configure.default handle IPv6 localhost better.
[exim.git] / doc / doc-txt / ChangeLog
1 Change log file for Exim from version 4.21
2 -------------------------------------------
4 Exim version 4.82
5 -----------------
7 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
9 PP/02 Make -n do something, by making it not do something.
10 When combined with -bP, the name of an option is not output.
12 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
13 by GnuTLS.
15 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
16 $sender_host_name and config options to manage this, and basic check
17 routines.
19 PP/05 DSCP support for outbound connections and control modifier for inbound.
21 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
22 (Only plugin which currently uses this is kerberos4, which nobody should
23 be using, but we should make it available and other future plugins might
24 conceivably use it, even though it would break NAT; stuff *should* be
25 using channel bindings instead).
27 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
28 name; added for Sendmail compatibility; requires admin caller.
29 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
30 just ignore it); requires trusted caller.
31 Also parse but ignore: -Ac -Am -X<logfile>
32 Bugzilla 1117.
34 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
36 TL/02 Add +smtp_confirmation as a default logging option.
38 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
39 Patch by Magnus Holmgren from 2007-02-20.
41 TL/04 Bugzilla 1281 - Spec typo.
42 Bugzilla 1283 - Spec typo.
43 Bugzilla 1290 - Spec grammar fixes.
45 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
47 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
49 JH/02 Support "G" suffix to numbers in ${if comparisons.
51 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
53 NM/01 Bugzilla 1197 - Spec typo
54 Bugzilla 1196 - Spec examples corrections
56 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
58 PP/09 Add gnutls_enable_pkcs11 option.
60 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
61 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
63 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
64 "acl {{name}{arg}...}", and optional args on acl condition
65 "acl = name arg..."
67 JH/05 Permit multiple router/transport headers_add/remove lines.
69 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
71 JH/07 Avoid using a waiting database for a single-message-only transport.
72 Performance patch from Paul Fisher. Bugzilla 1262.
74 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
75 Bugzilla 884.
77 JH/09 Add $headers_added variable, with content from use of ACL modifier
78 add_header (but not yet added to the message). Bugzilla 199.
80 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
81 Pulled from Bugzilla 817 by Wolfgang Breyha.
83 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
84 CVE-2012-5671
86 JH/11 Add A= logging on delivery lines, and a client_set_id option on
87 authenticators.
89 JH/12 Add optional authenticated_sender logging to A= and a log_selector
90 for control.
92 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
94 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
95 advertise SMTP AUTH mechanism to us, instead of a generic
96 protocol violation error. Also, make Exim more robust to bad
97 data from the Dovecot auth socket.
99 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
101 When a queue runner is handling a message, Exim first routes the
102 recipient addresses, during which it prunes them based on the retry
103 hints database. After that it attempts to deliver the message to
104 any remaining recipients. It then updates the hints database using
105 the retry rules.
107 So if a recipient address works intermittently, it can get repeatedly
108 deferred at routing time. The retry hints record remains fresh so the
109 address never reaches the final cutoff time.
111 This is a fairly common occurrence when a user is bumping up against
112 their storage quota. Exim had some logic in its local delivery code
113 to deal with this. However it did not apply to per-recipient defers
114 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
116 This change adds a proper retry rule check during routing so that the
117 final cutoff time is checked against the message's age. We only do
118 this check if there is an address retry record and there is not a
119 domain retry record; this implies that previous attempts to handle
120 the address had the retry_use_local_parts option turned on. We use
121 this as an approximation for the destination being like a local
122 delivery, as in LMTP.
124 I suspect this new check makes the old local delivery cutoff check
125 redundant, but I have not verified this so I left the code in place.
127 TF/02 Correct gecos expansion when From: is a prefix of the username.
129 Test 0254 submits a message to Exim with the header
131 Resent-From: f
133 When I ran the test suite under the user fanf2, Exim expanded
134 the header to contain my full name, whereas it should have added
135 a Resent-Sender: header. It erroneously treats any prefix of the
136 username as equal to the username.
138 This change corrects that bug.
140 GF/01 DCC debug and logging tidyup
141 Error conditions log to paniclog rather than rejectlog.
142 Debug lines prefixed by "DCC: " to remove any ambiguity.
144 TF/03 Avoid unnecessary rebuilds of lookup-related code.
146 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
147 Bug spotted by Jeremy Harris; was flawed since initial commit.
148 Would have resulted in OCSP responses post-SNI triggering an Exim
149 NULL dereference and crash.
151 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
154 Bug detection, analysis and fix by Samuel Thibault.
155 Bugzilla 1331, Debian bug #698092.
157 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
159 JH/14 SMTP PRDR (
160 Server implementation by Todd Lyons, client by JH.
161 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
162 config variable "prdr_enable" controls whether the server
163 advertises the facility. If the client requests PRDR a new
164 acl_data_smtp_prdr ACL is called once for each recipient, after
165 the body content is received and before the acl_smtp_data ACL.
166 The client is controlled by bolth of: a hosts_try_prdr option
167 on the smtp transport, and the server advertisement.
168 Default client logging of deliveries and rejections involving
169 PRDR are flagged with the string "PRDR".
171 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
172 fclose(). Diagnosis by Todd Lyons.
174 PP/17 Update configure.default to handle IPv6 localhost better.
175 Patch by Alain Williams (plus minor tweaks).
176 Bugzilla 880.
179 Exim version 4.80.1
180 -------------------
182 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
183 CVE-2012-5671
184 This, or similar/improved, will also be change PP/11 of 4.82.
187 Exim version 4.80
188 -----------------
190 PP/01 Handle short writes when writing local log-files.
191 In practice, only affects FreeBSD (8 onwards).
192 Bugzilla 1053, with thanks to Dmitry Isaikin.
194 NM/01 Bugzilla 949 - Documentation tweak
196 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
197 improved.
199 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
201 PP/02 Implemented gsasl authenticator.
203 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
205 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
206 `pkg-config foo` for cflags/libs.
208 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
209 with rest of GSASL and with heimdal_gssapi.
211 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
212 `pkg-config foo` for cflags/libs for the TLS implementation.
214 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
215 properties get this fed in as external SSF. A number of robustness
216 and debugging improvements to the cyrus_sasl authenticator.
218 PP/08 cyrus_sasl server now expands the server_realm option.
220 PP/09 Bugzilla 1214 - Log authentication information in reject log.
221 Patch by Jeremy Harris.
223 PP/10 Added dbmjz lookup type.
225 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
227 PP/12 MAIL args handles TAB as well as SP, for better interop with
228 non-compliant senders.
229 Analysis and variant patch by Todd Lyons.
231 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
232 Bug report from Lars Müller <> (via SUSE),
233 Patch from Dirk Mueller <>
235 PP/13 tls_peerdn now print-escaped for spool files.
236 Observed some $tls_peerdn in wild which contained \n, which resulted
237 in spool file corruption.
239 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
240 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
241 or write after TLS renegotiation, which otherwise led to messages
242 "Got SSL error 2".
244 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
245 as a tracking header (ie: a signed header comes before the signature).
246 Patch from Wolfgang Breyha.
248 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
249 comma-sep list; embedded commas doubled.
251 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
253 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
254 diagnostics.
255 Report and patch from Dmitry Banschikov.
257 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
258 Removed SSL_clear() after SSL_new() which led to protocol negotiation
259 failures. We appear to now support TLS1.1+ with Exim.
261 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
262 lets Exim select keys and certificates based upon TLS SNI from client.
263 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
264 before an outbound SMTP session. New log_selector, +tls_sni.
266 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
267 NULL dereference. Report and patch from Alun Jones.
269 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
270 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
271 Not seeing resolver debug output on NetBSD, but suspect this is a
272 resolver implementation change.
274 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
275 Left warnings. Added "eximon gdb" invocation mode.
277 PP/21 Defaulting "accept_8bitmime" to true, not false.
279 PP/22 Added -bw for inetd wait mode support.
281 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
282 locate the relevant includes and libraries. Made this the default.
284 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
285 Bugzilla 1246, report and most of solution from Tomasz Kusy.
287 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
288 This may cause build issues on older platforms.
290 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
291 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
292 gnutls_require_mac & gnutls_require_protocols (no longer supported).
293 Added SNI support via GnuTLS too.
294 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
296 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
298 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
299 Applied second patch from Janne, implementing suggestion to default
300 multiple-strings-in-record handling to match SPF spec.
302 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
304 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
305 read-only, out of scope).
306 Patch from Wolfgang Breyha, report from Stuart Northfield.
308 PP/29 Fix three issues highlighted by clang analyser static analysis.
309 Only crash-plausible issue would require the Cambridge-specific
310 iplookup router and a misconfiguration.
311 Report from Marcin Mirosław.
313 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
315 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
316 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
317 As part of this, removing so much warning spew let me fix some minor
318 real issues in debug logging.
320 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
321 assignment on my part. Fixed.
323 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
324 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
325 Janne Snabb (who went above and beyond: thank you).
327 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
328 string otherwise requires a connection and a bunch more work and it's
329 relatively easy to get wrong. Should also expose TLS library linkage
330 problems.
332 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
333 64-bit ${eval} (JH/03).
335 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
336 GNU libc to support some of the 64-bit stuff, should not lead to
337 conflicts. Defined before os.h is pulled in, so if a given platform
338 needs to override this, it can.
340 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
341 protection layer was required, which is not implemented.
342 Bugzilla 1254, patch from Wolfgang Breyha.
344 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
345 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
346 tls_dhparam take prime identifiers. Also unbreak combination of
347 OpenSSL+DH_params+TLSSNI.
349 PP/39 Disable SSLv2 by default in OpenSSL support.
352 Exim version 4.77
353 -----------------
355 PP/01 Solaris build fix for Oracle's LDAP libraries.
356 Bugzilla 1109, patch from Stephen Usher.
358 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
360 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
361 whitespace trailer
363 TF/02 Fix a couple more cases where we did not log the error message
364 when unlink() failed. See also change 4.74-TF/03.
366 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
367 lock up or crash if it happened to be inside a call to libc when it
368 got a SIGUSR1 from exiwhat.
370 The SIGUSR1 handler appends the current process status to the process
371 log which is later printed by exiwhat. It used to use the general
372 purpose logging code to do this, but several functions it calls are
373 not safe for signals.
375 The new output code in the SIGUSR1 handler is specific to the process
376 log, and simple enough that it's easy to inspect for signal safety.
377 Removing some special cases also simplifies the general logging code.
378 Removing the spurious timestamps from the process log simplifies
379 exiwhat.
381 TF/04 Improved ratelimit ACL condition.
383 The /noupdate option has been deprecated in favour of /readonly which
384 has clearer semantics. The /leaky, /strict, and /readonly update modes
385 are mutually exclusive. The update mode is no longer included in the
386 database key; it just determines when the database is updated. (This
387 means that when you upgrde Exim will forget old rate measurements.)
389 Exim now checks that the per_* options are used with an update mode that
390 makes sense for the current ACL. For example, when Exim is processing a
391 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
392 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
393 must specify per_mail/readonly. If you omit the update mode it defaults to
394 /leaky where that makes sense (as before) or /readonly where required.
396 The /noupdate option is now undocumented but still supported for
397 backwards compatibility. It is equivalent to /readonly except that in
398 ACLs where /readonly is required you may specify /leaky/noupdate or
399 /strict/noupdate which are treated the same as /readonly.
401 A useful new feature is the /count= option. This is a generalization
402 of the per_byte option, so that you can measure the throughput of other
403 aggregate values. For example, the per_byte option is now equivalent
404 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
406 The per_rcpt option has been generalized using the /count= mechanism
407 (though it's more complicated than the per_byte equivalence). When it is
408 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
409 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
410 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
411 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
412 Note that using per_rcpt with a non-readonly update mode in more than
413 one ACL will cause the recipients to be double-counted. (The per_mail
414 and per_byte options don't have this problem.)
416 The handling of very low rates has changed slightly. If the computed rate
417 is less than the event's count (usually one) then this event is the first
418 after a long gap. In this case the rate is set to the same as this event's
419 count, so that the first message of a spam run is counted properly.
421 The major new feature is a mechanism for counting the rate of unique
422 events. The new per_addr option counts the number of different
423 recipients that someone has sent messages to in the last time period. It
424 behaves like per_rcpt if all the recipient addresses are different, but
425 duplicate recipient addresses do not increase the measured rate. Like
426 the /count= option this is a general mechanism, so the per_addr option
427 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
428 example, measure the rate that a client uses different sender addresses
429 with the options per_mail/unique=$sender_address. There are further
430 details in the main documentation.
432 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
434 TF/06 Removed a few PCRE remnants.
436 TF/07 Automatically extract Exim's version number from tags in the git
437 repository when doing development or release builds.
439 PP/02 Raise smtp_cmd_buffer_size to 16kB.
440 Bugzilla 879. Patch from Paul Fisher.
442 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
443 Heavily based on revision 40f9a89a from Simon Arlott's tree.
444 Bugzilla 97.
446 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
448 PP/05 Variable $av_failed, true if the AV scanner deferred.
449 Bugzilla 1078. Patch from John Horne.
451 PP/06 Stop make process more reliably on build failure.
452 Bugzilla 1087. Patch from Heiko Schlittermann.
454 PP/07 Make maildir_use_size_file an _expandable_ boolean.
455 Bugzilla 1089. Patch from Heiko Schlittermann.
457 PP/08 Handle ${run} returning more data than OS pipe buffer size.
458 Bugzilla 1131. Patch from Holger Weiß.
460 PP/09 Handle IPv6 addresses with SPF.
461 Bugzilla 860. Patch from Wolfgang Breyha.
463 PP/10 GnuTLS: support TLS 1.2 & 1.1.
464 Bugzilla 1156.
465 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
466 Bugzilla 1095.
468 PP/11 match_* no longer expand right-hand-side by default.
469 New compile-time build option, EXPAND_LISTMATCH_RHS.
470 New expansion conditions, "inlist", "inlisti".
472 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
474 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
476 PP/14 fix log_write() format string regression from TF/03.
477 Bugzilla 1152. Patch from Dmitry Isaikin.
480 Exim version 4.76
481 -----------------
483 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
485 PP/02 Harmonised TLS library version reporting; only show if debugging.
486 Layout now matches that introduced for other libraries in 4.74 PP/03.
488 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
490 PP/04 New "dns_use_edns0" global option.
492 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
493 Bugzilla 1098.
495 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
496 nb: Exim is not vulnerable to
498 TK/01 Updated PolarSSL code to 0.14.2.
499 Bugzilla 1097. Patch from Andreas Metzler.
501 PP/07 Catch divide-by-zero in ${eval:...}.
502 Fixes bugzilla 1102.
504 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
505 Bugzilla 1104.
507 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
508 format-string attack -- SECURITY: remote arbitrary code execution.
510 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
511 time unintentionally subject to list matching rules, letting the header
512 cause arbitrary Exim lookups (of items which can occur in lists, *not*
513 arbitrary string expansion). This allowed for information disclosure.
515 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
516 INT_MIN/-1 -- value coerced to INT_MAX.
519 Exim version 4.75
520 -----------------
522 NM/01 Workround for PCRE version dependency in version reporting
523 Bugzilla 1073
525 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
526 This fixes portability to compilers other than gcc, notably
527 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
529 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
530 makefiles for portability to HP-UX and POSIX correctness.
532 PP/01 Permit LOOKUP_foo enabling on the make command-line.
533 Also via indented variable definition in the Makefile.
534 (Debugging by Oliver Heesakkers).
536 PP/02 Restore caching of spamd results with expanded spamd_address.
537 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
539 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
540 Improves build reliability. Fix from: Frank Elsner
542 NM/02 Fix wide character breakage in the rfc2047 coding
543 Fixes bug 1064. Patch from Andrey N. Oktyabrski
545 NM/03 Allow underscore in dnslist lookups
546 Fixes bug 1026. Patch from Graeme Fowler
548 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
549 Code patches from Adam Ciarcinski of NetBSD.
551 NM/04 Fixed exiqgrep to cope with mailq missing size issue
552 Fixes bug 943.
554 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
555 is logged, to avoid truncation. Patch from John Horne.
557 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
558 Patch from Jakob Hirsch.
560 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
561 SQL string expansion failure details.
562 Patch from Andrey Oktyabrski.
564 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
565 Patch from Simon Arlott.
567 PP/09 New lookups functionality failed to compile on old gcc which rejects
568 extern declarations in function scope.
569 Patch from Oliver Fleischmann
571 PP/10 Use sig_atomic_t for flags set from signal handlers.
572 Check getgroups() return and improve debugging.
573 Fixed developed for diagnosis in bug 927 (which turned out to be
574 a kernel bug).
576 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
577 Patch from Mark Zealey.
579 PP/12 Bugzilla 1056: Improved spamd server selection.
580 Patch from Mark Zealey.
582 PP/13 Bugzilla 1086: Deal with maildir quota file races.
583 Based on patch from Heiko Schlittermann.
585 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
586 Patch from Uwe Doering, sign-off by Michael Haardt.
588 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
589 variable declaration deep within a block. Bug and patch from
590 Dennis Davis.
592 PP/15 lookups-Makefile IRIX compatibilty coercion.
594 PP/16 Make DISABLE_DKIM build knob functional.
596 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
597 Patch by Simon Arlott
599 TF/03 Fix valgrind.h portability to C89 compilers that do not support
600 variable argument macros. Our copy now differs from upstream.
603 Exim version 4.74
604 -----------------
606 TF/01 Failure to get a lock on a hints database can have serious
607 consequences so log it to the panic log.
609 TF/02 Log LMTP confirmation messages in the same way as SMTP,
610 controlled using the smtp_confirmation log selector.
612 TF/03 Include the error message when we fail to unlink a spool file.
614 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
615 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
616 for maintaining out-of-tree patches for some time.
618 PP/01 Bugzilla 139: Documentation and portability issues.
619 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
620 Handle per-OS dynamic-module compilation flags.
622 PP/02 Let /dev/null have normal permissions.
623 The 4.73 fixes were a little too stringent and complained about the
624 permissions on /dev/null. Exempt it from some checks.
625 Reported by Andreas M. Kirchwitz.
627 PP/03 Report version information for many libraries, including
628 Exim version information for dynamically loaded libraries. Created
629 version.h, now support a version extension string for distributors
630 who patch heavily. Dynamic module ABI change.
632 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
633 privilege escalation vulnerability whereby the Exim run-time user
634 can cause root to append content of the attacker's choosing to
635 arbitrary files.
637 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
638 (Wolfgang Breyha)
640 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
641 If dropping privileges for untrusted macros, we disabled normal logging
642 on the basis that it would fail; for the Exim run-time user, this is not
643 the case, and it resulted in successful deliveries going unlogged.
644 Fixed. Reported by Andreas Metzler.
647 Exim version 4.73
648 -----------------
650 PP/01 Date: & Message-Id: revert to normally being appended to a message,
651 only prepend for the Resent-* case. Fixes regression introduced in
652 Exim 4.70 by NM/22 for Bugzilla 607.
654 PP/02 Include check_rfc2047_length in configure.default because we're seeing
655 increasing numbers of administrators be bitten by this.
657 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
659 PP/03 Bugzilla 994: added openssl_options main configuration option.
661 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
663 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
665 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
667 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
668 without a peer certificate, leading to a segfault because of an
669 assumption that peers always have certificates. Be a little more
670 paranoid. Problem reported by Martin Tscholak.
672 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
673 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
674 NB: ClamAV planning to remove STREAM in "middle of 2010".
675 CL also introduces -bmalware, various -d+acl logging additions and
676 more caution in buffer sizes.
678 PP/09 Implemented reverse_ip expansion operator.
680 PP/10 Bugzilla 937: provide a "debug" ACL control.
682 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
684 PP/12 Bugzilla 973: Implement --version.
686 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
688 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
690 PP/15 Bugzilla 816: support multiple condition rules on Routers.
692 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
693 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
694 ignore trailing whitespace.
696 JJ/02 prevent non-panic DKIM error from being sent to paniclog
698 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
699 "exim" to be used
701 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
702 Notification from Dr Andrew Aitchison.
704 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
705 ExtendedDetectionInfo response format.
706 Notification from John Horne.
708 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
709 compatible.
711 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
712 XSL and documented dependency on system catalogs, with examples of how
713 it normally works.
715 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
716 access.
718 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
719 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
720 configuration file which is writeable by the Exim user or group.
722 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
723 of configuration files to cover files specified with the -C option if
724 they are going to be used with root privileges, not just the default
725 configuration file.
727 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
728 option (effectively making it always true).
730 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
731 files to be used while preserving root privileges.
733 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
734 that rogue child processes cannot use them.
736 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
737 run-time user, instead of root.
739 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
740 Exim run-time user without dropping privileges.
742 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
743 result string, instead of calling string_vformat() twice with the same
744 arguments.
747 for other users. Others should always drop root privileges if they use
748 -C on the command line, even for a whitelisted configure file.
752 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
755 Exim version 4.72
756 -----------------
758 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
759 $data_path, and $header_path variables; fixed documentation bugs and
760 typos
762 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
763 exipick to access non-standard spools, including the "frozen" queue
764 (Finput)
766 NM/01 Bugzilla 965: Support mysql stored procedures.
767 Patch from Alain Williams
769 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
771 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
772 Patch from Andreas Metzler
774 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
775 Patch from Kirill Miazine
777 NM/05 Bugzilla 671: Added umask to procmail example.
779 JJ/03 installed exipick 20100323.0, fixing doc bug
781 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
782 directory. Notification and patch from Dan Rosenberg.
784 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
786 TK/02 Improve log output when DKIM signing operation fails.
788 MH/01 Treat the transport option dkim_domain as a colon separated
789 list, not as a single string, and sign the message with each element,
790 omitting multiple occurences of the same signer.
792 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
793 Bugzilla 985, 986. Patch by Simon Arlott
795 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
796 Patch by Simon Arlott
798 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
799 MBX locking. Notification from Dan Rosenberg.
802 Exim version 4.71
803 -----------------
805 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
807 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
809 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
811 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
813 NM/04 Bugzilla 915: Flag broken perl installation during build.
816 Exim version 4.70
817 -----------------
819 TK/01 Added patch by Johannes Berg that expands the main option
820 "spamd_address" if it starts with a dollar sign.
822 TK/02 Write list of recipients to X-Envelope-Sender header when building
823 the mbox-format spool file for content scanning (suggested by Jakob
824 Hirsch).
826 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
827 ( support via dccifd. Activated by
828 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
830 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
831 by Mark Daniel Reidel <>.
833 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
834 When building exim an external PCRE library is now needed -
835 PCRE is a system library on the majority of modern systems.
836 See entry on PCRE_LIBS in EDITME file.
838 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
839 conversation. Added nologin parameter to request.
840 Patch contributed by Kirill Miazine.
842 TF/01 Do not log submission mode rewrites if they do not change the address.
844 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
846 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
847 log files in place. Contributed by Roberto Lima.
849 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
851 TF/03 Bugzilla 615: When checking the local_parts router precondition
852 after a local_part_suffix or local_part_prefix option, Exim now
853 does not use the address's named list lookup cache, since this
854 contains cached lookups for the whole local part.
856 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
857 Robert Millan. Documentation is in experimental-spec.txt.
859 TF/04 Bugzilla 668: Fix parallel build (make -j).
861 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
863 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
864 Patch provided by Jan Srzednicki.
866 TF/05 Leading white space used to be stripped from $spam_report which
867 wrecked the formatting. Now it is preserved.
869 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
870 that they are available at delivery time.
872 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
874 TF/08 TLS error reporting now respects the incoming_interface and
875 incoming_port log selectors.
877 TF/09 Produce a more useful error message if an SMTP transport's hosts
878 setting expands to an empty string.
880 NM/06 Bugzilla 744: EXPN did not work under TLS.
881 Patch provided by Phil Pennock.
883 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
884 Patch provided by Richard Godbee.
886 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
887 acl_smtp_notquit, added index entry.
889 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
890 Patch provided by Eugene Bujak.
892 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
893 accept(). Patch provided by Maxim Dounin.
895 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
896 Patch provided by Phil Pennock.
898 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
900 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
901 Patch provided by Brad "anomie" Jorsch.
903 NM/14 Bugzilla 622: Added timeout setting to transport filter.
904 Patch provided by Dean Brooks.
906 TK/05 Add native DKIM support (does not depend on external libraries).
908 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
909 Patch provided by Graeme Fowler.
911 NM/16 Bugzilla 851: Documentation example syntax fix.
913 NM/17 Changed NOTICE file to remove references to embedded PCRE.
915 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
916 lsearch.
918 NM/19 Bugzilla 745: TLS version reporting.
919 Patch provided by Phil Pennock.
921 NM/20 Bugzilla 167: bool: condition support.
922 Patch provided by Phil Pennock.
924 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
925 clients. Patch provided by Phil Pennock.
927 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
928 Patch provided by Brad "anomie" Jorsch.
930 NM/23 Bugzilla 687: Fix misparses in eximstats.
931 Patch provided by Heiko Schlittermann.
933 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
934 Patch provided by Heiko Schlittermann.
936 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
937 plus update to original patch.
939 NM/26 Bugzilla 799: Documentation correction for ratelimit.
941 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
942 Patch provided by David Brownlee.
944 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
946 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
948 NM/30 Bugzilla 888: TLS documentation bugfixes.
950 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
952 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
953 Unlike the original bugzilla I have changed all shell scripts in src tree.
955 NM/33 Bugzilla 898: Transport filter timeout fix.
956 Patch by Todd Rinaldo.
958 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
959 Patch by Serge Demonchaux.
961 NM/35 Bugzilla 39: Base64 decode bug fixes.
962 Patch by Jakob Hirsch.
964 NM/36 Bugzilla 909: Correct connect() call in dcc code.
966 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
968 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
970 NM/39 Bugzilla 911: Fixed MakeLinks build script.
973 Exim version 4.69
974 -----------------
976 TK/01 Add preliminary DKIM support. Currently requires a forked version of
977 ALT-N's libdkim that I have put here:
980 Note to Michael Haardt: I had to rename some vars in sieve.c. They
981 were called 'true' and it seems that C99 defines that as a reserved
982 keyword to be used with 'bool' variable types. That means you could
983 not include C99-style headers which use bools without triggering
984 build errors in sieve.c.
986 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
987 as mailq or other aliases. Changed the --help handling significantly
988 to do whats expected. exim_usage() emits usage/help information.
990 SC/01 Added the -bylocaldomain option to eximstats.
992 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
994 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
996 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
999 Exim version 4.68
1000 -----------------
1002 PH/01 Another patch from the Sieve maintainer.
1004 PH/02 When an IPv6 address is converted to a string for single-key lookup
1005 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1006 dots are used instead of colons so that keys in lsearch files need not
1007 contain colons. This was done some time before quoting was made available
1008 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1009 (notated using the quote facility) so as to distinguish them from IPv4
1010 keys. This meant that lookups for IP addresses in host lists did not work
1011 for iplsearch lookups.
1013 This has been fixed by arranging for IPv6 addresses to be expressed with
1014 colons if the lookup type is iplsearch. This is not incompatible, because
1015 previously such lookups could never work.
1017 The situation is now rather anomolous, since one *can* have colons in
1018 ordinary lsearch keys. However, making the change in all cases is
1019 incompatible and would probably break a number of configurations.
1021 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1022 version.
1024 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1025 conversion specification without a maximum field width, thereby enabling
1026 a rogue spamd server to cause a buffer overflow. While nobody in their
1027 right mind would setup Exim to query an untrusted spamd server, an
1028 attacker that gains access to a server running spamd could potentially
1029 exploit this vulnerability to run arbitrary code as the Exim user.
1031 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1032 $primary_hostname instead of what libspf2 thinks the hosts name is.
1034 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1035 a directory entry by the name of the lookup key. Previously, if a
1036 symlink pointed to a non-existing file or a file in a directory that
1037 Exim lacked permissions to read, a lookup for a key matching that
1038 symlink would fail. Now it is enough that a matching directory entry
1039 exists, symlink or not. (Bugzilla 503.)
1041 PH/03 The body_linecount and body_zerocount variables are now exported in the
1042 local_scan API.
1044 PH/04 Added the $dnslist_matched variable.
1046 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1047 This means they are set thereafter only if the connection becomes
1048 encrypted.
1050 PH/06 Added the client_condition to authenticators so that some can be skipped
1051 by clients under certain conditions.
1053 PH/07 The error message for a badly-placed control=no_multiline_responses left
1054 "_responses" off the end of the name.
1056 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1058 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1059 (without spaces) instead of just copying the configuration text.
1061 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1063 PH/11 Added $max_received_linelength.
1065 PH/12 Added +ignore_defer and +include_defer to host lists.
1067 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1068 way in which PCRE > 7.0 is built.
1070 PH/14 Implemented queue_only_load_latch.
1072 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1073 MAIL command. The effect was to mangle the value on 64-bit systems.
1075 PH/16 Another patch from the Sieve maintainer.
1077 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1079 PH/18 If a system quota error occurred while trying to create the file for
1080 a maildir delivery, the message "Mailbox is full" was not appended to the
1081 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1082 only to a quota excession during the actual writing of the file.
1084 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1085 characters?) which causes problems in log lines. The DN values are now
1086 passed through string_printing() before being added to log lines.
1088 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1089 and InterBase are left for another time.)
1091 PH/21 Added message_body_newlines option.
1093 PH/22 Guard against possible overflow in moan_check_errorcopy().
1095 PH/23 POSIX allows open() to be a macro; guard against that.
1097 PH/24 If the recipient of an error message contained an @ in the local part
1098 (suitably quoted, of course), incorrect values were put in $domain and
1099 $local_part during the evaluation of errors_copy.
1102 Exim version 4.67
1103 -----------------
1105 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1106 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1107 Jan Srzednicki.
1109 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1110 issue a MAIL command.
1112 PH/02 In an ACL statement such as
1114 deny dnslists = X!= : X=
1116 if a client was not listed at all, or was listed with a value other than
1117, in the X list, but was listed with in the Y list,
1118 the condition was not true (as it should be), so access was not denied.
1119 The bug was that the ! inversion was incorrectly passed on to the second
1120 item. This has been fixed.
1122 PH/03 Added additional dnslists conditions == and =& which are different from
1123 = and & when the dns lookup returns more than one IP address.
1125 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1126 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1128 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1129 FSYNC, which compiles an option called disable_fsync that allows for
1130 bypassing fsync(). The documentation is heavily laced with warnings.
1132 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1134 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1135 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1136 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1137 including adding "make clean"; (3) Added -fPIC when compiling the test
1138 dynamically loaded module, to get rid of a warning.
1140 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1141 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1142 = 0s. The bug is otherwise harmless.
1144 PH/07 There was a bug in the dovecot authenticator such that the value of
1145 $auth1 could be overwritten, and so not correctly preserved, after a
1146 successful authentication. This usually meant that the value preserved by
1147 the server_setid option was incorrect.
1149 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1151 PH/09 Installed PCRE release 7.0.
1153 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1154 run for batched SMTP input. It is now run at the start of every message
1155 in the batch. While fixing this I discovered that the process information
1156 (output by running exiwhat) was not always getting set for -bs and -bS
1157 input. This is fixed, and it now also says "batched" for BSMTP.
1159 PH/11 Added control=no_pipelining.
1161 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1162 patch, slightly modified), and move the expansion of helo_data till after
1163 the connection is made in the smtp transport (so it can use these
1164 values).
1166 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1168 PH/14 Added log_selector = +pid.
1170 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1172 PH/16 Add ${if forany and ${if forall.
1174 PH/17 Added dsn_from option to vary the From: line in DSNs.
1176 PH/18 Flush SMTP output before performing a callout, unless control =
1177 no_callout_flush is set.
1179 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1180 was true (the default) a successful delivery failed to delete the retry
1181 item, thus causing premature timeout of the address. The bug is now
1182 fixed.
1184 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1186 PH/21 Long custom messages for fakedefer and fakereject are now split up
1187 into multiline reponses in the same way that messages for "deny" and
1188 other ACL rejections are.
1190 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1191 with slight modification.
1193 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1194 draft, changing the syntax and factoring some duplicate code".
1196 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1197 for deliveries of the second and subsequent messages over the same SMTP
1198 connection.
1200 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1201 ${reduce, with only minor "tidies".
1203 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1205 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1206 expansion side effects.
1208 PH/27 When a message times out after an over-quota error from an Exim-imposed
1209 quota, the bounce message says "mailbox is full". This message was not
1210 being given when it was a system quota that was exceeded. It now should
1211 be the same.
1213 MH/03 Made $recipients available in local_scan(). local_scan() already has
1214 better access to the recipient list through recipients_list[], but
1215 $recipients can be useful in postmaster-provided expansion strings.
1217 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1218 in the case of a MAIL command with additional options following the
1219 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1220 were accidentally chopped off.
1222 PH/29 SMTP synchronization checks are implemented when a command is read -
1223 there is a check that no more input is waiting when there shouldn't be
1224 any. However, for some commands, a delay in an ACL can mean that it is
1225 some time before the response is written. In this time, more input might
1226 arrive, invalidly. So now there are extra checks after an ACL has run for
1227 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1228 pipelining has not been advertised.
1230 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1232 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1233 returned in the SMTP response when smtp_return_error_details was false.
1234 This has been fixed.
1236 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1237 instead of the C I/O that was originally supplied, because problems were
1238 reported on Solaris.
1240 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1241 Exim which did not show up earlier: it was assuming that a call to
1242 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1243 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1244 was a macro that became an assignment, so it seemed to work. This has
1245 changed to a proper function call with a void return, hence the compile
1246 error. Exim's code has been fixed.
1248 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1249 cpus.
1251 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1253 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1255 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1256 because it turns out to be too big or there is a timeout) there is no
1257 "Completed" line in the log. When some messages of this type were
1258 selected by exigrep, they were listed as "not completed". Others were
1259 picked up by some special patterns. I have improved the selection
1260 criteria to be more general.
1262 PH/38 The host_find_failed option in the manualroute router can now be set
1263 to "ignore", to completely ignore a host whose IP address cannot be
1264 found. If all hosts are ignored, the behaviour is controlled by the new
1265 host_all_ignored option.
1267 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1268 homing or because of multiple MX records with /mx) generated more than
1269 one IP address, and the following item turned out to be the local host,
1270 all the secondary addresses of the first item were incorrectly removed
1271 from the list, along with the local host and any following hosts (which
1272 is what is supposed to happen).
1274 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1275 whoever called Exim into the -H file. In the case of the daemon it was
1276 behaving confusingly. When first started, it used values for whoever
1277 started the daemon, but after a SIGHUP it used the Exim user (because it
1278 calls itself on a restart). I have changed the code so that it now always
1279 uses the Exim user.
1281 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1282 message are rejected with the same error (e.g. no authentication or bad
1283 sender address), and a DATA command is nevertheless sent (as can happen
1284 with PIPELINING or a stupid MUA), the error message that was given to the
1285 RCPT commands is included in the rejection of the DATA command. This is
1286 intended to be helpful for MUAs that show only the final error to their
1287 users.
1289 PH/42 Another patch from the Sieve maintainer.
1291 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1292 Eximstats - Fixed some broken HTML links and added missing column headers
1293 (Jez Hancock).
1294 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1295 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1297 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1299 PH/43 Yet another patch from the Sieve maintainer.
1301 PH/44 I found a way to check for a TCP/IP connection going away before sending
1302 the response to the final '.' that terminates a message, but only in the
1303 case where the client has not sent further data following the '.'
1304 (unfortunately, this is allowed). However, in many cases there won't be
1305 any further data because there won't be any more messages to send. A call
1306 to select() can be used: if it shows that the input is "ready", there is
1307 either input waiting, or the socket has been closed. An attempt to read
1308 the next input character can distinguish the two cases. Previously, Exim
1309 would have sent an OK response which the client would never have see.
1310 This could lead to message repetition. This fix should cure that, at
1311 least in a lot of common cases.
1313 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1314 advertised in response to EHLO.
1317 Exim version 4.66
1318 -----------------
1320 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1321 fixed by 4.65/MH/01 (is this a record?) are fixed:
1323 (i) An empty string was always treated as zero by the numeric comparison
1324 operators. This behaviour has been restored.
1326 (ii) It is documented that the numeric comparison operators always treat
1327 their arguments as decimal numbers. This was broken in that numbers
1328 starting with 0 were being interpreted as octal.
1330 While fixing these problems I realized that there was another issue that
1331 hadn't been noticed. Values of message_size_limit (both the global option
1332 and the transport option) were treated as octal if they started with 0.
1333 The documentation was vague. These values are now always treated as
1334 decimal, and I will make that clear in the documentation.
1337 Exim version 4.65
1338 -----------------
1340 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1341 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1342 versions. (#438)
1344 MH/01 Don't check that the operands of numeric comparison operators are
1345 integers when their expansion is in "skipping" mode (fixes bug
1346 introduced by 4.64-PH/07).
1348 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1349 child addresses, Exim now panics and dies. Previously, because the count
1350 is held in a short int, deliveries were likely to be lost. As such a
1351 large number of recipients for a single message is ridiculous
1352 (performance will be very, very poor), I have chosen to impose a limit
1353 rather than extend the field.
1356 Exim version 4.64
1357 -----------------
1359 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1360 leftover -K file (the existence of which was triggered by #402).
1361 While we were at it, introduced process PID as part of the -K
1362 filename. This should rule out race conditions when creating
1363 these files.
1365 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1366 processing considerably. Previous code took too long for large mails,
1367 triggering a timeout which in turn triggers #401.
1369 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1370 in the DK code in transports.c. sendfile() is not really portable,
1371 hence the _LINUX specificness.
1373 TF/01 In the add_headers option to the mail command in an Exim filter,
1374 there was a bug that Exim would claim a syntax error in any
1375 header after the first one which had an odd number of characters
1376 in the field name.
1378 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1379 callout verification, Exim cached a "reject" for the entire domain. This
1380 is correct for most verifications, but it is not correct for a recipient
1381 verification with use_sender or use_postmaster set, because in that case
1382 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1383 case of MAIL FROM:<> rejection from other early rejections (e.g.
1384 rejection of HELO). When verifying a recipient using a non-null MAIL
1385 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1386 Whatever the result of the callout, the value of the domain cache is
1387 left unchanged (for any other kind of callout, getting as far as trying
1388 RCPT means that the domain itself is ok).
1390 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1391 gcc 4.1.1 threw up.
1393 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1394 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1395 session, the daemon ignores ECONNECT errors and logs others; it now
1396 ignores EPIPE as well.
1398 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1399 (quoted-printable decoding).
1401 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1402 later the small subsequent patch to fix an introduced bug.
1404 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1406 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1408 PH/08 An error is now given if message_size_limit is specified negative.
1410 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1411 to be given (somewhat) arbitrary names.
1413 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1414 in 4.64-PH/09.
1416 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1417 miscellaneous code fixes
1419 PH/10 Added the log_reject_target ACL modifier to specify where to log
1420 rejections.
1422 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1423 hostname. This is wrong, because it relates to the incoming message (and
1424 probably the interface on which it is arriving) and not to the outgoing
1425 callout (which could be using a different interface). This has been
1426 changed to use the value of the helo_data option from the smtp transport
1427 instead - this is what is used when a message is actually being sent. If
1428 there is no remote transport (possible with a router that sets up host
1429 addresses), $smtp_active_hostname is used.
1431 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1432 tweaks were necessary in order to get it to work (see also 21 below):
1433 (a) The code assumed that strncpy() returns a negative number on buffer
1434 overflow, which isn't the case. Replaced with Exim's string_format()
1435 function.
1436 (b) There were several signed/unsigned issues. I just did the minimum
1437 hacking in of casts. There is scope for a larger refactoring.
1438 (c) The code used strcasecmp() which is not a standard C function.
1439 Replaced with Exim's strcmpic() function.
1440 (d) The code set only $1; it now sets $auth1 as well.
1441 (e) A simple test gave the error "authentication client didn't specify
1442 service in request". It would seem that Dovecot has changed its
1443 interface. Fortunately there's a specification; I followed it and
1444 changed what the client sends and it appears to be working now.
1446 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1447 decoding.
1449 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1450 address A is aliased to B and C, where B exists and C does not. Without
1451 -v the output is "A verified" because verification stops after a
1452 successful redirection if more than one address is generated. However,
1453 with -v the child addresses are also verified. Exim was outputting "A
1454 failed to verify" and then showing the successful verification for C,
1455 with its parentage. It now outputs "B failed to verify", showing B's
1456 parentage before showing the successful verification of C.
1458 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1459 look up a TXT record in a specific list after matching in a combined
1460 list.
1462 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1463 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1464 they consult the DNS. I had assumed they would set it the way they
1465 wanted; and indeed my experiments on Linux seem to show that in some
1466 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1467 To be on the safe side, however, I have now made the interface to
1468 host_find_byname() similar to host_find_bydns(), with an argument
1469 containing the DNS resolver options. The host_find_byname() function now
1470 sets these options at its start, just as host_find_bydns() does. The smtp
1471 transport options dns_qualify_single and dns_search_parents are passed to
1472 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
1473 of host_find_byname() use the default settings of RES_DEFNAMES
1474 (qualify_single) but not RES_DNSRCH (search_parents).
1476 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
1477 spool_read_header() do less string testing, by means of a preliminary
1478 switch on the second character of optional "-foo" lines. (This is
1479 overdue, caused by the large number of possibilities that now exist.
1480 Originally there were few.) While I was there, I also converted the
1481 str(n)cmp tests so they don't re-test the leading "-" and the first
1482 character, in the hope this might squeeze out yet more improvement.
1484 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
1485 flag allowing group syntax was set by the header_syntax check but not
1486 turned off, possible causing trouble later; (2) The flag was not being
1487 set at all for the header_verify test, causing "group"-style headers to
1488 be rejected. I have now set it in this case, and also caused header_
1489 verify to ignore an empty address taken from a group. While doing this, I
1490 came across some other cases where the code for allowing group syntax
1491 while scanning a header line wasn't quite right (mostly, not resetting
1492 the flag correctly in the right place). These bugs could have caused
1493 trouble for malformed header lines. I hope it is now all correct.
1495 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
1496 with the "reply" argument non-NULL. The code, however (which originally
1497 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
1498 but it didn't always do it. This confused somebody who was copying the
1499 code for some other use. I have removed all the tests.
1501 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
1502 feature that was used to support insecure browsers during the U.S. crypto
1503 embargo. It requires special client support, and Exim is probably the
1504 only MTA that supported it -- and would never use it because real RSA is
1505 always available. This code has been removed, because it had the bad
1506 effect of slowing Exim down by computing (never used) parameters for the
1507 RSA_EXPORT functionality.
1509 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
1510 authenticator to fail if there's a tab character in the incoming data
1511 (there should never be unless someone is messing about, as it's supposed
1512 to be base64-encoded). Also added, on Timo's advice, the "secured" option
1513 if the connection is using TLS or if the remote IP is the same as the
1514 local IP, and the "valid-client-cert option" if a client certificate has
1515 been verified.
1517 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
1518 authenticators. This can be used for authorization after authentication
1519 succeeds. (In the case of plaintext, it servers for both authentication
1520 and authorization.)
1522 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1523 if any retry times were supplied.
1525 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1526 connection, where there is no client IP address to check. In this
1527 situation, the verify now always succeeds.
1529 PH/25 Applied John Jetmore's -Mset patch.
1531 PH/26 Added -bem to be like -Mset, but loading a message from a file.
1533 PH/27 In a string expansion for a processed (not raw) header when multiple
1534 headers of the same name were present, leading whitespace was being
1535 removed from all of them, but trailing whitespace was being removed only
1536 from the last one. Now trailing whitespace is removed from each header
1537 before concatenation. Completely empty headers in a concatenation (as
1538 before) are ignored.
1540 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1541 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1543 PH/29 [Removed. This was a change that I later backed out, and forgot to
1544 correct the ChangeLog entry (that I had efficiently created) before
1545 committing the later change.]
1547 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1548 address errors (4xx response to RCPT) over the same connection as other
1549 messages routed to the same hosts. Such deliveries are always "forced",
1550 so retry times are not inspected. This resulted in far too many retries
1551 for the affected addresses. The effect occurred only when there were more
1552 hosts than the hosts_max_try setting in the smtp transport when it had
1553 the 4xx errors. Those hosts that it had tried were not added to the list
1554 of hosts for which the message was waiting, so if all were tried, there
1555 was no problem. Two fixes have been applied:
1557 (i) If there are any address or message errors in an SMTP delivery, none
1558 of the hosts (tried or untried) are now added to the list of hosts
1559 for which the message is waiting, so the message should not be a
1560 candidate for sending over the same connection that was used for a
1561 successful delivery of some other message. This seems entirely
1562 reasonable: after all the message is NOT "waiting for some host".
1563 This is so "obvious" that I'm not sure why it wasn't done
1564 previously. Hope I haven't missed anything, but it can't do any
1565 harm, as the worst effect is to miss an optimization.
1567 (ii) If, despite (i), such a delivery is accidentally attempted, the
1568 routing retry time is respected, so at least it doesn't keep
1569 hammering the server.
1571 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1572 in ${readsocket because some servers need this prod.
1574 PH/32 Added some extra debug output when updating a wait-xxx database.
1576 PH/33 The hint "could be header name not terminated by colon", which has been
1577 given for certain expansion errors for a long time, was not being given
1578 for the ${if def:h_colon_omitted{... case.
1580 PH/34 The spec says: "With one important exception, whenever a domain list is
1581 being scanned, $domain contains the subject domain." There was at least
1582 one case where this was not true.
1584 PH/35 The error "getsockname() failed: connection reset by peer" was being
1585 written to the panic log as well as the main log, but it isn't really
1586 panic-worthy as it just means the connection died rather early on. I have
1587 removed the panic log writing for the ECONNRESET error when getsockname()
1588 fails.
1590 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1591 runs only) independently of the message's sender address. This meant
1592 that, if the 4xx error was in fact related to the sender, a different
1593 message to the same recipient with a different sender could confuse
1594 things. In particualar, this can happen when sending to a greylisting
1595 server, but other circumstances could also provoke similar problems.
1596 I have changed the default so that the retry time for these errors is now
1597 based a combination of the sender and recipient addresses. This change
1598 can be overridden by setting address_retry_include_sender=false in the
1599 smtp transport.
1601 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1602 remote server are returned as part of bounce messages. This was not
1603 happening for LMTP over a pipe (the lmtp transport), but now it is the
1604 same for both kinds of LMTP.
1606 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1607 in header lines that were in fact CNAMEs. This is no longer the case.
1609 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1610 and queue runs started by the daemon processed all messages. This has
1611 been fixed so that -R and -S can now usefully be given with -q<time>.
1613 PH/40 Import PCRE release 6.7 (fixes some bugs).
1615 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1617 PH/42 Give an error if -q is specified more than once.
1619 PH/43 Renamed the variables $interface_address and $interface_port as
1620 $received_ip_address and $received_port, to make it clear that these
1621 values apply to message reception, and not to the outgoing interface when
1622 a message is delivered. (The old names remain recognized, of course.)
1624 PH/44 There was no timeout on the connect() call when using a Unix domain
1625 socket in the ${readsocket expansion. There now is.
1627 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1628 be meaningful with "accept".
1630 SC/01 Eximstats V1.43
1631 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1633 SC/02 Eximstats V1.44
1634 Use a glob alias rather than an array ref in the generated
1635 parser. This improves both readability and performance.
1637 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1638 Collect SpamAssassin and rejection statistics.
1639 Don't display local sender or destination tables unless
1640 there is data to show.
1641 Added average volumes into the top table text output.
1643 SC/04 Eximstats V1.46
1644 Collect data on the number of addresses (recipients)
1645 as well as the number of messages.
1647 SC/05 Eximstats V1.47
1648 Added 'Message too big' to the list of mail rejection
1649 reasons (thanks to Marco Gaiarin).
1651 SC/06 Eximstats V1.48
1652 Mainlog lines which have GMT offsets and are too short to
1653 have a flag are now skipped.
1655 SC/07 Eximstats V1.49 (Alain Williams)
1656 Added the -emptyok flag.
1658 SC/08 Eximstats V1.50
1659 Fixes for obtaining the IP address from reject messages.
1661 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1662 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1663 whitesspace changes from 4.64-PH/27
1665 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1666 match 4.64-PH/13
1668 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1669 are found, allow negative numbers in numeric criteria)
1671 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1673 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1674 to match changes made in 4.64-PH/43
1676 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1678 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1679 "domain" is actually an IP address, and force a failure. This locks out
1680 those revolvers/nameservers that support "A-for-A" lookups, in
1681 contravention of the specifications.
1683 PH/48 When a host name was looked up from an IP address, and the subsequent
1684 forward lookup of the name timed out, the host name was left in
1685 $sender_host_name, contrary to the specification.
1687 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1688 restricted to single-key lookups, Exim was not diagnosing an error if
1689 * or *@ was used with a query-style lookup.
1691 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1693 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1694 long ago, but noone interested enough thought of it. Let's just say that
1695 the "1.1" means that there are some new functions that weren't there at
1696 some point in the past.
1698 PH/51 Error processing for expansion failure of helo_data from an smtp
1699 transport during callout processing was broken.
1701 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1702 tested/used via the -bh/-bhc/-bs options.
1704 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1705 bug, fixed in subsequent PCRE releases).
1707 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1708 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1710 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1713 Exim version 4.63
1714 -----------------
1716 SC/01 Use a glob alias rather than an array ref in eximstats generated
1717 parser. This improves both readability and performance.
1719 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1720 Don't display local sender or destination tables in eximstats unless
1721 there is data to show.
1722 Added average volumes into the eximstats top table text output.
1724 SC/03 Collect data on the number of addresses (recipients) as well
1725 as the number of messages in eximstats.
1727 TF/01 Correct an error in the documentation for the redirect router. Exim
1728 does (usually) call initgroups() when daemonizing.
1730 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1731 with consistent privilege compared to when running as a daemon.
1733 TF/03 Note in the spec that $authenticated_id is not set for local
1734 submissions from trusted users.
1736 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1737 Thanks to Dean Brooks <> for the patch.
1739 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1740 by adding some example configuration directives to the default
1741 configuration file. A little bit of work is required to uncomment the
1742 directives and define how usernames and passwords are checked, but
1743 there is now a framework to start from.
1745 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1746 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1747 without this. I don't know how relevant this is to other LDAP libraries.
1749 PH/02 Add the verb name to the "unknown ACL verb" error.
1751 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1753 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1755 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1756 directory not expanded when it should be if an expanded home directory
1757 was set for the address (which is overridden by the transport).
1759 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1760 libradius.
1762 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1763 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1764 because it is too late at that time, and has no effect.
1766 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1767 security issue with \' (bugzilla #107). I could not use the
1768 PQescapeStringConn() function, because it needs a PGconn value as one of
1769 its arguments.
1771 PH/08 When testing addresses using -bt, indicate those final addresses that
1772 are duplicates that would not cause an additional delivery. At least one
1773 person was confused, thinking that -bt output corresponded to deliveries.
1774 (Suppressing duplicates isn't a good idea as you lose the information
1775 about possibly different redirections that led to the duplicates.)
1777 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1778 systems where poll() doesn't work, in particular OS X.
1780 PH/10 Added more information to debugging output for retry time not reached.
1782 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1783 operations in malware.c.
1785 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1786 signatures.
1788 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1789 syslog_duplication set false, log lines that would normally be written
1790 both the the main log and to the reject log were not written to syslog at
1791 all.
1793 PH/14 In the default configuration, change the use of "message" in ACL warn
1794 statements to "add_header".
1796 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1797 not followed by a command (e.g. "seen endif").
1799 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1800 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1801 latter.
1803 PH/17 Added extra conditions to the default value of delay_warning_condition
1804 so that it is now:
1806 ${if or { \
1807 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1808 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1809 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1810 }{no}{yes}}
1812 The Auto-Submitted: and various List- headers are standardised, whereas I
1813 don't think Precedence: ever was.
1815 PH/18 Refactored debugging code in route_finduser() to show more information,
1816 in particular, the error code if getpwnam() issues one.
1818 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1819 This is apparently needed in addition to the PH/07 change above to avoid
1820 any possible encoding problems.
1822 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1823 but not after initializing Perl.
1825 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1826 output them only if debugging. By default they are written stderr,
1827 apparently, which is not desirable.
1829 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1830 queries.
1832 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1833 --not options
1835 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1837 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1838 authenticated or an ident call has been made. Suppress the default
1839 values for $authenticated_id and $authenticated_sender (but permit -oMai
1840 and -oMas) when testing with -bh.
1842 PH/24 Re-jigged the order of the tests in the default configuration so that the
1843 tests for valid domains and recipients precede the DNS black list and CSA
1844 tests, on the grounds that those ones are more expensive.
1846 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1847 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1848 This bug exists in every version of Exim that I still have, right back to
1849 0.12.
1851 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1852 However, an attempt to turn on case-sensitivity in a regex key by
1853 including (?-i) didn't work because the subject string was already
1854 lowercased, and the effects were non-intuitive. It turns out that a
1855 one-line patch can be used to allow (?-i) to work as expected.
1858 Exim version 4.62
1859 -----------------
1861 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1862 other effects) broke the use of negated acl sub-conditions.
1864 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1865 patch).
1867 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1868 "Deny" causes Exim to reject the incoming connection with a 554 error.
1869 Unfortunately, if there is a major crisis, such as a disk failure,
1870 tcp-wrappers gives "deny", whereas what one would like would be some
1871 kind of temporary error. A kludge has been added to help with this.
1872 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1873 554 error is used if errno is still zero or contains ENOENT (which occurs
1874 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1875 451 error is used.
1877 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1879 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1880 errors. Otherwise a message that provokes a temporary error (when other
1881 messages do not) can cause a whole host to time out.
1883 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1884 addresses were routed directly to files or pipes from a redirect router.
1885 File deliveries just didn't batch; pipe deliveries might have suffered
1886 odd errors.
1888 PH/06 A failure to get a lock for a hints database would erroneously always say
1889 "Failed to get write lock", even when it was really a read lock.
1891 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1892 of 0600. This has been changed to use the value of the lockfile_mode
1893 option (which defaults to 0600).
1895 PH/08 Applied small patch from the Sieve maintainer.
1897 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1898 folder from quota calculations, a direct delivery into this folder messed
1899 up the contents of the maildirsize file. This was because the regex was
1900 used only to exclude .Trash (or whatever) when the size of the mailbox
1901 was calculated. There was no check that a delivery was happening into an
1902 excluded directory. This bug has been fixed by ignoring all quota
1903 processing for deliveries into excluded directories.
1905 PH/10 Added the maildirfolder_create_regex option to appendfile.
1908 Exim version 4.61
1909 -----------------
1911 PH/01 The code for finding all the local interface addresses on a FreeBSD
1912 system running IPv6 was broken. This may well have applied to all BSD
1913 systems, as well as to others that have similar system calls. The broken
1914 code found IPv4 interfaces correctly, but gave incorrect values for the
1915 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1916 that it would not match correctly against @[] and not recognize the IPv6
1917 addresses as local.
1919 PH/02 The ipliteral router was not recognizing addresses of the form user@
1920 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1922 PH/03 Added disable_ipv6.
1924 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1925 decoded form, because it is most often used to construct To: headers
1926 lines in autoreplies, and the decoded form may well be syntactically
1927 invalid. However, $reply_address has leading white space removed, and all
1928 newlines turned into spaces so that the autoreply transport does not
1929 grumble.
1931 PH/05 If group was specified without a user on a router, and no group or user
1932 was specified on a transport, the group from the router was ignored.
1934 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1935 for visible compile-time settings that can be used to change these
1936 numbers, for those that want even more. Backwards compatibility with old
1937 spool files has been maintained. However, going back to a previous Exim
1938 release will lost any variables that are in spool files.
1940 PH/07 Two small changes when running in the test harness: increase delay when
1941 passing a TCP/IP connection to a new process, in case the original
1942 process has to generate a bounce, and remove special handling of
1943 (sic), which is no longer necessary.
1945 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1946 be the same on different OS.
1948 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1949 testing.
1951 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1952 whether --show-vars was specified or not
1954 JJ/02 exipick: Added support for new ACL variable spool format introduced
1955 in 4.61-PH/06
1957 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1958 syntactically invalid From: or Reply-to: line, and a filter used this to
1959 generate an autoreply, and therefore failed to obtain an address for the
1960 autoreply, Exim could try to deliver to a non-existent relative file
1961 name, causing unrelated and misleading errors. What now happens is that
1962 it logs this as a hard delivery error, but does not attempt to create a
1963 bounce message.
1965 PH/11 The exinext utility has a -C option for testing purposes, but although
1966 the given file was scanned by exinext itself; it wasn't being passed on
1967 when Exim was called.
1969 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1970 an end-of-file indication when reading a command response.
1972 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1973 compiled. In many other places in Exim, IPv6 addresses are always
1974 recognized, so I have changed this. It also means that IPv4 domain
1975 literals of the form [IPV4:n.n.n.n] are now always recognized.
1977 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1978 used if the router is not running as root, for example, when verifying at
1979 ACL time, or when using -bh. The debugging output from this situation was
1980 non-existent - all you got was a failure to exec. I have made two
1981 changes:
1983 (a) Failures to set uid/gid, the current directory, or a process leader
1984 in a subprocess such as that created by queryprogram now generate
1985 suitable debugging ouput when -d is set.
1987 (b) The queryprogram router detects when it is not running as root,
1988 outputs suitable debugging information if -d is set, and then runs
1989 the subprocess without attempting to change uid/gid.
1991 PH/15 Minor change to Makefile for building test_host (undocumented testing
1992 feature).
1994 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1995 additional section of a DNS packet that returns MX or SRV records.
1996 Instead, it always explicitly searches for A/AAAA records. This avoids
1997 major problems that occur when a DNS server includes only records of one
1998 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1999 fixed another bug: if SRV records were looked up and the corresponding
2000 address records were *not* found in the additional section, the port
2001 values from the SRV records were lost.
2003 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2004 using the correct key (the original address) when searching the retry
2005 rules in order to find which one to use for generating the retry hint.
2007 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2008 adding the default one. Similarly, if it contains a Reply-To: header, the
2009 errors_reply_to option, if set, is not used.
2011 PH/19 When calculating a retry time, Exim used to measure the "time since
2012 failure" by looking at the "first failed" field in the retry record. Now
2013 it does not use this if it is later than than the arrival time of the
2014 message. Instead it uses the arrival time. This makes for better
2015 behaviour in cases where some deliveries succeed, thus re-setting the
2016 "first failed" field. An example is a quota failure for a huge message
2017 when small messages continue to be delivered. Without this change, the
2018 "time since failure" will always be short, possible causing more frequent
2019 delivery attempts for the huge message than are intended.
2020 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2022 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2023 $1, $2, $3) because the numerical variables can be reset during some
2024 expansion items (e.g. "match"), thereby losing the authentication data.
2026 PH/21 Make -bV show the size of off_t variables so that the test suite can
2027 decide whether to run tests for quotas > 2G.
2029 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2030 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2031 filecount value is greater than 2G or if a quota value is greater than 2G
2032 on a system where the size of off_t is not greater than 4, a panic error
2033 is given.
2035 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2036 never match. The debug and -bh output now contains an explicit error
2037 message indicating a malformed IPv4 address or mask.
2039 PH/24 An host item such as was being treated as the IP address
2040 without a mask. Now it is not recognized as an IP address, and
2041 PH/23 above applies.
2043 PH/25 Do not write to syslog when running in the test harness. The only
2044 occasion when this arises is a failure to open the main or panic logs
2045 (for which there is an explicit test).
2047 PH/26 Added the /no_tell option to "control=freeze".
2049 PH/27 If a host name lookup failed very early in a connection, for example, if
2050 the IP address matched host_lookup and the reverse lookup yielded a name
2051 that did not have a forward lookup, an error message of the form "no IP
2052 address found for host (during SMTP connection from NULL)"
2053 could be logged. Now it outputs the IP address instead of "NULL".
2055 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2056 allows the sender and the authenticated sender to be set when
2057 submitting a message from within Exim. Since child_open_exim() is
2058 documented for local_scan(), the new function should be too.
2060 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2061 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2062 results in an empty string is now treated as unset.
2064 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2066 PH/31 Added sender_verify_fail logging option.
2068 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2069 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2070 tidied the source and removed it altogether.
2072 PH/33 When a queue run was abandoned because the load average was too high, a
2073 log line was always written; now it is written only if the queue_run log
2074 selector is set. In addition, the log line for abandonment now contains
2075 information about the queue run such as the pid. This is always present
2076 in "start" and "stop" lines but was omitted from the "abandon" line.
2078 PH/34 Omit spaces between a header name and the colon in the error message that
2079 is given when verify = headers_syntax fails (if there are lots of them,
2080 the message gets confusing).
2082 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2083 names, as there are now some PTR records that contain slashes. This check
2084 is only to protect against broken name servers that fall over on strange
2085 characters, so the fact that it applies to all lookups doesn't matter.
2087 PH/36 Now that the new test suite is complete, we can remove some of the
2088 special code in Exim that was needed for the old test suite. For example,
2089 sorting DNS records because real resolvers return them in an arbitrary
2090 order. The new test suite's fake resolver always returns records in the
2091 same order.
2093 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2094 bounces) except when queue_only is set, to avoid logging races between
2095 the different processes.
2097 PH/38 Panic-die if .include specifies a non-absolute path.
2099 PH/39 A tweak to the "H" retry rule from its user.
2101 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2102 a label. They prevented compilation on older perls.
2104 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2105 a warning to be raised on newish perls.
2107 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2108 on queue. Changes to match documented behaviour of showing count of
2109 messages matching specified criteria.
2111 PH/40 Changed the default ident timeout from 30s to 5s.
2113 PH/41 Added support for the use of login_cap features, on those BSD systems
2114 that have them, for controlling the resources used by pipe deliveries.
2116 PH/42 The content-scanning code uses fopen() to create files in which to put
2117 message data. Previously it was not paying any attention to the mode of
2118 the files. Exim runs with umask(0) because the rest of the code creates
2119 files with open(), and sets the required mode explicitly. Thus, these
2120 files were ending up world-writeable. This was not a big issue, because,
2121 being within the spool directory, they were not world-accessible. I have
2122 created a function called modefopen, which takes an additional mode
2123 argument. It sets umask(777), creates the file, chmods it to the required
2124 mode, then resets the umask. All the relevant calls to fopen() in the
2125 content scanning code have been changed to use this function.
2127 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2128 to 24 hours. This avoids potential overflow problems when processing G
2129 and H retry rules. I suspect nobody ever tinkers with this value.
2131 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2133 PH/45 When the plaintext authenticator is running as a client, the server's
2134 challenges are checked to ensure they are valid base64 strings. By
2135 default, the authentication attempt is cancelled if an invalid string is
2136 received. Setting client_ignore_invalid_base64 true ignores these errors.
2137 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2138 they are received. Thus, the responses can be made to depend on the
2139 challenges. If an invalid string is ignored, an empty string is placed in
2140 the variable.
2142 PH/46 Messages that are created by the autoreply transport now contains a
2143 References: header, in accordance with RFCs 2822 and 3834.
2145 PH/47 Added authenticated_sender_force to the smtp transport.
2147 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2149 PH/49 Installed latest patch from the Sieve maintainer.
2151 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2152 was also set, the appendfile transport was unnecessarily scanning a
2153 directory of message files (e.g. for maildir delivery) to find the count
2154 of files (along with the size), even though it did not need this
2155 information. It now does the scan only if it needs to find either the
2156 size of the count of files.
2158 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2160 PH/52 Two bugs concerned with error handling when the smtp transport is
2161 used in LMTP mode:
2163 (i) Exim was not creating retry information for temporary errors given
2164 for individual recipients after the DATA command when the smtp transport
2165 was used in LMTP mode. This meant that they could be retried too
2166 frequently, and not timed out correctly.
2168 (ii) Exim was setting the flag that allows error details to be returned
2169 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2170 recipients that were returned after the DATA command.
2172 PH/53 This is related to PH/52, but is more general: for any failing address,
2173 when detailed error information was permitted to be returned to the
2174 sender, but the error was temporary, then after the final timeout, only
2175 "retry timeout exceeded" was returned. Now it returns the full error as
2176 well as "retry timeout exceeded".
2178 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2179 do this, and (what is worse) MTAs that accept it.
2181 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2182 will now be deprecated.
2184 PH/56 New os.c-cygwin from the Cygwin maintainer.
2186 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2187 formats (previously only available in exim formats via -bpr, -bpru,
2188 and -bpra. Now also available in native and exiqgrep formats)
2190 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2191 with very large, slow to parse queues
2193 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2195 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2197 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2198 responses to authentication challenges, though it was showing the
2199 challenges; (ii) I've removed the CR characters from the debug output for
2200 SMTP output lines.
2202 PH/58 Allow for the insertion of a newline as well as a space when a string
2203 is turned into more than one encoded-word during RFC 2047 encoding. The
2204 Sieve code now uses this.
2206 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2207 data_4xx, lost_connection, tls_required.
2209 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2210 message was being sent as an SMTP response.
2212 PH/61 Add -l and -k options to exicyclog.
2214 PH/62 When verifying, if an address was redirected to one new address, so that
2215 verification continued, and the new address failed or deferred after
2216 having set something in $address_data, the value of $address_data was not
2217 passed back to the ACL. This was different to the case when no
2218 redirection occurred. The value is now passed back in both cases.
2220 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2221 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2222 use login_cap.h, so on its own it isn't the distinguishing feature. The
2223 new name refers directly to the setclassresources() function.
2225 PH/65 Added configuration files for NetBSD3.
2227 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2229 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2230 is preferred over IPv4.
2232 PH/68 The bounce_return_message and bounce_return_body options were not being
2233 honoured for bounces generated during the reception of non-SMTP messages.
2234 In particular, this applied to messages rejected by the ACL. This bug has
2235 been fixed. However, if bounce_return_message is true and bounce_return_
2236 body is false, the headers that are returned for a non-SMTP message
2237 include only those that have been read before the error was detected.
2238 (In the case of an ACL rejection, they have all been read.)
2240 PH/69 The HTML version of the specification is now built in a directory called
2241 spec_html instead of spec.html, because the latter looks like a path with
2242 a MIME-type, and this confuses some software.
2244 PH/70 Catch two compiler warnings in sieve.c.
2246 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2247 function verify_get_ident() calls ip_connect() to connect a socket, but
2248 if the "connect()" function timed out, ip_connect() used to close the
2249 socket. However, verify_get_ident() also closes the socket later, and in
2250 between Exim writes to the log, which may get opened at this point. When
2251 the socket was closed in ip_connect(), the log could get the same file
2252 descriptor number as the socket. This naturally causes chaos. The fix is
2253 not to close the socket in ip_connect(); the socket should be closed by
2254 the function that creates it. There was only one place in the code where
2255 this was missing, in the iplookup router, which I don't think anybody now
2256 uses, but I've fixed it anyway.
2258 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2259 well as to direct DNS lookups. Otherwise the handling of names in host
2260 lists is inconsistent and therefore confusing.
2263 Exim version 4.60
2264 -----------------
2266 PH/01 Two changes to the default runtime configuration:
2268 (1) Move the checks for relay_from_hosts and authenticated clients from
2269 after to before the (commented out) DNS black list checks.
2271 (2) Add control=submission to the relay_from_hosts and authenticated
2272 clients checks, on the grounds that messages accepted by these
2273 statements are most likely to be submissions.
2275 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2277 (1) Generate an error if the third argument for the ${prvs expansion is
2278 not a single digit.
2280 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2281 string.
2283 (3) Reset the variables that are obtained from the first argument of
2284 ${prvscheck and used in the second argument before leaving the code,
2285 because their memory is reclaimed, so using them afterwards may do
2286 silly things.
2288 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2289 one (it's much easier than Tom thought :-).
2291 (5) Because of (4), we can now allow for the use of $prvscheck_result
2292 inside the third argument.
2294 PH/03 For some reason, the default setting of PATH when running a command from
2295 a pipe transport was just "/usr/bin". I have changed it to
2296 "/bin:/usr/bin".
2299 anything to be listed in the output from -bV.
2301 PH/05 When a filter generated an autoreply, the entire To: header line was
2302 quoted in the delivery log line, like this:
2304 => >A.N.Other <ano@some.domain> <original@ddress> ...
2306 This has been changed so that it extracts the operative address. There
2307 may be more than one such address. If so, they are comma-separated, like
2308 this:
2310 => >ano@some.domain,ona@other.domain <original@ddress> ...
2312 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2313 command, (for example, EHLO []) and the client's IP address was
2314 not being looked up in the rDNS to get a host name, Exim was showing the
2315 IP address twice in Received: lines, even though the IP addresses were
2316 identical. For example:
2318 Received: from [] (helo=[])
2320 However, if the real host name was known, it was omitting the HELO data
2321 if it matched the actual IP address. This has been tidied up so that it
2322 doesn't show the same IP address twice.
2324 PH/07 When both +timestamp and +memory debugging was on, the value given by
2325 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2326 called by the debug printing, thereby overwriting the timestamp buffer.
2327 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2329 PH/08 When the original message was included in an autoreply transport, it
2330 always said "this is a copy of the message, including all the headers",
2331 even if body_only or headers_only was set. It now gives an appropriate
2332 message.
2334 PH/09 Applied a patch from the Sieve maintainer which:
2336 o fixes some comments
2337 o adds the (disabled) notify extension core
2338 o adds some debug output for the result of if/elsif tests
2339 o points to the current vacation draft in the documentation
2340 and documents the missing references header update
2342 and most important:
2344 o fixes a bug in processing the envelope test (when testing
2345 multiple envelope elements, the last element determinted the
2346 result)
2348 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2349 Electronic Mail") by including:
2351 Auto-submitted: auto-generated
2353 in the messages that it generates (bounce messages and others, such as
2354 warnings). In the case of bounce messages for non-SMTP mesages, there was
2355 also a typo: it was using "Auto_submitted" (underscore instead of
2356 hyphen). Since every message generated by Exim is necessarily in response
2357 to another message, thes have all been changed to:
2359 Auto-Submitted: auto-replied
2361 in accordance with these statements in the RFC:
2363 The auto-replied keyword:
2365 - SHOULD be used on messages sent in direct response to another
2366 message by an automatic process,
2368 - MUST NOT be used on manually-generated messages,
2370 - MAY be used on Delivery Status Notifications (DSNs) and Message
2371 Disposition Notifications (MDNs),
2373 - MUST NOT be used on messages generated by automatic or periodic
2374 processes, except for messages which are automatic responses to
2375 other messages.
2377 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2378 to the default Received: header definition.
2380 PH/12 Added log selector acl_warn_skipped (default on).
2382 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2383 variables because (a) they are in the wrong storage pool and (b) even if
2384 they were copied, it wouldn't work properly because of the caching.
2386 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2387 checking when decoding. Apparently there are clients that generate
2388 overlong encoded strings. Why am I not surprised?
2390 PH/15 If the first argument of "${if match_address" was not empty, but did not
2391 contain an "@" character, Exim crashed. Now it writes a panic log message
2392 and treats the condition as false.
2394 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2396 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2397 extension "envelope-auth". The code is finished and in agreement with
2398 other implementations, but there is no documentation so far and in fact,
2399 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2400 not changing the active code.
2402 Print executed "if" and "elsif" statements when debugging is used. This
2403 helps a great deal to understand what a filter does.
2405 Document more things not specified clearly in RFC3028. I had all this
2406 sorted out, when out of a sudden new issues came to my mind. Oops."
2408 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2409 (Bugzilla #53).
2411 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2412 canonical form (as documented). However, after a host name lookup from
2413 the IP address, check_host() was doing a simple string comparison with
2414 addresses acquired from the DNS when checking that the found name did
2415 have the original IP as one of its addresses. Since any found IPv6
2416 addresses are likely to be in abbreviated form, the comparison could
2417 fail. Luckily, there already exists a function for doing the comparison
2418 by converting both addresses to binary, so now that is used instead of
2419 the text comparison.
2421 PH/20 There was another similar case to PH/19, when a complete host name was
2422 given in a host list; looking up its IP address could give an abbreviated
2423 form, whereas the current host's name might or might not be abbreviated.
2424 The same fix has been applied.
2427 Exim version 4.54
2428 -----------------
2430 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2431 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2432 It now does.
2434 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2435 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2437 PH/03 Typo: missing ".o" in src/pcre/Makefile.
2439 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2440 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2441 Also, for "Auto-Submitted", treat anything other than "no" as
2442 non-personal, in accordance with RFC 3834. (Previously it treated
2443 anything starting "auto-" as non-personal.)
2445 TF/01 The control=submission/name=... option had a problem with syntax
2446 errors if the name included a slash character. The /name= option
2447 now slurps the rest of the string, so it can include any characters
2448 but it must come last in the list of options (after /sender_retain
2449 or /domain=).
2451 PH/05 Some modifications to the interface to the fake nameserver for the new
2452 testing suite.
2456 Exim version 4.53
2457 -----------------
2459 TK/01 Added the "success_on_redirect" address verification option. See
2460 NewStuff for rationale and an example.
2462 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2464 PH/02 Patch to exigrep to allow it to work on syslog lines.
2466 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2467 fread() to skip over the body file's header line, because in Cygwin the
2468 header line is locked and is inaccessible.
2470 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
2471 co-exist for some time) to make it clear that it is the Exim ID that is
2472 referenced, not the Message-ID: header line.
2474 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
2475 string_format() function, because snprintf() does not exist on all
2476 operating systems.
2478 PH/06 The use of forbid_filter_existstest now also locks out the use of the
2479 ${stat: expansion item.
2481 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
2482 protocol synchronization error", to keep the pedants happy.
2484 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
2485 well as for IRIX systems, when gcc is being used. See the host.c source
2486 file for comments.
2488 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
2490 PH/10 Named domain lists were not working if used in a queue_smtp_domains
2491 setting.
2493 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
2494 transport and to the smtp transport in LMTP mode.
2496 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
2498 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
2499 run a filter in a subprocess. This could lead to confusion in subsequent
2500 lookups in the parent process. There should also be a search_tidyup() at
2501 the end of the subprocess.
2503 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
2504 only if the host matched helo_try_verify_hosts, which caused the
2505 verification to occur when the EHLO/HELO command was issued. The ACL just
2506 tested the remembered result. Now, if a previous verification attempt has
2507 not happened, "verify = helo" does it there and then.
2509 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
2511 TK/03 Fix log output including CR from clamd.
2513 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
2514 exist provoked a memory error which could cause a segfault.
2516 PH/15 Installed PCRE 6.2
2518 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2520 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2521 of the problem. Specifically, suggested +O2 rather than +O1 for the
2522 HP-UX compiler.
2524 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2526 PH/20 If a delivery was routed to a non-standard port by means of an SRV
2527 record, the port was not correctly logged when the outgoing_port log
2528 selector was set (it logged the transort's default port).
2530 PH/21 Added support for host-specific ports to manualroute, queryprogram,
2531 fallback_hosts, and "hosts" in the smtp transport.
2533 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2534 host errors such as "Connection refused".
2536 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2537 authentication with radiusclient 0.4.9:
2539 - Error returned from rc_read_config was caught wrongly
2540 - Username/password not passed on to radius server due to wrong length.
2542 The presumption is that some radiusclient API changes for 4.51/PH/17
2543 were not taken care of correctly. The code is still untested by me (my
2544 Linux distribution still has 0.3.2 of radiusclient), but it was
2545 contributed by a Radius user.
2547 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2548 expanding the "port" option of the smtp transport.
2550 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2551 while reading a MIME header. Thanks to Tom Hughes for a patch.
2553 PH/24 Include config.h inside local_scan.h so that configuration settings are
2554 available.
2556 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2557 that in an ACL for RCPT (for example), you can examine exactly what was
2558 received.
2560 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2561 commands, but it was not correctly comparing the address with the actual
2562 client host address. Thus, it would show the EHLO address in Received:
2563 header lines when this was not necessary.
2565 PH/27 Added the % operator to ${eval:}.
2567 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2568 it should be ignoring failures (because with -C, for example, it has lost
2569 privilege). It wasn't ignoring creation failures other than "already
2570 exists".
2572 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2573 -bV or -d is used.
2575 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2576 because an input line was too long, either on its own, or by virtue of
2577 too many continuations, the temporary file was not being removed, and the
2578 return code was incorrect.
2580 PH/31 Missing "BOOL" in function definition in filtertest.c.
2582 PH/32 Applied Sieve patches from the maintainer.
2584 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2586 PH/33 Added "verify = not_blind".
2588 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2589 Local/Makefile (with some defaults set). These are used in built scripts
2590 such as exicyclog, but they have never been used in the exim_install
2591 script (though there are many overriding facilities there). I have
2592 arranged that the exim_install script now takes note of these two
2593 settings.
2595 PH/35 Installed configuration files for Dragonfly.
2597 PH/36 When a locally submitted message by a trusted user did not contain a
2598 From: header, and the sender address was obtained from -f or from an SMTP
2599 MAIL command, and the trusted user did not use -F to supply a sender
2600 name, $originator_name was incorrectly used when constructing a From:
2601 header. Furthermore, $originator_name was used for submission mode
2602 messages from external hosts without From: headers in a similar way,
2603 which is clearly wrong.
2605 PH/37 Added control=suppress_local_fixups.
2607 PH/38 When log_selector = +received_sender was set, and the addition of the
2608 sender made the log line's construction buffer exactly full, or one byte
2609 less than full, an overflow happened when the terminating "\n" was
2610 subsequently added.
2612 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2613 when the result of a list match is failure because a DNS lookup failed.
2615 PH/40 RM_COMMAND is now used in the building process.
2617 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2618 the "build-* directories that it finds.
2620 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2621 address in a domain literal was a prefix of an interface address.
2623 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2624 when verifying a sender address, unless rewrite_headers is false.
2626 PH/44 Wrote a long comment about why errors_to addresses are verified as
2627 recipients, not senders.
2629 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2630 the ratelimit ACL was added.
2632 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2634 PH/46 Added extra information about PostgreSQL errors to the error string.
2636 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2637 avoiding the need to install special zones in a real server. This is
2638 backwards compatible; if it can't find the fake resolver, it drops back.
2639 Thus, both old and new test suites can be run.
2641 TF/02 Added util/
2643 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2644 clock is set back in time.
2646 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2647 Candler <>.
2649 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2650 OK for addresses that are the result of redirections.
2652 PH/48 A number of further additions for the benefit of the new test suite,
2653 including a fake gethostbyname() that interfaces to the fake DNS resolver
2654 (see PH/47 above).
2656 TF/06 The fix for widen_domains has also been applied to qualify_single and
2657 search_parents which are the other dnslookup options that can cause
2658 header rewrites.
2660 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2661 type ("H").
2663 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2665 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2666 service" if it encountered an MX record with an empty target hostname.
2667 The message is now "an MX or SRV record indicated no SMTP service".
2669 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2670 if the DNS of the sending site is misconfigured. This is quite a
2671 common situation. This change restores the behaviour of treating a
2672 helo verification defer as a failure.
2674 PH/51 If self=fail was set on a router, the bounce message did not include the
2675 actual error message.
2678 Exim version 4.52
2679 -----------------
2681 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2683 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2684 command itself ended in error, the underlying message about the transport
2685 filter timeout was being overwritten with the pipe command error. Now the
2686 underlying error message should be appended to the second error message.
2688 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2690 PH/02 Reduce the amount of output that "make" produces by default. Full output
2691 can still be requested.
2693 PH/03 The warning log line about a condition test deferring for a "warn" verb
2694 was being output only once per connection, rather than after each
2695 occurrence (because it was using the same function as for successful
2696 "warn" verbs). This seems wrong, so I have changed it.
2698 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2699 it should not have, which might have caused a crash in the right
2700 circumstances, but probably never did.
2702 PH/04 Installed a modified version of Tony Finch's patch to make submission
2703 mode fix the return path as well as the Sender: header line, and to
2704 add a /name= option so that you can make the user's friendly name appear
2705 in the header line.
2707 TF/03 Added the control = fakedefer ACL modifier.
2709 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2710 Mark Lowes for thorough testing.
2712 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2714 TK/03 Merged latest SRS patch from Miles Wilton.
2716 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2717 with the definition in sysexits.h (which is #included earlier).
2718 Fortunately, Exim does not actually use EX_OK. The code used to try to
2719 preserve the sysexits.h value, by assumimg that macro definitions were
2720 scanned for macro replacements. I have been disabused of this notion,
2721 so now the code just undefines EX_OK before #including unistd.h.
2723 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2724 in the smtp transport. When a block could not be written in a single
2725 write() function, the timeout was being re-applied to each part-write.
2726 This seems wrong - if the receiver was accepting one byte at a time it
2727 would take for ever. The timeout is now adjusted when this happens. It
2728 doesn't have to be particularly precise.
2730 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2731 details. Thanks to Chris Webb <> for the patch!
2733 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2734 without a domain if the check to <postmaster@domain> fails.
2736 SC/01 Eximstats: added -xls and the ability to specify output files
2737 (patch written by Frank Heydlauf).
2739 SC/02 Eximstats: use FileHandles for outputing results.
2741 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2743 SC/04 Eximstats: fixed display of large numbers with -nvr option
2745 SC/05 Eximstats: fixed merging of reports with empty tables.
2747 SC/06 Eximstats: added the -include_original_destination flag
2749 SC/07 Eximstats: removed tabs and trailing whitespace.
2751 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2753 TK/06 MBOX spool code: Add real "From " MBOX separator line
2754 so the .eml file is really in mbox format (even though
2755 most programs do not really care). Patch from Alex Miller.
2757 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2758 The latter is generated from $received_to and is only set if the
2759 message has one envelope recipient. SA can use these headers,
2760 obviously out-of-the-box. Patch from Alex Miller.
2762 PH/08 The ${def test on a variable was returning false if the variable's
2763 value was "0", contrary to what the specification has always said!
2764 The result should be true unless the variable is empty.
2766 PH/09 The syntax error of a character other than { following "${if
2767 def:variable_name" (after optional whitespace) was not being diagnosed.
2768 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2769 accidental colon was present, for example, could give incorrect results.
2771 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2772 result is used (not including appendfile, where other changes are about
2773 to be made).
2775 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2776 This involved changing a lot of size variables from int to off_t. It
2777 should work with maildirs and everything.
2779 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2780 spamd dying while we are connected to it.
2782 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2783 <>
2785 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2786 function for BDB 4.3.
2788 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2790 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2791 the sources for PCRE have been re-arranged and more files are now
2792 involved.
2794 PH/15 The code I had for printing potentially long long variables in PH/11
2795 above was not the best (it lost precision). The length of off_t variables
2796 is now inspected at build time, and an appropriate printing format (%ld
2797 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2798 to be "long long int" or "long int". This is needed for the internal
2799 formatting function string_vformat().
2801 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2802 the configuration file to be ":syslog", then the script "guesses" where
2803 the logs files are, rather than using the compiled in default. In our
2804 case the guess is not the same as the compiled default, so the script
2805 suddenly stopped working when I started to use syslog. The patch checks
2806 to see if log_file_path is "". If so, it attempts to read it from exim
2807 with no configuration file to get the compiled in version, before it
2808 falls back to the previous guessing code."
2810 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2811 implementing BATV in an Exim configuration. See NewStuff for the gory
2812 details.
2814 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2815 Makefile that are specific to HP-UX.
2817 PH/18 If the "use_postmaster" option was set for a recipient callout together
2818 with the "random" option, the postmaster address was used as the MAIL
2819 FROM address for the random test, but not for the subsequent recipient
2820 test. It is now used for both.
2822 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2823 patch removes a few documentation additions to RFC 3028, because the
2824 latest draft now contains them. It adds the new en;ascii-case comparator
2825 and a new error check for 8bit text in MIME parts. Comparator and
2826 require names are now matched exactly. I enabled the subaddress
2827 extension, but it is not well tested yet (read: it works for me)."
2829 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2830 rework some of the code of TK/09 above to avoid the hardwired use of
2831 "%lld" and "long long". Replaced the call to snprintf() with a call to
2832 string_vformat().
2834 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2835 records point to non-existent hosts", "retry timeout exceeded", and
2836 "retry time not reached for any host after a long failure period".
2838 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2839 experimental DomainKeys support:
2841 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2842 (2) On an error such as an illegally used "control", the wrong name for
2843 the control was given.
2845 These problems did NOT occur unless DomainKeys support was compiled.
2847 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2849 PH/24 Added ${if match_ip condition.
2851 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2852 databases so that it will be absolutely obvious if a crash occurs in the
2853 DB library. This is a regular occurrence (often caused by mis-matched
2854 db.h files).
2856 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2857 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2858 picked up on a user's system that detects such things. There doesn't seem
2859 to be a gcc warning option for this - only an attribute that has to be
2860 put on the function's prototype. It seems that in Fedora Core 4 they have
2861 set this on a number of new functions. No doubt there will be more in due
2862 course.
2864 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2865 specify a transport. However, if an address that was verified by such a
2866 router was the subject of a callout, Exim crashed because it tried to
2867 read the rcpt_include_affixes from the non-existent transport. Now it
2868 just assumes that the setting of that option is false. This bug was
2869 introduced by 4.51/PH/31.
2871 PH/28 Changed -d+all to exclude +memory, because that information is very
2872 rarely of interest, but it makes the output a lot bigger. People tend to
2873 do -d+all out of habit.
2875 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2876 code in os-type was giving problems when lives in lib64, like on
2877 x86_64 Fedora Core.
2879 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2880 aren't the modern standard, and it seems that some systems' include files
2881 don't always have them. Exim was already checking for some of the newer
2882 ones like T_AAAA, and defining it itself. I've added checks for all the
2883 record types that Exim uses.
2885 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2886 not automatically generating a new one, as it is supposed to. This
2887 prevented TLS from working. If the file did exist, but contained invalid
2888 data, a new version was generated, as expected. It was only the case of a
2889 non-existent file that was broken.
2891 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2892 with a change in libdomainkeys > 0.64.
2894 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2895 from DNS. If the selector record carries the flag, it now has
2896 precedence over the domain-wide flag.
2898 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2900 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2901 the use of a port name that isn't defined in /etc/services) occurred, the
2902 message was deferred as in a normal delivery, and thus remained on the
2903 spool, instead of being failed because of the mua_wrapper setting. This
2904 is now fixed, and I tidied up some of the mua_wrapper messages at the
2905 same time.
2907 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2908 the messages in a hash of arrays rather than using individual hashes.
2909 This is a bit cleaner and results in dramatic memory savings, albeit
2910 at a slight CPU cost.
2912 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2913 as requested by Marc Sherman.
2915 SC/10 Eximstats: added histograms for user specified patterns as requested
2916 by Marc Sherman.
2918 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2920 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2921 fopen() in the content-scanning modules that did not already have it.
2924 Exim version 4.51
2925 -----------------
2927 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2928 doc/experimental-spec.txt for details. (
2930 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2932 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2934 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2935 to test Sieve filters that use "vacation".
2937 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2938 that changes the way the GnuTLS parameters are stored in the cache file.
2939 The new format can be generated externally. For backward compatibility,
2940 if the data in the cache doesn't make sense, Exim assumes it has read an
2941 old-format file, and it generates new data and writes a new file. This
2942 means that you can't go back to an older release without removing the
2943 file.
2945 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2946 work if there are any delivery delays because "one_time" forces the
2947 parent to be marked "delivered", so its unseen clone is never tried
2948 again. For this reason, Exim now forbids the simultaneous setting of
2949 these two options.
2951 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2952 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2953 entry if you want to know the details. The fix, however, neglected to
2954 consider the case when local delivery batching is involved. The test for
2955 "previously delivered" was not happening when checking to see if an
2956 address could be batched with a previous (undelivered) one; under
2957 certain circumstances this could lead to multiple deliveries to the same
2958 address.
2960 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2961 in its include files, and this causes problems building Exim.
2963 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2964 header_syntax) but Exim was just ignoring anything given after a slash.
2965 In particular, this caused confusion with an attempt to use "verify =
2966 reverse_host_lookup/defer_ok". An error is now given when options are
2967 supplied for verify items that do not have them. (Maybe reverse_host_
2968 lookup should have a defer_ok option, but that's a different point.)
2970 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2971 defined by RFC 821) to 2048, because there were problems with some AUTH
2972 commands, and RFC 1869 says the size should be increased for extended
2973 SMTP commands that take arguments.
2975 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2976 Finch).
2978 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2979 "unknown" error; now it says that the functionality isn't in the binary.
2981 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2982 an address' error message when a string expansion fails (syntax or
2983 whatever). Otherwise the password may appear in the log. Following change
2984 PH/42 below, there is no longer a chance of it appearing in a bounce
2985 message.
2987 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2989 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2990 of its addresses was ever tried. (Bugzilla bug #2.)
2992 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2993 the result incorrectly in the debug output. (It correctly added a newline
2994 to what was transported.)
2996 TF/01 Added $received_time.
2998 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2999 commented out examples of how to interface to a virus scanner and to
3000 SpamAssassin. Also added commented examples of av_scanner and
3001 spamd_address settings.
3003 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3004 and controls are allowed in which ACLs. There were a couple of minor
3005 errors. Some of the entries in the conditions table (which is a table of
3006 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3007 negation of where the condition IS allowed.
3009 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3011 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3012 header file does not have a version number, so I've had to invent a new
3013 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3014 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3015 radiusclient), but it was contributed by a Radius user.
3017 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3018 files or directories, for OpenSSL.
3020 PH/19 When an Exim process that is running as root has to create an Exim log
3021 file, it does so in a subprocess that runs as exim:exim so as to get the
3022 ownership right at creation (otherwise, other Exim processes might see
3023 the file with the wrong ownership). There was no test for failure of this
3024 fork() call, which would lead to the process getting stuck as it waited
3025 for a non-existent subprocess. Forks do occasionally fail when resources
3026 run out. I reviewed all the other calls to fork(); they all seem to check
3027 for failure.
3029 PH/20 When checking for unexpected SMTP input at connect time (before writing
3030 the banner), Exim was not dealing correctly with a non-positive return
3031 from the read() function. If the client had disconnected by this time,
3032 the result was a log entry for a synchronization error with an empty
3033 string after "input=" when read() returned zero. If read() returned -1
3034 (an event I could not check), uninitialized data bytes were printed.
3035 There were reports of junk text (parts of files, etc) appearing after
3036 "input=".
3038 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3040 PH/22 Added support for macro redefinition, and (re)definition in between
3041 driver and ACL definitions.
3043 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3044 forgetting to use the resulting value; it was using the unexpanded value.
3046 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3047 hadn't been configured. The fix is from Juergen Kreileder, who
3048 understands it better than I do:
3050 "Here's what I see happening with three configured cyrus_sasl
3051 authenticators configured (plain, login, cram-md5):
3053 On startup auth_cyrus_sasl_init() gets called for each of these.
3054 This means three calls to sasl_listmech() without a specified mech_list.
3055 => SASL tests which mechs of all available mechs actually work
3056 => three warnings about OTP not working
3057 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3059 With the patch, sasl_listmech() also gets called three times. But now
3060 SASL's mech_list option is set to the server_mech specified in the the
3061 authenticator. Or in other words, the answer from sasl_listmech()
3062 gets limited to just the mech you're testing for (which is different
3063 for each call.)
3064 => the return list contains just 'plain' or 'login', 'cram-md5' or
3065 nothing depending on the value of ob->server_mech.
3067 I've just tested the patch: Authentication still works fine,
3068 unavailable mechs specified in the exim configuration are still
3069 caught, and the auth.log warnings about OTP are gone."
3071 PH/25 When debugging is enabled, the contents of the command line are added
3072 to the debugging output, even when log_selector=+arguments is not
3073 specified.
3075 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3076 answer is "GNU", and only if the return is "GNU/something" is the answer
3077 "Linux".
3079 PH/27 $acl_verify_message is now set immediately after the failure of a
3080 verification in an ACL, and so is available in subsequent modifiers. In
3081 particular, the message can be preserved by coding like this:
3083 warn !verify = sender
3084 set acl_m0 = $acl_verify_message
3086 Previously, $acl_verify_message was set only while expanding "message"
3087 and "log_message" when a very denied access.
3089 PH/28 Modified OS/os.c-Linux with
3091 -#ifndef OS_LOAD_AVERAGE
3092 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3094 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3095 nomenclature these days.)
3097 PH/29 Installed patch from the Sieve maintainer that adds the options
3098 sieve_useraddress and sieve_subaddress to the redirect router.
3100 PH/30 In these circumstances:
3101 . Two addresses routed to the same list of hosts;
3102 . First host does not offer TLS;
3103 . First host accepts first address;
3104 . First host gives temporary error to second address;
3105 . Second host offers TLS and a TLS session is established;
3106 . Second host accepts second address.
3107 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3108 and peerdn, if requested) that were in fact used only for the second
3109 address.
3111 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3112 attention to any local part prefix or suffix that was matched by the
3113 router that accepted the address. It now behaves in the same way as it
3114 does for delivery: the affixes are removed from the local part unless
3115 rcpt_include_affixes is set on the transport.
3117 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3118 timeout during the DATA phase of an incoming message.
3120 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3121 applied a patch sanctioned by the Sieve maintainer.
3123 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3124 the uid or gid is negative. A case of a negative gid caused this to be
3125 noticed. The fix allows for either to be negative.
3127 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3128 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3129 overlooked.
3131 PH/36 The change PH/12 above was broken. Fixed it.
3133 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3134 the grounds that routing the same address twice would always produce the
3135 same answer. This might have been true once, but it is certainly no
3136 longer true now. Routing a child address may depend on the previous
3137 routing that produced that child. Some complicated redirection strategies
3138 went wrong when messages had multiple recipients, and made Exim's
3139 behaviour dependent on the order in which the addresses were given.
3141 I have moved the duplicate checking until after the routing is complete.
3142 Exim scans the addresses that are assigned to local and remote
3143 transports, and removes any duplicates. This means that more work will be
3144 done, as duplicates will always all be routed, but duplicates are
3145 presumably rare, so I don't expect this is of any significance.
3147 For deliveries to pipes, files, and autoreplies, the duplicate checking
3148 still happens during the routing process, since they are not going to be
3149 routed further.
3151 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3152 It corrects a timeout issue with spamd. This is Ian's comment: "The
3153 background is that sometimes spamd either never reads data from a
3154 connection it has accepted, or it never writes response data. The exiscan
3155 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3156 blindly assumes that writes won't block so it may never time out."
3158 PH/39 Allow G after quota size as well as K and M.
3160 PH/40 The value set for $authenticated_id in an authenticator may not contain
3161 binary zeroes or newlines because the value is written to log lines and
3162 to spool files. There was no check on this. Now the value is run through
3163 the string_printing() function so that such characters are converted to
3164 printable escape sequences.
3166 PH/41 $message_linecount is a new variable that contains the total number of
3167 lines in the message. Compare $body_linecount, which is the count for the
3168 body only.
3170 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3171 bounce and delay warning messages, except in certain special cases, which
3172 are as follows:
3174 (a) An SMTP error message from a remote host;
3175 (b) A message specified in a :fail: redirection;
3176 (c) A message specified in a "fail" command in a system filter;
3177 (d) A message specified in a FAIL return from the queryprogram router;
3178 (e) A message specified by the cannot_route_message router option.
3180 In these cases only, Exim does include the error details in bounce and
3181 warning messages. There are also a few cases where bland messages such
3182 as "unrouteable address" or "local delivery error" are given.
3184 PH/43 $value is now also set for the "else" part of a ${run expansion.
3186 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3187 being worked on, but at least Exim now implements the latest version to
3188 play with."
3190 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3191 process to complete was treated as a delivery failure, a timeout while
3192 writing the message to the pipe was logged, but erroneously treated as a
3193 successful delivery. Such timeouts include transport filter timeouts. For
3194 consistency with the overall process timeout, these timeouts are now
3195 treated as errors, giving rise to delivery failures by default. However,
3196 there is now a new Boolean option for the pipe transport called
3197 timeout_defer, which, if set TRUE, converts the failures into defers for
3198 both kinds of timeout. A transport filter timeout is now identified in
3199 the log output.
3201 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3202 systems where "make" and "gmake" are different, calling "gmake" at top
3203 level broke things. I've arranged for the value of $(MAKE) to be passed
3204 from the Makefile to this script so that it can call the same version of
3205 "make".
3208 A note about Exim versions 4.44 and 4.50
3209 ----------------------------------------
3211 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3212 changes of various kinds. As a consequence, a big documentation update was
3213 needed. This delayed the release for rather longer than seemed good, especially
3214 in the light of a couple of (minor) security issues. Therefore, the changes
3215 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3216 release. So 4.44 and 4.50 are in effect two different branches that both start
3217 from 4.43.
3219 I have left the 4.50 change log unchanged; it contains all the changes since
3220 4.43. The change log for 4.44 is below; many of its items are identical to
3221 those for 4.50. This seems to be the most sensible way to preserve the
3222 historical information.
3225 Exim version 4.50
3226 -----------------
3228 1. Minor wording change to the doc/README.SIEVE file.
3230 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3231 computation of the current number of files was incorrect.
3233 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3234 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3235 place.
3237 4. Give more explanation in the error message when the command for a transport
3238 filter fails to execute.
3240 5. There are several places where Exim runs a non-Exim command in a
3241 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3242 was being done only for the command run by the queryprogram router. It is
3243 now done for all such subprocesses. The other cases are: ${run, transport
3244 filters, and the commands run by the lmtp and pipe transports.
3246 6. Added CONFIGURE_GROUP build-time option.
3248 7. Some older OS have a limit of 256 on the maximum number of file
3249 descriptors. Exim was using setrlimit() to set 1000 as a large value
3250 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3251 systems. I've change it so that if it can't get 1000, it tries for 256.
3253 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3254 was an oversight, and furthermore, ever since the addition of extra
3255 controls (e.g. 4.43/32), the checks on when to allow different forms of
3256 "control" were broken. There should now be diagnostics for all cases when a
3257 control that does not make sense is encountered.
3259 9. Added the /retain_sender option to "control=submission".
3261 10. $recipients is now available in the predata ACL (oversight).
3263 11. Tidy the search cache before the fork to do a delivery from a message
3264 received from the command line. Otherwise the child will trigger a lookup
3265 failure and thereby defer the delivery if it tries to use (for example) a
3266 cached ldap connection that the parent has called unbind on.
3268 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3269 of $address_data from the recipient verification was clobbered by the
3270 sender verification.
3272 13. The value of address_data from a sender verification is now available in
3273 $sender_address_data in subsequent conditions in the ACL statement.
3275 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3277 15. Added a new option "connect=<time>" to callout options, to set a different
3278 connection timeout.
3280 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3281 was its contents. (It was OK if the option was not defined at all.)
3283 17. A "Completed" log line is now written for messages that are removed from
3284 the spool by the -Mrm option.
3286 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3287 information about exactly what failed.
3289 19. Added -dd to debug only the daemon process.
3291 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3292 handles timeouts, both on the server side and network timeouts. Renamed the
3293 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3295 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3296 It is now set to "smtps".
3298 22. $host_address is now set to the target address during the checking of
3299 ignore_target_hosts.
3301 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3302 being passed; this would have caused $sender_host_name to have been used if
3303 matching the list had actually called for a host name (not very likely,
3304 since this list is usually IP addresses). A host name is now passed as
3305 "[x.x.x.x]".
3307 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3308 code that specifies a non-restarting handler (typically sigaction() in
3309 modern systems) in an attempt to fix a rare and obscure crash bug.
3311 25. Narrowed the window for a race in the daemon that could cause it to ignore
3312 SIGCHLD signals. This is not a major problem, because they are used only to
3313 wake it up if nothing else does.
3315 26. A malformed maildirsize file could cause Exim to calculate negative values
3316 for the mailbox size or file count. Odd effects could occur as a result.
3317 The maildirsize information is now recalculated if the size or filecount
3318 end up negative.
3320 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3321 support for a long time. Removed HAVE_SYS_VFS_H.
3323 28. Installed the latest version of exipick from John Jetmore.
3325 29. In an address list, if the pattern was not a regular expression, an empty
3326 subject address (from a bounce message) matched only if the pattern was an
3327 empty string. Non-empty patterns were not even tested. This was the wrong
3328 because it is perfectly reasonable to use an empty address as part of a
3329 database query. An empty address is now tested by patterns that are
3330 lookups. However, all the other forms of pattern expect the subject to
3331 contain a local part and a domain, and therefore, for them, an empty
3332 address still always fails if the pattern is not itself empty.
3334 30. Exim went into a mad DNS loop when attempting to do a callout where the
3335 host was specified on an smtp transport, and looking it up yielded more
3336 than one IP address.
3338 31. Re-factored the code for checking spool and log partition space into a
3339 function that finds that data and another that does the check. The former
3340 is then used to implement four new variables: $spool_space, $log_space,
3341 $spool_inodes, and $log_inodes.
3343 32. The RFC2047 encoding function was originally intended for short strings
3344 such as real names; it was not keeping to the 75-character limit for
3345 encoded words that the RFC imposes. It now respects the limit, and
3346 generates multiple encoded words if necessary. To be on the safe side, I
3347 have increased the buffer size for the ${rfc2047: expansion operator from
3348 1024 to 2048 bytes.
3350 33. It is now permitted to omit both strings after an "if" condition; if the
3351 condition is true, the result is "true". As before, when the second string
3352 is omitted, a false condition yields an empty string. This makes it less
3353 cumbersome to write custom ACL and router conditions.
3355 34. Failure to deliver a bounce message always caused it to be frozen, even if
3356 there was an errors_to setting on the router. The errors_to setting is now
3357 respected.
3359 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3360 canonical form (fully expanded) before being placed in
3361 $sender_host_address.
3363 36. The table in the code that translates DNS record types into text (T_A to
3364 "A" for instance) was missing entries for NS and CNAME. It is just possible
3365 that this could have caused confusion if both these types were looked up
3366 for the same domain, because the text type is used as part of Exim's
3367 per-process caching. But the chance of anyone hitting this buglet seems
3368 very small.
3370 37. The dnsdb lookup has been extended in a number of ways.
3372 (1) There is a new type, "zns", which walks up the domain tree until it
3373 finds some nameserver records. It should be used with care.
3375 (2) There is a new type, "mxh", which is like "mx" except that it returns
3376 just the host names, not the priorities.
3378 (3) It is now possible to give a list of domains (or IP addresses) to be
3379 looked up. The behaviour when one of the lookups defers can be
3380 controlled by a keyword.
3382 (4) It is now possible to specify the separator character for use when
3383 multiple records are returned.
3385 38. The dnslists ACL condition has been extended: it is now possible to supply
3386 a list of IP addresses and/or domains to be looked up in a particular DNS
3387 domain.
3389 39. Added log_selector=+queue_time_overall.
3391 40. When running the queue in the test harness, wait just a tad after forking a
3392 delivery process, to get repeatability of debugging output.
3394 41. Include certificate and key file names in error message when GnuTLS fails
3395 to set them up, because the GnuTLS error message doesn't include the name
3396 of the failing file when there is a problem reading it.
3398 42. Allow both -bf and -bF in the same test run.
3400 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3402 44. The "Exiscan patch" is now merged into the mainline Exim source.
3404 45. Sometimes the final signoff response after QUIT could fail to get
3405 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3406 < 0 before doing a fflush(). This bug looks as though it goes back to the
3407 introduction of TLS in release 3.20, but "sometimes" must have been rare
3408 because the tests only now provoked it.
3410 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3411 (this can affect the format of dates).
3413 47. exim_tidydb, when checking for the continued existence of a message for
3414 which it has found a message-specific retry record, was not finding
3415 messages that were in split spool directories. Consequently, it was
3416 deleting retry records that should have stayed in existence.
3418 48. Steve fixed some bugs in eximstats.
3420 49. The SPA authentication driver was not abandoning authentication and moving
3421 on to the next authenticator when an expansion was forced to fail,
3422 contradicting the general specification for all authenticators. Instead it
3423 was generating a temporary error. It now behaves as specified.
3425 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3426 (the order specifies the preference for clients). The order is now AES256,
3427 AES128, 3DES, ARCFOUR128.
3429 51. Small patch to Sieve code - explicitly set From: when generating an
3430 autoreply.
3432 52. Exim crashed if a remote delivery caused a very long error message to be
3433 recorded - for instance if somebody sent an entire SpamAssassin report back
3434 as a large number of 550 error lines. This bug was coincidentally fixed by
3435 increasing the size of one of Exim's internal buffers (big_buffer) that
3436 happened as part of the Exiscan merge. However, to be on the safe side, I
3437 have made the code more robust (and fixed the comments that describe what
3438 is going on).
3440 53. Now that there can be additional text after "Completed" in log lines (if
3441 the queue_time_overall log selector is set), a one-byte patch to exigrep
3442 was needed to allow it to recognize "Completed" as not the last thing in
3443 the line.
3445 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3446 patch that reportedly fixes this has been added. I am not expert enough to
3447 create a test for it. This is what the patch creator wrote:
3449 "I found a little strange behaviour of ldap code when working with
3450 Windows 2003 AD Domain, where users was placed in more than one
3451 Organization Units. When I tried to give exim partial DN, the exit code
3452 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
3453 But simultaneously result of request was absolutely normal ldap result,
3454 so I produce this patch..."
3456 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
3457 REFERENCE, so I have modified the code to exclude the patch when that macro
3458 is not defined.
3460 55. Some experimental protocols are using DNS PTR records for new purposes. The
3461 keys for these records are domain names, not reversed IP addresses. The
3462 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3463 leaves it alone. Component reversal etc. now happens only for IP addresses.
3464 CAN-2005-0021
3466 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
3468 57. Double the size of the debug message buffer (to 2048) so that more of very
3469 long debug lines gets shown.
3471 58. The exicyclog utility now does better if the number of log files to keep
3472 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
3474 59. Two changes related to the smtp_active_hostname option:
3476 (1) $smtp_active_hostname is now available as a variable.
3477 (2) The default for smtp_banner uses $smtp_active_hostname instead
3478 of $primary_hostname.
3480 60. The host_aton() function is supposed to be passed a string that is known
3481 to be a valid IP address. However, in the case of IPv6 addresses, it was
3482 not checking this. This is a hostage to fortune. Exim now panics and dies
3483 if the condition is not met. A case was found where this could be provoked
3484 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3485 components; fortuitously, this particular loophole had already been fixed
3486 by change 4.50/55 above.
3488 If there are any other similar loopholes, the new check in host_aton()
3489 itself should stop them being exploited. The report I received stated that
3490 data on the command line could provoke the exploit when Exim was running as
3491 exim, but did not say which command line option was involved. All I could
3492 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3493 running as the user.
3494 CAN-2005-0021
3496 61. There was a buffer overflow vulnerability in the SPA authentication code
3497 (which came originally from the Samba project). I have added a test to the
3498 spa_base64_to_bits() function which I hope fixes it.
3499 CAN-2005-0022
3501 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
3502 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
3504 63. The daemon start-up calls getloadavg() while still root for those OS that
3505 need the first call to be done as root, but it missed one case: when
3506 deliver_queue_load_max is set with deliver_drop_privilege. This is
3507 necessary for the benefit of the queue runner, because there is no re-exec
3508 when deliver_drop_privilege is set.
3510 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
3511 This has been fixed.
3513 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3514 was in use, was not putting the data itself into the right store pool;
3515 consequently, it could be overwritten for a subsequent message in the same
3516 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3517 the caching.)
3519 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
3521 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
3522 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
3523 the function were treating the return as a boolean value, which happened to
3524 work because 0=false and not-0=true, but is not correct code.
3526 68. The host_aton() function was not handling scoped IPv6 addresses (those
3527 with, for example, "%eth0" on the end) correctly.
3529 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
3530 negated items (that is, ~something) in unsigned ints. Some compilers
3531 apparently mutter when there is no cast.
3533 70. If an address verification called from an ACL failed, and did not produce a
3534 user-specific message (i.e. there was only a "system" message), nothing was
3535 put in $acl_verify_message. In this situation, it now puts the system
3536 message there.
3538 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3539 session; change 4.31/43 added the unwanted input to the log line - except
3540 that it did not do this in the start of session case. It now does.
3542 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3543 This is wrong and can cause the other end to generate a synchronization
3544 error if it is another Exim or anything else that does the synchronization
3545 check. A QUIT command is no longer sent after a timeout.
3547 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3548 during host lookups.
3550 74. The defer_ok option of callout verification was not working if it was used
3551 when verifying addresses in header lines, that is, for this case:
3553 verify = header_sender/callout=defer_ok
3555 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3556 those file descriptors could be used for SMTP connections. If anything
3557 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3558 could be sent to the SMTP client, causing chaos. The daemon now opens
3559 stdin, stdout, and stderr to /dev/null when it puts itself into the
3560 background.
3562 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3563 log by default. The user can override this with suitable Perl magic.
3565 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3566 the log message when discard triggers, was not working for the DATA ACL or
3567 for the non-SMTP ACL.
3569 78. Error message wording change in sieve.c.
3571 79. If smtp_accept_max_per_host was set, the number of connections could be
3572 restricted to fewer than expected, because the daemon was trying to set up
3573 a new connection before checking whether the processes handling previous
3574 connections had finished. The check for completed processes is now done
3575 earlier. On busy systems, this bug wouldn't be noticed because something
3576 else would have woken the daemon, and it would have reaped the completed
3577 process earlier.
3579 80. If a message was submitted locally by a user whose login name contained one
3580 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3581 It caused a spool format error. I have fixed the spool reading code. A
3582 related problem was that the "from" clause in the Received: line became
3583 illegal because of the space(s). It is now covered by ${quote_local_part.
3585 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3586 tables).
3588 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3590 83. Patch from Sieve maintainer to fix unterminated string problem in
3591 "vacation" handling.
3593 84. Some minor changes to the Linux configuration files to help with other
3594 OS variants using glibc.
3596 85. One more patch for Sieve to update vacation handling to latest spec.
3599 ----------------------------------------------------
3600 See the note above about the 4.44 and 4.50 releases.
3601 ----------------------------------------------------
3604 Exim version 4.44
3605 -----------------
3607 1. Change 4.43/35 introduced a bug that caused file counts to be
3608 incorrectly computed when quota_filecount was set in an appendfile
3609 transport
3611 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3612 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3613 place.
3615 3. Give more explanation in the error message when the command for a transport
3616 filter fails to execute.
3618 4. There are several places where Exim runs a non-Exim command in a
3619 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3620 was being done only for the command run by the queryprogram router. It is
3621 now done for all such subprocesses. The other cases are: ${run, transport
3622 filters, and the commands run by the lmtp and pipe transports.
3624 5. Some older OS have a limit of 256 on the maximum number of file
3625 descriptors. Exim was using setrlimit() to set 1000 as a large value
3626 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3627 systems. I've change it so that if it can't get 1000, it tries for 256.
3629 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3630 was an oversight, and furthermore, ever since the addition of extra
3631 controls (e.g. 4.43/32), the checks on when to allow different forms of
3632 "control" were broken. There should now be diagnostics for all cases when a
3633 control that does not make sense is encountered.
3635 7. $recipients is now available in the predata ACL (oversight).
3637 8. Tidy the search cache before the fork to do a delivery from a message
3638 received from the command line. Otherwise the child will trigger a lookup
3639 failure and thereby defer the delivery if it tries to use (for example) a
3640 cached ldap connection that the parent has called unbind on.
3642 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3643 of $address_data from the recipient verification was clobbered by the
3644 sender verification.
3646 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3647 was its contents. (It was OK if the option was not defined at all.)
3649 11. A "Completed" log line is now written for messages that are removed from
3650 the spool by the -Mrm option.
3652 12. $host_address is now set to the target address during the checking of
3653 ignore_target_hosts.
3655 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3656 being passed; this would have caused $sender_host_name to have been used if
3657 matching the list had actually called for a host name (not very likely,
3658 since this list is usually IP addresses). A host name is now passed as
3659 "[x.x.x.x]".
3661 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3662 code that specifies a non-restarting handler (typically sigaction() in
3663 modern systems) in an attempt to fix a rare and obscure crash bug.
3665 15. Narrowed the window for a race in the daemon that could cause it to ignore
3666 SIGCHLD signals. This is not a major problem, because they are used only to
3667 wake it up if nothing else does.
3669 16. A malformed maildirsize file could cause Exim to calculate negative values
3670 for the mailbox size or file count. Odd effects could occur as a result.
3671 The maildirsize information is now recalculated if the size or filecount
3672 end up negative.
3674 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3675 support for a long time. Removed HAVE_SYS_VFS_H.
3677 18. Updated exipick to current release from John Jetmore.
3679 19. Allow an empty sender to be matched against a lookup in an address list.
3680 Previously the only cases considered were a regular expression, or an
3681 empty pattern.
3683 20. Exim went into a mad DNS lookup loop when doing a callout where the
3684 host was specified on the transport, if the DNS lookup yielded more than
3685 one IP address.
3687 21. The RFC2047 encoding function was originally intended for short strings
3688 such as real names; it was not keeping to the 75-character limit for
3689 encoded words that the RFC imposes. It now respects the limit, and
3690 generates multiple encoded words if necessary. To be on the safe side, I
3691 have increased the buffer size for the ${rfc2047: expansion operator from
3692 1024 to 2048 bytes.
3694 22. Failure to deliver a bounce message always caused it to be frozen, even if
3695 there was an errors_to setting on the router. The errors_to setting is now
3696 respected.
3698 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3699 canonical form (fully expanded) before being placed in
3700 $sender_host_address.
3702 24. Updated eximstats to version 1.33
3704 25. Include certificate and key file names in error message when GnuTLS fails
3705 to set them up, because the GnuTLS error message doesn't include the name
3706 of the failing file when there is a problem reading it.
3708 26. Expand error message when OpenSSL has problems setting up cert/key files.
3709 As per change 25.
3711 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3712 (this can affect the format of dates).
3714 28. exim_tidydb, when checking for the continued existence of a message for
3715 which it has found a message-specific retry record, was not finding
3716 messages that were in split spool directories. Consequently, it was
3717 deleting retry records that should have stayed in existence.
3719 29. eximstats updated to version 1.35
3720 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3721 1.35 - bugfix such that pie charts by volume are generated correctly
3723 30. The SPA authentication driver was not abandoning authentication and moving
3724 on to the next authenticator when an expansion was forced to fail,
3725 contradicting the general specification for all authenticators. Instead it
3726 was generating a temporary error. It now behaves as specified.
3728 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3729 (the order specifies the preference for clients). The order is now AES256,
3730 AES128, 3DES, ARCFOUR128.
3732 31. Small patch to Sieve code - explicitly set From: when generating an
3733 autoreply.
3735 32. Exim crashed if a remote delivery caused a very long error message to be
3736 recorded - for instance if somebody sent an entire SpamAssassin report back
3737 as a large number of 550 error lines. This bug was coincidentally fixed by
3738 increasing the size of one of Exim's internal buffers (big_buffer) that
3739 happened as part of the Exiscan merge. However, to be on the safe side, I
3740 have made the code more robust (and fixed the comments that describe what
3741 is going on).
3743 33. Some experimental protocols are using DNS PTR records for new purposes. The
3744 keys for these records are domain names, not reversed IP addresses. The
3745 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3746 leaves it alone. Component reversal etc. now happens only for IP addresses.
3747 CAN-2005-0021
3749 34. The host_aton() function is supposed to be passed a string that is known
3750 to be a valid IP address. However, in the case of IPv6 addresses, it was
3751 not checking this. This is a hostage to fortune. Exim now panics and dies
3752 if the condition is not met. A case was found where this could be provoked
3753 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3754 components; fortuitously, this particular loophole had already been fixed
3755 by change 4.50/55 or 4.44/33 above.
3757 If there are any other similar loopholes, the new check in host_aton()
3758 itself should stop them being exploited. The report I received stated that
3759 data on the command line could provoke the exploit when Exim was running as
3760 exim, but did not say which command line option was involved. All I could
3761 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3762 running as the user.
3763 CAN-2005-0021
3765 35. There was a buffer overflow vulnerability in the SPA authentication code
3766 (which came originally from the Samba project). I have added a test to the
3767 spa_base64_to_bits() function which I hope fixes it.
3768 CAN-2005-0022
3770 36. The daemon start-up calls getloadavg() while still root for those OS that
3771 need the first call to be done as root, but it missed one case: when
3772 deliver_queue_load_max is set with deliver_drop_privilege. This is
3773 necessary for the benefit of the queue runner, because there is no re-exec
3774 when deliver_drop_privilege is set.
3776 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3777 was in use, was not putting the data itself into the right store pool;
3778 consequently, it could be overwritten for a subsequent message in the same
3779 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3780 the caching.)
3782 38. Sometimes the final signoff response after QUIT could fail to get
3783 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3784 < 0 before doing a fflush(). This bug looks as though it goes back to the
3785 introduction of TLS in release 3.20, but "sometimes" must have been rare
3786 because the tests only now provoked it.
3789 Exim version 4.43
3790 -----------------
3792 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3793 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3794 Now it returns an integer. A number of calls were still expecting a T/F
3795 return. Fortuitously, in all cases, the tests worked in OK situations,
3796 which is the norm. However, things would have gone wrong on any write
3797 failures on the smtp file descriptor. This function is used when sending
3798 messages over SMTP and also when doing verify callouts.
3800 2. When Exim is called to do synchronous delivery of a locally submitted
3801 message (the -odf or -odi options), it no longer closes stderr before doing
3802 the delivery.
3804 3. Implemented the mua_wrapper option.
3806 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3808 5. Implemented the functions header_remove(), header_testname(),
3809 header_add_at_position(), and receive_remove_recipient(), and exported them
3810 to local_scan().
3812 6. If an ACL "warn" statement specified the addition of headers, Exim already
3813 inserted X-ACL-Warn: at the start if there was no header name. However, it
3814 was not making this test for the second and subsequent header lines if
3815 there were newlines in the string. This meant that an invalid header could
3816 be inserted if Exim was badly configured.
3818 7. Allow an ACL "warn" statement to add header lines at the start or after all
3819 the Received: headers, as well as at the end.
3821 8. Added the rcpt_4xx retry error code.
3823 9. Added postmaster_mailfrom=xxx to callout verification option.
3825 10. Added mailfrom=xxxx to the callout verification option, for verify=
3826 header_sender only.
3828 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3829 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3831 12. Inserted some casts to stop certain compilers warning when using pointer
3832 differences as field lengths or precisions in printf-type calls (mostly
3833 affecting debugging statements).
3835 13. Added optional readline() support for -be (dynamically loaded).
3837 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3838 same clock tick as a message's arrival, so that its received time was the
3839 same as the "first fail" time on the retry record, and that message
3840 remained on the queue past the ultimate address timeout, every queue runner
3841 would try a delivery (because it was past the ultimate address timeout) but
3842 after another failure, the ultimate address timeout, which should have then
3843 bounced the address, did not kick in. This was a "< instead of <=" error;
3844 in most cases the first failure would have been in the next clock tick
3845 after the received time, and all would be well.
3847 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3848 being recognized when the domain list was tested by the match_domain
3849 condition in an expansion string.
3851 16. Added the ${str2b64: operator.
3853 17. Exim was always calling setrlimit() to set a large limit for the number of
3854 processes, without checking whether the existing limit was already
3855 adequate. (It did check for the limit on file descriptors.) Furthermore,
3856 errors from getrlimit() and setrlimit() were being ignored. Now they are
3857 logged to the main and panic logs, but Exim does carry on, to try to do its
3858 job under whatever limits there are.
3860 18. Imported PCRE 5.0.
3862 19. Trivial typo in log message " temporarily refused connection" (the leading
3863 space).
3865 20. If the log selector return_path_on_delivery was set and an address was
3866 redirected to /dev/null, the delivery process crashed because it assumed
3867 that a return path would always be set for a "successful" delivery. In this
3868 case, the whole delivery is bypassed as an optimization, and therefore no
3869 return path is set.
3871 21. Internal re-arrangement: the function for sending a challenge and reading
3872 a response while authentication was assuming a zero-terminated challenge
3873 string. It's now changed to take a pointer and a length, to allow for
3874 binary data in such strings.
3876 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3878 23. Exim was not respecting finduser_retries when seeking the login of the
3879 uid under which it was called; it was always trying 10 times. (The default
3880 setting of finduser_retries is zero.) Also, it was sleeping after the final
3881 failure, which is pointless.
3883 24. Implemented tls_on_connect_ports.
3885 25. Implemented acl_smtp_predata.
3887 26. If the domain in control=submission is set empty, Exim assumes that the
3888 authenticated id is a complete email address when it generates From: or
3889 Sender: header lines.
3891 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3892 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3893 chown and chgrp in /bin and hostname in /usr/bin.
3895 28. Exim was keeping the "process log" file open after each use, just as it
3896 does for the main log. This opens the possibility of it remaining open for
3897 long periods when the USR1 signal hits a daemon. Occasional processlog
3898 errors were reported, that could have been caused by this. Anyway, it seems
3899 much more sensible not to leave this file open at all, so that is what now
3900 happens.
3902 29. The long-running daemon process does not normally write to the log once it
3903 has entered its main loop, and it closes the log before doing so. This is
3904 so that log files can straightforwardly be renamed and moved. However,
3905 there are a couple of unusual error situations where the daemon does write
3906 log entries, and I had neglected to close the log afterwards.
3908 30. The text of an SMTP error response that was received during a remote
3909 delivery was being truncated at 512 bytes. This is too short for some of
3910 the long messages that one sometimes sees. I've increased the limit to
3911 1024.
3913 31. It is now possible to make retry rules that apply only when a message has a
3914 specific sender, in particular, an empty sender.
3916 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3917 it possible to be selective about when SMTP synchronization is enforced.
3919 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3921 32. Implemented hosts_connection_nolog.
3923 33. Added an ACL for QUIT.
3925 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3926 syntax error.
3928 35. Added mailbox_size and mailbox_filecount to appendfile.
3930 36. Added control = no_multiline_responses to ACLs.
3932 37. There was a bug in the logic of the code that waits for the clock to tick
3933 in the case where the clock went backwards by a substantial amount such
3934 that the microsecond fraction of "now" was more than the microsecond
3935 fraction of "then" (but the whole seconds number was less).
3937 38. Added support for the libradius Radius client library this is found on
3938 FreeBSD (previously only the radiusclient library was supported).
3941 Exim version 4.42
3942 -----------------
3944 1. When certain lookups returned multiple values in the form name=value, the
3945 quoting of the values was not always being done properly. Specifically:
3946 (a) If the value started with a double quote, but contained no whitespace,
3947 it was not quoted.
3948 (b) If the value contained whitespace other than a space character (i.e.
3949 tabs or newlines or carriage returns) it was not quoted.
3950 This fix has been applied to the mysql and pgsql lookups by writing a
3951 separate quoting function and calling it from the lookup code. The fix
3952 should probably also be applied to nisplus, ibase and oracle lookups, but
3953 since I cannot test any of those, I have not disturbed their existing code.
3955 2. A hit in the callout cache for a specific address caused a log line with no
3956 reason for rejecting RCPT. Now it says "Previous (cached) callout
3957 verification failure".
3959 3. There was an off-by-one bug in the queryprogram router. An over-long
3960 return line was truncated at 256 instead of 255 characters, thereby
3961 overflowing its buffer with the terminating zero. As well as fixing this, I
3962 have increased the buffer size to 1024 (and made a note to document this).
3964 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3965 when Exim is waiting for an SMTP response from a remote server, Exim
3966 restarts its select() call on the socket, thereby resetting its timeout.
3967 This is not a problem when such interrupts are rare. Somebody set up a cron
3968 job to run exiwhat every 2 minutes, which is less than the normal select()
3969 timeout (5 or 10 minutes). This meant that the select() timeout never
3970 kicked in because it was always reset. I have fixed this by comparing the
3971 time when an interrupt arrives with the time at the start of the first call
3972 to select(). If more time than the timeout has elapsed, the interrupt is
3973 treated as a timeout.
3975 5. Some internal re-factoring in preparation for the addition of Sieve
3976 extensions (by MH). In particular, the "personal" test is moved to a
3977 separate function, and given an option for scanning Cc: and Bcc: (which is
3978 not set for Exim filters).
3980 6. When Exim created an email address using the login of the caller as the
3981 local part (e.g. when creating a From: or Sender: header line), it was not
3982 quoting the local part when it contained special characters such as @.
3984 7. Installed new OpenBSD configuration files.
3986 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3987 try to make them clearer.
3989 9. Callout options, other than the timeout value, were being ignored when
3990 verifying sender addresses in header lines. For example, when using
3992 verify = header_sender/callout=no_cache
3994 the cache was (incorrectly) being used.
3996 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3997 only the Cygwin environment.
3999 11. When return_path_on_delivery was set as a log selector, if different remote
4000 addresses in the same message used different return paths and parallel
4001 remote delivery occurred, the wrong values would sometimes be logged.
4002 (Whenever a remote delivery process finished, the return path value from
4003 the most recently started remote delivery process was logged.)
4005 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4006 lines when AUTH and/or TLS are in use. This is the "received protocol"
4007 field. Exim used to use "asmtp" for authenticated SMTP, without any
4008 indication (in the protocol name) for TLS use. Now it follows the RFC and
4009 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4010 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4011 names appear in log lines as well as in Received: header lines.
4013 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4014 extensions, and comparison tests, and to fix some bugs.
4016 14. Changes to the "personal" filter test:
4018 (1) The test was buggy in that it was just doing the equivalent of
4019 "contains" tests on header lines. For example, if a user's address was
4020 anne@some.where, the "personal" test would incorrectly be true for
4022 To: susanne@some.where
4024 This test is now done by extracting each address from the header in turn,
4025 and checking the entire address. Other tests that are part of "personal"
4026 are now done using regular expressions (for example, to check local parts
4027 of addresses in From: header lines).
4029 (2) The list of non-personal local parts in From: addresses has been
4030 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4031 taken from the Sieve specification recommendations.
4033 (3) If the message contains any header line starting with "List-" it is
4034 treated as non-personal.
4036 (4) The test for "circular" in the Subject: header line has been removed
4037 because it now seems ill-conceived.
4039 15. Minor typos in src/EDITME comments corrected.
4041 16. Installed latest exipick from John Jetmore.
4043 17. If headers_add on a router specified a text string that was too long for
4044 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4045 of string_sprintf() is now avoided.
4047 18. $message_body_size was not set (it was always zero) when running the DATA
4048 ACL and the local_scan() function.
4050 19. For the "mail" command in an Exim filter, no default was being set for
4051 the once_repeat time, causing a random time value to be used if "once" was
4052 specified. (If the value happened to be <= 0, no repeat happened.) The
4053 default is now 0s, meaning "never repeat". The "vacation" command was OK
4054 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4055 (I found it when inspecting the code).
4057 20. There is now an overall timeout for performing a callout verification. It
4058 defaults to 4 times the callout timeout, which applies to individual SMTP
4059 commands during the callout. The overall timeout applies when there is more
4060 than one host that can be tried. The timeout is checked before trying the
4061 next host. This prevents very long delays if there are a large number of
4062 hosts and all are timing out (e.g. when the network connections are timing
4063 out). The value of the overall timeout can be changed by specifying an
4064 additional sub-option for "callout", called "maxwait". For example:
4066 verify = sender/callout=5s,maxwait=20s
4068 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4069 to the end before writing, but this should make it even safer).
4071 22. Exim was forgetting that it had advertised PIPELINING for the second and
4072 subsequent messages on an SMTP connection. It was also not resetting its
4073 memory on STARTTLS and an internal HELO.
4075 23. When Exim logs an SMTP synchronization error within a session, it now
4076 records whether PIPELINING has been advertised or not.
4078 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4079 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4080 rather than long int.
4082 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4084 26. Added the never_mail option to autoreply.
4087 Exim version 4.41
4088 -----------------
4090 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4091 crash if the getsockname() call failed; this can happen if a connection is
4092 closed very soon after it is established. The problem was simply in the
4093 order in which certain operations were done, causing Exim to try to write
4094 to the SMTP stream before it had set up the file descriptor. The bug has
4095 been fixed by making things happen in the correct order.
4098 Exim version 4.40
4099 -----------------
4101 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4102 before the connection was closed, thus losing the rejection response.
4104 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4105 some early Solaris releases, but causes trouble in current releases where
4106 socklen_t is defined.
4108 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4109 always exist.
4111 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4112 configured.
4114 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4115 that releasing the top of it at the end releases what was used for sub-
4116 expansions (unless the block got too big). However, discard this block if
4117 the first thing is a variable or header, so that we can use its block when
4118 it is dynamic (useful for very large $message_headers, for example).
4120 6. Lookups now cache *every* query, not just the most recent. A new, separate
4121 store pool is used for this. It can be recovered when all lookup caches are
4122 flushed. Lookups now release memory at the end of their result strings.
4123 This has involved some general refactoring of the lookup sources.
4125 7. Some code has been added to the store_xxx() functions to reduce the amount
4126 of flapping under certain conditions.
4128 8. log_incoming_interface used to affect only the <= reception log lines. Now
4129 it causes the local interface and port to be added to several more SMTP log
4130 lines, for example "SMTP connection from", and rejection lines.
4132 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4134 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4136 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4137 could be overwritten at the end of the current message (or the start of a
4138 new message if it was set in a HELO ACL). The value is now preserved for
4139 the duration of the SMTP connection.
4141 12. If a transport had a headers_rewrite setting, and a matching header line
4142 contained an unqualified address, that address was qualified, even if it
4143 did not match any rewriting rules. The underlying bug was that the values
4144 of the flags that permit the existence of unqualified sender and recipient
4145 addresses in header lines (set by {sender,recipient}_unqualified_hosts for