Revert "Lower EXIM_CLIENT_DH_MIN_BITS 1024 -> 512."
[exim.git] / doc / doc-txt / ChangeLog
CommitLineData
495ae4b0
PH
1Change log file for Exim from version 4.21
2-------------------------------------------
3
b1770b6e 4Exim version 4.80
0599f9cf
PP
5-----------------
6
7PP/01 Handle short writes when writing local log-files.
8 In practice, only affects FreeBSD (8 onwards).
9 Bugzilla 1053, with thanks to Dmitry Isaikin.
10
23c7e742
NM
11NM/01 Bugzilla 949 - Documentation tweak
12
b322aac8
NM
13NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
14 improved.
15
4a891427
NM
16NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
17
c1e794ba 18PP/02 Implemented gsasl authenticator.
b322aac8 19
97753960
PP
20PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
21
22PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
23 `pkg-config foo` for cflags/libs.
24
df6303fa
PP
25PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
26 with rest of GSASL and with heimdal_gssapi.
27
7e6a8985
PP
28PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
29 `pkg-config foo` for cflags/libs for the TLS implementation.
30
f1e05cc7 31PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
20aa9dbd
PP
32 properties get this fed in as external SSF. A number of robustness
33 and debugging improvements to the cyrus_sasl authenticator.
b322aac8 34
4c287009
PP
35PP/08 cyrus_sasl server now expands the server_realm option.
36
b98bb9ac
PP
37PP/09 Bugzilla 1214 - Log authentication information in reject log.
38 Patch by Jeremy Harris.
39
4a6a987a
PP
40PP/10 Added dbmjz lookup type.
41
c45dd180 42PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
c7955b11 43
7db8d074
PP
44PP/12 MAIL args handles TAB as well as SP, for better interop with
45 non-compliant senders.
46 Analysis and variant patch by Todd Lyons.
47
eae0036b 48NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
e0df1c83
DM
49 Bug report from Lars Müller <lars@samba.org> (via SUSE),
50 Patch from Dirk Mueller <dmueller@suse.com>
51
dec5017e
PP
52PP/13 tls_peerdn now print-escaped for spool files.
53 Observed some $tls_peerdn in wild which contained \n, which resulted
54 in spool file corruption.
55
c80c5570
PP
56PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
57 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
58 or write after TLS renegotiation, which otherwise led to messages
59 "Got SSL error 2".
60
076b11e2
PP
61TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
62 as a tracking header (ie: a signed header comes before the signature).
63 Patch from Wolfgang Breyha.
64
5407bfff
JH
65JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
66 comma-sep list; embedded commas doubled.
67
9e45c72b
PP
68JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
69
e74376d8
PP
70PP/15 LDAP: Check for errors of TLS initialisation, to give correct
71 diagnostics.
72 Report and patch from Dmitry Banschikov.
73
da3ad30d
PP
74PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
75 Removed SSL_clear() after SSL_new() which led to protocol negotiation
76 failures. We appear to now support TLS1.1+ with Exim.
77
7be682ca
PP
78PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
79 lets Exim select keys and certificates based upon TLS SNI from client.
3f0945ff
PP
80 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
81 before an outbound SMTP session. New log_selector, +tls_sni.
7be682ca 82
ef840681
PP
83PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
84 NULL dereference. Report and patch from Alun Jones.
85
5bfb4cdf
PP
86PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
87 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
88 Not seeing resolver debug output on NetBSD, but suspect this is a
89 resolver implementation change.
90
c6e95d22
PP
91PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
92 Left warnings. Added "eximon gdb" invocation mode.
93
9cbad13b
PP
94PP/21 Defaulting "accept_8bitmime" to true, not false.
95
9ee44efb
PP
96PP/22 Added -bw for inetd wait mode support.
97
6a6084f8
PP
98PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
99 locate the relevant includes and libraries. Made this the default.
100
12dd53c7
PP
101PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
102 Bugzilla 1246, report and most of solution from Tomasz Kusy.
103
9e45c72b 104JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
97d17305
JH
105 This may cause build issues on older platforms.
106
17c76198
PP
107PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
108 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
109 gnutls_require_mac & gnutls_require_protocols (no longer supported).
110 Added SNI support via GnuTLS too.
af3498d6 111 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
17c76198 112
53947857 113PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
3f7eeb86 114
eae0036b 115PP/27 Applied dnsdb SPF support patch from Janne Snabb.
8ee4b30e
PP
116 Applied second patch from Janne, implementing suggestion to default
117 multiple-strings-in-record handling to match SPF spec.
eae0036b 118
9e45c72b 119JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
2605c55b 120
7390e768
PP
121PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
122 read-only, out of scope).
123 Patch from Wolfgang Breyha, report from Stuart Northfield.
124
08488c86
PP
125PP/29 Fix three issues highlighted by clang analyser static analysis.
126 Only crash-plausible issue would require the Cambridge-specific
127 iplookup router and a misconfiguration.
128 Report from Marcin Mirosław.
129
6475bd82
PP
130PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
131
81f91683
PP
132PP/31 %D in printf continues to cause issues (-Wformat=security), so for
133 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
134 As part of this, removing so much warning spew let me fix some minor
135 real issues in debug logging.
136
5779e6aa
PP
137PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
138 assignment on my part. Fixed.
139
3375e053
PP
140PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
141 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
142 Janne Snabb (who went above and beyond: thank you).
143
144PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
145 string otherwise requires a connection and a bunch more work and it's
78e0c7a3
PP
146 relatively easy to get wrong. Should also expose TLS library linkage
147 problems.
3375e053 148
9d26b8c0
PP
149PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
150 64-bit ${eval} (JH/03).
151
57eb9e91 152PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
b87a6e0e
PP
153 GNU libc to support some of the 64-bit stuff, should not lead to
154 conflicts. Defined before os.h is pulled in, so if a given platform
155 needs to override this, it can.
156
16880d1a
PP
157PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
158 protection layer was required, which is not implemented.
159 Bugzilla 1254, patch from Wolfgang Breyha.
160
a799883d
PP
161PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
162 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
163 tls_dhparam take prime identifiers. Also unbreak combination of
164 OpenSSL+DH_params+TLSSNI.
165
3ecab157 166PP/39 Disable SSLv2 by default in OpenSSL support.
f0f5a555 167
0599f9cf 168
867fcbf5
PP
169Exim version 4.77
170-----------------
171
172PP/01 Solaris build fix for Oracle's LDAP libraries.
173 Bugzilla 1109, patch from Stephen Usher.
174
f1a29782
TF
175TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
176
ab42bd23
TK
177TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
178 whitespace trailer
867fcbf5 179
0ca0cf52
TF
180TF/02 Fix a couple more cases where we did not log the error message
181 when unlink() failed. See also change 4.74-TF/03.
182
921b12ca
TF
183TF/03 Make the exiwhat support code safe for signals. Previously Exim might
184 lock up or crash if it happened to be inside a call to libc when it
185 got a SIGUSR1 from exiwhat.
186
187 The SIGUSR1 handler appends the current process status to the process
188 log which is later printed by exiwhat. It used to use the general
189 purpose logging code to do this, but several functions it calls are
190 not safe for signals.
191
192 The new output code in the SIGUSR1 handler is specific to the process
193 log, and simple enough that it's easy to inspect for signal safety.
194 Removing some special cases also simplifies the general logging code.
195 Removing the spurious timestamps from the process log simplifies
196 exiwhat.
197
c99ce5c9
TF
198TF/04 Improved ratelimit ACL condition.
199
200 The /noupdate option has been deprecated in favour of /readonly which
201 has clearer semantics. The /leaky, /strict, and /readonly update modes
202 are mutually exclusive. The update mode is no longer included in the
203 database key; it just determines when the database is updated. (This
204 means that when you upgrde Exim will forget old rate measurements.)
205
206 Exim now checks that the per_* options are used with an update mode that
207 makes sense for the current ACL. For example, when Exim is processing a
208 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
209 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
210 must specify per_mail/readonly. If you omit the update mode it defaults to
211 /leaky where that makes sense (as before) or /readonly where required.
212
213 The /noupdate option is now undocumented but still supported for
214 backwards compatibility. It is equivalent to /readonly except that in
215 ACLs where /readonly is required you may specify /leaky/noupdate or
216 /strict/noupdate which are treated the same as /readonly.
217
218 A useful new feature is the /count= option. This is a generalization
219 of the per_byte option, so that you can measure the throughput of other
220 aggregate values. For example, the per_byte option is now equivalent
221 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
222
223 The per_rcpt option has been generalized using the /count= mechanism
224 (though it's more complicated than the per_byte equivalence). When it is
225 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
226 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
227 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
228 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
229 Note that using per_rcpt with a non-readonly update mode in more than
230 one ACL will cause the recipients to be double-counted. (The per_mail
231 and per_byte options don't have this problem.)
232
233 The handling of very low rates has changed slightly. If the computed rate
234 is less than the event's count (usually one) then this event is the first
235 after a long gap. In this case the rate is set to the same as this event's
236 count, so that the first message of a spam run is counted properly.
237
238 The major new feature is a mechanism for counting the rate of unique
239 events. The new per_addr option counts the number of different
240 recipients that someone has sent messages to in the last time period. It
241 behaves like per_rcpt if all the recipient addresses are different, but
242 duplicate recipient addresses do not increase the measured rate. Like
243 the /count= option this is a general mechanism, so the per_addr option
244 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
245 example, measure the rate that a client uses different sender addresses
246 with the options per_mail/unique=$sender_address. There are further
247 details in the main documentation.
248
3634fc25
TF
249TF/05 Removed obsolete $Cambridge$ CVS revision strings.
250
792e8a19
TF
251TF/06 Removed a few PCRE remnants.
252
5901f0ab
TF
253TF/07 Automatically extract Exim's version number from tags in the git
254 repository when doing development or release builds.
255
7f2a2a43
PP
256PP/02 Raise smtp_cmd_buffer_size to 16kB.
257 Bugzilla 879. Patch from Paul Fisher.
e2ca7082 258
061b7ebd
PP
259PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
260 Heavily based on revision 40f9a89a from Simon Arlott's tree.
261 Bugzilla 97.
262
e12f8c32
PP
263PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
264
9e949f00 265PP/05 Variable $av_failed, true if the AV scanner deferred.
7f2a2a43
PP
266 Bugzilla 1078. Patch from John Horne.
267
268PP/06 Stop make process more reliably on build failure.
269 Bugzilla 1087. Patch from Heiko Schlittermann.
9e949f00 270
555ae6af 271PP/07 Make maildir_use_size_file an _expandable_ boolean.
ac53fcda
PP
272 Bugzilla 1089. Patch from Heiko Schlittermann.
273
274PP/08 Handle ${run} returning more data than OS pipe buffer size.
275 Bugzilla 1131. Patch from Holger Weiß.
555ae6af 276
6f7fe114
PP
277PP/09 Handle IPv6 addresses with SPF.
278 Bugzilla 860. Patch from Wolfgang Breyha.
279
c566dd90
PP
280PP/10 GnuTLS: support TLS 1.2 & 1.1.
281 Bugzilla 1156.
89f897c3
PP
282 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
283 Bugzilla 1095.
c566dd90 284
d6cc7c78 285PP/11 match_* no longer expand right-hand-side by default.
39257585
PP
286 New compile-time build option, EXPAND_LISTMATCH_RHS.
287 New expansion conditions, "inlist", "inlisti".
288
0d0e4455
PP
289PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
290
3399bb60 291PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
d690cbdc
PP
292
293PP/14 fix log_write() format string regression from TF/03.
294 Bugzilla 1152. Patch from Dmitry Isaikin.
295
0ca0cf52 296
10906672
PP
297Exim version 4.76
298-----------------
299
300PP/01 The new ldap_require_cert option would segfault if used. Fixed.
301
754a0503
PP
302PP/02 Harmonised TLS library version reporting; only show if debugging.
303 Layout now matches that introduced for other libraries in 4.74 PP/03.
304
c0c7b2da
PP
305PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
306
e97d1f08
PP
307PP/04 New "dns_use_edns0" global option.
308
084c1d8c
PP
309PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
310 Bugzilla 1098.
311
4e7ee012
PP
312PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
313 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
da80c2a8 314
c8d52a00
PP
315TK/01 Updated PolarSSL code to 0.14.2.
316 Bugzilla 1097. Patch from Andreas Metzler.
317
54e7ce4a
PP
318PP/07 Catch divide-by-zero in ${eval:...}.
319 Fixes bugzilla 1102.
320
5ee6f336
PP
321PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
322 Bugzilla 1104.
323
c8d52a00 324TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
6ea4a851
PP
325 format-string attack -- SECURITY: remote arbitrary code execution.
326
327TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
328 time unintentionally subject to list matching rules, letting the header
329 cause arbitrary Exim lookups (of items which can occur in lists, *not*
330 arbitrary string expansion). This allowed for information disclosure.
331
332PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
333 INT_MIN/-1 -- value coerced to INT_MAX.
c8d52a00 334
10906672 335
aa097c4c
NM
336Exim version 4.75
337-----------------
338
4c9ef03a 339NM/01 Workround for PCRE version dependency in version reporting
aa097c4c
NM
340 Bugzilla 1073
341
7f3d9eff
TF
342TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
343 This fixes portability to compilers other than gcc, notably
344 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
345
159f52d2
TF
346TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
347 makefiles for portability to HP-UX and POSIX correctness.
348
0cc9542a
PP
349PP/01 Permit LOOKUP_foo enabling on the make command-line.
350 Also via indented variable definition in the Makefile.
351 (Debugging by Oliver Heesakkers).
352
f7274286
PP
353PP/02 Restore caching of spamd results with expanded spamd_address.
354 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
355
7b797365
PP
356PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
357 Improves build reliability. Fix from: Frank Elsner
358
caacae52
NM
359NM/02 Fix wide character breakage in the rfc2047 coding
360 Fixes bug 1064. Patch from Andrey N. Oktyabrski
361
09dcaba9
NM
362NM/03 Allow underscore in dnslist lookups
363 Fixes bug 1026. Patch from Graeme Fowler
364
bc19a55b
PP
365PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
366 Code patches from Adam Ciarcinski of NetBSD.
caacae52 367
bd4c9759
NM
368NM/04 Fixed exiqgrep to cope with mailq missing size issue
369 Fixes bug 943.
370
b72aab72
PP
371PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
372 is logged, to avoid truncation. Patch from John Horne.
373
2fe76745
PP
374PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
375 Patch from Jakob Hirsch.
376
76aa570c
PP
377PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
378 SQL string expansion failure details.
379 Patch from Andrey Oktyabrski.
380
f1e5fef5
PP
381PP/08 Bugzilla 486: implement %M datestamping in log filenames.
382 Patch from Simon Arlott.
383
4d805ee9
PP
384PP/09 New lookups functionality failed to compile on old gcc which rejects
385 extern declarations in function scope.
386 Patch from Oliver Fleischmann
387
cd59ab18
PP
388PP/10 Use sig_atomic_t for flags set from signal handlers.
389 Check getgroups() return and improve debugging.
390 Fixed developed for diagnosis in bug 927 (which turned out to be
391 a kernel bug).
392
332f5cf3
PP
393PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
394 Patch from Mark Zealey.
395
29cfeb94
PP
396PP/12 Bugzilla 1056: Improved spamd server selection.
397 Patch from Mark Zealey.
398
660242ad
PP
399PP/13 Bugzilla 1086: Deal with maildir quota file races.
400 Based on patch from Heiko Schlittermann.
401
bc4bc4c5
PP
402PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
403 Patch from Uwe Doering, sign-off by Michael Haardt.
404
2e64baa9
NM
405NM/05 Fix to spam.c to accommodate older gcc versions which dislike
406 variable declaration deep within a block. Bug and patch from
407 Dennis Davis.
408
bddd7526
PP
409PP/15 lookups-Makefile IRIX compatibilty coercion.
410
6bac1a9a
PP
411PP/16 Make DISABLE_DKIM build knob functional.
412
552193f0
NM
413NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
414 Patch by Simon Arlott
baeee2c1 415
1b587e48
TF
416TF/03 Fix valgrind.h portability to C89 compilers that do not support
417 variable argument macros. Our copy now differs from upstream.
418
aa097c4c 419
8c07b69f
TF
420Exim version 4.74
421-----------------
422
423TF/01 Failure to get a lock on a hints database can have serious
424 consequences so log it to the panic log.
425
c0ea85ab
TF
426TF/02 Log LMTP confirmation messages in the same way as SMTP,
427 controlled using the smtp_confirmation log selector.
428
0761d44e
TF
429TF/03 Include the error message when we fail to unlink a spool file.
430
0a349494
PP
431DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
432 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
433 for maintaining out-of-tree patches for some time.
434
435PP/01 Bugzilla 139: Documentation and portability issues.
436 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
437 Handle per-OS dynamic-module compilation flags.
438
fea24b2e
PP
439PP/02 Let /dev/null have normal permissions.
440 The 4.73 fixes were a little too stringent and complained about the
441 permissions on /dev/null. Exempt it from some checks.
442 Reported by Andreas M. Kirchwitz.
443
6545de78
PP
444PP/03 Report version information for many libraries, including
445 Exim version information for dynamically loaded libraries. Created
446 version.h, now support a version extension string for distributors
447 who patch heavily. Dynamic module ABI change.
448
1670ef10
PP
449PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
450 privilege escalation vulnerability whereby the Exim run-time user
451 can cause root to append content of the attacker's choosing to
452 arbitrary files.
453
c0886197
PP
454PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
455 (Wolfgang Breyha)
456
b7487bce
PP
457PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
458 If dropping privileges for untrusted macros, we disabled normal logging
459 on the basis that it would fail; for the Exim run-time user, this is not
460 the case, and it resulted in successful deliveries going unlogged.
461 Fixed. Reported by Andreas Metzler.
462
8c07b69f 463
97fd1e48 464Exim version 4.73
ed7f7860 465-----------------
97fd1e48
PP
466
467PP/01 Date: & Message-Id: revert to normally being appended to a message,
468 only prepend for the Resent-* case. Fixes regression introduced in
469 Exim 4.70 by NM/22 for Bugzilla 607.
470
6901c596
PP
471PP/02 Include check_rfc2047_length in configure.default because we're seeing
472 increasing numbers of administrators be bitten by this.
473
a8c8d6b5
JJ
474JJ/01 Added DISABLE_DKIM and comment to src/EDITME
475
77bb000f
PP
476PP/03 Bugzilla 994: added openssl_options main configuration option.
477
a29e5231
PP
478PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
479
ec5a0394 480PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
a29e5231 481
55c75993
PP
482PP/06 Adjust NTLM authentication to handle SASL Initial Response.
483
453a6645 484PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
ec5a0394
PP
485 without a peer certificate, leading to a segfault because of an
486 assumption that peers always have certificates. Be a little more
453a6645
PP
487 paranoid. Problem reported by Martin Tscholak.
488
8544e77a
PP
489PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
490 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
491 NB: ClamAV planning to remove STREAM in "middle of 2010".
3346ab01
PP
492 CL also introduces -bmalware, various -d+acl logging additions and
493 more caution in buffer sizes.
8544e77a 494
83e029d5
PP
495PP/09 Implemented reverse_ip expansion operator.
496
ed7f7860
PP
497PP/10 Bugzilla 937: provide a "debug" ACL control.
498
7d9f747b
PP
499PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
500
4b2241d2
PP
501PP/12 Bugzilla 973: Implement --version.
502
10385c15
PP
503PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
504
dbc4b90d
PP
505PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
506
532be449
PP
507PP/15 Bugzilla 816: support multiple condition rules on Routers.
508
6a8de854 509PP/16 Add bool_lax{} expansion operator and use that for combining multiple
71265ae9
PP
510 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
511 ignore trailing whitespace.
6a8de854 512
5dc43717
JJ
513JJ/02 prevent non-panic DKIM error from being sent to paniclog
514
515JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
516 "exim" to be used
55c75993 517
3346ab01
PP
518PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
519 Notification from Dr Andrew Aitchison.
520
491fab4c
PP
521PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
522 ExtendedDetectionInfo response format.
523 Notification from John Horne.
524
13eb9497
PP
525PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
526 compatible.
527
528PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
529 XSL and documented dependency on system catalogs, with examples of how
530 it normally works.
531
7f36d675
DW
532DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
533 access.
534
c1d94452
DW
535DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
536 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
537 configuration file which is writeable by the Exim user or group.
538
e2f5dc15
DW
539DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
540 of configuration files to cover files specified with the -C option if
541 they are going to be used with root privileges, not just the default
542 configuration file.
543
cd25e41d
DW
544DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
545 option (effectively making it always true).
546
261dc43e
DW
547DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
548 files to be used while preserving root privileges.
549
fa32850b
DW
550DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
551 that rogue child processes cannot use them.
552
79d4bc3d
PP
553PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
554 run-time user, instead of root.
555
43236f35 556PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2cfd3221
PP
557 Exim run-time user without dropping privileges.
558
fb08281f
DW
559DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
560 result string, instead of calling string_vformat() twice with the same
561 arguments.
3346ab01 562
74935b98
DW
563DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
564 for other users. Others should always drop root privileges if they use
565 -C on the command line, even for a whitelisted configure file.
566
90b6341f
DW
567DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
568
57730b52
ML
569NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
570
66581d1e 571
465e92cf
JJ
572Exim version 4.72
573-----------------
574
453a6645
PP
575JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
576 $data_path, and $header_path variables; fixed documentation bugs and
577 typos
465e92cf 578
453a6645
PP
579JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
580 exipick to access non-standard spools, including the "frozen" queue
581 (Finput)
edae0343 582
9bd3e22c
NM
583NM/01 Bugzilla 965: Support mysql stored procedures.
584 Patch from Alain Williams
585
bb576ff7
NM
586NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
587
5a1a5845
NM
588NM/03 Bugzilla 955: Documentation fix for max_rcpts.
589 Patch from Andreas Metzler
590
981a9fad
NM
591NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
592 Patch from Kirill Miazine
593
7fc497ee
NM
594NM/05 Bugzilla 671: Added umask to procmail example.
595
1a41defa
JJ
596JJ/03 installed exipick 20100323.0, fixing doc bug
597
a466095c 598NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
b26eacf1 599 directory. Notification and patch from Dan Rosenberg.
a466095c 600
94a6bd0b
NM
601TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
602
603TK/02 Improve log output when DKIM signing operation fails.
604
605MH/01 Treat the transport option dkim_domain as a colon separated
606 list, not as a single string, and sign the message with each element,
607 omitting multiple occurences of the same signer.
608
c1b141a8
NM
609NM/07 Null terminate DKIM strings, Null initialise DKIM variable
610 Bugzilla 985, 986. Patch by Simon Arlott
94a6bd0b 611
b26eacf1 612NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
0d0c6357
NM
613 Patch by Simon Arlott
614
179c5980 615PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
b26eacf1 616 MBX locking. Notification from Dan Rosenberg.
179c5980 617
9bd3e22c 618
7c6d71af
NM
619Exim version 4.71
620-----------------
621
7d9f747b 622TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
7c6d71af 623
f013fb92
NM
624NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
625
0eb8eedd
NM
626NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
627
663ee6d9
NM
628NM/03 Bugzilla 847: Enable DNSDB lookup by default.
629
177ebd9b
NM
630NM/04 Bugzilla 915: Flag broken perl installation during build.
631
7c6d71af 632
210f147e
NM
633Exim version 4.70
634-----------------
635
cdd3bb85 636TK/01 Added patch by Johannes Berg that expands the main option
e739e3d9 637 "spamd_address" if it starts with a dollar sign.
cdd3bb85
TK
638
639TK/02 Write list of recipients to X-Envelope-Sender header when building
640 the mbox-format spool file for content scanning (suggested by Jakob
7d9f747b 641 Hirsch).
cdd3bb85
TK
642
643TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
644 (http://www.dcc-servers.net/) support via dccifd. Activated by
e739e3d9 645 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
cdd3bb85
TK
646
647TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
648 by Mark Daniel Reidel <mr@df.eu>.
649
210f147e
NM
650NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
651 When building exim an external PCRE library is now needed -
652 PCRE is a system library on the majority of modern systems.
653 See entry on PCRE_LIBS in EDITME file.
654
deafd5b3
NM
655NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
656 conversation. Added nologin parameter to request.
7d9f747b 657 Patch contributed by Kirill Miazine.
deafd5b3 658
089793a4
TF
659TF/01 Do not log submission mode rewrites if they do not change the address.
660
5f16ca82
TF
661TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
662
dae9d94e 663NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
7d9f747b 664 log files in place. Contributed by Roberto Lima.
dae9d94e 665
7d9f747b 666NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
3f0da4d0 667
06864c44
TF
668TF/03 Bugzilla 615: When checking the local_parts router precondition
669 after a local_part_suffix or local_part_prefix option, Exim now
670 does not use the address's named list lookup cache, since this
671 contains cached lookups for the whole local part.
672
65a7d8c3 673NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
7d9f747b 674 Robert Millan. Documentation is in experimental-spec.txt.
65a7d8c3 675
23510047 676TF/04 Bugzilla 668: Fix parallel build (make -j).
65a7d8c3 677
7d9f747b 678NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
5f28a6e8 679
7d8eec3a 680NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
7d9f747b 681 Patch provided by Jan Srzednicki.
6c588e74 682
89dec7b6
TF
683TF/05 Leading white space used to be stripped from $spam_report which
684 wrecked the formatting. Now it is preserved.
5f28a6e8 685
a99de90c
TF
686TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
687 that they are available at delivery time.
688
e2803e40
TF
689TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
690
7199e1ee
TF
691TF/08 TLS error reporting now respects the incoming_interface and
692 incoming_port log selectors.
693
e276e04b
TF
694TF/09 Produce a more useful error message if an SMTP transport's hosts
695 setting expands to an empty string.
696
ce552449 697NM/06 Bugzilla 744: EXPN did not work under TLS.
7d9f747b 698 Patch provided by Phil Pennock.
ce552449 699
e765a0f1 700NM/07 Bugzilla 769: Extraneous comma in usage fprintf
7d9f747b 701 Patch provided by Richard Godbee.
e765a0f1 702
4f054c63 703NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
447de4b0 704 acl_smtp_notquit, added index entry.
4f054c63 705
7d9f747b
PP
706NM/09 Bugzilla 787: Potential buffer overflow in string_format.
707 Patch provided by Eugene Bujak.
24c929a2 708
7d9f747b
PP
709NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
710 accept(). Patch provided by Maxim Dounin.
cf73943b 711
b52bc06e 712NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
7d9f747b 713 Patch provided by Phil Pennock.
b52bc06e 714
447de4b0
NM
715NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
716
4c69d561 717NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
7d9f747b 718 Patch provided by Brad "anomie" Jorsch.
4c69d561 719
d5c39246 720NM/14 Bugzilla 622: Added timeout setting to transport filter.
7d9f747b 721 Patch provided by Dean Brooks.
9b989985 722
0b23848a
TK
723TK/05 Add native DKIM support (does not depend on external libraries).
724
8f3414a1 725NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
7d9f747b 726 Patch provided by Graeme Fowler.
e2aacdfd 727
fb6f955d
NM
728NM/16 Bugzilla 851: Documentation example syntax fix.
729
730NM/17 Changed NOTICE file to remove references to embedded PCRE.
8f3414a1 731
7d9f747b
PP
732NM/18 Bugzilla 894: Fix issue with very long lines including comments in
733 lsearch.
dbb0bf41 734
7d9f747b
PP
735NM/19 Bugzilla 745: TLS version reporting.
736 Patch provided by Phil Pennock.
f3766eb5 737
7d9f747b
PP
738NM/20 Bugzilla 167: bool: condition support.
739 Patch provided by Phil Pennock.
36f12725 740
7d9f747b
PP
741NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
742 clients. Patch provided by Phil Pennock.
e6060e2c 743
7d9f747b
PP
744NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
745 Patch provided by Brad "anomie" Jorsch.
5eb690a1 746
7d9f747b
PP
747NM/23 Bugzilla 687: Fix misparses in eximstats.
748 Patch provided by Heiko Schlittermann.
d5c13d66 749
7d9f747b
PP
750NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
751 Patch provided by Heiko Schlittermann.
b2335c0b 752
7d9f747b 753NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1da77999 754 plus update to original patch.
f4cd9433 755
7d9f747b 756NM/26 Bugzilla 799: Documentation correction for ratelimit.
dc988b7e 757
7d9f747b
PP
758NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
759 Patch provided by David Brownlee.
8dc71ab3 760
7d9f747b 761NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
400eda43 762
7d9f747b 763NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
ec5a421b 764
7d9f747b 765NM/30 Bugzilla 888: TLS documentation bugfixes.
07af267e 766
7d9f747b 767NM/31 Bugzilla 896: Dovecot buffer overrun fix.
51473862 768
17792b53 769NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
7d9f747b 770 Unlike the original bugzilla I have changed all shell scripts in src tree.
17792b53 771
7d9f747b
PP
772NM/33 Bugzilla 898: Transport filter timeout fix.
773 Patch by Todd Rinaldo.
52383f8f 774
7d9f747b
PP
775NM/34 Bugzilla 901: Fix sign/unsigned and UTF mistmatches.
776 Patch by Serge Demonchaux.
5ca6d115 777
7d9f747b
PP
778NM/35 Bugzilla 39: Base64 decode bug fixes.
779 Patch by Jakob Hirsch.
baee9eee 780
7d9f747b 781NM/36 Bugzilla 909: Correct connect() call in dcc code.
e93a964c 782
7d9f747b 783NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
9bf3d68f 784
7d9f747b 785NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
96535b98 786
7d9f747b 787NM/39 Bugzilla 911: Fixed MakeLinks build script.
30339e0f 788
deafd5b3 789
47db1125
NM
790Exim version 4.69
791-----------------
792
4b3504d0
TK
793TK/01 Add preliminary DKIM support. Currently requires a forked version of
794 ALT-N's libdkim that I have put here:
795 http://duncanthrax.net/exim-experimental/
796
797 Note to Michael Haardt: I had to rename some vars in sieve.c. They
798 were called 'true' and it seems that C99 defines that as a reserved
799 keyword to be used with 'bool' variable types. That means you could
800 not include C99-style headers which use bools without triggering
801 build errors in sieve.c.
802
81ea09ca
NM
803NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
804 as mailq or other aliases. Changed the --help handling significantly
805 to do whats expected. exim_usage() emits usage/help information.
806
f13cddcb
SC
807SC/01 Added the -bylocaldomain option to eximstats.
808
7d9f747b 809NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
8ad076b2 810
7d9f747b 811NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
a843aaa6 812
7d9f747b 813NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
47db1125
NM
814
815
eb4c0de6
PH
816Exim version 4.68
817-----------------
818
819PH/01 Another patch from the Sieve maintainer.
820
6a3bceb1
PH
821PH/02 When an IPv6 address is converted to a string for single-key lookup
822 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
823 dots are used instead of colons so that keys in lsearch files need not
824 contain colons. This was done some time before quoting was made available
825 in lsearch files. However, iplsearch files do require colons in IPv6 keys
826 (notated using the quote facility) so as to distinguish them from IPv4
827 keys. This meant that lookups for IP addresses in host lists did not work
828 for iplsearch lookups.
829
830 This has been fixed by arranging for IPv6 addresses to be expressed with
831 colons if the lookup type is iplsearch. This is not incompatible, because
832 previously such lookups could never work.
833
834 The situation is now rather anomolous, since one *can* have colons in
835 ordinary lsearch keys. However, making the change in all cases is
836 incompatible and would probably break a number of configurations.
837
2e30fa9d
TK
838TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
839 version.
840
0806a9c5
MH
841MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
842 conversion specification without a maximum field width, thereby enabling
843 a rogue spamd server to cause a buffer overflow. While nobody in their
844 right mind would setup Exim to query an untrusted spamd server, an
845 attacker that gains access to a server running spamd could potentially
846 exploit this vulnerability to run arbitrary code as the Exim user.
847
ae276964
TK
848TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
849 $primary_hostname instead of what libspf2 thinks the hosts name is.
850
0f2cbd1b
MH
851MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
852 a directory entry by the name of the lookup key. Previously, if a
853 symlink pointed to a non-existing file or a file in a directory that
854 Exim lacked permissions to read, a lookup for a key matching that
855 symlink would fail. Now it is enough that a matching directory entry
856 exists, symlink or not. (Bugzilla 503.)
857
2b85bce7
PH
858PH/03 The body_linecount and body_zerocount variables are now exported in the
859 local_scan API.
860
93655c46
PH
861PH/04 Added the $dnslist_matched variable.
862
6c512171
PH
863PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
864 This means they are set thereafter only if the connection becomes
865 encrypted.
866
867PH/06 Added the client_condition to authenticators so that some can be skipped
868 by clients under certain conditions.
869
aa6dc513
PH
870PH/07 The error message for a badly-placed control=no_multiline_responses left
871 "_responses" off the end of the name.
872
a96603a0
PH
873PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
874
8f240103
PH
875PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
876 (without spaces) instead of just copying the configuration text.
877
878PH/10 Added the /noupdate option to the ratelimit ACL condition.
879
d677b2f2
PH
880PH/11 Added $max_received_linelength.
881
d52120f2
PH
882PH/12 Added +ignore_defer and +include_defer to host lists.
883
64f2600a
PH
884PH/13 Installed PCRE version 7.2. This needed some changes because of the new
885 way in which PCRE > 7.0 is built.
886
8669f003
PH
887PH/14 Implemented queue_only_load_latch.
888
a4dc33a8
PH
889PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
890 MAIL command. The effect was to mangle the value on 64-bit systems.
891
d6a60c0f
PH
892PH/16 Another patch from the Sieve maintainer.
893
8f128379
PH
894PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
895
8932dffe
PH
896PH/18 If a system quota error occurred while trying to create the file for
897 a maildir delivery, the message "Mailbox is full" was not appended to the
898 bounce if the delivery eventually timed out. Change 4.67/27 below applied
899 only to a quota excession during the actual writing of the file.
d6a60c0f 900
ddea74fa 901PH/19 It seems that peer DN values may contain newlines (and other non-printing
48ed62d9
PH
902 characters?) which causes problems in log lines. The DN values are now
903 passed through string_printing() before being added to log lines.
904
ddea74fa 905PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
b7670459
PH
906 and InterBase are left for another time.)
907
ddea74fa
PH
908PH/21 Added message_body_newlines option.
909
ce9f225c
PH
910PH/22 Guard against possible overflow in moan_check_errorcopy().
911
19897d52
PH
912PH/23 POSIX allows open() to be a macro; guard against that.
913
bc64a74d
PH
914PH/24 If the recipient of an error message contained an @ in the local part
915 (suitably quoted, of course), incorrect values were put in $domain and
916 $local_part during the evaluation of errors_copy.
917
eb4c0de6 918
b4ed4da0
PH
919Exim version 4.67
920-----------------
921
22ad45c9
MH
922MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
923 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
924 Jan Srzednicki.
925
b4ed4da0
PH
926PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
927 issue a MAIL command.
928
431b7361
PH
929PH/02 In an ACL statement such as
930
931 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
932
933 if a client was not listed at all, or was listed with a value other than
934 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
935 the condition was not true (as it should be), so access was not denied.
936 The bug was that the ! inversion was incorrectly passed on to the second
937 item. This has been fixed.
938
939PH/03 Added additional dnslists conditions == and =& which are different from
940 = and & when the dns lookup returns more than one IP address.
941
83da1223
PH
942PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
943 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
944
54fc8428
PH
945PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
946 FSYNC, which compiles an option called disable_fsync that allows for
947 bypassing fsync(). The documentation is heavily laced with warnings.
948
34c5e8dd
SC
949SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
950
bbe15da8
PH
951PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
952 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
953 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
954 including adding "make clean"; (3) Added -fPIC when compiling the test
955 dynamically loaded module, to get rid of a warning.
956
0e8a9471
MH
957MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
958 message fails, move_frozen_messages = true and ignore_bounce_errors_after
959 = 0s. The bug is otherwise harmless.
960
f0872424
PH
961PH/07 There was a bug in the dovecot authenticator such that the value of
962 $auth1 could be overwritten, and so not correctly preserved, after a
963 successful authentication. This usually meant that the value preserved by
964 the server_setid option was incorrect.
965
b01dd148
PH
966PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
967
6bf342e1
PH
968PH/09 Installed PCRE release 7.0.
969
273f34d0
PH
970PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
971 run for batched SMTP input. It is now run at the start of every message
972 in the batch. While fixing this I discovered that the process information
973 (output by running exiwhat) was not always getting set for -bs and -bS
974 input. This is fixed, and it now also says "batched" for BSMTP.
975
cf8b11a5
PH
976PH/11 Added control=no_pipelining.
977
41c7c167
PH
978PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
979 patch, slightly modified), and move the expansion of helo_data till after
980 the connection is made in the smtp transport (so it can use these
981 values).
982
9c57cbc0
PH
983PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
984
f3f065bb
PH
985PH/14 Added log_selector = +pid.
986
047bdd8c
PH
987PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
988
0ce9abe6
PH
989PH/16 Add ${if forany and ${if forall.
990
0e22dfd1
PH
991PH/17 Added dsn_from option to vary the From: line in DSNs.
992
4c590bd1
PH
993PH/18 Flush SMTP output before performing a callout, unless control =
994 no_callout_flush is set.
995
09945f1e
PH
996PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
997 was true (the default) a successful delivery failed to delete the retry
998 item, thus causing premature timeout of the address. The bug is now
999 fixed.
1000
c51b8e75
PH
1001PH/20 Added hosts_avoid_pipelining to the smtp transport.
1002
e28326d8
PH
1003PH/21 Long custom messages for fakedefer and fakereject are now split up
1004 into multiline reponses in the same way that messages for "deny" and
1005 other ACL rejections are.
1006
75b1493f
PH
1007PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1008 with slight modification.
1009
7c5214ec
PH
1010PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1011 draft, changing the syntax and factoring some duplicate code".
1012
4311097e
PH
1013PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1014 for deliveries of the second and subsequent messages over the same SMTP
1015 connection.
1016
29f89cad
PH
1017PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1018 ${reduce, with only minor "tidies".
1019
5e687460
SC
1020SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1021
c3611384
PH
1022PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1023 expansion side effects.
1024
5a11a7b4
PH
1025PH/27 When a message times out after an over-quota error from an Exim-imposed
1026 quota, the bounce message says "mailbox is full". This message was not
1027 being given when it was a system quota that was exceeded. It now should
1028 be the same.
1029
0e20aff9
MH
1030MH/03 Made $recipients available in local_scan(). local_scan() already has
1031 better access to the recipient list through recipients_list[], but
1032 $recipients can be useful in postmaster-provided expansion strings.
1033
ca86f471
PH
1034PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1035 in the case of a MAIL command with additional options following the
1036 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1037 were accidentally chopped off.
1038
a14e5636
PH
1039PH/29 SMTP synchronization checks are implemented when a command is read -
1040 there is a check that no more input is waiting when there shouldn't be
1041 any. However, for some commands, a delay in an ACL can mean that it is
1042 some time before the response is written. In this time, more input might
1043 arrive, invalidly. So now there are extra checks after an ACL has run for
1044 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1045 pipelining has not been advertised.
1046
ec95d1a6
PH
1047PH/30 MH's patch to allow iscntrl() characters to be list separators.
1048
42855d71
PH
1049PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1050 returned in the SMTP response when smtp_return_error_details was false.
1051 This has been fixed.
1052
57c2c631
PH
1053PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1054 instead of the C I/O that was originally supplied, because problems were
1055 reported on Solaris.
1056
58c01c94
PH
1057PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1058 Exim which did not show up earlier: it was assuming that a call to
1059 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1060 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1061 was a macro that became an assignment, so it seemed to work. This has
1062 changed to a proper function call with a void return, hence the compile
1063 error. Exim's code has been fixed.
1064
dee5a20a
PH
1065PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1066 cpus.
1067
d2ee6114
PH
1068PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1069
b2d5182b
PH
1070PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1071
79749a79
PH
1072PH/37 If a message is not accepted after it has had an id assigned (e.g.
1073 because it turns out to be too big or there is a timeout) there is no
3ce62588
PH
1074 "Completed" line in the log. When some messages of this type were
1075 selected by exigrep, they were listed as "not completed". Others were
1076 picked up by some special patterns. I have improved the selection
1077 criteria to be more general.
79749a79 1078
c456d9bb
PH
1079PH/38 The host_find_failed option in the manualroute router can now be set
1080 to "ignore", to completely ignore a host whose IP address cannot be
1081 found. If all hosts are ignored, the behaviour is controlled by the new
1082 host_all_ignored option.
1083
cd9868ec
PH
1084PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1085 homing or because of multiple MX records with /mx) generated more than
1086 one IP address, and the following item turned out to be the local host,
1087 all the secondary addresses of the first item were incorrectly removed
1088 from the list, along with the local host and any following hosts (which
1089 is what is supposed to happen).
1090
ebeaf996
PH
1091PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1092 whoever called Exim into the -H file. In the case of the daemon it was
1093 behaving confusingly. When first started, it used values for whoever
1094 started the daemon, but after a SIGHUP it used the Exim user (because it
1095 calls itself on a restart). I have changed the code so that it now always
1096 uses the Exim user.
1097
2679d413
PH
1098PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1099 message are rejected with the same error (e.g. no authentication or bad
1100 sender address), and a DATA command is nevertheless sent (as can happen
1101 with PIPELINING or a stupid MUA), the error message that was given to the
1102 RCPT commands is included in the rejection of the DATA command. This is
1103 intended to be helpful for MUAs that show only the final error to their
1104 users.
1105
84024b72
PH
1106PH/42 Another patch from the Sieve maintainer.
1107
8005d38e
SC
1108SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1109 Eximstats - Fixed some broken HTML links and added missing column headers
1110 (Jez Hancock).
1111 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1112 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1113
3298c6c6
SC
1114SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1115
a43a27c5
PH
1116PH/43 Yet another patch from the Sieve maintainer.
1117
58eb016e 1118PH/44 I found a way to check for a TCP/IP connection going away before sending
563b63fa
PH
1119 the response to the final '.' that terminates a message, but only in the
1120 case where the client has not sent further data following the '.'
1121 (unfortunately, this is allowed). However, in many cases there won't be
1122 any further data because there won't be any more messages to send. A call
1123 to select() can be used: if it shows that the input is "ready", there is
1124 either input waiting, or the socket has been closed. An attempt to read
1125 the next input character can distinguish the two cases. Previously, Exim
58eb016e 1126 would have sent an OK response which the client would never have see.
563b63fa
PH
1127 This could lead to message repetition. This fix should cure that, at
1128 least in a lot of common cases.
58eb016e 1129
b43a74ea
PH
1130PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1131 advertised in response to EHLO.
1132
b4ed4da0 1133
5dd1517f
PH
1134Exim version 4.66
1135-----------------
1136
1137PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1138 fixed by 4.65/MH/01 (is this a record?) are fixed:
1139
1140 (i) An empty string was always treated as zero by the numeric comparison
1141 operators. This behaviour has been restored.
1142
1143 (ii) It is documented that the numeric comparison operators always treat
1144 their arguments as decimal numbers. This was broken in that numbers
1145 starting with 0 were being interpreted as octal.
1146
1147 While fixing these problems I realized that there was another issue that
1148 hadn't been noticed. Values of message_size_limit (both the global option
1149 and the transport option) were treated as octal if they started with 0.
1150 The documentation was vague. These values are now always treated as
1151 decimal, and I will make that clear in the documentation.
1152
1153
93cfa765
TK
1154Exim version 4.65
1155-----------------
1156
1157TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1158 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1159 versions. (#438)
1160
d6066548
MH
1161MH/01 Don't check that the operands of numeric comparison operators are
1162 integers when their expansion is in "skipping" mode (fixes bug
1163 introduced by 4.64-PH/07).
1164
4362ff0d
PH
1165PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1166 child addresses, Exim now panics and dies. Previously, because the count
1167 is held in a short int, deliveries were likely to be lost. As such a
1168 large number of recipients for a single message is ridiculous
1169 (performance will be very, very poor), I have chosen to impose a limit
1170 rather than extend the field.
1171
93cfa765 1172
944e9e9c
TF
1173Exim version 4.64
1174-----------------
aa41d2de 1175
21d74bd9
TK
1176TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1177 leftover -K file (the existence of which was triggered by #402).
1178 While we were at it, introduced process PID as part of the -K
1179 filename. This should rule out race conditions when creating
1180 these files.
1181
1182TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1183 processing considerably. Previous code took too long for large mails,
1184 triggering a timeout which in turn triggers #401.
1185
1186TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1187 in the DK code in transports.c. sendfile() is not really portable,
1188 hence the _LINUX specificness.
944e9e9c
TF
1189
1190TF/01 In the add_headers option to the mail command in an Exim filter,
1191 there was a bug that Exim would claim a syntax error in any
1192 header after the first one which had an odd number of characters
1193 in the field name.
1194
2b1c6e3a
PH
1195PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1196 callout verification, Exim cached a "reject" for the entire domain. This
1197 is correct for most verifications, but it is not correct for a recipient
1198 verification with use_sender or use_postmaster set, because in that case
1199 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1200 case of MAIL FROM:<> rejection from other early rejections (e.g.
1201 rejection of HELO). When verifying a recipient using a non-null MAIL
1202 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1203 Whatever the result of the callout, the value of the domain cache is
1204 left unchanged (for any other kind of callout, getting as far as trying
1205 RCPT means that the domain itself is ok).
1206
1f872c80
PH
1207PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1208 gcc 4.1.1 threw up.
1209
1210PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1211 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1212 session, the daemon ignores ECONNECT errors and logs others; it now
1213 ignores EPIPE as well.
1214
d203e649
PH
1215PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1216 (quoted-printable decoding).
1217
cc2ed8f7 1218PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
21a04aa3 1219 later the small subsequent patch to fix an introduced bug.
f951fd57 1220
ddfcd446
PH
1221PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1222
d45b1de8
PH
1223PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1224
1225PH/08 An error is now given if message_size_limit is specified negative.
1226
38a0a95f 1227PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
641cb756 1228 to be given (somewhat) arbitrary names.
38a0a95f 1229
a2405d83
JJ
1230JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1231 in 4.64-PH/09.
1232
1233JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1234 miscellaneous code fixes
1235
6ea85e9a
PH
1236PH/10 Added the log_reject_target ACL modifier to specify where to log
1237 rejections.
1238
26da7e20
PH
1239PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1240 hostname. This is wrong, because it relates to the incoming message (and
1241 probably the interface on which it is arriving) and not to the outgoing
1242 callout (which could be using a different interface). This has been
1243 changed to use the value of the helo_data option from the smtp transport
1244 instead - this is what is used when a message is actually being sent. If
1245 there is no remote transport (possible with a router that sets up host
1246 addresses), $smtp_active_hostname is used.
6ea85e9a 1247
14aa5a05 1248PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
7befa435 1249 tweaks were necessary in order to get it to work (see also 21 below):
14aa5a05
PH
1250 (a) The code assumed that strncpy() returns a negative number on buffer
1251 overflow, which isn't the case. Replaced with Exim's string_format()
1252 function.
1253 (b) There were several signed/unsigned issues. I just did the minimum
1254 hacking in of casts. There is scope for a larger refactoring.
1255 (c) The code used strcasecmp() which is not a standard C function.
1256 Replaced with Exim's strcmpic() function.
1257 (d) The code set only $1; it now sets $auth1 as well.
1258 (e) A simple test gave the error "authentication client didn't specify
1259 service in request". It would seem that Dovecot has changed its
1260 interface. Fortunately there's a specification; I followed it and
1261 changed what the client sends and it appears to be working now.
1262
ff75a1f7
PH
1263PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1264 decoding.
1265
e6f6568e
PH
1266PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1267 address A is aliased to B and C, where B exists and C does not. Without
1268 -v the output is "A verified" because verification stops after a
1269 successful redirection if more than one address is generated. However,
1270 with -v the child addresses are also verified. Exim was outputting "A
1271 failed to verify" and then showing the successful verification for C,
1272 with its parentage. It now outputs "B failed to verify", showing B's
1273 parentage before showing the successful verification of C.
1274
d6f6e0dc
PH
1275PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1276 look up a TXT record in a specific list after matching in a combined
1277 list.
1278
322050c2
PH
1279PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1280 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1281 they consult the DNS. I had assumed they would set it the way they
1282 wanted; and indeed my experiments on Linux seem to show that in some
1283 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1284 To be on the safe side, however, I have now made the interface to
1285 host_find_byname() similar to host_find_bydns(), with an argument
1286 containing the DNS resolver options. The host_find_byname() function now
1287 sets these options at its start, just as host_find_bydns() does. The smtp
1288 transport options dns_qualify_single and dns_search_parents are passed to
1289 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
1290 of host_find_byname() use the default settings of RES_DEFNAMES
1291 (qualify_single) but not RES_DNSRCH (search_parents).
1292
08955dd3
PH
1293PH/17 Applied (a modified version of) Nico Erfurth's patch to make
1294 spool_read_header() do less string testing, by means of a preliminary
1295 switch on the second character of optional "-foo" lines. (This is
1296 overdue, caused by the large number of possibilities that now exist.
1297 Originally there were few.) While I was there, I also converted the
1298 str(n)cmp tests so they don't re-test the leading "-" and the first
1299 character, in the hope this might squeeze out yet more improvement.
1300
1eccaa59
PH
1301PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
1302 flag allowing group syntax was set by the header_syntax check but not
1303 turned off, possible causing trouble later; (2) The flag was not being
1304 set at all for the header_verify test, causing "group"-style headers to
1305 be rejected. I have now set it in this case, and also caused header_
1306 verify to ignore an empty address taken from a group. While doing this, I
1307 came across some other cases where the code for allowing group syntax
1308 while scanning a header line wasn't quite right (mostly, not resetting
1309 the flag correctly in the right place). These bugs could have caused
1310 trouble for malformed header lines. I hope it is now all correct.
1311
602e59e5
PH
1312PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
1313 with the "reply" argument non-NULL. The code, however (which originally
1314 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
1315 but it didn't always do it. This confused somebody who was copying the
1316 code for some other use. I have removed all the tests.
1317
411ef850
PH
1318PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
1319 feature that was used to support insecure browsers during the U.S. crypto
1320 embargo. It requires special client support, and Exim is probably the
1321 only MTA that supported it -- and would never use it because real RSA is
1322 always available. This code has been removed, because it had the bad
1323 effect of slowing Exim down by computing (never used) parameters for the
1324 RSA_EXPORT functionality.
1325
7befa435
PH
1326PH/21 On the advice of Timo Sirainen, added a check to the dovecot
1327 authenticator to fail if there's a tab character in the incoming data
1328 (there should never be unless someone is messing about, as it's supposed
1329 to be base64-encoded). Also added, on Timo's advice, the "secured" option
1330 if the connection is using TLS or if the remote IP is the same as the
1331 local IP, and the "valid-client-cert option" if a client certificate has
1332 been verified.
1333
48da4259 1334PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
16ff981e
PH
1335 authenticators. This can be used for authorization after authentication
1336 succeeds. (In the case of plaintext, it servers for both authentication
1337 and authorization.)
1338
48da4259
PH
1339PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1340 if any retry times were supplied.
1341
d1d5595c
PH
1342PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1343 connection, where there is no client IP address to check. In this
1344 situation, the verify now always succeeds.
1345
0ef732d9
PH
1346PH/25 Applied John Jetmore's -Mset patch.
1347
328895cc
PH
1348PH/26 Added -bem to be like -Mset, but loading a message from a file.
1349
fd700877
PH
1350PH/27 In a string expansion for a processed (not raw) header when multiple
1351 headers of the same name were present, leading whitespace was being
1352 removed from all of them, but trailing whitespace was being removed only
1353 from the last one. Now trailing whitespace is removed from each header
f6c332bd
PH
1354 before concatenation. Completely empty headers in a concatenation (as
1355 before) are ignored.
fd700877 1356
8dce1a6f
PH
1357PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1358 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1359
17af4a17
PH
1360PH/29 [Removed. This was a change that I later backed out, and forgot to
1361 correct the ChangeLog entry (that I had efficiently created) before
1362 committing the later change.]
f6c332bd
PH
1363
1364PH/30 Exim was sometimes attempting to deliver messages that had suffered
1365 address errors (4xx response to RCPT) over the same connection as other
1366 messages routed to the same hosts. Such deliveries are always "forced",
1367 so retry times are not inspected. This resulted in far too many retries
1368 for the affected addresses. The effect occurred only when there were more
1369 hosts than the hosts_max_try setting in the smtp transport when it had
1370 the 4xx errors. Those hosts that it had tried were not added to the list
1371 of hosts for which the message was waiting, so if all were tried, there
1372 was no problem. Two fixes have been applied:
1373
1374 (i) If there are any address or message errors in an SMTP delivery, none
1375 of the hosts (tried or untried) are now added to the list of hosts
1376 for which the message is waiting, so the message should not be a
1377 candidate for sending over the same connection that was used for a
1378 successful delivery of some other message. This seems entirely
1379 reasonable: after all the message is NOT "waiting for some host".
1380 This is so "obvious" that I'm not sure why it wasn't done
1381 previously. Hope I haven't missed anything, but it can't do any
1382 harm, as the worst effect is to miss an optimization.
1383
1384 (ii) If, despite (i), such a delivery is accidentally attempted, the
1385 routing retry time is respected, so at least it doesn't keep
1386 hammering the server.
1387
c1114884
PH
1388PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1389 in ${readsocket because some servers need this prod.
1390
7a0743eb
PH
1391PH/32 Added some extra debug output when updating a wait-xxx database.
1392
0d85fa3f
PH
1393PH/33 The hint "could be header name not terminated by colon", which has been
1394 given for certain expansion errors for a long time, was not being given
1395 for the ${if def:h_colon_omitted{... case.
1396
1bf43b78
PH
1397PH/34 The spec says: "With one important exception, whenever a domain list is
1398 being scanned, $domain contains the subject domain." There was at least
1399 one case where this was not true.
1400
520de300
PH
1401PH/35 The error "getsockname() failed: connection reset by peer" was being
1402 written to the panic log as well as the main log, but it isn't really
1403 panic-worthy as it just means the connection died rather early on. I have
1404 removed the panic log writing for the ECONNRESET error when getsockname()
1405 fails.
1406
48c7f9e2
PH
1407PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1408 runs only) independently of the message's sender address. This meant
1409 that, if the 4xx error was in fact related to the sender, a different
1410 message to the same recipient with a different sender could confuse
1411 things. In particualar, this can happen when sending to a greylisting
1412 server, but other circumstances could also provoke similar problems.
1413 I have changed the default so that the retry time for these errors is now
1414 based a combination of the sender and recipient addresses. This change
1415 can be overridden by setting address_retry_include_sender=false in the
1416 smtp transport.
1417
99ea1c86
PH
1418PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1419 remote server are returned as part of bounce messages. This was not
1420 happening for LMTP over a pipe (the lmtp transport), but now it is the
1421 same for both kinds of LMTP.
1422
a2042e78
PH
1423PH/38 Despite being documented as not happening, Exim was rewriting addresses
1424 in header lines that were in fact CNAMEs. This is no longer the case.
1425
4fbcfc2e
PH
1426PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1427 and queue runs started by the daemon processed all messages. This has
1428 been fixed so that -R and -S can now usefully be given with -q<time>.
1429
aa41d2de
PH
1430PH/40 Import PCRE release 6.7 (fixes some bugs).
1431
af561417
PH
1432PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1433
3cc66b45
PH
1434PH/42 Give an error if -q is specified more than once.
1435
194cc0e4
PH
1436PH/43 Renamed the variables $interface_address and $interface_port as
1437 $received_ip_address and $received_port, to make it clear that these
1438 values apply to message reception, and not to the outgoing interface when
1439 a message is delivered. (The old names remain recognized, of course.)
1440
a401ddaa
PH
1441PH/44 There was no timeout on the connect() call when using a Unix domain
1442 socket in the ${readsocket expansion. There now is.
1443
4e88a19f
PH
1444PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1445 be meaningful with "accept".
1446
d7d7b289
SC
1447SC/01 Eximstats V1.43
1448 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1449
1450SC/02 Eximstats V1.44
1451 Use a glob alias rather than an array ref in the generated
1452 parser. This improves both readability and performance.
1453
1454SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1455 Collect SpamAssassin and rejection statistics.
1456 Don't display local sender or destination tables unless
1457 there is data to show.
1458 Added average volumes into the top table text output.
1459
1460SC/04 Eximstats V1.46
1461 Collect data on the number of addresses (recipients)
1462 as well as the number of messages.
1463
1464SC/05 Eximstats V1.47
1465 Added 'Message too big' to the list of mail rejection
1466 reasons (thanks to Marco Gaiarin).
1467
1468SC/06 Eximstats V1.48
1469 Mainlog lines which have GMT offsets and are too short to
1470 have a flag are now skipped.
1471
1472SC/07 Eximstats V1.49 (Alain Williams)
1473 Added the -emptyok flag.
1474
1475SC/08 Eximstats V1.50
1476 Fixes for obtaining the IP address from reject messages.
1477
0ea2a468
JJ
1478JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1479 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1480 whitesspace changes from 4.64-PH/27
1481
1482JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1483 match 4.64-PH/13
1484
1485JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1486 are found, allow negative numbers in numeric criteria)
1487
1488JJ/06 exipick.20061117.2, added new $message_body_missing variable
1489
1490JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1491 to match changes made in 4.64-PH/43
1492
8a10f5a4
PH
1493PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1494
30e18802
PH
1495PH/47 Put in an explicit test for a DNS lookup of an address record where the
1496 "domain" is actually an IP address, and force a failure. This locks out
1497 those revolvers/nameservers that support "A-for-A" lookups, in
1498 contravention of the specifications.
1499
55728a4f
PH
1500PH/48 When a host name was looked up from an IP address, and the subsequent
1501 forward lookup of the name timed out, the host name was left in
1502 $sender_host_name, contrary to the specification.
d7d7b289 1503
d7837193
PH
1504PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1505 restricted to single-key lookups, Exim was not diagnosing an error if
1506 * or *@ was used with a query-style lookup.
1507
87054a31
PH
1508PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1509
ea2c01d2
MH
1510MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1511 long ago, but noone interested enough thought of it. Let's just say that
1512 the "1.1" means that there are some new functions that weren't there at
1513 some point in the past.
1514
e4fa6968
PH
1515PH/51 Error processing for expansion failure of helo_data from an smtp
1516 transport during callout processing was broken.
1517
56f5d9bd
PH
1518PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1519 tested/used via the -bh/-bhc/-bs options.
1520
922e1c28
PH
1521PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1522 bug, fixed in subsequent PCRE releases).
1523
21eb6e72
PH
1524PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1525 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1526
a0540757
PH
1527PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1528
944e9e9c 1529
478be7b0
SC
1530Exim version 4.63
1531-----------------
1532
1533SC/01 Use a glob alias rather than an array ref in eximstats generated
1534 parser. This improves both readability and performance.
1535
1536SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1537 Don't display local sender or destination tables in eximstats unless
1538 there is data to show.
1539 Added average volumes into the eximstats top table text output.
1540
1541SC/03 Collect data on the number of addresses (recipients) as well
1542 as the number of messages in eximstats.
1543
2b965a65
TF
1544TF/01 Correct an error in the documentation for the redirect router. Exim
1545 does (usually) call initgroups() when daemonizing.
478be7b0 1546
45b91596
PH
1547TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1548 with consistent privilege compared to when running as a daemon.
478be7b0 1549
c59f5781
TF
1550TF/03 Note in the spec that $authenticated_id is not set for local
1551 submissions from trusted users.
1552
90fc3069
TF
1553TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1554 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1555
6083aca0
TF
1556TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1557 by adding some example configuration directives to the default
1558 configuration file. A little bit of work is required to uncomment the
1559 directives and define how usernames and passwords are checked, but
1560 there is now a framework to start from.
1561
765b530f
PH
1562PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1563 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1564 without this. I don't know how relevant this is to other LDAP libraries.
1565
4e167a8c
PH
1566PH/02 Add the verb name to the "unknown ACL verb" error.
1567
4608d683
PH
1568PH/03 Magnus Holmgren's patch for filter_prepend_home.
1569
b8dc3e4a
PH
1570PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1571
5418e93b
PH
1572PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1573 directory not expanded when it should be if an expanded home directory
1574 was set for the address (which is overridden by the transport).
1575
b4a9bda2
PH
1576PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1577 libradius.
1578
45b91596
PH
1579PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1580 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1581 because it is too late at that time, and has no effect.
1582
5547e2c5
PH
1583PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1584 security issue with \' (bugzilla #107). I could not use the
1585 PQescapeStringConn() function, because it needs a PGconn value as one of
1586 its arguments.
1587
dbcef0ea
PH
1588PH/08 When testing addresses using -bt, indicate those final addresses that
1589 are duplicates that would not cause an additional delivery. At least one
1590 person was confused, thinking that -bt output corresponded to deliveries.
1591 (Suppressing duplicates isn't a good idea as you lose the information
1592 about possibly different redirections that led to the duplicates.)
1593
25257489
PH
1594PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1595 systems where poll() doesn't work, in particular OS X.
1596
c816d124
PH
1597PH/10 Added more information to debugging output for retry time not reached.
1598
a9ccd69a
PH
1599PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1600 operations in malware.c.
1601
75fa1910
PH
1602PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1603 signatures.
1604
a7d7aa58
PH
1605PH/13 If write_rejectlog was set false when logging was sent to syslog with
1606 syslog_duplication set false, log lines that would normally be written
1607 both the the main log and to the reject log were not written to syslog at
1608 all.
1609
42119b09
PH
1610PH/14 In the default configuration, change the use of "message" in ACL warn
1611 statements to "add_header".
1612
41609df5
PH
1613PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1614 not followed by a command (e.g. "seen endif").
1615
a5bd321b
PH
1616PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1617 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1618 latter.
1619
e85a7ad5 1620PH/17 Added extra conditions to the default value of delay_warning_condition
5dff5817
PH
1621 so that it is now:
1622
e85a7ad5
PH
1623 ${if or { \
1624 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1625 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1626 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
5dff5817
PH
1627 }{no}{yes}}
1628
e85a7ad5
PH
1629 The Auto-Submitted: and various List- headers are standardised, whereas I
1630 don't think Precedence: ever was.
5dff5817 1631
d8fe1c03
PH
1632PH/18 Refactored debugging code in route_finduser() to show more information,
1633 in particular, the error code if getpwnam() issues one.
1634
16282d2b
PH
1635PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1636 This is apparently needed in addition to the PH/07 change above to avoid
1637 any possible encoding problems.
1638
35d40a98
PH
1639PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1640 but not after initializing Perl.
1641
034d99ab
PH
1642PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1643 output them only if debugging. By default they are written stderr,
1644 apparently, which is not desirable.
1645
6ec97b1b
PH
1646PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1647 queries.
1648
e22ca4ac
JJ
1649JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1650 --not options
1651
1652JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1653
33d73e3b
PH
1654PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1655 authenticated or an ident call has been made. Suppress the default
1656 values for $authenticated_id and $authenticated_sender (but permit -oMai
1657 and -oMas) when testing with -bh.
1658
9ecb03f3
PH
1659PH/24 Re-jigged the order of the tests in the default configuration so that the
1660 tests for valid domains and recipients precede the DNS black list and CSA
1661 tests, on the grounds that those ones are more expensive.
1662
084efe8d
PH
1663PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1664 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1665 This bug exists in every version of Exim that I still have, right back to
1666 0.12.
1667
366fc9f0
PH
1668PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1669 However, an attempt to turn on case-sensitivity in a regex key by
1670 including (?-i) didn't work because the subject string was already
1671 lowercased, and the effects were non-intuitive. It turns out that a
1672 one-line patch can be used to allow (?-i) to work as expected.
1673
c59f5781 1674
c887c79e
TF
1675Exim version 4.62
1676-----------------
1677
1678TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1679 other effects) broke the use of negated acl sub-conditions.
1680
1cce3af8
PH
1681PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1682 patch).
1683
afb3eaaf
PH
1684PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1685 "Deny" causes Exim to reject the incoming connection with a 554 error.
1686 Unfortunately, if there is a major crisis, such as a disk failure,
1687 tcp-wrappers gives "deny", whereas what one would like would be some
1688 kind of temporary error. A kludge has been added to help with this.
1689 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1690 554 error is used if errno is still zero or contains ENOENT (which occurs
1691 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1692 451 error is used.
1693
e173618b
PH
1694PH/03 Add -lutil to the default FreeBSD LIBS setting.
1695
dd16e114
PH
1696PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1697 errors. Otherwise a message that provokes a temporary error (when other
1698 messages do not) can cause a whole host to time out.
1699
f7fd3850
PH
1700PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1701 addresses were routed directly to files or pipes from a redirect router.
1702 File deliveries just didn't batch; pipe deliveries might have suffered
1703 odd errors.
1704
d87df92c
PH
1705PH/06 A failure to get a lock for a hints database would erroneously always say
1706 "Failed to get write lock", even when it was really a read lock.
1707
7e9f683d
PH
1708PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1709 of 0600. This has been changed to use the value of the lockfile_mode
1710 option (which defaults to 0600).
1711
bfad5236
PH
1712PH/08 Applied small patch from the Sieve maintainer.
1713
01c490df
PH
1714PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1715 folder from quota calculations, a direct delivery into this folder messed
1716 up the contents of the maildirsize file. This was because the regex was
1717 used only to exclude .Trash (or whatever) when the size of the mailbox
1718 was calculated. There was no check that a delivery was happening into an
1719 excluded directory. This bug has been fixed by ignoring all quota
1720 processing for deliveries into excluded directories.
1721
d6629cdc
PH
1722PH/10 Added the maildirfolder_create_regex option to appendfile.
1723
1cce3af8 1724
214e2000
PH
1725Exim version 4.61
1726-----------------
1727
1728PH/01 The code for finding all the local interface addresses on a FreeBSD
1729 system running IPv6 was broken. This may well have applied to all BSD
1730 systems, as well as to others that have similar system calls. The broken
1731 code found IPv4 interfaces correctly, but gave incorrect values for the
1732 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1733 that it would not match correctly against @[] and not recognize the IPv6
1734 addresses as local.
1735
f9daeae0
PH
1736PH/02 The ipliteral router was not recognizing addresses of the form user@
1737 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1738
7e66e54d
PH
1739PH/03 Added disable_ipv6.
1740
c8ea1597
PH
1741PH/04 Changed $reply_address to use the raw form of the headers instead of the
1742 decoded form, because it is most often used to construct To: headers
1743 lines in autoreplies, and the decoded form may well be syntactically
1744 invalid. However, $reply_address has leading white space removed, and all
1745 newlines turned into spaces so that the autoreply transport does not
1746 grumble.
1747
911f6fde
PH
1748PH/05 If group was specified without a user on a router, and no group or user
1749 was specified on a transport, the group from the router was ignored.
1750
47ca6d6c
PH
1751PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1752 for visible compile-time settings that can be used to change these
1753 numbers, for those that want even more. Backwards compatibility with old
1754 spool files has been maintained. However, going back to a previous Exim
1755 release will lost any variables that are in spool files.
1756
ed0e9820
PH
1757PH/07 Two small changes when running in the test harness: increase delay when
1758 passing a TCP/IP connection to a new process, in case the original
1759 process has to generate a bounce, and remove special handling of
1760 127.0.0.2 (sic), which is no longer necessary.
1761
eff37e47
PH
1762PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1763 be the same on different OS.
1764
1921d2ea
PH
1765PH/09 Moved a debug statement in filter processing to avoid a race problem when
1766 testing.
1767
b3f69ca8
JJ
1768JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1769 whether --show-vars was specified or not
1770
1771JJ/02 exipick: Added support for new ACL variable spool format introduced
1772 in 4.61-PH/06
1773
424a1c63
PH
1774PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1775 syntactically invalid From: or Reply-to: line, and a filter used this to
1776 generate an autoreply, and therefore failed to obtain an address for the
1777 autoreply, Exim could try to deliver to a non-existent relative file
1778 name, causing unrelated and misleading errors. What now happens is that
1779 it logs this as a hard delivery error, but does not attempt to create a
1780 bounce message.
1781
7a100415
PH
1782PH/11 The exinext utility has a -C option for testing purposes, but although
1783 the given file was scanned by exinext itself; it wasn't being passed on
1784 when Exim was called.
1785
19b9dc85
PH
1786PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1787 an end-of-file indication when reading a command response.
1788
309bd837
PH
1789PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1790 compiled. In many other places in Exim, IPv6 addresses are always
1791 recognized, so I have changed this. It also means that IPv4 domain
1792 literals of the form [IPV4:n.n.n.n] are now always recognized.
1793
59e82a2a
PH
1794PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1795 used if the router is not running as root, for example, when verifying at
1796 ACL time, or when using -bh. The debugging output from this situation was
1797 non-existent - all you got was a failure to exec. I have made two
1798 changes:
1799
1800 (a) Failures to set uid/gid, the current directory, or a process leader
1801 in a subprocess such as that created by queryprogram now generate
1802 suitable debugging ouput when -d is set.
1803
1804 (b) The queryprogram router detects when it is not running as root,
1805 outputs suitable debugging information if -d is set, and then runs
1806 the subprocess without attempting to change uid/gid.
1807
9edc04ce
PH
1808PH/15 Minor change to Makefile for building test_host (undocumented testing
1809 feature).
1810
1349e1e5
PH
1811PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1812 additional section of a DNS packet that returns MX or SRV records.
1813 Instead, it always explicitly searches for A/AAAA records. This avoids
1814 major problems that occur when a DNS server includes only records of one
1815 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1816 fixed another bug: if SRV records were looked up and the corresponding
1817 address records were *not* found in the additional section, the port
1818 values from the SRV records were lost.
1819
ea49d0e1
PH
1820PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1821 using the correct key (the original address) when searching the retry
1822 rules in order to find which one to use for generating the retry hint.
1823
064a94c9
PH
1824PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1825 adding the default one. Similarly, if it contains a Reply-To: header, the
1826 errors_reply_to option, if set, is not used.
1827
727071f8
PH
1828PH/19 When calculating a retry time, Exim used to measure the "time since
1829 failure" by looking at the "first failed" field in the retry record. Now
1830 it does not use this if it is later than than the arrival time of the
1831 message. Instead it uses the arrival time. This makes for better
1832 behaviour in cases where some deliveries succeed, thus re-setting the
1833 "first failed" field. An example is a quota failure for a huge message
1834 when small messages continue to be delivered. Without this change, the
1835 "time since failure" will always be short, possible causing more frequent
1836 delivery attempts for the huge message than are intended.
dd16e114 1837 [Note: This change was subsequently modified - see PH/04 for 4.62.]
727071f8 1838
f78eb7c6
PH
1839PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1840 $1, $2, $3) because the numerical variables can be reset during some
1841 expansion items (e.g. "match"), thereby losing the authentication data.
1842
21c28500
PH
1843PH/21 Make -bV show the size of off_t variables so that the test suite can
1844 decide whether to run tests for quotas > 2G.
1845
1846PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1847 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1848 filecount value is greater than 2G or if a quota value is greater than 2G
1849 on a system where the size of off_t is not greater than 4, a panic error
1850 is given.
1851
1688f43b
PH
1852PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1853 never match. The debug and -bh output now contains an explicit error
1854 message indicating a malformed IPv4 address or mask.
1855
1856PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1857 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1858 PH/23 above applies.
1859
9675b384
PH
1860PH/25 Do not write to syslog when running in the test harness. The only
1861 occasion when this arises is a failure to open the main or panic logs
1862 (for which there is an explicit test).
1863
6a3f1455
PH
1864PH/26 Added the /no_tell option to "control=freeze".
1865
dac79d3e
PH
1866PH/27 If a host name lookup failed very early in a connection, for example, if
1867 the IP address matched host_lookup and the reverse lookup yielded a name
1868 that did not have a forward lookup, an error message of the form "no IP
1869 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1870 could be logged. Now it outputs the IP address instead of "NULL".
1349e1e5 1871
5977a0b3
PH
1872PH/28 An enabling patch from MH: add new function child_open_exim2() which
1873 allows the sender and the authenticated sender to be set when
1874 submitting a message from within Exim. Since child_open_exim() is
1875 documented for local_scan(), the new function should be too.
1876
c91535f3
PH
1877PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1878 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1879 results in an empty string is now treated as unset.
1880
0d46a8c8
PH
1881PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1882
278c6e6c
PH
1883PH/31 Added sender_verify_fail logging option.
1884
2cbb4081
PH
1885PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1886 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1887 tidied the source and removed it altogether.
1888
3eef829e
PH
1889PH/33 When a queue run was abandoned because the load average was too high, a
1890 log line was always written; now it is written only if the queue_run log
1891 selector is set. In addition, the log line for abandonment now contains
1892 information about the queue run such as the pid. This is always present
1893 in "start" and "stop" lines but was omitted from the "abandon" line.
1894
1ab95fa6
PH
1895PH/34 Omit spaces between a header name and the colon in the error message that
1896 is given when verify = headers_syntax fails (if there are lots of them,
1897 the message gets confusing).
1898
230205fc
PH
1899PH/35 Change the default for dns_check_names_pattern to allow slashes within
1900 names, as there are now some PTR records that contain slashes. This check
1901 is only to protect against broken name servers that fall over on strange
1902 characters, so the fact that it applies to all lookups doesn't matter.
1903
75e0e026
PH
1904PH/36 Now that the new test suite is complete, we can remove some of the
1905 special code in Exim that was needed for the old test suite. For example,
1906 sorting DNS records because real resolvers return them in an arbitrary
1907 order. The new test suite's fake resolver always returns records in the
1908 same order.
1909
1910PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1911 bounces) except when queue_only is set, to avoid logging races between
1912 the different processes.
1913
145396a6
PH
1914PH/38 Panic-die if .include specifies a non-absolute path.
1915
3cd34f13
PH
1916PH/39 A tweak to the "H" retry rule from its user.
1917
11121d3d
JJ
1918JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1919 a label. They prevented compilation on older perls.
1920
1921JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1922 a warning to be raised on newish perls.
1923
1924JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1925 on queue. Changes to match documented behaviour of showing count of
1926 messages matching specified criteria.
1927
8def5aaf
PH
1928PH/40 Changed the default ident timeout from 30s to 5s.
1929
929ba01c
PH
1930PH/41 Added support for the use of login_cap features, on those BSD systems
1931 that have them, for controlling the resources used by pipe deliveries.
1932
2632889e
PH
1933PH/42 The content-scanning code uses fopen() to create files in which to put
1934 message data. Previously it was not paying any attention to the mode of
1935 the files. Exim runs with umask(0) because the rest of the code creates
1936 files with open(), and sets the required mode explicitly. Thus, these
1937 files were ending up world-writeable. This was not a big issue, because,
1938 being within the spool directory, they were not world-accessible. I have
1939 created a function called modefopen, which takes an additional mode
1940 argument. It sets umask(777), creates the file, chmods it to the required
1941 mode, then resets the umask. All the relevant calls to fopen() in the
1942 content scanning code have been changed to use this function.
1943
944a9c55
PH
1944PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1945 to 24 hours. This avoids potential overflow problems when processing G
1946 and H retry rules. I suspect nobody ever tinkers with this value.
1947
4a23603b
PH
1948PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1949
4730f942
PH
1950PH/45 When the plaintext authenticator is running as a client, the server's
1951 challenges are checked to ensure they are valid base64 strings. By
1952 default, the authentication attempt is cancelled if an invalid string is
1953 received. Setting client_ignore_invalid_base64 true ignores these errors.
1954 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1955 they are received. Thus, the responses can be made to depend on the
1956 challenges. If an invalid string is ignored, an empty string is placed in
1957 the variable.
1958
30dba1e6
PH
1959PH/46 Messages that are created by the autoreply transport now contains a
1960 References: header, in accordance with RFCs 2822 and 3834.
1961
382afc6b
PH
1962PH/47 Added authenticated_sender_force to the smtp transport.
1963
a86229cf
PH
1964PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1965
50c99ba6
PH
1966PH/49 Installed latest patch from the Sieve maintainer.
1967
d35e429d
PH
1968PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1969 was also set, the appendfile transport was unnecessarily scanning a
1970 directory of message files (e.g. for maildir delivery) to find the count
1971 of files (along with the size), even though it did not need this
1972 information. It now does the scan only if it needs to find either the
1973 size of the count of files.
1974
f90d018c
PH
1975PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1976
75def545
PH
1977PH/52 Two bugs concerned with error handling when the smtp transport is
1978 used in LMTP mode:
1979
1980 (i) Exim was not creating retry information for temporary errors given
1981 for individual recipients after the DATA command when the smtp transport
1982 was used in LMTP mode. This meant that they could be retried too
1983 frequently, and not timed out correctly.
1984
1985 (ii) Exim was setting the flag that allows error details to be returned
1986 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1987 recipients that were returned after the DATA command.
1988
1989PH/53 This is related to PH/52, but is more general: for any failing address,
1990 when detailed error information was permitted to be returned to the
1991 sender, but the error was temporary, then after the final timeout, only
1992 "retry timeout exceeded" was returned. Now it returns the full error as
1993 well as "retry timeout exceeded".
1994
c46782ef
PH
1995PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1996 do this, and (what is worse) MTAs that accept it.
1997
71fafd95
PH
1998PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1999 will now be deprecated.
2000
2c5db4fd
PH
2001PH/56 New os.c-cygwin from the Cygwin maintainer.
2002
9cf6b11a
JJ
2003JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2004 formats (previously only available in exim formats via -bpr, -bpru,
2005 and -bpra. Now also available in native and exiqgrep formats)
2006
2007JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2008 with very large, slow to parse queues
2009
2010JJ/08 exipick: added ! as generic prefix to negate any criteria format
2011
2012JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2013
898d150f
PH
2014PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2015 responses to authentication challenges, though it was showing the
2016 challenges; (ii) I've removed the CR characters from the debug output for
2017 SMTP output lines.
2018
46218253
PH
2019PH/58 Allow for the insertion of a newline as well as a space when a string
2020 is turned into more than one encoded-word during RFC 2047 encoding. The
2021 Sieve code now uses this.
2022
e97957bc
PH
2023PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2024 data_4xx, lost_connection, tls_required.
2025
81e509d7
PH
2026PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2027 message was being sent as an SMTP response.
2028
3d240ff7
PH
2029PH/61 Add -l and -k options to exicyclog.
2030
b37c4101
PH
2031PH/62 When verifying, if an address was redirected to one new address, so that
2032 verification continued, and the new address failed or deferred after
2033 having set something in $address_data, the value of $address_data was not
2034 passed back to the ACL. This was different to the case when no
2035 redirection occurred. The value is now passed back in both cases.
2036
79378e0f
PH
2037PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2038 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2039 use login_cap.h, so on its own it isn't the distinguishing feature. The
2040 new name refers directly to the setclassresources() function.
2041
e49c7bb4
PH
2042PH/65 Added configuration files for NetBSD3.
2043
d114ec46
PH
2044PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2045
f3d7df6c
PH
2046PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2047 is preferred over IPv4.
2048
715ab376
PH
2049PH/68 The bounce_return_message and bounce_return_body options were not being
2050 honoured for bounces generated during the reception of non-SMTP messages.
2051 In particular, this applied to messages rejected by the ACL. This bug has
2052 been fixed. However, if bounce_return_message is true and bounce_return_
2053 body is false, the headers that are returned for a non-SMTP message
2054 include only those that have been read before the error was detected.
2055 (In the case of an ACL rejection, they have all been read.)
2056
6b31b150
PH
2057PH/69 The HTML version of the specification is now built in a directory called
2058 spec_html instead of spec.html, because the latter looks like a path with
2059 a MIME-type, and this confuses some software.
2060
2061PH/70 Catch two compiler warnings in sieve.c.
2062
d515a917
PH
2063PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2064 function verify_get_ident() calls ip_connect() to connect a socket, but
2065 if the "connect()" function timed out, ip_connect() used to close the
2066 socket. However, verify_get_ident() also closes the socket later, and in
2067 between Exim writes to the log, which may get opened at this point. When
2068 the socket was closed in ip_connect(), the log could get the same file
2069 descriptor number as the socket. This naturally causes chaos. The fix is
2070 not to close the socket in ip_connect(); the socket should be closed by
2071 the function that creates it. There was only one place in the code where
2072 this was missing, in the iplookup router, which I don't think anybody now
2073 uses, but I've fixed it anyway.
2074
9b8fadde
PH
2075PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2076 well as to direct DNS lookups. Otherwise the handling of names in host
2077 lists is inconsistent and therefore confusing.
2078
214e2000 2079
5de37277
PH
2080Exim version 4.60
2081-----------------
2082
cc38ddbf
PH
2083PH/01 Two changes to the default runtime configuration:
2084
2085 (1) Move the checks for relay_from_hosts and authenticated clients from
2086 after to before the (commented out) DNS black list checks.
2087
2088 (2) Add control=submission to the relay_from_hosts and authenticated
2089 clients checks, on the grounds that messages accepted by these
2090 statements are most likely to be submissions.
5de37277 2091
72fdd6ae
PH
2092PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2093
2094 (1) Generate an error if the third argument for the ${prvs expansion is
2095 not a single digit.
2096
2097 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2098 string.
2099
2100 (3) Reset the variables that are obtained from the first argument of
2101 ${prvscheck and used in the second argument before leaving the code,
2102 because their memory is reclaimed, so using them afterwards may do
2103 silly things.
2104
2105 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2106 one (it's much easier than Tom thought :-).
2107
2108 (5) Because of (4), we can now allow for the use of $prvscheck_result
2109 inside the third argument.
cb9328de 2110
cb741023
PH
2111PH/03 For some reason, the default setting of PATH when running a command from
2112 a pipe transport was just "/usr/bin". I have changed it to
2113 "/bin:/usr/bin".
2114
f174f16e
PH
2115PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2116 anything to be listed in the output from -bV.
b2f5a032 2117
c25242d7
PH
2118PH/05 When a filter generated an autoreply, the entire To: header line was
2119 quoted in the delivery log line, like this:
2120
2121 => >A.N.Other <ano@some.domain> <original@ddress> ...
2122
2123 This has been changed so that it extracts the operative address. There
2124 may be more than one such address. If so, they are comma-separated, like
2125 this:
2126
2127 => >ano@some.domain,ona@other.domain <original@ddress> ...
2128
82c19f95
PH
2129PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2130 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2131 not being looked up in the rDNS to get a host name, Exim was showing the
2132 IP address twice in Received: lines, even though the IP addresses were
2133 identical. For example:
2134
2135 Received: from [1.2.3.4] (helo=[1.2.3.4])
2136
2137 However, if the real host name was known, it was omitting the HELO data
2138 if it matched the actual IP address. This has been tidied up so that it
2139 doesn't show the same IP address twice.
2140
d7ffbc12
PH
2141PH/07 When both +timestamp and +memory debugging was on, the value given by
2142 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2143 called by the debug printing, thereby overwriting the timestamp buffer.
2144 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2145
9f526266
PH
2146PH/08 When the original message was included in an autoreply transport, it
2147 always said "this is a copy of the message, including all the headers",
2148 even if body_only or headers_only was set. It now gives an appropriate
2149 message.
2150
87fcc8b9
PH
2151PH/09 Applied a patch from the Sieve maintainer which:
2152
2153 o fixes some comments
2154 o adds the (disabled) notify extension core
2155 o adds some debug output for the result of if/elsif tests
2156 o points to the current vacation draft in the documentation
2157 and documents the missing references header update
2158
2159 and most important:
2160
2161 o fixes a bug in processing the envelope test (when testing
2162 multiple envelope elements, the last element determinted the
2163 result)
2164
456682f5
PH
2165PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2166 Electronic Mail") by including:
2167
2168 Auto-submitted: auto-generated
2169
2170 in the messages that it generates (bounce messages and others, such as
2171 warnings). In the case of bounce messages for non-SMTP mesages, there was
2172 also a typo: it was using "Auto_submitted" (underscore instead of
2173 hyphen). Since every message generated by Exim is necessarily in response
2174 to another message, thes have all been changed to:
2175
2176 Auto-Submitted: auto-replied
2177
2178 in accordance with these statements in the RFC:
2179
2180 The auto-replied keyword:
2181
2182 - SHOULD be used on messages sent in direct response to another
2183 message by an automatic process,
2184
2185 - MUST NOT be used on manually-generated messages,
2186
2187 - MAY be used on Delivery Status Notifications (DSNs) and Message
2188 Disposition Notifications (MDNs),
2189
2190 - MUST NOT be used on messages generated by automatic or periodic
2191 processes, except for messages which are automatic responses to
2192 other messages.
2193
3e46c1aa
PH
2194PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2195 to the default Received: header definition.
456682f5 2196
49826d12
PH
2197PH/12 Added log selector acl_warn_skipped (default on).
2198
eba0c039
PH
2199PH/13 After a successful wildlsearch lookup, discard the values of numeric
2200 variables because (a) they are in the wrong storage pool and (b) even if
2201 they were copied, it wouldn't work properly because of the caching.
2202
a0d6ba8a
PH
2203PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2204 checking when decoding. Apparently there are clients that generate
2205 overlong encoded strings. Why am I not surprised?
2206
f0917727
PH
2207PH/15 If the first argument of "${if match_address" was not empty, but did not
2208 contain an "@" character, Exim crashed. Now it writes a panic log message
2209 and treats the condition as false.
2210
096fee00
PH
2211PH/16 In autoreply, treat an empty string for "once" the same as unset.
2212
024bd3c2
PH
2213PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2214 extension "envelope-auth". The code is finished and in agreement with
2215 other implementations, but there is no documentation so far and in fact,
2216 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2217 not changing the active code.
2218
2219 Print executed "if" and "elsif" statements when debugging is used. This
2220 helps a great deal to understand what a filter does.
2221
2222 Document more things not specified clearly in RFC3028. I had all this
2223 sorted out, when out of a sudden new issues came to my mind. Oops."
2224
df199fec
PH
2225PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2226 (Bugzilla #53).
2227
d27f1df3
PH
2228PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2229 canonical form (as documented). However, after a host name lookup from
2230 the IP address, check_host() was doing a simple string comparison with
2231 addresses acquired from the DNS when checking that the found name did
2232 have the original IP as one of its addresses. Since any found IPv6
2233 addresses are likely to be in abbreviated form, the comparison could
2234 fail. Luckily, there already exists a function for doing the comparison
2235 by converting both addresses to binary, so now that is used instead of
2236 the text comparison.
2237
96776534
PH
2238PH/20 There was another similar case to PH/19, when a complete host name was
2239 given in a host list; looking up its IP address could give an abbreviated
2240 form, whereas the current host's name might or might not be abbreviated.
2241 The same fix has been applied.
2242
5de37277 2243
9a799bc0
PH
2244Exim version 4.54
2245-----------------
2246
2247PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2248 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2249 It now does.
2250
99a4b039
PH
2251PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2252 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2253
2254PH/03 Typo: missing ".o" in src/pcre/Makefile.
2255
4b233853
PH
2256PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2257 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2258 Also, for "Auto-Submitted", treat anything other than "no" as
2259 non-personal, in accordance with RFC 3834. (Previously it treated
2260 anything starting "auto-" as non-personal.)
2261
8857ccfd
PH
2262TF/01 The control=submission/name=... option had a problem with syntax
2263 errors if the name included a slash character. The /name= option
2264 now slurps the rest of the string, so it can include any characters
2265 but it must come last in the list of options (after /sender_retain
2266 or /domain=).
2267
433a2980
PH
2268PH/05 Some modifications to the interface to the fake nameserver for the new
2269 testing suite.
2270
3e46c1aa 2271
9a799bc0 2272
e3a311ba
TK
2273Exim version 4.53
2274-----------------
2275
2276TK/01 Added the "success_on_redirect" address verification option. See
2277 NewStuff for rationale and an example.
2278
13b685f9
PH
2279PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2280
395ff96d
PH
2281PH/02 Patch to exigrep to allow it to work on syslog lines.
2282
5b68f6e4
PH
2283PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2284 fread() to skip over the body file's header line, because in Cygwin the
2285 header line is locked and is inaccessible.
2286
1ab52c69
PH
2287PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
2288 co-exist for some time) to make it clear that it is the Exim ID that is
2289 referenced, not the Message-ID: header line.
2290
b07e6aa3
PH
2291PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
2292 string_format() function, because snprintf() does not exist on all
2293 operating systems.
2294
254e032f
PH
2295PH/06 The use of forbid_filter_existstest now also locks out the use of the
2296 ${stat: expansion item.
2297
3af76a81
PH
2298PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
2299 protocol synchronization error", to keep the pedants happy.
2300
2548ba04
PH
2301PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
2302 well as for IRIX systems, when gcc is being used. See the host.c source
2303 file for comments.
2304
b6c6011d
PH
2305PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
2306
cf39cf57
PH
2307PH/10 Named domain lists were not working if used in a queue_smtp_domains
2308 setting.
2309
f1513293
PH
2310PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
2311 transport and to the smtp transport in LMTP mode.
2312
727549a4
PH
2313TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
2314
af46795e
PH
2315PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
2316 run a filter in a subprocess. This could lead to confusion in subsequent
2317 lookups in the parent process. There should also be a search_tidyup() at
2318 the end of the subprocess.
2319
d7b47fd0
PH
2320PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
2321 only if the host matched helo_try_verify_hosts, which caused the
2322 verification to occur when the EHLO/HELO command was issued. The ACL just
2323 tested the remembered result. Now, if a previous verification attempt has
2324 not happened, "verify = helo" does it there and then.
2325
ee744174
JJ
2326JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
2327
b582ab87
PH
2328TK/03 Fix log output including CR from clamd.
2329
41a13e0a
PH
2330PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
2331 exist provoked a memory error which could cause a segfault.
2332
f625cc5a
PH
2333PH/15 Installed PCRE 6.2
2334
2335PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2336
21f7af35
PH
2337PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2338 of the problem. Specifically, suggested +O2 rather than +O1 for the
2339 HP-UX compiler.
2340
31480e42
PH
2341PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2342
2d280592
PH
2343PH/20 If a delivery was routed to a non-standard port by means of an SRV
2344 record, the port was not correctly logged when the outgoing_port log
2345 selector was set (it logged the transort's default port).
2346
7cd1141b
PH
2347PH/21 Added support for host-specific ports to manualroute, queryprogram,
2348 fallback_hosts, and "hosts" in the smtp transport.
2349
2350PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2351 host errors such as "Connection refused".
2352
750af86e
PH
2353PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2354 authentication with radiusclient 0.4.9:
2355
2356 - Error returned from rc_read_config was caught wrongly
2357 - Username/password not passed on to radius server due to wrong length.
2358
2359 The presumption is that some radiusclient API changes for 4.51/PH/17
2360 were not taken care of correctly. The code is still untested by me (my
2361 Linux distribution still has 0.3.2 of radiusclient), but it was
2362 contributed by a Radius user.
2363
2364PH/24 When doing a callout, the value of $domain wasn't set correctly when
2365 expanding the "port" option of the smtp transport.
2366
4304270b
TK
2367TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2368 while reading a MIME header. Thanks to Tom Hughes for a patch.
2369
750af86e
PH
2370PH/24 Include config.h inside local_scan.h so that configuration settings are
2371 available.
2372
64ffc24f
PH
2373PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2374 that in an ACL for RCPT (for example), you can examine exactly what was
2375 received.
2376
5dd9625b
PH
2377PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2378 commands, but it was not correctly comparing the address with the actual
2379 client host address. Thus, it would show the EHLO address in Received:
2380 header lines when this was not necessary.
2381
5591031b
PH
2382PH/27 Added the % operator to ${eval:}.
2383
ba18e66a
PH
2384PH/28 Exim tries to create and chdir to its spool directory when it starts;
2385 it should be ignoring failures (because with -C, for example, it has lost
2386 privilege). It wasn't ignoring creation failures other than "already
2387 exists".
2388
9cec981f
PH
2389PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2390 -bV or -d is used.
2391
aa2b5c79
PH
2392PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2393 because an input line was too long, either on its own, or by virtue of
1509d3a8
PH
2394 too many continuations, the temporary file was not being removed, and the
2395 return code was incorrect.
aa2b5c79 2396
48a53b7f
PH
2397PH/31 Missing "BOOL" in function definition in filtertest.c.
2398
1c59d63b
PH
2399PH/32 Applied Sieve patches from the maintainer.
2400
671012da
TK
2401TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2402
1509d3a8
PH
2403PH/33 Added "verify = not_blind".
2404
2405PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2406 Local/Makefile (with some defaults set). These are used in built scripts
2407 such as exicyclog, but they have never been used in the exim_install
2408 script (though there are many overriding facilities there). I have
2409 arranged that the exim_install script now takes note of these two
2410 settings.
2411
2412PH/35 Installed configuration files for Dragonfly.
2413
2fe1a124
PH
2414PH/36 When a locally submitted message by a trusted user did not contain a
2415 From: header, and the sender address was obtained from -f or from an SMTP
2416 MAIL command, and the trusted user did not use -F to supply a sender
2417 name, $originator_name was incorrectly used when constructing a From:
2418 header. Furthermore, $originator_name was used for submission mode
2419 messages from external hosts without From: headers in a similar way,
2420 which is clearly wrong.
2421
8800895a
PH
2422PH/37 Added control=suppress_local_fixups.
2423
ccfdb010
PH
2424PH/38 When log_selector = +received_sender was set, and the addition of the
2425 sender made the log line's construction buffer exactly full, or one byte
2426 less than full, an overflow happened when the terminating "\n" was
2427 subsequently added.
2428
1130bfb0
PH
2429PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2430 when the result of a list match is failure because a DNS lookup failed.
2431
ebcb507f
PH
2432PH/40 RM_COMMAND is now used in the building process.
2433
c35e155c
PH
2434PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2435 the "build-* directories that it finds.
2436
95d1f782
PH
2437PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2438 address in a domain literal was a prefix of an interface address.
2439
fd6de02e
PH
2440PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2441 when verifying a sender address, unless rewrite_headers is false.
2442
58de37c5
PH
2443PH/44 Wrote a long comment about why errors_to addresses are verified as
2444 recipients, not senders.
2445
261cf466
TF
2446TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2447 the ratelimit ACL was added.
2448
3ee512ff
PH
2449PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2450
e08c430f
PH
2451PH/46 Added extra information about PostgreSQL errors to the error string.
2452
bef5a11f
PH
2453PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2454 avoiding the need to install special zones in a real server. This is
2455 backwards compatible; if it can't find the fake resolver, it drops back.
2456 Thus, both old and new test suites can be run.
2457
7546de58
TF
2458TF/02 Added util/ratelimit.pl
2459
e5d5a95f
TF
2460TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2461 clock is set back in time.
2462
2e88a017
TF
2463TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2464 Candler <B.Candler@pobox.com>.
2465
a5f65aa4
TF
2466TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2467 OK for addresses that are the result of redirections.
2468
e7726cbf
PH
2469PH/48 A number of further additions for the benefit of the new test suite,
2470 including a fake gethostbyname() that interfaces to the fake DNS resolver
2471 (see PH/47 above).
2472
a7fdad5b
TF
2473TF/06 The fix for widen_domains has also been applied to qualify_single and
2474 search_parents which are the other dnslookup options that can cause
2475 header rewrites.
2476
6af56900
PH
2477PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2478 type ("H").
2479
0925ede6
PH
2480PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2481
66afa403
TF
2482TF/07 Exim produced the error message "an SRV record indicated no SMTP
2483 service" if it encountered an MX record with an empty target hostname.
2484 The message is now "an MX or SRV record indicated no SMTP service".
2485
0154e85a
TF
2486TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2487 if the DNS of the sending site is misconfigured. This is quite a
2488 common situation. This change restores the behaviour of treating a
2489 helo verification defer as a failure.
2490
16f12c76
PH
2491PH/51 If self=fail was set on a router, the bounce message did not include the
2492 actual error message.
2493
bbe902f0 2494
e5a9dba6
PH
2495Exim version 4.52
2496-----------------
2497
2498TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2499
22c3b60b
PH
2500PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2501 command itself ended in error, the underlying message about the transport
2502 filter timeout was being overwritten with the pipe command error. Now the
2503 underlying error message should be appended to the second error message.
2504
06a9b4b5
PH
2505TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2506
c1ac6996
PH
2507PH/02 Reduce the amount of output that "make" produces by default. Full output
2508 can still be requested.
2509
9c7a242c
PH
2510PH/03 The warning log line about a condition test deferring for a "warn" verb
2511 was being output only once per connection, rather than after each
2512 occurrence (because it was using the same function as for successful
2513 "warn" verbs). This seems wrong, so I have changed it.
2514
87ba3f5f
PH
2515TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2516 it should not have, which might have caused a crash in the right
2517 circumstances, but probably never did.
2518
2519PH/04 Installed a modified version of Tony Finch's patch to make submission
2520 mode fix the return path as well as the Sender: header line, and to
2521 add a /name= option so that you can make the user's friendly name appear
2522 in the header line.
2523
29aba418
TF
2524TF/03 Added the control = fakedefer ACL modifier.
2525
fe0dab11
TF
2526TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2527 Mark Lowes for thorough testing.
870f6ba8 2528
11d337a4
TK
2529TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2530
2531TK/03 Merged latest SRS patch from Miles Wilton.
2532
415c8f3b
PH
2533PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2534 with the definition in sysexits.h (which is #included earlier).
2535 Fortunately, Exim does not actually use EX_OK. The code used to try to
2536 preserve the sysexits.h value, by assumimg that macro definitions were
2537 scanned for macro replacements. I have been disabused of this notion,
2538 so now the code just undefines EX_OK before #including unistd.h.
11d337a4 2539
958541e9
PH
2540PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2541 in the smtp transport. When a block could not be written in a single
2542 write() function, the timeout was being re-applied to each part-write.
2543 This seems wrong - if the receiver was accepting one byte at a time it
2544 would take for ever. The timeout is now adjusted when this happens. It
2545 doesn't have to be particularly precise.
2546
c206415f
TK
2547TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2548 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2549
2a4be8f9
PH
2550PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2551 without a domain if the check to <postmaster@domain> fails.
2552
1cba11c5
SC
2553SC/01 Eximstats: added -xls and the ability to specify output files
2554 (patch written by Frank Heydlauf).
2555
2556SC/02 Eximstats: use FileHandles for outputing results.
2557
2558SC/03 Eximstats: allow any combination of xls, txt, and html output.
2559
2560SC/04 Eximstats: fixed display of large numbers with -nvr option
2561
2562SC/05 Eximstats: fixed merging of reports with empty tables.
2563
2564SC/06 Eximstats: added the -include_original_destination flag
2565
2566SC/07 Eximstats: removed tabs and trailing whitespace.
2567
1005d00e
TK
2568TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2569
2570TK/06 MBOX spool code: Add real "From " MBOX separator line
2571 so the .eml file is really in mbox format (even though
2572 most programs do not really care). Patch from Alex Miller.
2573
2574TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2575 The latter is generated from $received_to and is only set if the
2576 message has one envelope recipient. SA can use these headers,
2577 obviously out-of-the-box. Patch from Alex Miller.
2578
9b4768fa
PH
2579PH/08 The ${def test on a variable was returning false if the variable's
2580 value was "0", contrary to what the specification has always said!
2581 The result should be true unless the variable is empty.
2582
2583PH/09 The syntax error of a character other than { following "${if
2584 def:variable_name" (after optional whitespace) was not being diagnosed.
2585 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2586 accidental colon was present, for example, could give incorrect results.
2587
0d7eb84a
PH
2588PH/10 Tidied the code in a number of places where the st_size field of a stat()
2589 result is used (not including appendfile, where other changes are about
2590 to be made).
2591
2592PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2593 This involved changing a lot of size variables from int to off_t. It
2594 should work with maildirs and everything.
2595
40727bee
TK
2596TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2597 spamd dying while we are connected to it.
2598
554d2369
TF
2599TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2600 <jgh@wizmail.org>
2601
1f922db1
PH
2602PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2603 function for BDB 4.3.
2604
ef213c3b
PH
2605PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2606
8ac170f3
PH
2607PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2608 the sources for PCRE have been re-arranged and more files are now
2609 involved.
2610
b1c749bb
PH
2611PH/15 The code I had for printing potentially long long variables in PH/11
2612 above was not the best (it lost precision). The length of off_t variables
2613 is now inspected at build time, and an appropriate printing format (%ld
c6c2dc1d
PH
2614 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2615 to be "long long int" or "long int". This is needed for the internal
2616 formatting function string_vformat().
b1c749bb 2617
4aac9b49
PH
2618PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2619 the configuration file to be ":syslog", then the script "guesses" where
2620 the logs files are, rather than using the compiled in default. In our
2621 case the guess is not the same as the compiled default, so the script
2622 suddenly stopped working when I started to use syslog. The patch checks
2623 to see if log_file_path is "". If so, it attempts to read it from exim
2624 with no configuration file to get the compiled in version, before it
2625 falls back to the previous guessing code."
2626
294520c8
TK
2627TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2628 implementing BATV in an Exim configuration. See NewStuff for the gory
2629 details.
2630
5bd022fe
PH
2631PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2632 Makefile that are specific to HP-UX.
2633
90e9ce59
PH
2634PH/18 If the "use_postmaster" option was set for a recipient callout together
2635 with the "random" option, the postmaster address was used as the MAIL
2636 FROM address for the random test, but not for the subsequent recipient
2637 test. It is now used for both.
2638
5ea81592
PH
2639PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2640 patch removes a few documentation additions to RFC 3028, because the
2641 latest draft now contains them. It adds the new en;ascii-case comparator
2642 and a new error check for 8bit text in MIME parts. Comparator and
2643 require names are now matched exactly. I enabled the subaddress
2644 extension, but it is not well tested yet (read: it works for me)."
2645
c6c2dc1d
PH
2646PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2647 rework some of the code of TK/09 above to avoid the hardwired use of
2648 "%lld" and "long long". Replaced the call to snprintf() with a call to
2649 string_vformat().
2650
fffffe4c
PH
2651PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2652 records point to non-existent hosts", "retry timeout exceeded", and
2653 "retry time not reached for any host after a long failure period".
ca02eafb 2654
9a26b6b2
PH
2655PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2656 experimental DomainKeys support:
2657
2658 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2659 (2) On an error such as an illegally used "control", the wrong name for
2660 the control was given.
2661
2662 These problems did NOT occur unless DomainKeys support was compiled.
2663
4aee0225
PH
2664PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2665
32d668a5
PH
2666PH/24 Added ${if match_ip condition.
2667
8187c3f3
PH
2668PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2669 databases so that it will be absolutely obvious if a crash occurs in the
2670 DB library. This is a regular occurrence (often caused by mis-matched
2671 db.h files).
2672
ff790e47 2673PH/26 Insert a lot of missing (void) casts for functions such as chown(),
f1e894f3
PH
2674 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2675 picked up on a user's system that detects such things. There doesn't seem
2676 to be a gcc warning option for this - only an attribute that has to be
2677 put on the function's prototype. It seems that in Fedora Core 4 they have
2678 set this on a number of new functions. No doubt there will be more in due
2679 course.
ff790e47 2680
5417f6d1
PH
2681PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2682 specify a transport. However, if an address that was verified by such a
2683 router was the subject of a callout, Exim crashed because it tried to
2684 read the rcpt_include_affixes from the non-existent transport. Now it
2685 just assumes that the setting of that option is false. This bug was
2686 introduced by 4.51/PH/31.
2687
59cf8544
PH
2688PH/28 Changed -d+all to exclude +memory, because that information is very
2689 rarely of interest, but it makes the output a lot bigger. People tend to
2690 do -d+all out of habit.
2691
e7ad8a65
PH
2692PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2693 code in os-type was giving problems when libc.so lives in lib64, like on
2694 x86_64 Fedora Core.
2695
ade42478
PH
2696PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2697 aren't the modern standard, and it seems that some systems' include files
2698 don't always have them. Exim was already checking for some of the newer
2699 ones like T_AAAA, and defining it itself. I've added checks for all the
2700 record types that Exim uses.
2701
182ad5cf
PH
2702PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2703 not automatically generating a new one, as it is supposed to. This
2704 prevented TLS from working. If the file did exist, but contained invalid
2705 data, a new version was generated, as expected. It was only the case of a
2706 non-existent file that was broken.
2707
b0d9fc80
TK
2708TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2709 with a change in libdomainkeys > 0.64.
2710
2711TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2712 from DNS. If the selector record carries the flag, it now has
2713 precedence over the domain-wide flag.
2714
2715TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2716
47c7a64a
PH
2717PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2718 the use of a port name that isn't defined in /etc/services) occurred, the
2719 message was deferred as in a normal delivery, and thus remained on the
2720 spool, instead of being failed because of the mua_wrapper setting. This
2721 is now fixed, and I tidied up some of the mua_wrapper messages at the
2722 same time.
2723
a388bce4
SC
2724SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2725 the messages in a hash of arrays rather than using individual hashes.
2726 This is a bit cleaner and results in dramatic memory savings, albeit
2727 at a slight CPU cost.
2728
2729SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2730 as requested by Marc Sherman.
2731
2732SC/10 Eximstats: added histograms for user specified patterns as requested
2733 by Marc Sherman.
2734
0793e4ed
SC
2735SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2736
c58b88df
PH
2737PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2738 fopen() in the content-scanning modules that did not already have it.
2739
e7ad8a65 2740
7982096b
PH
2741Exim version 4.51
2742-----------------
2743
1a46a8c5
PH
2744TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2745 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2746
2f079f46 2747TK/02 Fix ACL "control" statement not being available in MIME ACL.
1a46a8c5
PH
2748
2749TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2750
2751PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2752 to test Sieve filters that use "vacation".
2753
2754PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2755 that changes the way the GnuTLS parameters are stored in the cache file.
2756 The new format can be generated externally. For backward compatibility,
2757 if the data in the cache doesn't make sense, Exim assumes it has read an
2758 old-format file, and it generates new data and writes a new file. This
2759 means that you can't go back to an older release without removing the
2760 file.
2761
2762PH/03 A redirect router that has both "unseen" and "one_time" set does not
2763 work if there are any delivery delays because "one_time" forces the
2764 parent to be marked "delivered", so its unseen clone is never tried
2765 again. For this reason, Exim now forbids the simultaneous setting of
2766 these two options.
2767
2768PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2769 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2770 entry if you want to know the details. The fix, however, neglected to
2771 consider the case when local delivery batching is involved. The test for
2772 "previously delivered" was not happening when checking to see if an
2773 address could be batched with a previous (undelivered) one; under
2774 certain circumstances this could lead to multiple deliveries to the same
c2c19e9d 2775 address.
1a46a8c5
PH
2776
2777PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2778 in its include files, and this causes problems building Exim.
2779
2780PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2781 header_syntax) but Exim was just ignoring anything given after a slash.
2782 In particular, this caused confusion with an attempt to use "verify =
2783 reverse_host_lookup/defer_ok". An error is now given when options are
2784 supplied for verify items that do not have them. (Maybe reverse_host_
2785 lookup should have a defer_ok option, but that's a different point.)
2786
2787PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2788 defined by RFC 821) to 2048, because there were problems with some AUTH
2789 commands, and RFC 1869 says the size should be increased for extended
2790 SMTP commands that take arguments.
2791
2792PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2793 Finch).
2794
2795PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2796 "unknown" error; now it says that the functionality isn't in the binary.
8d67ada3 2797
49c2d5ea
PH
2798PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2799 an address' error message when a string expansion fails (syntax or
f331f3b6
PH
2800 whatever). Otherwise the password may appear in the log. Following change
2801 PH/42 below, there is no longer a chance of it appearing in a bounce
2802 message.
49c2d5ea 2803
bf759a8b
PH
2804PH/11 Installed exipick version 20050225.0 from John Jetmore.
2805
83364d30
PH
2806PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2807 of its addresses was ever tried. (Bugzilla bug #2.)
2808
7999bbd7
PH
2809PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2810 the result incorrectly in the debug output. (It correctly added a newline
2811 to what was transported.)
2812
7dbf77c9
PH
2813TF/01 Added $received_time.
2814
74e0617f
PH
2815PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2816 commented out examples of how to interface to a virus scanner and to
2817 SpamAssassin. Also added commented examples of av_scanner and
2818 spamd_address settings.
2819
2f079f46
PH
2820PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2821 and controls are allowed in which ACLs. There were a couple of minor
2822 errors. Some of the entries in the conditions table (which is a table of
2823 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2824 negation of where the condition IS allowed.
2825
8c841523
PH
2826PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2827
7766a4f0
PH
2828PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2829 header file does not have a version number, so I've had to invent a new
2830 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2831 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2832 radiusclient), but it was contributed by a Radius user.
2833
8b417f2c
PH
2834PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2835 files or directories, for OpenSSL.
2836
901f42cb
PH
2837PH/19 When an Exim process that is running as root has to create an Exim log
2838 file, it does so in a subprocess that runs as exim:exim so as to get the
2839 ownership right at creation (otherwise, other Exim processes might see
2840 the file with the wrong ownership). There was no test for failure of this
2841 fork() call, which would lead to the process getting stuck as it waited
2842 for a non-existent subprocess. Forks do occasionally fail when resources
2843 run out. I reviewed all the other calls to fork(); they all seem to check
2844 for failure.
2845
f9b9210e
PH
2846PH/20 When checking for unexpected SMTP input at connect time (before writing
2847 the banner), Exim was not dealing correctly with a non-positive return
2848 from the read() function. If the client had disconnected by this time,
2849 the result was a log entry for a synchronization error with an empty
2850 string after "input=" when read() returned zero. If read() returned -1
2851 (an event I could not check), uninitialized data bytes were printed.
2852 There were reports of junk text (parts of files, etc) appearing after
2853 "input=".
2854
54cdb463
PH
2855PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2856
cf00dad6
PH
2857PH/22 Added support for macro redefinition, and (re)definition in between
2858 driver and ACL definitions.
2859
acb1b346
PH
2860PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2861 forgetting to use the resulting value; it was using the unexpanded value.
2862
c5ddb310
PH
2863PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2864 hadn't been configured. The fix is from Juergen Kreileder, who
2865 understands it better than I do:
2866
2867 "Here's what I see happening with three configured cyrus_sasl
2868 authenticators configured (plain, login, cram-md5):
2869
2870 On startup auth_cyrus_sasl_init() gets called for each of these.
2871 This means three calls to sasl_listmech() without a specified mech_list.
2872 => SASL tests which mechs of all available mechs actually work
2873 => three warnings about OTP not working
2874 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2875
2876 With the patch, sasl_listmech() also gets called three times. But now
2877 SASL's mech_list option is set to the server_mech specified in the the
2878 authenticator. Or in other words, the answer from sasl_listmech()
2879 gets limited to just the mech you're testing for (which is different
2880 for each call.)
2881 => the return list contains just 'plain' or 'login', 'cram-md5' or
2882 nothing depending on the value of ob->server_mech.
2883
2884 I've just tested the patch: Authentication still works fine,
2885 unavailable mechs specified in the exim configuration are still
2886 caught, and the auth.log warnings about OTP are gone."
2887
31619da6
PH
2888PH/25 When debugging is enabled, the contents of the command line are added
2889 to the debugging output, even when log_selector=+arguments is not
2890 specified.
2891
bebaf0fc
PH
2892PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2893 answer is "GNU", and only if the return is "GNU/something" is the answer
2894 "Linux".
2895
475fe28a
PH
2896PH/27 $acl_verify_message is now set immediately after the failure of a
2897 verification in an ACL, and so is available in subsequent modifiers. In
2898 particular, the message can be preserved by coding like this:
2899
2900 warn !verify = sender
2901 set acl_m0 = $acl_verify_message
2902
2903 Previously, $acl_verify_message was set only while expanding "message"
2904 and "log_message" when a very denied access.
2905
7e8bec7a
PH
2906PH/28 Modified OS/os.c-Linux with
2907
2908 -#ifndef OS_LOAD_AVERAGE
2909 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2910
2911 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2912 nomenclature these days.)
2913
e4a89c47
PH
2914PH/29 Installed patch from the Sieve maintainer that adds the options
2915 sieve_useraddress and sieve_subaddress to the redirect router.
2916
5ca2a9a1
PH
2917PH/30 In these circumstances:
2918 . Two addresses routed to the same list of hosts;
2919 . First host does not offer TLS;
2920 . First host accepts first address;
2921 . First host gives temporary error to second address;
2922 . Second host offers TLS and a TLS session is established;
2923 . Second host accepts second address.
2924 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2925 and peerdn, if requested) that were in fact used only for the second
2926 address.
7e8bec7a 2927
c688b954
PH
2928PH/31 When doing a callout as part of verifying an address, Exim was not paying
2929 attention to any local part prefix or suffix that was matched by the
2930 router that accepted the address. It now behaves in the same way as it
2931 does for delivery: the affixes are removed from the local part unless
2932 rcpt_include_affixes is set on the transport.
2933
fed77020
PH
2934PH/32 Add the sender address, as F=<...>, to the log line when logging a
2935 timeout during the DATA phase of an incoming message.
2936
7fe1560f
PH
2937PH/33 Sieve envelope tests were broken for match types other than :is. I have
2938 applied a patch sanctioned by the Sieve maintainer.
c688b954 2939
ebb6e6d5
PH
2940PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2941 the uid or gid is negative. A case of a negative gid caused this to be
2942 noticed. The fix allows for either to be negative.
2943
9c4e8f60
PH
2944PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2945 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2946 overlooked.
2947
2948PH/36 The change PH/12 above was broken. Fixed it.
2949
d7174846
PH
2950PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2951 the grounds that routing the same address twice would always produce the
2952 same answer. This might have been true once, but it is certainly no
2953 longer true now. Routing a child address may depend on the previous
2954 routing that produced that child. Some complicated redirection strategies
2955 went wrong when messages had multiple recipients, and made Exim's
2956 behaviour dependent on the order in which the addresses were given.
2957
2958 I have moved the duplicate checking until after the routing is complete.
2959 Exim scans the addresses that are assigned to local and remote
2960 transports, and removes any duplicates. This means that more work will be
2961 done, as duplicates will always all be routed, but duplicates are
2962 presumably rare, so I don't expect this is of any significance.
2963
2964 For deliveries to pipes, files, and autoreplies, the duplicate checking
2965 still happens during the routing process, since they are not going to be
2966 routed further.
2967
cfe75fc3
PH
2968PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2969 It corrects a timeout issue with spamd. This is Ian's comment: "The
2970 background is that sometimes spamd either never reads data from a
2971 connection it has accepted, or it never writes response data. The exiscan
2972 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2973 blindly assumes that writes won't block so it may never time out."
2974
be22d70e
PH
2975PH/39 Allow G after quota size as well as K and M.
2976
0612b098
PH
2977PH/40 The value set for $authenticated_id in an authenticator may not contain
2978 binary zeroes or newlines because the value is written to log lines and
2979 to spool files. There was no check on this. Now the value is run through
2980 the string_printing() function so that such characters are converted to
2981 printable escape sequences.
2982
2e0c1448
PH
2983PH/41 $message_linecount is a new variable that contains the total number of
2984 lines in the message. Compare $body_linecount, which is the count for the
2985 body only.
2986
447d236c
PH
2987PH/42 Exim no longer gives details of delivery errors for specific addresses in
2988 bounce and delay warning messages, except in certain special cases, which
2989 are as follows:
2990
2991 (a) An SMTP error message from a remote host;
2992 (b) A message specified in a :fail: redirection;
2993 (c) A message specified in a "fail" command in a system filter;
2994 (d) A message specified in a FAIL return from the queryprogram router;
2995 (e) A message specified by the cannot_route_message router option.
2996
2997 In these cases only, Exim does include the error details in bounce and
2998 warning messages. There are also a few cases where bland messages such
2999 as "unrouteable address" or "local delivery error" are given.
3000
d20976dc
PH
3001PH/43 $value is now also set for the "else" part of a ${run expansion.
3002
f656d135
PH
3003PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3004 being worked on, but at least Exim now implements the latest version to
3005 play with."
3006
2e2a30b4
PH
3007PH/45 In a pipe transport, although a timeout while waiting for the pipe
3008 process to complete was treated as a delivery failure, a timeout while
3009 writing the message to the pipe was logged, but erroneously treated as a
3010 successful delivery. Such timeouts include transport filter timeouts. For
3011 consistency with the overall process timeout, these timeouts are now
3012 treated as errors, giving rise to delivery failures by default. However,
3013 there is now a new Boolean option for the pipe transport called
3014 timeout_defer, which, if set TRUE, converts the failures into defers for
3015 both kinds of timeout. A transport filter timeout is now identified in
3016 the log output.
3017
9176e9f0
PH
3018PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3019 systems where "make" and "gmake" are different, calling "gmake" at top
3020 level broke things. I've arranged for the value of $(MAKE) to be passed
3021 from the Makefile to this script so that it can call the same version of
3022 "make".
3023
7982096b 3024
bbe902f0
PH
3025A note about Exim versions 4.44 and 4.50
3026----------------------------------------
3027
3028Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3029changes of various kinds. As a consequence, a big documentation update was
3030needed. This delayed the release for rather longer than seemed good, especially
3031in the light of a couple of (minor) security issues. Therefore, the changes
3032that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3033release. So 4.44 and 4.50 are in effect two different branches that both start
3034from 4.43.
3035
3036I have left the 4.50 change log unchanged; it contains all the changes since
30374.43. The change log for 4.44 is below; many of its items are identical to
3038those for 4.50. This seems to be the most sensible way to preserve the
3039historical information.
3040
3041
f7b63901 3042Exim version 4.50
495ae4b0
PH
3043-----------------
3044
5fe762f6
PH
3045 1. Minor wording change to the doc/README.SIEVE file.
3046
139059f6 3047 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
5fe762f6 3048 computation of the current number of files was incorrect.
495ae4b0 3049
7086e875
PH
3050 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3051 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3052 place.
3053
35af9f61
PH
3054 4. Give more explanation in the error message when the command for a transport
3055 filter fails to execute.
3056
b668c215
PH
3057 5. There are several places where Exim runs a non-Exim command in a
3058 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3059 was being done only for the command run by the queryprogram router. It is
3060 now done for all such subprocesses. The other cases are: ${run, transport
3061 filters, and the commands run by the lmtp and pipe transports.
3062
a494b1e1
PH
3063 6. Added CONFIGURE_GROUP build-time option.
3064
3065 7. Some older OS have a limit of 256 on the maximum number of file
3066 descriptors. Exim was using setrlimit() to set 1000 as a large value
3067 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3068 systems. I've change it so that if it can't get 1000, it tries for 256.
35edf2ff 3069
c5fcb476
PH
3070 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3071 was an oversight, and furthermore, ever since the addition of extra
3072 controls (e.g. 4.43/32), the checks on when to allow different forms of
3073 "control" were broken. There should now be diagnostics for all cases when a
3074 control that does not make sense is encountered.
3075
69358f02
PH
3076 9. Added the /retain_sender option to "control=submission".
3077
5be20824
PH
307810. $recipients is now available in the predata ACL (oversight).
3079
eb2c0248
PH
308011. Tidy the search cache before the fork to do a delivery from a message
3081 received from the command line. Otherwise the child will trigger a lookup
3082 failure and thereby defer the delivery if it tries to use (for example) a
3083 cached ldap connection that the parent has called unbind on.
3084
2a3eea10
PH
308512. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3086 of $address_data from the recipient verification was clobbered by the
3087 sender verification.
3088
308913. The value of address_data from a sender verification is now available in
3090 $sender_address_data in subsequent conditions in the ACL statement.
3091
23c7ff99
PH
309214. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3093
4deaf07d
PH
309415. Added a new option "connect=<time>" to callout options, to set a different
3095 connection timeout.
3096
926e1192
PH
309716. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3098 was its contents. (It was OK if the option was not defined at all.)
3099
650edc6f
PH
310017. A "Completed" log line is now written for messages that are removed from
3101 the spool by the -Mrm option.
3102
2c7db3f5
PH
310318. New variables $sender_verify_failure and $recipient_verify_failure contain
3104 information about exactly what failed.
3105
3d235903
PH
310619. Added -dd to debug only the daemon process.
3107
7c7ad977
PH
310820. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3109 handles timeouts, both on the server side and network timeouts. Renamed the
3110 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3111
981756db
PH
311221. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3113 It is now set to "smtps".
3114
d4eb88df
PH
311522. $host_address is now set to the target address during the checking of
3116 ignore_target_hosts.
3117
311823. When checking ignore_target_hosts for an ipliteral router, no host name was
3119 being passed; this would have caused $sender_host_name to have been used if
3120 matching the list had actually called for a host name (not very likely,
3121 since this list is usually IP addresses). A host name is now passed as
3122 "[x.x.x.x]".
3123
7d468ab8
PH
312424. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3125 code that specifies a non-restarting handler (typically sigaction() in
3126 modern systems) in an attempt to fix a rare and obscure crash bug.
3127
312825. Narrowed the window for a race in the daemon that could cause it to ignore
3129 SIGCHLD signals. This is not a major problem, because they are used only to
3130 wake it up if nothing else does.
3131
62c0818f
PH
313226. A malformed maildirsize file could cause Exim to calculate negative values
3133 for the mailbox size or file count. Odd effects could occur as a result.
3134 The maildirsize information is now recalculated if the size or filecount
3135 end up negative.
3136
26034054
PH
313727. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3138 support for a long time. Removed HAVE_SYS_VFS_H.
3139
af66f652
PH
314028. Installed the latest version of exipick from John Jetmore.
3141
90af77f4
PH
314229. In an address list, if the pattern was not a regular expression, an empty
3143 subject address (from a bounce message) matched only if the pattern was an
3144 empty string. Non-empty patterns were not even tested. This was the wrong
3145 because it is perfectly reasonable to use an empty address as part of a
3146 database query. An empty address is now tested by patterns that are
3147 lookups. However, all the other forms of pattern expect the subject to
3148 contain a local part and a domain, and therefore, for them, an empty
3149 address still always fails if the pattern is not itself empty.
3150
d8ef3577
PH
315130. Exim went into a mad DNS loop when attempting to do a callout where the
3152 host was specified on an smtp transport, and looking it up yielded more
3153 than one IP address.
3154
5cb8cbc6
PH
315531. Re-factored the code for checking spool and log partition space into a
3156 function that finds that data and another that does the check. The former
3157 is then used to implement four new variables: $spool_space, $log_space,
3158 $spool_inodes, and $log_inodes.
3159
14702f5b
PH
316032. The RFC2047 encoding function was originally intended for short strings
3161 such as real names; it was not keeping to the 75-character limit for
3162 encoded words that the RFC imposes. It now respects the limit, and
3163 generates multiple encoded words if necessary. To be on the safe side, I
3164 have increased the buffer size for the ${rfc2047: expansion operator from
3165 1024 to 2048 bytes.
3166
063b1e99
PH
316733. It is now permitted to omit both strings after an "if" condition; if the
3168 condition is true, the result is "true". As before, when the second string
3169 is omitted, a false condition yields an empty string. This makes it less
3170 cumbersome to write custom ACL and router conditions.
3171
652e1b65
PH
317234. Failure to deliver a bounce message always caused it to be frozen, even if
3173 there was an errors_to setting on the router. The errors_to setting is now
3174 respected.
3175
6f0c9a4f
PH
317635. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3177 canonical form (fully expanded) before being placed in
3178 $sender_host_address.
3179
33397d19
PH
318036. The table in the code that translates DNS record types into text (T_A to
3181 "A" for instance) was missing entries for NS and CNAME. It is just possible
3182 that this could have caused confusion if both these types were looked up
3183 for the same domain, because the text type is used as part of Exim's
3184 per-process caching. But the chance of anyone hitting this buglet seems
3185 very small.
3186
7bb56e1f
PH
318737. The dnsdb lookup has been extended in a number of ways.
3188
3189 (1) There is a new type, "zns", which walks up the domain tree until it
3190 finds some nameserver records. It should be used with care.
3191
ea3bc19b
PH
3192 (2) There is a new type, "mxh", which is like "mx" except that it returns
3193 just the host names, not the priorities.
3194
3195 (3) It is now possible to give a list of domains (or IP addresses) to be
ff4dbb19
PH
3196 looked up. The behaviour when one of the lookups defers can be
3197 controlled by a keyword.
7bb56e1f 3198
ea3bc19b 3199 (4) It is now possible to specify the separator character for use when
7bb56e1f 3200 multiple records are returned.
33397d19 3201
0bcb2a0e
PH
320238. The dnslists ACL condition has been extended: it is now possible to supply
3203 a list of IP addresses and/or domains to be looked up in a particular DNS
3204 domain.
3205
2ac0e484
PH
320639. Added log_selector=+queue_time_overall.
3207
4e1fde53
PH
320840. When running the queue in the test harness, wait just a tad after forking a
3209 delivery process, to get repeatability of debugging output.
3210
de365ded
PH
321141. Include certificate and key file names in error message when GnuTLS fails
3212 to set them up, because the GnuTLS error message doesn't include the name
3213 of the failing file when there is a problem reading it.
3214
f05da2e8
PH
321542. Allow both -bf and -bF in the same test run.
3216
d6453af2
PH
321743. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3218
f7b63901
PH
321944. The "Exiscan patch" is now merged into the mainline Exim source.
3220
322145. Sometimes the final signoff response after QUIT could fail to get
3222 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3223 < 0 before doing a fflush(). This bug looks as though it goes back to the
3224 introduction of TLS in release 3.20, but "sometimes" must have been rare
3225 because the tests only now provoked it.
3226
a444213a
PH
322746. Reset the locale to "C" after calling embedded Perl, in case it was changed
3228 (this can affect the format of dates).
3229
0ec020ea
PH
323047. exim_tidydb, when checking for the continued existence of a message for
3231 which it has found a message-specific retry record, was not finding
3232 messages that were in split spool directories. Consequently, it was
3233 deleting retry records that should have stayed in existence.
3234
b1206957
PH
323548. Steve fixed some bugs in eximstats.
3236
323749. The SPA authentication driver was not abandoning authentication and moving
3238 on to the next authenticator when an expansion was forced to fail,
3239 contradicting the general specification for all authenticators. Instead it
3240 was generating a temporary error. It now behaves as specified.
3241
26dd5a95
PH
324250. The default ordering of permitted cipher suites for GnuTLS was pessimal
3243 (the order specifies the preference for clients). The order is now AES256,
3244 AES128, 3DES, ARCFOUR128.
3245
343b2385
PH
324651. Small patch to Sieve code - explicitly set From: when generating an
3247 autoreply.
3248
1c5466b9
PH
324952. Exim crashed if a remote delivery caused a very long error message to be
3250 recorded - for instance if somebody sent an entire SpamAssassin report back
3251 as a large number of 550 error lines. This bug was coincidentally fixed by
3252 increasing the size of one of Exim's internal buffers (big_buffer) that
3253 happened as part of the Exiscan merge. However, to be on the safe side, I
3254 have made the code more robust (and fixed the comments that describe what
3255 is going on).
3256
55ee9ee3
PH
325753. Now that there can be additional text after "Completed" in log lines (if
3258 the queue_time_overall log selector is set), a one-byte patch to exigrep
3259 was needed to allow it to recognize "Completed" as not the last thing in
3260 the line.
3261
d38f8232
PH
326254. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3263 patch that reportedly fixes this has been added. I am not expert enough to
3264 create a test for it. This is what the patch creator wrote:
3265
3266 "I found a little strange behaviour of ldap code when working with
3267 Windows 2003 AD Domain, where users was placed in more than one
3268 Organization Units. When I tried to give exim partial DN, the exit code
3269 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
3270 But simultaneously result of request was absolutely normal ldap result,
3271 so I produce this patch..."
3272
3295e65b
PH
3273 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
3274 REFERENCE, so I have modified the code to exclude the patch when that macro
3275 is not defined.
3276
7102e136
PH
327755. Some experimental protocols are using DNS PTR records for new purposes. The
3278 keys for these records are domain names, not reversed IP addresses. The
b975ba52
PH
3279 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3280 leaves it alone. Component reversal etc. now happens only for IP addresses.
ea3a6f44 3281 CAN-2005-0021
7102e136 3282
3ca0ba97
PH
328356. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
3284
c2bcbe20
PH
328557. Double the size of the debug message buffer (to 2048) so that more of very
3286 long debug lines gets shown.
3287
18ce445d
PH
328858. The exicyclog utility now does better if the number of log files to keep
3289 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
3290
1f5b4c3d
PH
329159. Two changes related to the smtp_active_hostname option:
3292
3293 (1) $smtp_active_hostname is now available as a variable.
3294 (2) The default for smtp_banner uses $smtp_active_hostname instead
3295 of $primary_hostname.
3296
b975ba52
PH
329760. The host_aton() function is supposed to be passed a string that is known
3298 to be a valid IP address. However, in the case of IPv6 addresses, it was
3299 not checking this. This is a hostage to fortune. Exim now panics and dies
3300 if the condition is not met. A case was found where this could be provoked
85b87bc2
PH
3301 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3302 components; fortuitously, this particular loophole had already been fixed
3303 by change 4.50/55 above.
3304
3305 If there are any other similar loopholes, the new check in host_aton()
3306 itself should stop them being exploited. The report I received stated that
3307 data on the command line could provoke the exploit when Exim was running as
3308 exim, but did not say which command line option was involved. All I could
3309 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3310 running as the user.
ea3a6f44 3311 CAN-2005-0021
85b87bc2
PH
3312
331361. There was a buffer overflow vulnerability in the SPA authentication code
3314 (which came originally from the Samba project). I have added a test to the
3315 spa_base64_to_bits() function which I hope fixes it.
ea3a6f44 3316 CAN-2005-0022
b975ba52 3317
17ffcae7
PH
331862. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
3319 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
3320
d95f9fdb
PH
332163. The daemon start-up calls getloadavg() while still root for those OS that
3322 need the first call to be done as root, but it missed one case: when
3323 deliver_queue_load_max is set with deliver_drop_privilege. This is
3324 necessary for the benefit of the queue runner, because there is no re-exec
3325 when deliver_drop_privilege is set.
3326
86b8287f
PH
332764. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
3328 This has been fixed.
3329
60dc5e56
PH
333065. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3331 was in use, was not putting the data itself into the right store pool;
3332 consequently, it could be overwritten for a subsequent message in the same
3333 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3334 the caching.)
3335
533244af
PH
333666. Added hosts_max_try_hardlimit to the smtp transport, default 50.
3337
a5a28604
PH
333867. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
3339 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
3340 the function were treating the return as a boolean value, which happened to
3341 work because 0=false and not-0=true, but is not correct code.
3342
7e634d24
PH
334368. The host_aton() function was not handling scoped IPv6 addresses (those
3344 with, for example, "%eth0" on the end) correctly.
3345
3e11c26b
PH
334669. Fixed some compiler warnings in acl.c for the bitmaps specified with
3347 negated items (that is, ~something) in unsigned ints. Some compilers
3348 apparently mutter when there is no cast.
3349
6729cf78
PH
335070. If an address verification called from an ACL failed, and did not produce a
3351 user-specific message (i.e. there was only a "system" message), nothing was
3352 put in $acl_verify_message. In this situation, it now puts the system
3353 message there.
3354
00f00ca5
PH
335571. Change 4.23/11 added synchronization checking at the start of an SMTP
3356 session; change 4.31/43 added the unwanted input to the log line - except
3357 that it did not do this in the start of session case. It now does.
3358
c9bdd01c
PH
335972. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3360 This is wrong and can cause the other end to generate a synchronization
3361 error if it is another Exim or anything else that does the synchronization
3362 check. A QUIT command is no longer sent after a timeout.
3363
d43194df
PH
336473. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3365 during host lookups.
3366
fe5b5d0b
PH
336774. The defer_ok option of callout verification was not working if it was used
3368 when verifying addresses in header lines, that is, for this case:
3369
3370 verify = header_sender/callout=defer_ok
3371
76a2d7ba
PH
337275. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3373 those file descriptors could be used for SMTP connections. If anything
3374 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3375 could be sent to the SMTP client, causing chaos. The daemon now opens
3376 stdin, stdout, and stderr to /dev/null when it puts itself into the
3377 background.
3378
337976. Arrange for output from Perl's "warn" command to be written to Exim's main
3380 log by default. The user can override this with suitable Perl magic.
3381
04f7d5b9
PH
338277. The use of log_message on a "discard" ACL verb, which is supposed to add to
3383 the log message when discard triggers, was not working for the DATA ACL or
3384 for the non-SMTP ACL.
3385
bc60667e
PH
338678. Error message wording change in sieve.c.
3387
bb6e88ff
PH
338879. If smtp_accept_max_per_host was set, the number of connections could be
3389 restricted to fewer than expected, because the daemon was trying to set up
3390 a new connection before checking whether the processes handling previous
3391 connec