9f747b
PP
1155NM/33 Bugzilla 898: Transport filter timeout fix.
1156 Patch by Todd Rinaldo.
52383f8f 1157
91576cec 1158NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
7d9f747b 1159 Patch by Serge Demonchaux.
5ca6d115 1160
7d9f747b
PP
1161NM/35 Bugzilla 39: Base64 decode bug fixes.
1162 Patch by Jakob Hirsch.
baee9eee 1163
7d9f747b 1164NM/36 Bugzilla 909: Correct connect() call in dcc code.
e93a964c 1165
7d9f747b 1166NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
9bf3d68f 1167
7d9f747b 1168NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
96535b98 1169
7d9f747b 1170NM/39 Bugzilla 911: Fixed MakeLinks build script.
30339e0f 1171
deafd5b3 1172
47db1125
NM
1173Exim version 4.69
1174-----------------
1175
4b3504d0
TK
1176TK/01 Add preliminary DKIM support. Currently requires a forked version of
1177 ALT-N's libdkim that I have put here:
1178 http://duncanthrax.net/exim-experimental/
1179
1180 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1181 were called 'true' and it seems that C99 defines that as a reserved
1182 keyword to be used with 'bool' variable types. That means you could
1183 not include C99-style headers which use bools without triggering
1184 build errors in sieve.c.
1185
81ea09ca
NM
1186NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1187 as mailq or other aliases. Changed the --help handling significantly
1188 to do whats expected. exim_usage() emits usage/help information.
1189
f13cddcb
SC
1190SC/01 Added the -bylocaldomain option to eximstats.
1191
7d9f747b 1192NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
8ad076b2 1193
7d9f747b 1194NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
a843aaa6 1195
7d9f747b 1196NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
47db1125
NM
1197
1198
eb4c0de6
PH
1199Exim version 4.68
1200-----------------
1201
1202PH/01 Another patch from the Sieve maintainer.
1203
6a3bceb1
PH
1204PH/02 When an IPv6 address is converted to a string for single-key lookup
1205 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1206 dots are used instead of colons so that keys in lsearch files need not
1207 contain colons. This was done some time before quoting was made available
1208 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1209 (notated using the quote facility) so as to distinguish them from IPv4
1210 keys. This meant that lookups for IP addresses in host lists did not work
1211 for iplsearch lookups.
1212
1213 This has been fixed by arranging for IPv6 addresses to be expressed with
1214 colons if the lookup type is iplsearch. This is not incompatible, because
1215 previously such lookups could never work.
1216
1217 The situation is now rather anomolous, since one *can* have colons in
1218 ordinary lsearch keys. However, making the change in all cases is
1219 incompatible and would probably break a number of configurations.
1220
2e30fa9d
TK
1221TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1222 version.
1223
0806a9c5
MH
1224MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1225 conversion specification without a maximum field width, thereby enabling
1226 a rogue spamd server to cause a buffer overflow. While nobody in their
1227 right mind would setup Exim to query an untrusted spamd server, an
1228 attacker that gains access to a server running spamd could potentially
1229 exploit this vulnerability to run arbitrary code as the Exim user.
1230
ae276964
TK
1231TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1232 $primary_hostname instead of what libspf2 thinks the hosts name is.
1233
0f2cbd1b
MH
1234MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1235 a directory entry by the name of the lookup key. Previously, if a
1236 symlink pointed to a non-existing file or a file in a directory that
1237 Exim lacked permissions to read, a lookup for a key matching that
1238 symlink would fail. Now it is enough that a matching directory entry
1239 exists, symlink or not. (Bugzilla 503.)
1240
2b85bce7
PH
1241PH/03 The body_linecount and body_zerocount variables are now exported in the
1242 local_scan API.
1243
93655c46
PH
1244PH/04 Added the $dnslist_matched variable.
1245
6c512171
PH
1246PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1247 This means they are set thereafter only if the connection becomes
1248 encrypted.
1249
1250PH/06 Added the client_condition to authenticators so that some can be skipped
1251 by clients under certain conditions.
1252
aa6dc513
PH
1253PH/07 The error message for a badly-placed control=no_multiline_responses left
1254 "_responses" off the end of the name.
1255
a96603a0
PH
1256PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1257
8f240103
PH
1258PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1259 (without spaces) instead of just copying the configuration text.
1260
1261PH/10 Added the /noupdate option to the ratelimit ACL condition.
1262
d677b2f2
PH
1263PH/11 Added $max_received_linelength.
1264
d52120f2
PH
1265PH/12 Added +ignore_defer and +include_defer to host lists.
1266
64f2600a
PH
1267PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1268 way in which PCRE > 7.0 is built.
1269
8669f003
PH
1270PH/14 Implemented queue_only_load_latch.
1271
a4dc33a8
PH
1272PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1273 MAIL command. The effect was to mangle the value on 64-bit systems.
1274
d6a60c0f
PH
1275PH/16 Another patch from the Sieve maintainer.
1276
8f128379
PH
1277PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1278
8932dffe
PH
1279PH/18 If a system quota error occurred while trying to create the file for
1280 a maildir delivery, the message "Mailbox is full" was not appended to the
1281 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1282 only to a quota excession during the actual writing of the file.
d6a60c0f 1283
ddea74fa 1284PH/19 It seems that peer DN values may contain newlines (and other non-printing
48ed62d9
PH
1285 characters?) which causes problems in log lines. The DN values are now
1286 passed through string_printing() before being added to log lines.
1287
ddea74fa 1288PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
b7670459
PH
1289 and InterBase are left for another time.)
1290
ddea74fa
PH
1291PH/21 Added message_body_newlines option.
1292
ce9f225c
PH
1293PH/22 Guard against possible overflow in moan_check_errorcopy().
1294
19897d52
PH
1295PH/23 POSIX allows open() to be a macro; guard against that.
1296
bc64a74d
PH
1297PH/24 If the recipient of an error message contained an @ in the local part
1298 (suitably quoted, of course), incorrect values were put in $domain and
1299 $local_part during the evaluation of errors_copy.
1300
eb4c0de6 1301
b4ed4da0
PH
1302Exim version 4.67
1303-----------------
1304
22ad45c9
MH
1305MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1306 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1307 Jan Srzednicki.
1308
b4ed4da0
PH
1309PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1310 issue a MAIL command.
1311
431b7361
PH
1312PH/02 In an ACL statement such as
1313
1314 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1315
1316 if a client was not listed at all, or was listed with a value other than
1317 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1318 the condition was not true (as it should be), so access was not denied.
1319 The bug was that the ! inversion was incorrectly passed on to the second
1320 item. This has been fixed.
1321
1322PH/03 Added additional dnslists conditions == and =& which are different from
1323 = and & when the dns lookup returns more than one IP address.
1324
83da1223
PH
1325PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1326 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1327
54fc8428
PH
1328PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1329 FSYNC, which compiles an option called disable_fsync that allows for
1330 bypassing fsync(). The documentation is heavily laced with warnings.
1331
34c5e8dd
SC
1332SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1333
bbe15da8
PH
1334PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1335 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1336 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1337 including adding "make clean"; (3) Added -fPIC when compiling the test
1338 dynamically loaded module, to get rid of a warning.
1339
0e8a9471
MH
1340MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1341 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1342 = 0s. The bug is otherwise harmless.
1343
f0872424
PH
1344PH/07 There was a bug in the dovecot authenticator such that the value of
1345 $auth1 could be overwritten, and so not correctly preserved, after a
1346 successful authentication. This usually meant that the value preserved by
1347 the server_setid option was incorrect.
1348
b01dd148
PH
1349PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1350
6bf342e1
PH
1351PH/09 Installed PCRE release 7.0.
1352
273f34d0
PH
1353PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1354 run for batched SMTP input. It is now run at the start of every message
1355 in the batch. While fixing this I discovered that the process information
1356 (output by running exiwhat) was not always getting set for -bs and -bS
1357 input. This is fixed, and it now also says "batched" for BSMTP.
1358
cf8b11a5
PH
1359PH/11 Added control=no_pipelining.
1360
41c7c167
PH
1361PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1362 patch, slightly modified), and move the expansion of helo_data till after
1363 the connection is made in the smtp transport (so it can use these
1364 values).
1365
9c57cbc0
PH
1366PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1367
f3f065bb
PH
1368PH/14 Added log_selector = +pid.
1369
047bdd8c
PH
1370PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1371
0ce9abe6
PH
1372PH/16 Add ${if forany and ${if forall.
1373
0e22dfd1
PH
1374PH/17 Added dsn_from option to vary the From: line in DSNs.
1375
4c590bd1
PH
1376PH/18 Flush SMTP output before performing a callout, unless control =
1377 no_callout_flush is set.
1378
09945f1e
PH
1379PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1380 was true (the default) a successful delivery failed to delete the retry
1381 item, thus causing premature timeout of the address. The bug is now
1382 fixed.
1383
c51b8e75
PH
1384PH/20 Added hosts_avoid_pipelining to the smtp transport.
1385
e28326d8
PH
1386PH/21 Long custom messages for fakedefer and fakereject are now split up
1387 into multiline reponses in the same way that messages for "deny" and
1388 other ACL rejections are.
1389
75b1493f
PH
1390PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1391 with slight modification.
1392
7c5214ec
PH
1393PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1394 draft, changing the syntax and factoring some duplicate code".
1395
4311097e
PH
1396PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1397 for deliveries of the second and subsequent messages over the same SMTP
1398 connection.
1399
29f89cad
PH
1400PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1401 ${reduce, with only minor "tidies".
1402
5e687460
SC
1403SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1404
c3611384
PH
1405PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1406 expansion side effects.
1407
5a11a7b4
PH
1408PH/27 When a message times out after an over-quota error from an Exim-imposed
1409 quota, the bounce message says "mailbox is full". This message was not
1410 being given when it was a system quota that was exceeded. It now should
1411 be the same.
1412
0e20aff9
MH
1413MH/03 Made $recipients available in local_scan(). local_scan() already has
1414 better access to the recipient list through recipients_list[], but
1415 $recipients can be useful in postmaster-provided expansion strings.
1416
ca86f471
PH
1417PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1418 in the case of a MAIL command with additional options following the
1419 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1420 were accidentally chopped off.
1421
a14e5636
PH
1422PH/29 SMTP synchronization checks are implemented when a command is read -
1423 there is a check that no more input is waiting when there shouldn't be
1424 any. However, for some commands, a delay in an ACL can mean that it is
1425 some time before the response is written. In this time, more input might
1426 arrive, invalidly. So now there are extra checks after an ACL has run for
1427 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1428 pipelining has not been advertised.
1429
ec95d1a6
PH
1430PH/30 MH's patch to allow iscntrl() characters to be list separators.
1431
42855d71
PH
1432PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1433 returned in the SMTP response when smtp_return_error_details was false.
1434 This has been fixed.
1435
57c2c631
PH
1436PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1437 instead of the C I/O that was originally supplied, because problems were
1438 reported on Solaris.
1439
58c01c94
PH
1440PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1441 Exim which did not show up earlier: it was assuming that a call to
1442 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1443 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1444 was a macro that became an assignment, so it seemed to work. This has
1445 changed to a proper function call with a void return, hence the compile
1446 error. Exim's code has been fixed.
1447
dee5a20a
PH
1448PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1449 cpus.
1450
d2ee6114
PH
1451PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1452
b2d5182b
PH
1453PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1454
79749a79
PH
1455PH/37 If a message is not accepted after it has had an id assigned (e.g.
1456 because it turns out to be too big or there is a timeout) there is no
3ce62588
PH
1457 "Completed" line in the log. When some messages of this type were
1458 selected by exigrep, they were listed as "not completed". Others were
1459 picked up by some special patterns. I have improved the selection
1460 criteria to be more general.
79749a79 1461
c456d9bb
PH
1462PH/38 The host_find_failed option in the manualroute router can now be set
1463 to "ignore", to completely ignore a host whose IP address cannot be
1464 found. If all hosts are ignored, the behaviour is controlled by the new
1465 host_all_ignored option.
1466
cd9868ec
PH
1467PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1468 homing or because of multiple MX records with /mx) generated more than
1469 one IP address, and the following item turned out to be the local host,
1470 all the secondary addresses of the first item were incorrectly removed
1471 from the list, along with the local host and any following hosts (which
1472 is what is supposed to happen).
1473
ebeaf996
PH
1474PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1475 whoever called Exim into the -H file. In the case of the daemon it was
1476 behaving confusingly. When first started, it used values for whoever
1477 started the daemon, but after a SIGHUP it used the Exim user (because it
1478 calls itself on a restart). I have changed the code so that it now always
1479 uses the Exim user.
1480
2679d413
PH
1481PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1482 message are rejected with the same error (e.g. no authentication or bad
1483 sender address), and a DATA command is nevertheless sent (as can happen
1484 with PIPELINING or a stupid MUA), the error message that was given to the
1485 RCPT commands is included in the rejection of the DATA command. This is
1486 intended to be helpful for MUAs that show only the final error to their
1487 users.
1488
84024b72
PH
1489PH/42 Another patch from the Sieve maintainer.
1490
8005d38e
SC
1491SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1492 Eximstats - Fixed some broken HTML links and added missing column headers
1493 (Jez Hancock).
1494 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1495 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1496
3298c6c6
SC
1497SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1498
a43a27c5
PH
1499PH/43 Yet another patch from the Sieve maintainer.
1500
58eb016e 1501PH/44 I found a way to check for a TCP/IP connection going away before sending
563b63fa
PH
1502 the response to the final '.' that terminates a message, but only in the
1503 case where the client has not sent further data following the '.'
1504 (unfortunately, this is allowed). However, in many cases there won't be
1505 any further data because there won't be any more messages to send. A call
1506 to select() can be used: if it shows that the input is "ready", there is
1507 either input waiting, or the socket has been closed. An attempt to read
1508 the next input character can distinguish the two cases. Previously, Exim
58eb016e 1509 would have sent an OK response which the client would never have see.
563b63fa
PH
1510 This could lead to message repetition. This fix should cure that, at
1511 least in a lot of common cases.
58eb016e 1512
b43a74ea
PH
1513PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1514 advertised in response to EHLO.
1515
b4ed4da0 1516
5dd1517f
PH
1517Exim version 4.66
1518-----------------
1519
1520PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1521 fixed by 4.65/MH/01 (is this a record?) are fixed:
1522
1523 (i) An empty string was always treated as zero by the numeric comparison
1524 operators. This behaviour has been restored.
1525
1526 (ii) It is documented that the numeric comparison operators always treat
1527 their arguments as decimal numbers. This was broken in that numbers
1528 starting with 0 were being interpreted as octal.
1529
1530 While fixing these problems I realized that there was another issue that
1531 hadn't been noticed. Values of message_size_limit (both the global option
1532 and the transport option) were treated as octal if they started with 0.
1533 The documentation was vague. These values are now always treated as
1534 decimal, and I will make that clear in the documentation.
1535
1536
93cfa765
TK
1537Exim version 4.65
1538-----------------
1539
1540TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1541 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1542 versions. (#438)
1543
d6066548
MH
1544MH/01 Don't check that the operands of numeric comparison operators are
1545 integers when their expansion is in "skipping" mode (fixes bug
1546 introduced by 4.64-PH/07).
1547
4362ff0d
PH
1548PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1549 child addresses, Exim now panics and dies. Previously, because the count
1550 is held in a short int, deliveries were likely to be lost. As such a
1551 large number of recipients for a single message is ridiculous
1552 (performance will be very, very poor), I have chosen to impose a limit
1553 rather than extend the field.
1554
93cfa765 1555
944e9e9c
TF
1556Exim version 4.64
1557-----------------
aa41d2de 1558
21d74bd9
TK
1559TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1560 leftover -K file (the existence of which was triggered by #402).
1561 While we were at it, introduced process PID as part of the -K
1562 filename. This should rule out race conditions when creating
1563 these files.
1564
1565TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1566 processing considerably. Previous code took too long for large mails,
1567 triggering a timeout which in turn triggers #401.
1568
1569TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1570 in the DK code in transports.c. sendfile() is not really portable,
1571 hence the _LINUX specificness.
944e9e9c
TF
1572
1573TF/01 In the add_headers option to the mail command in an Exim filter,
1574 there was a bug that Exim would claim a syntax error in any
1575 header after the first one which had an odd number of characters
1576 in the field name.
1577
2b1c6e3a
PH
1578PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1579 callout verification, Exim cached a "reject" for the entire domain. This
1580 is correct for most verifications, but it is not correct for a recipient
1581 verification with use_sender or use_postmaster set, because in that case
1582 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1583 case of MAIL FROM:<> rejection from other early rejections (e.g.
1584 rejection of HELO). When verifying a recipient using a non-null MAIL
1585 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1586 Whatever the result of the callout, the value of the domain cache is
1587 left unchanged (for any other kind of callout, getting as far as trying
1588 RCPT means that the domain itself is ok).
1589
1f872c80
PH
1590PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1591 gcc 4.1.1 threw up.
1592
1593PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1594 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1595 session, the daemon ignores ECONNECT errors and logs others; it now
1596 ignores EPIPE as well.
1597
d203e649
PH
1598PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1599 (quoted-printable decoding).
1600
cc2ed8f7 1601PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
21a04aa3 1602 later the small subsequent patch to fix an introduced bug.
f951fd57 1603
ddfcd446
PH
1604PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1605
d45b1de8
PH
1606PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1607
1608PH/08 An error is now given if message_size_limit is specified negative.
1609
38a0a95f 1610PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
641cb756 1611 to be given (somewhat) arbitrary names.
38a0a95f 1612
a2405d83
JJ
1613JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1614 in 4.64-PH/09.
1615
1616JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1617 miscellaneous code fixes
1618
6ea85e9a
PH
1619PH/10 Added the log_reject_target ACL modifier to specify where to log
1620 rejections.
1621
26da7e20
PH
1622PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1623 hostname. This is wrong, because it relates to the incoming message (and
1624 probably the interface on which it is arriving) and not to the outgoing
1625 callout (which could be using a different interface). This has been
1626 changed to use the value of the helo_data option from the smtp transport
1627 instead - this is what is used when a message is actually being sent. If
1628 there is no remote transport (possible with a router that sets up host
1629 addresses), $smtp_active_hostname is used.
6ea85e9a 1630
14aa5a05 1631PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
7befa435 1632 tweaks were necessary in order to get it to work (see also 21 below):
14aa5a05
PH
1633 (a) The code assumed that strncpy() returns a negative number on buffer
1634 overflow, which isn't the case. Replaced with Exim's string_format()
1635 function.
1636 (b) There were several signed/unsigned issues. I just did the minimum
1637 hacking in of casts. There is scope for a larger refactoring.
1638 (c) The code used strcasecmp() which is not a standard C function.
1639 Replaced with Exim's strcmpic() function.
1640 (d) The code set only $1; it now sets $auth1 as well.
1641 (e) A simple test gave the error "authentication client didn't specify
1642 service in request". It would seem that Dovecot has changed its
1643 interface. Fortunately there's a specification; I followed it and
1644 changed what the client sends and it appears to be working now.
1645
ff75a1f7
PH
1646PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1647 decoding.
1648
e6f6568e
PH
1649PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1650 address A is aliased to B and C, where B exists and C does not. Without
1651 -v the output is "A verified" because verification stops after a
1652 successful redirection if more than one address is generated. However,
1653 with -v the child addresses are also verified. Exim was outputting "A
1654 failed to verify" and then showing the successful verification for C,
1655 with its parentage. It now outputs "B failed to verify", showing B's
1656 parentage before showing the successful verification of C.
1657
d6f6e0dc
PH
1658PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1659 look up a TXT record in a specific list after matching in a combined
1660 list.
1661
322050c2
PH
1662PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1663 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1664 they consult the DNS. I had assumed they would set it the way they
1665 wanted; and indeed my experiments on Linux seem to show that in some
1666 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1667 To be on the safe side, however, I have now made the interface to
1668 host_find_byname() similar to host_find_bydns(), with an argument
1669 containing the DNS resolver options. The host_find_byname() function now
1670 sets these options at its start, just as host_find_bydns() does. The smtp
1671 transport options dns_qualify_single and dns_search_parents are passed to
1672 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
1673 of host_find_byname() use the default settings of RES_DEFNAMES
1674 (qualify_single) but not RES_DNSRCH (search_parents).
1675
08955dd3
PH
1676PH/17 Applied (a modified version of) Nico Erfurth's patch to make
1677 spool_read_header() do less string testing, by means of a preliminary
1678 switch on the second character of optional "-foo" lines. (This is
1679 overdue, caused by the large number of possibilities that now exist.
1680 Originally there were few.) While I was there, I also converted the
1681 str(n)cmp tests so they don't re-test the leading "-" and the first
1682 character, in the hope this might squeeze out yet more improvement.
1683
1eccaa59
PH
1684PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
1685 flag allowing group syntax was set by the header_syntax check but not
1686 turned off, possible causing trouble later; (2) The flag was not being
1687 set at all for the header_verify test, causing "group"-style headers to
1688 be rejected. I have now set it in this case, and also caused header_
1689 verify to ignore an empty address taken from a group. While doing this, I
1690 came across some other cases where the code for allowing group syntax
1691 while scanning a header line wasn't quite right (mostly, not resetting
1692 the flag correctly in the right place). These bugs could have caused
1693 trouble for malformed header lines. I hope it is now all correct.
1694
602e59e5
PH
1695PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
1696 with the "reply" argument non-NULL. The code, however (which originally
1697 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
1698 but it didn't always do it. This confused somebody who was copying the
1699 code for some other use. I have removed all the tests.
1700
411ef850
PH
1701PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
1702 feature that was used to support insecure browsers during the U.S. crypto
1703 embargo. It requires special client support, and Exim is probably the
1704 only MTA that supported it -- and would never use it because real RSA is
1705 always available. This code has been removed, because it had the bad
1706 effect of slowing Exim down by computing (never used) parameters for the
1707 RSA_EXPORT functionality.
1708
7befa435
PH
1709PH/21 On the advice of Timo Sirainen, added a check to the dovecot
1710 authenticator to fail if there's a tab character in the incoming data
1711 (there should never be unless someone is messing about, as it's supposed
1712 to be base64-encoded). Also added, on Timo's advice, the "secured" option
1713 if the connection is using TLS or if the remote IP is the same as the
1714 local IP, and the "valid-client-cert option" if a client certificate has
1715 been verified.
1716
48da4259 1717PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
16ff981e
PH
1718 authenticators. This can be used for authorization after authentication
1719 succeeds. (In the case of plaintext, it servers for both authentication
1720 and authorization.)
1721
48da4259
PH
1722PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1723 if any retry times were supplied.
1724
d1d5595c
PH
1725PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1726 connection, where there is no client IP address to check. In this
1727 situation, the verify now always succeeds.
1728
0ef732d9
PH
1729PH/25 Applied John Jetmore's -Mset patch.
1730
328895cc
PH
1731PH/26 Added -bem to be like -Mset, but loading a message from a file.
1732
fd700877
PH
1733PH/27 In a string expansion for a processed (not raw) header when multiple
1734 headers of the same name were present, leading whitespace was being
1735 removed from all of them, but trailing whitespace was being removed only
1736 from the last one. Now trailing whitespace is removed from each header
f6c332bd
PH
1737 before concatenation. Completely empty headers in a concatenation (as
1738 before) are ignored.
fd700877 1739
8dce1a6f
PH
1740PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1741 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1742
17af4a17
PH
1743PH/29 [Removed. This was a change that I later backed out, and forgot to
1744 correct the ChangeLog entry (that I had efficiently created) before
1745 committing the later change.]
f6c332bd
PH
1746
1747PH/30 Exim was sometimes attempting to deliver messages that had suffered
1748 address errors (4xx response to RCPT) over the same connection as other
1749 messages routed to the same hosts. Such deliveries are always "forced",
1750 so retry times are not inspected. This resulted in far too many retries
1751 for the affected addresses. The effect occurred only when there were more
1752 hosts than the hosts_max_try setting in the smtp transport when it had
1753 the 4xx errors. Those hosts that it had tried were not added to the list
1754 of hosts for which the message was waiting, so if all were tried, there
1755 was no problem. Two fixes have been applied:
1756
1757 (i) If there are any address or message errors in an SMTP delivery, none
1758 of the hosts (tried or untried) are now added to the list of hosts
1759 for which the message is waiting, so the message should not be a
1760 candidate for sending over the same connection that was used for a
1761 successful delivery of some other message. This seems entirely
1762 reasonable: after all the message is NOT "waiting for some host".
1763 This is so "obvious" that I'm not sure why it wasn't done
1764 previously. Hope I haven't missed anything, but it can't do any
1765 harm, as the worst effect is to miss an optimization.
1766
1767 (ii) If, despite (i), such a delivery is accidentally attempted, the
1768 routing retry time is respected, so at least it doesn't keep
1769 hammering the server.
1770
c1114884
PH
1771PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1772 in ${readsocket because some servers need this prod.
1773
7a0743eb
PH
1774PH/32 Added some extra debug output when updating a wait-xxx database.
1775
0d85fa3f
PH
1776PH/33 The hint "could be header name not terminated by colon", which has been
1777 given for certain expansion errors for a long time, was not being given
1778 for the ${if def:h_colon_omitted{... case.
1779
1bf43b78
PH
1780PH/34 The spec says: "With one important exception, whenever a domain list is
1781 being scanned, $domain contains the subject domain." There was at least
1782 one case where this was not true.
1783
520de300
PH
1784PH/35 The error "getsockname() failed: connection reset by peer" was being
1785 written to the panic log as well as the main log, but it isn't really
1786 panic-worthy as it just means the connection died rather early on. I have
1787 removed the panic log writing for the ECONNRESET error when getsockname()
1788 fails.
1789
48c7f9e2
PH
1790PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1791 runs only) independently of the message's sender address. This meant
1792 that, if the 4xx error was in fact related to the sender, a different
1793 message to the same recipient with a different sender could confuse
1794 things. In particualar, this can happen when sending to a greylisting
1795 server, but other circumstances could also provoke similar problems.
1796 I have changed the default so that the retry time for these errors is now
1797 based a combination of the sender and recipient addresses. This change
1798 can be overridden by setting address_retry_include_sender=false in the
1799 smtp transport.
1800
99ea1c86
PH
1801PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1802 remote server are returned as part of bounce messages. This was not
1803 happening for LMTP over a pipe (the lmtp transport), but now it is the
1804 same for both kinds of LMTP.
1805
a2042e78
PH
1806PH/38 Despite being documented as not happening, Exim was rewriting addresses
1807 in header lines that were in fact CNAMEs. This is no longer the case.
1808
4fbcfc2e
PH
1809PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1810 and queue runs started by the daemon processed all messages. This has
1811 been fixed so that -R and -S can now usefully be given with -q<time>.
1812
aa41d2de
PH
1813PH/40 Import PCRE release 6.7 (fixes some bugs).
1814
af561417
PH
1815PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1816
3cc66b45
PH
1817PH/42 Give an error if -q is specified more than once.
1818
194cc0e4
PH
1819PH/43 Renamed the variables $interface_address and $interface_port as
1820 $received_ip_address and $received_port, to make it clear that these
1821 values apply to message reception, and not to the outgoing interface when
1822 a message is delivered. (The old names remain recognized, of course.)
1823
a401ddaa
PH
1824PH/44 There was no timeout on the connect() call when using a Unix domain
1825 socket in the ${readsocket expansion. There now is.
1826
4e88a19f
PH
1827PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1828 be meaningful with "accept".
1829
d7d7b289
SC
1830SC/01 Eximstats V1.43
1831 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1832
1833SC/02 Eximstats V1.44
1834 Use a glob alias rather than an array ref in the generated
1835 parser. This improves both readability and performance.
1836
1837SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1838 Collect SpamAssassin and rejection statistics.
1839 Don't display local sender or destination tables unless
1840 there is data to show.
1841 Added average volumes into the top table text output.
1842
1843SC/04 Eximstats V1.46
1844 Collect data on the number of addresses (recipients)
1845 as well as the number of messages.
1846
1847SC/05 Eximstats V1.47
1848 Added 'Message too big' to the list of mail rejection
1849 reasons (thanks to Marco Gaiarin).
1850
1851SC/06 Eximstats V1.48
1852 Mainlog lines which have GMT offsets and are too short to
1853 have a flag are now skipped.
1854
1855SC/07 Eximstats V1.49 (Alain Williams)
1856 Added the -emptyok flag.
1857
1858SC/08 Eximstats V1.50
1859 Fixes for obtaining the IP address from reject messages.
1860
0ea2a468
JJ
1861JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1862 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1863 whitesspace changes from 4.64-PH/27
1864
1865JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1866 match 4.64-PH/13
1867
1868JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1869 are found, allow negative numbers in numeric criteria)
1870
1871JJ/06 exipick.20061117.2, added new $message_body_missing variable
1872
1873JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1874 to match changes made in 4.64-PH/43
1875
8a10f5a4
PH
1876PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1877
30e18802
PH
1878PH/47 Put in an explicit test for a DNS lookup of an address record where the
1879 "domain" is actually an IP address, and force a failure. This locks out
1880 those revolvers/nameservers that support "A-for-A" lookups, in
1881 contravention of the specifications.
1882
55728a4f
PH
1883PH/48 When a host name was looked up from an IP address, and the subsequent
1884 forward lookup of the name timed out, the host name was left in
1885 $sender_host_name, contrary to the specification.
d7d7b289 1886
d7837193
PH
1887PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1888 restricted to single-key lookups, Exim was not diagnosing an error if
1889 * or *@ was used with a query-style lookup.
1890
87054a31
PH
1891PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1892
ea2c01d2
MH
1893MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1894 long ago, but noone interested enough thought of it. Let's just say that
1895 the "1.1" means that there are some new functions that weren't there at
1896 some point in the past.
1897
e4fa6968
PH
1898PH/51 Error processing for expansion failure of helo_data from an smtp
1899 transport during callout processing was broken.
1900
56f5d9bd
PH
1901PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1902 tested/used via the -bh/-bhc/-bs options.
1903
922e1c28
PH
1904PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1905 bug, fixed in subsequent PCRE releases).
1906
21eb6e72
PH
1907PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1908 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1909
a0540757
PH
1910PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1911
944e9e9c 1912
478be7b0
SC
1913Exim version 4.63
1914-----------------
1915
1916SC/01 Use a glob alias rather than an array ref in eximstats generated
1917 parser. This improves both readability and performance.
1918
1919SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1920 Don't display local sender or destination tables in eximstats unless
1921 there is data to show.
1922 Added average volumes into the eximstats top table text output.
1923
1924SC/03 Collect data on the number of addresses (recipients) as well
1925 as the number of messages in eximstats.
1926
2b965a65
TF
1927TF/01 Correct an error in the documentation for the redirect router. Exim
1928 does (usually) call initgroups() when daemonizing.
478be7b0 1929
45b91596
PH
1930TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1931 with consistent privilege compared to when running as a daemon.
478be7b0 1932
c59f5781
TF
1933TF/03 Note in the spec that $authenticated_id is not set for local
1934 submissions from trusted users.
1935
90fc3069
TF
1936TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1937 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1938
6083aca0
TF
1939TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1940 by adding some example configuration directives to the default
1941 configuration file. A little bit of work is required to uncomment the
1942 directives and define how usernames and passwords are checked, but
1943 there is now a framework to start from.
1944
765b530f
PH
1945PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1946 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1947 without this. I don't know how relevant this is to other LDAP libraries.
1948
4e167a8c
PH
1949PH/02 Add the verb name to the "unknown ACL verb" error.
1950
4608d683
PH
1951PH/03 Magnus Holmgren's patch for filter_prepend_home.
1952
b8dc3e4a
PH
1953PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1954
5418e93b
PH
1955PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1956 directory not expanded when it should be if an expanded home directory
1957 was set for the address (which is overridden by the transport).
1958
b4a9bda2
PH
1959PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1960 libradius.
1961
45b91596
PH
1962PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1963 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1964 because it is too late at that time, and has no effect.
1965
5547e2c5
PH
1966PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1967 security issue with \' (bugzilla #107). I could not use the
1968 PQescapeStringConn() function, because it needs a PGconn value as one of
1969 its arguments.
1970
dbcef0ea
PH
1971PH/08 When testing addresses using -bt, indicate those final addresses that
1972 are duplicates that would not cause an additional delivery. At least one
1973 person was confused, thinking that -bt output corresponded to deliveries.
1974 (Suppressing duplicates isn't a good idea as you lose the information
1975 about possibly different redirections that led to the duplicates.)
1976
25257489
PH
1977PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1978 systems where poll() doesn't work, in particular OS X.
1979
c816d124
PH
1980PH/10 Added more information to debugging output for retry time not reached.
1981
a9ccd69a
PH
1982PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1983 operations in malware.c.
1984
75fa1910
PH
1985PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1986 signatures.
1987
a7d7aa58
PH
1988PH/13 If write_rejectlog was set false when logging was sent to syslog with
1989 syslog_duplication set false, log lines that would normally be written
1990 both the the main log and to the reject log were not written to syslog at
1991 all.
1992
42119b09
PH
1993PH/14 In the default configuration, change the use of "message" in ACL warn
1994 statements to "add_header".
1995
41609df5
PH
1996PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1997 not followed by a command (e.g. "seen endif").
1998
a5bd321b
PH
1999PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2000 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2001 latter.
2002
e85a7ad5 2003PH/17 Added extra conditions to the default value of delay_warning_condition
5dff5817
PH
2004 so that it is now:
2005
e85a7ad5
PH
2006 ${if or { \
2007 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2008 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2009 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
5dff5817
PH
2010 }{no}{yes}}
2011
e85a7ad5
PH
2012 The Auto-Submitted: and various List- headers are standardised, whereas I
2013 don't think Precedence: ever was.
5dff5817 2014
d8fe1c03
PH
2015PH/18 Refactored debugging code in route_finduser() to show more information,
2016 in particular, the error code if getpwnam() issues one.
2017
16282d2b
PH
2018PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2019 This is apparently needed in addition to the PH/07 change above to avoid
2020 any possible encoding problems.
2021
35d40a98
PH
2022PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2023 but not after initializing Perl.
2024
034d99ab
PH
2025PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2026 output them only if debugging. By default they are written stderr,
2027 apparently, which is not desirable.
2028
6ec97b1b
PH
2029PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2030 queries.
2031
e22ca4ac
JJ
2032JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2033 --not options
2034
2035JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2036
33d73e3b
PH
2037PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2038 authenticated or an ident call has been made. Suppress the default
2039 values for $authenticated_id and $authenticated_sender (but permit -oMai
2040 and -oMas) when testing with -bh.
2041
9ecb03f3
PH
2042PH/24 Re-jigged the order of the tests in the default configuration so that the
2043 tests for valid domains and recipients precede the DNS black list and CSA
2044 tests, on the grounds that those ones are more expensive.
2045
084efe8d
PH
2046PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2047 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2048 This bug exists in every version of Exim that I still have, right back to
2049 0.12.
2050
366fc9f0
PH
2051PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2052 However, an attempt to turn on case-sensitivity in a regex key by
2053 including (?-i) didn't work because the subject string was already
2054 lowercased, and the effects were non-intuitive. It turns out that a
2055 one-line patch can be used to allow (?-i) to work as expected.
2056
c59f5781 2057
c887c79e
TF
2058Exim version 4.62
2059-----------------
2060
2061TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2062 other effects) broke the use of negated acl sub-conditions.
2063
1cce3af8
PH
2064PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2065 patch).
2066
afb3eaaf
PH
2067PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2068 "Deny" causes Exim to reject the incoming connection with a 554 error.
2069 Unfortunately, if there is a major crisis, such as a disk failure,
2070 tcp-wrappers gives "deny", whereas what one would like would be some
2071 kind of temporary error. A kludge has been added to help with this.
2072 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2073 554 error is used if errno is still zero or contains ENOENT (which occurs
2074 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2075 451 error is used.
2076
e173618b
PH
2077PH/03 Add -lutil to the default FreeBSD LIBS setting.
2078
dd16e114
PH
2079PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2080 errors. Otherwise a message that provokes a temporary error (when other
2081 messages do not) can cause a whole host to time out.
2082
f7fd3850
PH
2083PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2084 addresses were routed directly to files or pipes from a redirect router.
2085 File deliveries just didn't batch; pipe deliveries might have suffered
2086 odd errors.
2087
d87df92c
PH
2088PH/06 A failure to get a lock for a hints database would erroneously always say
2089 "Failed to get write lock", even when it was really a read lock.
2090
7e9f683d
PH
2091PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2092 of 0600. This has been changed to use the value of the lockfile_mode
2093 option (which defaults to 0600).
2094
bfad5236
PH
2095PH/08 Applied small patch from the Sieve maintainer.
2096
01c490df
PH
2097PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2098 folder from quota calculations, a direct delivery into this folder messed
2099 up the contents of the maildirsize file. This was because the regex was
2100 used only to exclude .Trash (or whatever) when the size of the mailbox
2101 was calculated. There was no check that a delivery was happening into an
2102 excluded directory. This bug has been fixed by ignoring all quota
2103 processing for deliveries into excluded directories.
2104
d6629cdc
PH
2105PH/10 Added the maildirfolder_create_regex option to appendfile.
2106
1cce3af8 2107
214e2000
PH
2108Exim version 4.61
2109-----------------
2110
2111PH/01 The code for finding all the local interface addresses on a FreeBSD
2112 system running IPv6 was broken. This may well have applied to all BSD
2113 systems, as well as to others that have similar system calls. The broken
2114 code found IPv4 interfaces correctly, but gave incorrect values for the
2115 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2116 that it would not match correctly against @[] and not recognize the IPv6
2117 addresses as local.
2118
f9daeae0
PH
2119PH/02 The ipliteral router was not recognizing addresses of the form user@
2120 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2121
7e66e54d
PH
2122PH/03 Added disable_ipv6.
2123
c8ea1597
PH
2124PH/04 Changed $reply_address to use the raw form of the headers instead of the
2125 decoded form, because it is most often used to construct To: headers
2126 lines in autoreplies, and the decoded form may well be syntactically
2127 invalid. However, $reply_address has leading white space removed, and all
2128 newlines turned into spaces so that the autoreply transport does not
2129 grumble.
2130
911f6fde
PH
2131PH/05 If group was specified without a user on a router, and no group or user
2132 was specified on a transport, the group from the router was ignored.
2133
47ca6d6c
PH
2134PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2135 for visible compile-time settings that can be used to change these
2136 numbers, for those that want even more. Backwards compatibility with old
2137 spool files has been maintained. However, going back to a previous Exim
2138 release will lost any variables that are in spool files.
2139
ed0e9820
PH
2140PH/07 Two small changes when running in the test harness: increase delay when
2141 passing a TCP/IP connection to a new process, in case the original
2142 process has to generate a bounce, and remove special handling of
2143 127.0.0.2 (sic), which is no longer necessary.
2144
eff37e47
PH
2145PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2146 be the same on different OS.
2147
1921d2ea
PH
2148PH/09 Moved a debug statement in filter processing to avoid a race problem when
2149 testing.
2150
b3f69ca8
JJ
2151JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2152 whether --show-vars was specified or not
2153
2154JJ/02 exipick: Added support for new ACL variable spool format introduced
2155 in 4.61-PH/06
2156
424a1c63
PH
2157PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2158 syntactically invalid From: or Reply-to: line, and a filter used this to
2159 generate an autoreply, and therefore failed to obtain an address for the
2160 autoreply, Exim could try to deliver to a non-existent relative file
2161 name, causing unrelated and misleading errors. What now happens is that
2162 it logs this as a hard delivery error, but does not attempt to create a
2163 bounce message.
2164
7a100415
PH
2165PH/11 The exinext utility has a -C option for testing purposes, but although
2166 the given file was scanned by exinext itself; it wasn't being passed on
2167 when Exim was called.
2168
19b9dc85
PH
2169PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2170 an end-of-file indication when reading a command response.
2171
309bd837
PH
2172PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2173 compiled. In many other places in Exim, IPv6 addresses are always
2174 recognized, so I have changed this. It also means that IPv4 domain
2175 literals of the form [IPV4:n.n.n.n] are now always recognized.
2176
59e82a2a
PH
2177PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2178 used if the router is not running as root, for example, when verifying at
2179 ACL time, or when using -bh. The debugging output from this situation was
2180 non-existent - all you got was a failure to exec. I have made two
2181 changes:
2182
2183 (a) Failures to set uid/gid, the current directory, or a process leader
2184 in a subprocess such as that created by queryprogram now generate
2185 suitable debugging ouput when -d is set.
2186
2187 (b) The queryprogram router detects when it is not running as root,
2188 outputs suitable debugging information if -d is set, and then runs
2189 the subprocess without attempting to change uid/gid.
2190
9edc04ce
PH
2191PH/15 Minor change to Makefile for building test_host (undocumented testing
2192 feature).
2193
1349e1e5
PH
2194PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2195 additional section of a DNS packet that returns MX or SRV records.
2196 Instead, it always explicitly searches for A/AAAA records. This avoids
2197 major problems that occur when a DNS server includes only records of one
2198 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2199 fixed another bug: if SRV records were looked up and the corresponding
2200 address records were *not* found in the additional section, the port
2201 values from the SRV records were lost.
2202
ea49d0e1
PH
2203PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2204 using the correct key (the original address) when searching the retry
2205 rules in order to find which one to use for generating the retry hint.
2206
064a94c9
PH
2207PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2208 adding the default one. Similarly, if it contains a Reply-To: header, the
2209 errors_reply_to option, if set, is not used.
2210
727071f8
PH
2211PH/19 When calculating a retry time, Exim used to measure the "time since
2212 failure" by looking at the "first failed" field in the retry record. Now
2213 it does not use this if it is later than than the arrival time of the
2214 message. Instead it uses the arrival time. This makes for better
2215 behaviour in cases where some deliveries succeed, thus re-setting the
2216 "first failed" field. An example is a quota failure for a huge message
2217 when small messages continue to be delivered. Without this change, the
2218 "time since failure" will always be short, possible causing more frequent
2219 delivery attempts for the huge message than are intended.
dd16e114 2220 [Note: This change was subsequently modified - see PH/04 for 4.62.]
727071f8 2221
f78eb7c6
PH
2222PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2223 $1, $2, $3) because the numerical variables can be reset during some
2224 expansion items (e.g. "match"), thereby losing the authentication data.
2225
21c28500
PH
2226PH/21 Make -bV show the size of off_t variables so that the test suite can
2227 decide whether to run tests for quotas > 2G.
2228
2229PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2230 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2231 filecount value is greater than 2G or if a quota value is greater than 2G
2232 on a system where the size of off_t is not greater than 4, a panic error
2233 is given.
2234
1688f43b
PH
2235PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2236 never match. The debug and -bh output now contains an explicit error
2237 message indicating a malformed IPv4 address or mask.
2238
2239PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2240 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2241 PH/23 above applies.
2242
9675b384
PH
2243PH/25 Do not write to syslog when running in the test harness. The only
2244 occasion when this arises is a failure to open the main or panic logs
2245 (for which there is an explicit test).
2246
6a3f1455
PH
2247PH/26 Added the /no_tell option to "control=freeze".
2248
dac79d3e
PH
2249PH/27 If a host name lookup failed very early in a connection, for example, if
2250 the IP address matched host_lookup and the reverse lookup yielded a name
2251 that did not have a forward lookup, an error message of the form "no IP
2252 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2253 could be logged. Now it outputs the IP address instead of "NULL".
1349e1e5 2254
5977a0b3
PH
2255PH/28 An enabling patch from MH: add new function child_open_exim2() which
2256 allows the sender and the authenticated sender to be set when
2257 submitting a message from within Exim. Since child_open_exim() is
2258 documented for local_scan(), the new function should be too.
2259
c91535f3
PH
2260PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2261 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2262 results in an empty string is now treated as unset.
2263
0d46a8c8
PH
2264PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2265
278c6e6c
PH
2266PH/31 Added sender_verify_fail logging option.
2267
2cbb4081
PH
2268PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2269 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2270 tidied the source and removed it altogether.
2271
3eef829e
PH
2272PH/33 When a queue run was abandoned because the load average was too high, a
2273 log line was always written; now it is written only if the queue_run log
2274 selector is set. In addition, the log line for abandonment now contains
2275 information about the queue run such as the pid. This is always present
2276 in "start" and "stop" lines but was omitted from the "abandon" line.
2277
1ab95fa6
PH
2278PH/34 Omit spaces between a header name and the colon in the error message that
2279 is given when verify = headers_syntax fails (if there are lots of them,
2280 the message gets confusing).
2281
230205fc
PH
2282PH/35 Change the default for dns_check_names_pattern to allow slashes within
2283 names, as there are now some PTR records that contain slashes. This check
2284 is only to protect against broken name servers that fall over on strange
2285 characters, so the fact that it applies to all lookups doesn't matter.
2286
75e0e026
PH
2287PH/36 Now that the new test suite is complete, we can remove some of the
2288 special code in Exim that was needed for the old test suite. For example,
2289 sorting DNS records because real resolvers return them in an arbitrary
2290 order. The new test suite's fake resolver always returns records in the
2291 same order.
2292
2293PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2294 bounces) except when queue_only is set, to avoid logging races between
2295 the different processes.
2296
145396a6
PH
2297PH/38 Panic-die if .include specifies a non-absolute path.
2298
3cd34f13
PH
2299PH/39 A tweak to the "H" retry rule from its user.
2300
11121d3d
JJ
2301JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2302 a label. They prevented compilation on older perls.
2303
2304JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2305 a warning to be raised on newish perls.
2306
2307JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2308 on queue. Changes to match documented behaviour of showing count of
2309 messages matching specified criteria.
2310
8def5aaf
PH
2311PH/40 Changed the default ident timeout from 30s to 5s.
2312
929ba01c
PH
2313PH/41 Added support for the use of login_cap features, on those BSD systems
2314 that have them, for controlling the resources used by pipe deliveries.
2315
2632889e
PH
2316PH/42 The content-scanning code uses fopen() to create files in which to put
2317 message data. Previously it was not paying any attention to the mode of
2318 the files. Exim runs with umask(0) because the rest of the code creates
2319 files with open(), and sets the required mode explicitly. Thus, these
2320 files were ending up world-writeable. This was not a big issue, because,
2321 being within the spool directory, they were not world-accessible. I have
2322 created a function called modefopen, which takes an additional mode
2323 argument. It sets umask(777), creates the file, chmods it to the required
2324 mode, then resets the umask. All the relevant calls to fopen() in the
2325 content scanning code have been changed to use this function.
2326
944a9c55
PH
2327PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2328 to 24 hours. This avoids potential overflow problems when processing G
2329 and H retry rules. I suspect nobody ever tinkers with this value.
2330
4a23603b
PH
2331PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2332
4730f942
PH
2333PH/45 When the plaintext authenticator is running as a client, the server's
2334 challenges are checked to ensure they are valid base64 strings. By
2335 default, the authentication attempt is cancelled if an invalid string is
2336 received. Setting client_ignore_invalid_base64 true ignores these errors.
2337 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2338 they are received. Thus, the responses can be made to depend on the
2339 challenges. If an invalid string is ignored, an empty string is placed in
2340 the variable.
2341
30dba1e6
PH
2342PH/46 Messages that are created by the autoreply transport now contains a
2343 References: header, in accordance with RFCs 2822 and 3834.
2344
382afc6b
PH
2345PH/47 Added authenticated_sender_force to the smtp transport.
2346
a86229cf
PH
2347PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2348
50c99ba6
PH
2349PH/49 Installed latest patch from the Sieve maintainer.
2350
d35e429d
PH
2351PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2352 was also set, the appendfile transport was unnecessarily scanning a
2353 directory of message files (e.g. for maildir delivery) to find the count
2354 of files (along with the size), even though it did not need this
2355 information. It now does the scan only if it needs to find either the
2356 size of the count of files.
2357
f90d018c
PH
2358PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2359
75def545
PH
2360PH/52 Two bugs concerned with error handling when the smtp transport is
2361 used in LMTP mode:
2362
2363 (i) Exim was not creating retry information for temporary errors given
2364 for individual recipients after the DATA command when the smtp transport
2365 was used in LMTP mode. This meant that they could be retried too
2366 frequently, and not timed out correctly.
2367
2368 (ii) Exim was setting the flag that allows error details to be returned
2369 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2370 recipients that were returned after the DATA command.
2371
2372PH/53 This is related to PH/52, but is more general: for any failing address,
2373 when detailed error information was permitted to be returned to the
2374 sender, but the error was temporary, then after the final timeout, only
2375 "retry timeout exceeded" was returned. Now it returns the full error as
2376 well as "retry timeout exceeded".
2377
c46782ef
PH
2378PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2379 do this, and (what is worse) MTAs that accept it.
2380
71fafd95
PH
2381PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2382 will now be deprecated.
2383
2c5db4fd
PH
2384PH/56 New os.c-cygwin from the Cygwin maintainer.
2385
9cf6b11a
JJ
2386JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2387 formats (previously only available in exim formats via -bpr, -bpru,
2388 and -bpra. Now also available in native and exiqgrep formats)
2389
2390JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2391 with very large, slow to parse queues
2392
2393JJ/08 exipick: added ! as generic prefix to negate any criteria format
2394
2395JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2396
898d150f
PH
2397PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2398 responses to authentication challenges, though it was showing the
2399 challenges; (ii) I've removed the CR characters from the debug output for
2400 SMTP output lines.
2401
46218253
PH
2402PH/58 Allow for the insertion of a newline as well as a space when a string
2403 is turned into more than one encoded-word during RFC 2047 encoding. The
2404 Sieve code now uses this.
2405
e97957bc
PH
2406PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2407 data_4xx, lost_connection, tls_required.
2408
81e509d7
PH
2409PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2410 message was being sent as an SMTP response.
2411
3d240ff7
PH
2412PH/61 Add -l and -k options to exicyclog.
2413
b37c4101
PH
2414PH/62 When verifying, if an address was redirected to one new address, so that
2415 verification continued, and the new address failed or deferred after
2416 having set something in $address_data, the value of $address_data was not
2417 passed back to the ACL. This was different to the case when no
2418 redirection occurred. The value is now passed back in both cases.
2419
79378e0f
PH
2420PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2421 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2422 use login_cap.h, so on its own it isn't the distinguishing feature. The
2423 new name refers directly to the setclassresources() function.
2424
e49c7bb4
PH
2425PH/65 Added configuration files for NetBSD3.
2426
d114ec46
PH
2427PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2428
f3d7df6c
PH
2429PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2430 is preferred over IPv4.
2431
715ab376
PH
2432PH/68 The bounce_return_message and bounce_return_body options were not being
2433 honoured for bounces generated during the reception of non-SMTP messages.
2434 In particular, this applied to messages rejected by the ACL. This bug has
2435 been fixed. However, if bounce_return_message is true and bounce_return_
2436 body is false, the headers that are returned for a non-SMTP message
2437 include only those that have been read before the error was detected.
2438 (In the case of an ACL rejection, they have all been read.)
2439
6b31b150
PH
2440PH/69 The HTML version of the specification is now built in a directory called
2441 spec_html instead of spec.html, because the latter looks like a path with
2442 a MIME-type, and this confuses some software.
2443
2444PH/70 Catch two compiler warnings in sieve.c.
2445
d515a917
PH
2446PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2447 function verify_get_ident() calls ip_connect() to connect a socket, but
2448 if the "connect()" function timed out, ip_connect() used to close the
2449 socket. However, verify_get_ident() also closes the socket later, and in
2450 between Exim writes to the log, which may get opened at this point. When
2451 the socket was closed in ip_connect(), the log could get the same file
2452 descriptor number as the socket. This naturally causes chaos. The fix is
2453 not to close the socket in ip_connect(); the socket should be closed by
2454 the function that creates it. There was only one place in the code where
2455 this was missing, in the iplookup router, which I don't think anybody now
2456 uses, but I've fixed it anyway.
2457
9b8fadde
PH
2458PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2459 well as to direct DNS lookups. Otherwise the handling of names in host
2460 lists is inconsistent and therefore confusing.
2461
214e2000 2462
5de37277
PH
2463Exim version 4.60
2464-----------------
2465
cc38ddbf
PH
2466PH/01 Two changes to the default runtime configuration:
2467
2468 (1) Move the checks for relay_from_hosts and authenticated clients from
2469 after to before the (commented out) DNS black list checks.
2470
2471 (2) Add control=submission to the relay_from_hosts and authenticated
2472 clients checks, on the grounds that messages accepted by these
2473 statements are most likely to be submissions.
5de37277 2474
72fdd6ae
PH
2475PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2476
2477 (1) Generate an error if the third argument for the ${prvs expansion is
2478 not a single digit.
2479
2480 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2481 string.
2482
2483 (3) Reset the variables that are obtained from the first argument of
2484 ${prvscheck and used in the second argument before leaving the code,
2485 because their memory is reclaimed, so using them afterwards may do
2486 silly things.
2487
2488 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2489 one (it's much easier than Tom thought :-).
2490
2491 (5) Because of (4), we can now allow for the use of $prvscheck_result
2492 inside the third argument.
cb9328de 2493
cb741023
PH
2494PH/03 For some reason, the default setting of PATH when running a command from
2495 a pipe transport was just "/usr/bin". I have changed it to
2496 "/bin:/usr/bin".
2497
f174f16e
PH
2498PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2499 anything to be listed in the output from -bV.
b2f5a032 2500
c25242d7
PH
2501PH/05 When a filter generated an autoreply, the entire To: header line was
2502 quoted in the delivery log line, like this:
2503
2504 => >A.N.Other <ano@some.domain> <original@ddress> ...
2505
2506 This has been changed so that it extracts the operative address. There
2507 may be more than one such address. If so, they are comma-separated, like
2508 this:
2509
2510 => >ano@some.domain,ona@other.domain <original@ddress> ...
2511
82c19f95
PH
2512PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2513 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2514 not being looked up in the rDNS to get a host name, Exim was showing the
2515 IP address twice in Received: lines, even though the IP addresses were
2516 identical. For example:
2517
2518 Received: from [1.2.3.4] (helo=[1.2.3.4])
2519
2520 However, if the real host name was known, it was omitting the HELO data
2521 if it matched the actual IP address. This has been tidied up so that it
2522 doesn't show the same IP address twice.
2523
d7ffbc12
PH
2524PH/07 When both +timestamp and +memory debugging was on, the value given by
2525 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2526 called by the debug printing, thereby overwriting the timestamp buffer.
2527 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2528
9f526266
PH
2529PH/08 When the original message was included in an autoreply transport, it
2530 always said "this is a copy of the message, including all the headers",
2531 even if body_only or headers_only was set. It now gives an appropriate
2532 message.
2533
87fcc8b9
PH
2534PH/09 Applied a patch from the Sieve maintainer which:
2535
2536 o fixes some comments
2537 o adds the (disabled) notify extension core
2538 o adds some debug output for the result of if/elsif tests
2539 o points to the current vacation draft in the documentation
2540 and documents the missing references header update
2541
2542 and most important:
2543
2544 o fixes a bug in processing the envelope test (when testing
2545 multiple envelope elements, the last element determinted the
2546 result)
2547
456682f5
PH
2548PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2549 Electronic Mail") by including:
2550
2551 Auto-submitted: auto-generated
2552
2553 in the messages that it generates (bounce messages and others, such as
2554 warnings). In the case of bounce messages for non-SMTP mesages, there was
2555 also a typo: it was using "Auto_submitted" (underscore instead of
2556 hyphen). Since every message generated by Exim is necessarily in response
2557 to another message, thes have all been changed to:
2558
2559 Auto-Submitted: auto-replied
2560
2561 in accordance with these statements in the RFC:
2562
2563 The auto-replied keyword:
2564
2565 - SHOULD be used on messages sent in direct response to another
2566 message by an automatic process,
2567
2568 - MUST NOT be used on manually-generated messages,
2569
2570 - MAY be used on Delivery Status Notifications (DSNs) and Message
2571 Disposition Notifications (MDNs),
2572
2573 - MUST NOT be used on messages generated by automatic or periodic
2574 processes, except for messages which are automatic responses to
2575 other messages.
2576
3e46c1aa
PH
2577PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2578 to the default Received: header definition.
456682f5 2579
49826d12
PH
2580PH/12 Added log selector acl_warn_skipped (default on).
2581
eba0c039
PH
2582PH/13 After a successful wildlsearch lookup, discard the values of numeric
2583 variables because (a) they are in the wrong storage pool and (b) even if
2584 they were copied, it wouldn't work properly because of the caching.
2585
a0d6ba8a
PH
2586PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2587 checking when decoding. Apparently there are clients that generate
2588 overlong encoded strings. Why am I not surprised?
2589
f0917727
PH
2590PH/15 If the first argument of "${if match_address" was not empty, but did not
2591 contain an "@" character, Exim crashed. Now it writes a panic log message
2592 and treats the condition as false.
2593
096fee00
PH
2594PH/16 In autoreply, treat an empty string for "once" the same as unset.
2595
024bd3c2
PH
2596PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2597 extension "envelope-auth". The code is finished and in agreement with
2598 other implementations, but there is no documentation so far and in fact,
2599 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2600 not changing the active code.
2601
2602 Print executed "if" and "elsif" statements when debugging is used. This
2603 helps a great deal to understand what a filter does.
2604
2605 Document more things not specified clearly in RFC3028. I had all this
2606 sorted out, when out of a sudden new issues came to my mind. Oops."
2607
df199fec
PH
2608PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2609 (Bugzilla #53).
2610
d27f1df3
PH
2611PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2612 canonical form (as documented). However, after a host name lookup from
2613 the IP address, check_host() was doing a simple string comparison with
2614 addresses acquired from the DNS when checking that the found name did
2615 have the original IP as one of its addresses. Since any found IPv6
2616 addresses are likely to be in abbreviated form, the comparison could
2617 fail. Luckily, there already exists a function for doing the comparison
2618 by converting both addresses to binary, so now that is used instead of
2619 the text comparison.
2620
96776534
PH
2621PH/20 There was another similar case to PH/19, when a complete host name was
2622 given in a host list; looking up its IP address could give an abbreviated
2623 form, whereas the current host's name might or might not be abbreviated.
2624 The same fix has been applied.
2625
5de37277 2626
9a799bc0
PH
2627Exim version 4.54
2628-----------------
2629
2630PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2631 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2632 It now does.
2633
99a4b039
PH
2634PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2635 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2636
2637PH/03 Typo: missing ".o" in src/pcre/Makefile.
2638
4b233853
PH
2639PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2640 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2641 Also, for "Auto-Submitted", treat anything other than "no" as
2642 non-personal, in accordance with RFC 3834. (Previously it treated
2643 anything starting "auto-" as non-personal.)
2644
8857ccfd
PH
2645TF/01 The control=submission/name=... option had a problem with syntax
2646 errors if the name included a slash character. The /name= option
2647 now slurps the rest of the string, so it can include any characters
2648 but it must come last in the list of options (after /sender_retain
2649 or /domain=).
2650
433a2980
PH
2651PH/05 Some modifications to the interface to the fake nameserver for the new
2652 testing suite.
2653
3e46c1aa 2654
9a799bc0 2655
e3a311ba
TK
2656Exim version 4.53
2657-----------------
2658
2659TK/01 Added the "success_on_redirect" address verification option. See
2660 NewStuff for rationale and an example.
2661
13b685f9
PH
2662PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2663
395ff96d
PH
2664PH/02 Patch to exigrep to allow it to work on syslog lines.
2665
5b68f6e4
PH
2666PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2667 fread() to skip over the body file's header line, because in Cygwin the
2668 header line is locked and is inaccessible.
2669
1ab52c69
PH
2670PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
2671 co-exist for some time) to make it clear that it is the Exim ID that is
2672 referenced, not the Message-ID: header line.
2673
b07e6aa3
PH
2674PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
2675 string_format() function, because snprintf() does not exist on all
2676 operating systems.
2677
254e032f
PH
2678PH/06 The use of forbid_filter_existstest now also locks out the use of the
2679 ${stat: expansion item.
2680
3af76a81
PH
2681PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
2682 protocol synchronization error", to keep the pedants happy.
2683
2548ba04
PH
2684PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
2685 well as for IRIX systems, when gcc is being used. See the host.c source
2686 file for comments.
2687
b6c6011d
PH
2688PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
2689
cf39cf57
PH
2690PH/10 Named domain lists were not working if used in a queue_smtp_domains
2691 setting.
2692
f1513293
PH
2693PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
2694 transport and to the smtp transport in LMTP mode.
2695
727549a4
PH
2696TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
2697
af46795e
PH
2698PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
2699 run a filter in a subprocess. This could lead to confusion in subsequent
2700 lookups in the parent process. There should also be a search_tidyup() at
2701 the end of the subprocess.
2702
d7b47fd0
PH
2703PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
2704 only if the host matched helo_try_verify_hosts, which caused the
2705 verification to occur when the EHLO/HELO command was issued. The ACL just
2706 tested the remembered result. Now, if a previous verification attempt has
2707 not happened, "verify = helo" does it there and then.
2708
ee744174
JJ
2709JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
2710
b582ab87
PH
2711TK/03 Fix log output including CR from clamd.
2712
41a13e0a
PH
2713PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
2714 exist provoked a memory error which could cause a segfault.
2715
f625cc5a
PH
2716PH/15 Installed PCRE 6.2
2717
2718PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2719
21f7af35
PH
2720PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2721 of the problem. Specifically, suggested +O2 rather than +O1 for the
2722 HP-UX compiler.
2723
31480e42
PH
2724PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2725
2d280592
PH
2726PH/20 If a delivery was routed to a non-standard port by means of an SRV
2727 record, the port was not correctly logged when the outgoing_port log
2728 selector was set (it logged the transort's default port).
2729
7cd1141b
PH
2730PH/21 Added support for host-specific ports to manualroute, queryprogram,
2731 fallback_hosts, and "hosts" in the smtp transport.
2732
2733PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2734 host errors such as "Connection refused".
2735
750af86e
PH
2736PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2737 authentication with radiusclient 0.4.9:
2738
2739 - Error returned from rc_read_config was caught wrongly
2740 - Username/password not passed on to radius server due to wrong length.
2741
2742 The presumption is that some radiusclient API changes for 4.51/PH/17
2743 were not taken care of correctly. The code is still untested by me (my
2744 Linux distribution still has 0.3.2 of radiusclient), but it was
2745 contributed by a Radius user.
2746
2747PH/24 When doing a callout, the value of $domain wasn't set correctly when
2748 expanding the "port" option of the smtp transport.
2749
4304270b
TK
2750TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2751 while reading a MIME header. Thanks to Tom Hughes for a patch.
2752
750af86e
PH
2753PH/24 Include config.h inside local_scan.h so that configuration settings are
2754 available.
2755
64ffc24f
PH
2756PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2757 that in an ACL for RCPT (for example), you can examine exactly what was
2758 received.
2759
5dd9625b
PH
2760PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2761 commands, but it was not correctly comparing the address with the actual
2762 client host address. Thus, it would show the EHLO address in Received:
2763 header lines when this was not necessary.
2764
5591031b
PH
2765PH/27 Added the % operator to ${eval:}.
2766
ba18e66a
PH
2767PH/28 Exim tries to create and chdir to its spool directory when it starts;
2768 it should be ignoring failures (because with -C, for example, it has lost
2769 privilege). It wasn't ignoring creation failures other than "already
2770 exists".
2771
9cec981f
PH
2772PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2773 -bV or -d is used.
2774
aa2b5c79
PH
2775PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2776 because an input line was too long, either on its own, or by virtue of
1509d3a8
PH
2777 too many continuations, the temporary file was not being removed, and the
2778 return code was incorrect.
aa2b5c79 2779
48a53b7f
PH
2780PH/31 Missing "BOOL" in function definition in filtertest.c.
2781
1c59d63b
PH
2782PH/32 Applied Sieve patches from the maintainer.
2783
671012da
TK
2784TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2785
1509d3a8
PH
2786PH/33 Added "verify = not_blind".
2787
2788PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2789 Local/Makefile (with some defaults set). These are used in built scripts
2790 such as exicyclog, but they have never been used in the exim_install
2791 script (though there are many overriding facilities there). I have
2792 arranged that the exim_install script now takes note of these two
2793 settings.
2794
2795PH/35 Installed configuration files for Dragonfly.
2796
2fe1a124
PH
2797PH/36 When a locally submitted message by a trusted user did not contain a
2798 From: header, and the sender address was obtained from -f or from an SMTP
2799 MAIL command, and the trusted user did not use -F to supply a sender
2800 name, $originator_name was incorrectly used when constructing a From:
2801 header. Furthermore, $originator_name was used for submission mode
2802 messages from external hosts without From: headers in a similar way,
2803 which is clearly wrong.
2804
8800895a
PH
2805PH/37 Added control=suppress_local_fixups.
2806
ccfdb010
PH
2807PH/38 When log_selector = +received_sender was set, and the addition of the
2808 sender made the log line's construction buffer exactly full, or one byte
2809 less than full, an overflow happened when the terminating "\n" was
2810 subsequently added.
2811
1130bfb0
PH
2812PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2813 when the result of a list match is failure because a DNS lookup failed.
2814
ebcb507f
PH
2815PH/40 RM_COMMAND is now used in the building process.
2816
c35e155c
PH
2817PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2818 the "build-* directories that it finds.
2819
95d1f782
PH
2820PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2821 address in a domain literal was a prefix of an interface address.
2822
fd6de02e
PH
2823PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2824 when verifying a sender address, unless rewrite_headers is false.
2825
58de37c5
PH
2826PH/44 Wrote a long comment about why errors_to addresses are verified as
2827 recipients, not senders.
2828
261cf466
TF
2829TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2830 the ratelimit ACL was added.
2831
3ee512ff
PH
2832PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2833
e08c430f
PH
2834PH/46 Added extra information about PostgreSQL errors to the error string.
2835
bef5a11f
PH
2836PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2837 avoiding the need to install special zones in a real server. This is
2838 backwards compatible; if it can't find the fake resolver, it drops back.
2839 Thus, both old and new test suites can be run.
2840
7546de58
TF
2841TF/02 Added util/ratelimit.pl
2842
e5d5a95f
TF
2843TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2844 clock is set back in time.
2845
2e88a017
TF
2846TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2847 Candler <B.Candler@pobox.com>.
2848
a5f65aa4
TF
2849TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2850 OK for addresses that are the result of redirections.
2851
e7726cbf
PH
2852PH/48 A number of further additions for the benefit of the new test suite,
2853 including a fake gethostbyname() that interfaces to the fake DNS resolver
2854 (see PH/47 above).
2855
a7fdad5b
TF
2856TF/06 The fix for widen_domains has also been applied to qualify_single and
2857 search_parents which are the other dnslookup options that can cause
2858 header rewrites.
2859
6af56900
PH
2860PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2861 type ("H").
2862
0925ede6
PH
2863PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2864
66afa403
TF
2865TF/07 Exim produced the error message "an SRV record indicated no SMTP
2866 service" if it encountered an MX record with an empty target hostname.
2867 The message is now "an MX or SRV record indicated no SMTP service".
2868
0154e85a
TF
2869TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2870 if the DNS of the sending site is misconfigured. This is quite a
2871 common situation. This change restores the behaviour of treating a
2872 helo verification defer as a failure.
2873
16f12c76
PH
2874PH/51 If self=fail was set on a router, the bounce message did not include the
2875 actual error message.
2876
bbe902f0 2877
e5a9dba6
PH
2878Exim version 4.52
2879-----------------
2880
2881TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2882
22c3b60b
PH
2883PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2884 command itself ended in error, the underlying message about the transport
2885 filter timeout was being overwritten with the pipe command error. Now the
2886 underlying error message should be appended to the second error message.
2887
06a9b4b5
PH
2888TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2889
c1ac6996
PH
2890PH/02 Reduce the amount of output that "make" produces by default. Full output
2891 can still be requested.
2892
9c7a242c
PH
2893PH/03 The warning log line about a condition test deferring for a "warn" verb
2894 was being output only once per connection, rather than after each
2895 occurrence (because it was using the same function as for successful
2896 "warn" verbs). This seems wrong, so I have changed it.
2897
87ba3f5f
PH
2898TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2899 it should not have, which might have caused a crash in the right
2900 circumstances, but probably never did.
2901
2902PH/04 Installed a modified version of Tony Finch's patch to make submission
2903 mode fix the return path as well as the Sender: header line, and to
2904 add a /name= option so that you can make the user's friendly name appear
2905 in the header line.
2906
29aba418
TF
2907TF/03 Added the control = fakedefer ACL modifier.
2908
fe0dab11
TF
2909TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2910 Mark Lowes for thorough testing.
870f6ba8 2911
11d337a4
TK
2912TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2913
2914TK/03 Merged latest SRS patch from Miles Wilton.
2915
415c8f3b
PH
2916PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2917 with the definition in sysexits.h (which is #included earlier).
2918 Fortunately, Exim does not actually use EX_OK. The code used to try to
2919 preserve the sysexits.h value, by assumimg that macro definitions were
2920 scanned for macro replacements. I have been disabused of this notion,
2921 so now the code just undefines EX_OK before #including unistd.h.
11d337a4 2922
958541e9
PH
2923PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2924 in the smtp transport. When a block could not be written in a single
2925 write() function, the timeout was being re-applied to each part-write.
2926 This seems wrong - if the receiver was accepting one byte at a time it
2927 would take for ever. The timeout is now adjusted when this happens. It
2928 doesn't have to be particularly precise.
2929
c206415f
TK
2930TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2931 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2932
2a4be8f9
PH
2933PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2934 without a domain if the check to <postmaster@domain> fails.
2935
1cba11c5
SC
2936SC/01 Eximstats: added -xls and the ability to specify output files
2937 (patch written by Frank Heydlauf).
2938
2939SC/02 Eximstats: use FileHandles for outputing results.
2940
2941SC/03 Eximstats: allow any combination of xls, txt, and html output.
2942
2943SC/04 Eximstats: fixed display of large numbers with -nvr option
2944
2945SC/05 Eximstats: fixed merging of reports with empty tables.
2946
2947SC/06 Eximstats: added the -include_original_destination flag
2948
2949SC/07 Eximstats: removed tabs and trailing whitespace.
2950
1005d00e
TK
2951TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2952
2953TK/06 MBOX spool code: Add real "From " MBOX separator line
2954 so the .eml file is really in mbox format (even though
2955 most programs do not really care). Patch from Alex Miller.
2956
2957TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2958 The latter is generated from $received_to and is only set if the
2959 message has one envelope recipient. SA can use these headers,
2960 obviously out-of-the-box. Patch from Alex Miller.
2961
9b4768fa
PH
2962PH/08 The ${def test on a variable was returning false if the variable's
2963 value was "0", contrary to what the specification has always said!
2964 The result should be true unless the variable is empty.
2965
2966PH/09 The syntax error of a character other than { following "${if
2967 def:variable_name" (after optional whitespace) was not being diagnosed.
2968 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2969 accidental colon was present, for example, could give incorrect results.
2970
0d7eb84a
PH
2971PH/10 Tidied the code in a number of places where the st_size field of a stat()
2972 result is used (not including appendfile, where other changes are about
2973 to be made).
2974
2975PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2976 This involved changing a lot of size variables from int to off_t. It
2977 should work with maildirs and everything.
2978
40727bee
TK
2979TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2980 spamd dying while we are connected to it.
2981
554d2369
TF
2982TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2983 <jgh@wizmail.org>
2984
1f922db1
PH
2985PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2986 function for BDB 4.3.
2987
ef213c3b
PH
2988PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2989
8ac170f3
PH
2990PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2991 the sources for PCRE have been re-arranged and more files are now
2992 involved.
2993
b1c749bb
PH
2994PH/15 The code I had for printing potentially long long variables in PH/11
2995 above was not the best (it lost precision). The length of off_t variables
2996 is now inspected at build time, and an appropriate printing format (%ld
c6c2dc1d
PH
2997 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2998 to be "long long int" or "long int". This is needed for the internal
2999 formatting function string_vformat().
b1c749bb 3000
4aac9b49
PH
3001PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3002 the configuration file to be ":syslog", then the script "guesses" where
3003 the logs files are, rather than using the compiled in default. In our
3004 case the guess is not the same as the compiled default, so the script
3005 suddenly stopped working when I started to use syslog. The patch checks
3006 to see if log_file_path is "". If so, it attempts to read it from exim
3007 with no configuration file to get the compiled in version, before it
3008 falls back to the previous guessing code."
3009
294520c8
TK
3010TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3011 implementing BATV in an Exim configuration. See NewStuff for the gory
3012 details.
3013
5bd022fe
PH
3014PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3015 Makefile that are specific to HP-UX.
3016
90e9ce59
PH
3017PH/18 If the "use_postmaster" option was set for a recipient callout together
3018 with the "random" option, the postmaster address was used as the MAIL
3019 FROM address for the random test, but not for the subsequent recipient
3020 test. It is now used for both.
3021
5ea81592
PH
3022PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3023 patch removes a few documentation additions to RFC 3028, because the
3024 latest draft now contains them. It adds the new en;ascii-case comparator
3025 and a new error check for 8bit text in MIME parts. Comparator and
3026 require names are now matched exactly. I enabled the subaddress
3027 extension, but it is not well tested yet (read: it works for me)."
3028
c6c2dc1d
PH
3029PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3030 rework some of the code of TK/09 above to avoid the hardwired use of
3031 "%lld" and "long long". Replaced the call to snprintf() with a call to
3032 string_vformat().
3033
fffffe4c
PH
3034PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3035 records point to non-existent hosts", "retry timeout exceeded", and
3036 "retry time not reached for any host after a long failure period".
ca02eafb 3037
9a26b6b2
PH
3038PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3039 experimental DomainKeys support:
3040
3041 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3042 (2) On an error such as an illegally used "control", the wrong name for
3043 the control was given.
3044
3045 These problems did NOT occur unless DomainKeys support was compiled.
3046
4aee0225
PH
3047PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3048
32d668a5
PH
3049PH/24 Added ${if match_ip condition.
3050
8187c3f3
PH
3051PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3052 databases so that it will be absolutely obvious if a crash occurs in the
3053 DB library. This is a regular occurrence (often caused by mis-matched
3054 db.h files).
3055
ff790e47 3056PH/26 Insert a lot of missing (void) casts for functions such as chown(),
f1e894f3
PH
3057 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3058 picked up on a user's system that detects such things. There doesn't seem
3059 to be a gcc warning option for this - only an attribute that has to be
3060 put on the function's prototype. It seems that in Fedora Core 4 they have
3061 set this on a number of new functions. No doubt there will be more in due
3062 course.
ff790e47 3063
5417f6d1
PH
3064PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3065 specify a transport. However, if an address that was verified by such a
3066 router was the subject of a callout, Exim crashed because it tried to
3067 read the rcpt_include_affixes from the non-existent transport. Now it
3068 just assumes that the setting of that option is false. This bug was
3069 introduced by 4.51/PH/31.
3070
59cf8544
PH
3071PH/28 Changed -d+all to exclude +memory, because that information is very
3072 rarely of interest, but it makes the output a lot bigger. People tend to
3073 do -d+all out of habit.
3074
e7ad8a65
PH
3075PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3076 code in os-type was giving problems when libc.so lives in lib64, like on
3077 x86_64 Fedora Core.
3078
ade42478
PH
3079PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3080 aren't the modern standard, and it seems that some systems' include files
3081 don't always have them. Exim was already checking for some of the newer
3082 ones like T_AAAA, and defining it itself. I've added checks for all the
3083 record types that Exim uses.
3084
182ad5cf
PH
3085PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3086 not automatically generating a new one, as it is supposed to. This
3087 prevented TLS from working. If the file did exist, but contained invalid
3088 data, a new version was generated, as expected. It was only the case of a
3089 non-existent file that was broken.
3090
b0d9fc80
TK
3091TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3092 with a change in libdomainkeys > 0.64.
3093
3094TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3095 from DNS. If the selector record carries the flag, it now has
3096 precedence over the domain-wide flag.
3097
3098TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3099
47c7a64a
PH
3100PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3101 the use of a port name that isn't defined in /etc/services) occurred, the
3102 message was deferred as in a normal delivery, and thus remained on the
3103 spool, instead of being failed because of the mua_wrapper setting. This
3104 is now fixed, and I tidied up some of the mua_wrapper messages at the
3105 same time.
3106
a388bce4
SC
3107SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3108 the messages in a hash of arrays rather than using individual hashes.
3109 This is a bit cleaner and results in dramatic memory savings, albeit
3110 at a slight CPU cost.
3111
3112SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3113 as requested by Marc Sherman.
3114
3115SC/10 Eximstats: added histograms for user specified patterns as requested
3116 by Marc Sherman.
3117
0793e4ed
SC
3118SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3119
c58b88df
PH
3120PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3121 fopen() in the content-scanning modules that did not already have it.
3122
e7ad8a65 3123
7982096b
PH
3124Exim version 4.51
3125-----------------
3126
1a46a8c5
PH
3127TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3128 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3129
2f079f46 3130TK/02 Fix ACL "control" statement not being available in MIME ACL.
1a46a8c5
PH
3131
3132TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3133
3134PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3135 to test Sieve filters that use "vacation".
3136
3137PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3138 that changes the way the GnuTLS parameters are stored in the cache file.
3139 The new format can be generated externally. For backward compatibility,
3140 if the data in the cache doesn't make sense, Exim assumes it has read an
3141 old-format file, and it generates new data and writes a new file. This
3142 means that you can't go back to an older release without removing the
3143 file.
3144
3145PH/03 A redirect router that has both "unseen" and "one_time" set does not
3146 work if there are any delivery delays because "one_time" forces the
3147 parent to be marked "delivered", so its unseen clone is never tried
3148 again. For this reason, Exim now forbids the simultaneous setting of
3149 these two options.
3150
3151PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3152 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3153 entry if you want to know the details. The fix, however, neglected to
3154 consider the case when local delivery batching is involved. The test for
3155 "previously delivered" was not happening when checking to see if an
3156 address could be batched with a previous (undelivered) one; under
3157 certain circumstances this could lead to multiple deliveries to the same
c2c19e9d 3158 address.
1a46a8c5
PH
3159
3160PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3161 in its include files, and this causes problems building Exim.
3162
3163PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3164 header_syntax) but Exim was just ignoring anything given after a slash.
3165 In particular, this caused confusion with an attempt to use "verify =
3166 reverse_host_lookup/defer_ok". An error is now given when options are
3167 supplied for verify items that do not have them. (Maybe reverse_host_
3168 lookup should have a defer_ok option, but that's a different point.)
3169
3170PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3171 defined by RFC 821) to 2048, because there were problems with some AUTH
3172 commands, and RFC 1869 says the size should be increased for extended
3173 SMTP commands that take arguments.
3174
3175PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3176 Finch).
3177
3178PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3179 "unknown" error; now it says that the functionality isn't in the binary.
8d67ada3 3180
49c2d5ea
PH
3181PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3182 an address' error message when a string expansion fails (syntax or
f331f3b6
PH
3183 whatever). Otherwise the password may appear in the log. Following change
3184 PH/42 below, there is no longer a chance of it appearing in a bounce
3185 message.
49c2d5ea 3186
bf759a8b
PH
3187PH/11 Installed exipick version 20050225.0 from John Jetmore.
3188
83364d30
PH
3189PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3190 of its addresses was ever tried. (Bugzilla bug #2.)
3191
7999bbd7
PH
3192PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3193 the result incorrectly in the debug output. (It correctly added a newline
3194 to what was transported.)
3195
7dbf77c9
PH
3196TF/01 Added $received_time.
3197
74e0617f
PH
3198PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3199 commented out examples of how to interface to a virus scanner and to
3200 SpamAssassin. Also added commented examples of av_scanner and
3201 spamd_address settings.
3202
2f079f46
PH
3203PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3204 and controls are allowed in which ACLs. There were a couple of minor
3205 errors. Some of the entries in the conditions table (which is a table of
3206 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3207 negation of where the condition IS allowed.
3208
8c841523
PH
3209PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3210
7766a4f0
PH
3211PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3212 header file does not have a version number, so I've had to invent a new
3213 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3214 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3215 radiusclient), but it was contributed by a Radius user.
3216
8b417f2c
PH
3217PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3218 files or directories, for OpenSSL.
3219
901f42cb
PH
3220PH/19 When an Exim process that is running as root has to create an Exim log
3221 file, it does so in a subprocess that runs as exim:exim so as to get the
3222 ownership right at creation (otherwise, other Exim processes might see
3223 the file with the wrong ownership). There was no test for failure of this
3224 fork() call, which would lead to the process getting stuck as it waited
3225 for a non-existent subprocess. Forks do occasionally fail when resources
3226 run out. I reviewed all the other calls to fork(); they all seem to check
3227 for failure.
3228
f9b9210e
PH
3229PH/20 When checking for unexpected SMTP input at connect time (before writing
3230 the banner), Exim was not dealing correctly with a non-positive return
3231 from the read() function. If the client had disconnected by this time,
3232 the result was a log entry for a synchronization error with an empty
3233 string after "input=" when read() returned zero. If read() returned -1
3234 (an event I could not check), uninitialized data bytes were printed.
3235 There were reports of junk text (parts of files, etc) appearing after
3236 "input=".
3237
54cdb463
PH
3238PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3239
cf00dad6
PH
3240PH/22 Added support for macro redefinition, and (re)definition in between
3241 driver and ACL definitions.
3242
acb1b346
PH
3243PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3244 forgetting to use the resulting value; it was using the unexpanded value.
3245
c5ddb310
PH
3246PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3247 hadn't been configured. The fix is from Juergen Kreileder, who
3248 understands it better than I do:
3249
3250 "Here's what I see happening with three configured cyrus_sasl
3251 authenticators configured (plain, login, cram-md5):
3252
3253 On startup auth_cyrus_sasl_init() gets called for each of these.
3254 This means three calls to sasl_listmech() without a specified mech_list.
3255 => SASL tests which mechs of all available mechs actually work
3256 => three warnings about OTP not working
3257 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3258
3259 With the patch, sasl_listmech() also gets called three times. But now
3260 SASL's mech_list option is set to the server_mech specified in the the
3261 authenticator. Or in other words, the answer from sasl_listmech()
3262 gets limited to just the mech you're testing for (which is different
3263 for each call.)
3264 => the return list contains just 'plain' or 'login', 'cram-md5' or
3265 nothing depending on the value of ob->server_mech.
3266
3267 I've just tested the patch: Authentication still works fine,
3268 unavailable mechs specified in the exim configuration are still
3269 caught, and the auth.log warnings about OTP are gone."
3270
31619da6
PH
3271PH/25 When debugging is enabled, the contents of the command line are added
3272 to the debugging output, even when log_selector=+arguments is not
3273 specified.
3274
bebaf0fc
PH
3275PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3276 answer is "GNU", and only if the return is "GNU/something" is the answer
3277 "Linux".
3278
475fe28a
PH
3279PH/27 $acl_verify_message is now set immediately after the failure of a
3280 verification in an ACL, and so is available in subsequent modifiers. In
3281 particular, the message can be preserved by coding like this:
3282
3283 warn !verify = sender
3284 set acl_m0 = $acl_verify_message
3285
3286 Previously, $acl_verify_message was set only while expanding "message"
3287 and "log_message" when a very denied access.
3288
7e8bec7a
PH
3289PH/28 Modified OS/os.c-Linux with
3290
3291 -#ifndef OS_LOAD_AVERAGE
3292 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3293
3294 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3295 nomenclature these days.)
3296
e4a89c47
PH
3297PH/29 Installed patch from the Sieve maintainer that adds the options
3298 sieve_useraddress and sieve_subaddress to the redirect router.
3299
5ca2a9a1
PH
3300PH/30 In these circumstances:
3301 . Two addresses routed to the same list of hosts;
3302 . First host does not offer TLS;
3303 . First host accepts first address;
3304 . First host gives temporary error to second address;
3305 . Second host offers TLS and a TLS session is established;
3306 . Second host accepts second address.
3307 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3308 and peerdn, if requested) that were in fact used only for the second
3309 address.
7e8bec7a 3310
c688b954
PH
3311PH/31 When doing a callout as part of verifying an address, Exim was not paying
3312 attention to any local part prefix or suffix that was matched by the
3313 router that accepted the address. It now behaves in the same way as it
3314 does for delivery: the affixes are removed from the local part unless
3315 rcpt_include_affixes is set on the transport.
3316
fed77020
PH
3317PH/32 Add the sender address, as F=<...>, to the log line when logging a
3318 timeout during the DATA phase of an incoming message.
3319
7fe1560f
PH
3320PH/33 Sieve envelope tests were broken for match types other than :is. I have
3321 applied a patch sanctioned by the Sieve maintainer.
c688b954 3322
ebb6e6d5
PH
3323PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3324 the uid or gid is negative. A case of a negative gid caused this to be
3325 noticed. The fix allows for either to be negative.
3326
9c4e8f60
PH
3327PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3328 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3329 overlooked.
3330
3331PH/36 The change PH/12 above was broken. Fixed it.
3332
d7174846
PH
3333PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3334 the grounds that routing the same address twice would always produce the
3335 same answer. This might have been true once, but it is certainly no
3336 longer true now. Routing a child address may depend on the previous
3337 routing that produced that child. Some complicated redirection strategies
3338 went wrong when messages had multiple recipients, and made Exim's
3339 behaviour dependent on the order in which the addresses were given.
3340
3341 I have moved the duplicate checking until after the routing is complete.
3342 Exim scans the addresses that are assigned to local and remote
3343 transports, and removes any duplicates. This means that more work will be
3344 done, as duplicates will always all be routed, but duplicates are
3345 presumably rare, so I don't expect this is of any significance.
3346
3347 For deliveries to pipes, files, and autoreplies, the duplicate checking
3348 still happens during the routing process, since they are not going to be
3349 routed further.
3350
cfe75fc3
PH
3351PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3352 It corrects a timeout issue with spamd. This is Ian's comment: "The
3353 background is that sometimes spamd either never reads data from a
3354 connection it has accepted, or it never writes response data. The exiscan
3355 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3356 blindly assumes that writes won't block so it may never time out."
3357
be22d70e
PH
3358PH/39 Allow G after quota size as well as K and M.
3359
0612b098
PH
3360PH/40 The value set for $authenticated_id in an authenticator may not contain
3361 binary zeroes or newlines because the value is written to log lines and
3362 to spool files. There was no check on this. Now the value is run through
3363 the string_printing() function so that such characters are converted to
3364 printable escape sequences.
3365
2e0c1448
PH
3366PH/41 $message_linecount is a new variable that contains the total number of
3367 lines in the message. Compare $body_linecount, which is the count for the
3368 body only.
3369
447d236c
PH
3370PH/42 Exim no longer gives details of delivery errors for specific addresses in
3371 bounce and delay warning messages, except in certain special cases, which
3372 are as follows: