1 Change log file for Exim from version 4.21
2 -------------------------------------------
8 TF/01 Correctly close the server side of TLS when forking for delivery.
10 When a message was received over SMTP with TLS, Exim failed to clear up
11 the incoming connection properly after forking off the child process to
12 deliver the message. In some situations the subsequent outgoing
13 delivery connection happened to have the same fd number as the incoming
14 connection previously had. Exim would try to use TLS and fail, logging
15 a "Bad file descriptor" error.
17 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
18 utilities have not been installed.
20 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
21 temporary space as the ACL may create new global variables.
23 TL/01 LDAP support uses per connection or global context settings, depending
24 upon the detected version of the libraries at build time.
26 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
27 to extract and use the src ip:port in logging and expansions as if it
28 were a direct connection from the outside internet.
30 JH/02 Add ${listextract {number}{list}{success}{fail}}.
32 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
33 Properly escape header and check for NULL return.
35 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
38 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
40 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
41 characters in header names, implemented as a verify condition.
42 Contributed by Michael Fischer v. Mollard.
44 TL/05 Rename SPF condition results err_perm and err_temp to standardized
45 results permerror and temperror. Is a backward incompatibility if
46 the ACL tests for either of these two results. Patch contributed by
47 user bes-internal on the mailing list.
49 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
51 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
52 selectors, in both main and reject logs.
57 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
59 PP/02 Make -n do something, by making it not do something.
60 When combined with -bP, the name of an option is not output.
62 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
65 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
66 $sender_host_name and config options to manage this, and basic check
69 PP/05 DSCP support for outbound connections and control modifier for inbound.
71 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
72 (Only plugin which currently uses this is kerberos4, which nobody should
73 be using, but we should make it available and other future plugins might
74 conceivably use it, even though it would break NAT; stuff *should* be
75 using channel bindings instead).
77 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
78 name; added for Sendmail compatibility; requires admin caller.
79 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
80 just ignore it); requires trusted caller.
81 Also parse but ignore: -Ac -Am -X<logfile>
84 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
86 TL/02 Add +smtp_confirmation as a default logging option.
88 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
89 Patch by Magnus Holmgren from 2007-02-20.
91 TL/04 Bugzilla 1281 - Spec typo.
92 Bugzilla 1283 - Spec typo.
93 Bugzilla 1290 - Spec grammar fixes.
95 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
97 TL/06 Add Experimental DMARC support using libopendmarc libraries.
99 TL/07 Fix an out of order global option causing a segfault. Reported to dev
100 mailing list by by Dmitry Isaikin.
102 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
104 JH/02 Support "G" suffix to numbers in ${if comparisons.
106 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
108 NM/01 Bugzilla 1197 - Spec typo
109 Bugzilla 1196 - Spec examples corrections
111 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
113 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
114 gnutls_enable_pkcs11, but renamed to more accurately indicate its
117 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
118 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
120 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
121 "acl {{name}{arg}...}", and optional args on acl condition
124 JH/05 Permit multiple router/transport headers_add/remove lines.
126 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
128 JH/07 Avoid using a waiting database for a single-message-only transport.
129 Performance patch from Paul Fisher. Bugzilla 1262.
131 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
134 JH/09 Add $headers_added variable, with content from use of ACL modifier
135 add_header (but not yet added to the message). Bugzilla 199.
137 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
138 Pulled from Bugzilla 817 by Wolfgang Breyha.
140 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
142 (nb: this is the same fix as in Exim 4.80.1)
144 JH/11 Add A= logging on delivery lines, and a client_set_id option on
147 JH/12 Add optional authenticated_sender logging to A= and a log_selector
150 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
152 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
153 advertise SMTP AUTH mechanism to us, instead of a generic
154 protocol violation error. Also, make Exim more robust to bad
155 data from the Dovecot auth socket.
157 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
159 When a queue runner is handling a message, Exim first routes the
160 recipient addresses, during which it prunes them based on the retry
161 hints database. After that it attempts to deliver the message to
162 any remaining recipients. It then updates the hints database using
165 So if a recipient address works intermittently, it can get repeatedly
166 deferred at routing time. The retry hints record remains fresh so the
167 address never reaches the final cutoff time.
169 This is a fairly common occurrence when a user is bumping up against
170 their storage quota. Exim had some logic in its local delivery code
171 to deal with this. However it did not apply to per-recipient defers
172 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
174 This change adds a proper retry rule check during routing so that the
175 final cutoff time is checked against the message's age. We only do
176 this check if there is an address retry record and there is not a
177 domain retry record; this implies that previous attempts to handle
178 the address had the retry_use_local_parts option turned on. We use
179 this as an approximation for the destination being like a local
180 delivery, as in LMTP.
182 I suspect this new check makes the old local delivery cutoff check
183 redundant, but I have not verified this so I left the code in place.
185 TF/02 Correct gecos expansion when From: is a prefix of the username.
187 Test 0254 submits a message to Exim with the header
191 When I ran the test suite under the user fanf2, Exim expanded
192 the header to contain my full name, whereas it should have added
193 a Resent-Sender: header. It erroneously treats any prefix of the
194 username as equal to the username.
196 This change corrects that bug.
198 GF/01 DCC debug and logging tidyup
199 Error conditions log to paniclog rather than rejectlog.
200 Debug lines prefixed by "DCC: " to remove any ambiguity.
202 TF/03 Avoid unnecessary rebuilds of lookup-related code.
204 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
205 Bug spotted by Jeremy Harris; was flawed since initial commit.
206 Would have resulted in OCSP responses post-SNI triggering an Exim
207 NULL dereference and crash.
209 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
211 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
212 Bug detection, analysis and fix by Samuel Thibault.
213 Bugzilla 1331, Debian bug #698092.
215 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
217 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
218 Server implementation by Todd Lyons, client by JH.
219 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
220 config variable "prdr_enable" controls whether the server
221 advertises the facility. If the client requests PRDR a new
222 acl_data_smtp_prdr ACL is called once for each recipient, after
223 the body content is received and before the acl_smtp_data ACL.
224 The client is controlled by bolth of: a hosts_try_prdr option
225 on the smtp transport, and the server advertisement.
226 Default client logging of deliveries and rejections involving
227 PRDR are flagged with the string "PRDR".
229 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
230 fclose(). Diagnosis by Todd Lyons.
232 PP/17 Update configure.default to handle IPv6 localhost better.
233 Patch by Alain Williams (plus minor tweaks).
236 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
237 This is now consistent with GnuTLS, and is now documented: the
238 previous undocumented portable approach to treating the option as
239 unset was to force an expansion failure. That still works, and
240 an empty string is now equivalent.
242 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
243 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
244 not performing validation itself.
246 PP/20 Added force_command boolean option to pipe transport.
247 Patch from Nick Koston, of cPanel Inc.
249 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
252 TF/04 Added udpsend ACL modifer and hexquote expansion operator
254 PP/21 Fix eximon continuous updating with timestamped log-files.
255 Broken in a format-string cleanup in 4.80, missed when I repaired the
256 other false fix of the same issue.
257 Report and fix from Heiko Schlichting.
260 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
261 Report from Prashanth Katuri.
263 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
264 It's SecureTransport, so affects any MacOS clients which use the
265 system-integrated TLS libraries, including email clients.
267 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
268 using a MIME ACL for non-SMTP local injection.
269 Report and assistance in diagnosis by Warren Baker.
271 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
273 JH/16 Fix comparisons for 64b. Bugzilla 1385.
275 TL/09 Add expansion variable $authenticated_fail_id to keep track of
276 last id that failed so it may be referenced in subsequent ACL's.
278 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
281 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
282 ldap library initialization, allowing self-signed CA's to be
283 used. Also properly sets require_cert option later in code by
284 using NULL (global ldap config) instead of ldap handle (per
285 session). Bug diagnosis and testing by alxgomz.
287 TL/12 Enhanced documentation in the ratelimit.pl script provided in
288 the src/util/ subdirectory.
290 TL/13 Bug 1301 - Imported transport SQL logging patch from Axel Rau
291 renamed to Transport Post Delivery Action by Jeremy Harris, as
294 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
295 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
296 redis_servers = needs to be configured which will be used by the redis
297 lookup. Patch from Warren Baker, of The Packet Hub.
299 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
301 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
302 hostname or reverse DNS when processing a host list. Used suggestions
303 from multiple comments on this bug.
305 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
307 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
308 Missed a few lines, added it to make the runtest require no keyboard
311 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
312 contains upper case chars. Make router use caseful_local_part.
314 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
315 support when GnuTLS has been built with p11-kit.
321 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
323 This, or similar/improved, will also be change PP/11 of 4.82.
329 PP/01 Handle short writes when writing local log-files.
330 In practice, only affects FreeBSD (8 onwards).
331 Bugzilla 1053, with thanks to Dmitry Isaikin.
333 NM/01 Bugzilla 949 - Documentation tweak
335 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
338 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
340 PP/02 Implemented gsasl authenticator.
342 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
344 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
345 `pkg-config foo` for cflags/libs.
347 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
348 with rest of GSASL and with heimdal_gssapi.
350 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
351 `pkg-config foo` for cflags/libs for the TLS implementation.
353 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
354 properties get this fed in as external SSF. A number of robustness
355 and debugging improvements to the cyrus_sasl authenticator.
357 PP/08 cyrus_sasl server now expands the server_realm option.
359 PP/09 Bugzilla 1214 - Log authentication information in reject log.
360 Patch by Jeremy Harris.
362 PP/10 Added dbmjz lookup type.
364 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
366 PP/12 MAIL args handles TAB as well as SP, for better interop with
367 non-compliant senders.
368 Analysis and variant patch by Todd Lyons.
370 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
371 Bug report from Lars Müller <lars@samba.org> (via SUSE),
372 Patch from Dirk Mueller <dmueller@suse.com>
374 PP/13 tls_peerdn now print-escaped for spool files.
375 Observed some $tls_peerdn in wild which contained \n, which resulted
376 in spool file corruption.
378 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
379 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
380 or write after TLS renegotiation, which otherwise led to messages
383 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
384 as a tracking header (ie: a signed header comes before the signature).
385 Patch from Wolfgang Breyha.
387 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
388 comma-sep list; embedded commas doubled.
390 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
392 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
394 Report and patch from Dmitry Banschikov.
396 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
397 Removed SSL_clear() after SSL_new() which led to protocol negotiation
398 failures. We appear to now support TLS1.1+ with Exim.
400 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
401 lets Exim select keys and certificates based upon TLS SNI from client.
402 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
403 before an outbound SMTP session. New log_selector, +tls_sni.
405 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
406 NULL dereference. Report and patch from Alun Jones.
408 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
409 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
410 Not seeing resolver debug output on NetBSD, but suspect this is a
411 resolver implementation change.
413 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
414 Left warnings. Added "eximon gdb" invocation mode.
416 PP/21 Defaulting "accept_8bitmime" to true, not false.
418 PP/22 Added -bw for inetd wait mode support.
420 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
421 locate the relevant includes and libraries. Made this the default.
423 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
424 Bugzilla 1246, report and most of solution from Tomasz Kusy.
426 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
427 This may cause build issues on older platforms.
429 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
430 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
431 gnutls_require_mac & gnutls_require_protocols (no longer supported).
432 Added SNI support via GnuTLS too.
433 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
435 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
437 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
438 Applied second patch from Janne, implementing suggestion to default
439 multiple-strings-in-record handling to match SPF spec.
441 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
443 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
444 read-only, out of scope).
445 Patch from Wolfgang Breyha, report from Stuart Northfield.
447 PP/29 Fix three issues highlighted by clang analyser static analysis.
448 Only crash-plausible issue would require the Cambridge-specific
449 iplookup router and a misconfiguration.
450 Report from Marcin Mirosław.
452 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
454 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
455 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
456 As part of this, removing so much warning spew let me fix some minor
457 real issues in debug logging.
459 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
460 assignment on my part. Fixed.
462 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
463 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
464 Janne Snabb (who went above and beyond: thank you).
466 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
467 string otherwise requires a connection and a bunch more work and it's
468 relatively easy to get wrong. Should also expose TLS library linkage
471 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
472 64-bit ${eval} (JH/03).
474 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
475 GNU libc to support some of the 64-bit stuff, should not lead to
476 conflicts. Defined before os.h is pulled in, so if a given platform
477 needs to override this, it can.
479 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
480 protection layer was required, which is not implemented.
481 Bugzilla 1254, patch from Wolfgang Breyha.
483 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
484 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
485 tls_dhparam take prime identifiers. Also unbreak combination of
486 OpenSSL+DH_params+TLSSNI.
488 PP/39 Disable SSLv2 by default in OpenSSL support.
494 PP/01 Solaris build fix for Oracle's LDAP libraries.
495 Bugzilla 1109, patch from Stephen Usher.
497 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
499 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
502 TF/02 Fix a couple more cases where we did not log the error message
503 when unlink() failed. See also change 4.74-TF/03.
505 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
506 lock up or crash if it happened to be inside a call to libc when it
507 got a SIGUSR1 from exiwhat.
509 The SIGUSR1 handler appends the current process status to the process
510 log which is later printed by exiwhat. It used to use the general
511 purpose logging code to do this, but several functions it calls are
512 not safe for signals.
514 The new output code in the SIGUSR1 handler is specific to the process
515 log, and simple enough that it's easy to inspect for signal safety.
516 Removing some special cases also simplifies the general logging code.
517 Removing the spurious timestamps from the process log simplifies
520 TF/04 Improved ratelimit ACL condition.
522 The /noupdate option has been deprecated in favour of /readonly which
523 has clearer semantics. The /leaky, /strict, and /readonly update modes
524 are mutually exclusive. The update mode is no longer included in the
525 database key; it just determines when the database is updated. (This
526 means that when you upgrde Exim will forget old rate measurements.)
528 Exim now checks that the per_* options are used with an update mode that
529 makes sense for the current ACL. For example, when Exim is processing a
530 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
531 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
532 must specify per_mail/readonly. If you omit the update mode it defaults to
533 /leaky where that makes sense (as before) or /readonly where required.
535 The /noupdate option is now undocumented but still supported for
536 backwards compatibility. It is equivalent to /readonly except that in
537 ACLs where /readonly is required you may specify /leaky/noupdate or
538 /strict/noupdate which are treated the same as /readonly.
540 A useful new feature is the /count= option. This is a generalization
541 of the per_byte option, so that you can measure the throughput of other
542 aggregate values. For example, the per_byte option is now equivalent
543 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
545 The per_rcpt option has been generalized using the /count= mechanism
546 (though it's more complicated than the per_byte equivalence). When it is
547 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
548 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
549 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
550 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
551 Note that using per_rcpt with a non-readonly update mode in more than
552 one ACL will cause the recipients to be double-counted. (The per_mail
553 and per_byte options don't have this problem.)
555 The handling of very low rates has changed slightly. If the computed rate
556 is less than the event's count (usually one) then this event is the first
557 after a long gap. In this case the rate is set to the same as this event's
558 count, so that the first message of a spam run is counted properly.
560 The major new feature is a mechanism for counting the rate of unique
561 events. The new per_addr option counts the number of different
562 recipients that someone has sent messages to in the last time period. It
563 behaves like per_rcpt if all the recipient addresses are different, but
564 duplicate recipient addresses do not increase the measured rate. Like
565 the /count= option this is a general mechanism, so the per_addr option
566 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
567 example, measure the rate that a client uses different sender addresses
568 with the options per_mail/unique=$sender_address. There are further
569 details in the main documentation.
571 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
573 TF/06 Removed a few PCRE remnants.
575 TF/07 Automatically extract Exim's version number from tags in the git
576 repository when doing development or release builds.
578 PP/02 Raise smtp_cmd_buffer_size to 16kB.
579 Bugzilla 879. Patch from Paul Fisher.
581 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
582 Heavily based on revision 40f9a89a from Simon Arlott's tree.
585 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
587 PP/05 Variable $av_failed, true if the AV scanner deferred.
588 Bugzilla 1078. Patch from John Horne.
590 PP/06 Stop make process more reliably on build failure.
591 Bugzilla 1087. Patch from Heiko Schlittermann.
593 PP/07 Make maildir_use_size_file an _expandable_ boolean.
594 Bugzilla 1089. Patch from Heiko Schlittermann.
596 PP/08 Handle ${run} returning more data than OS pipe buffer size.
597 Bugzilla 1131. Patch from Holger Weiß.
599 PP/09 Handle IPv6 addresses with SPF.
600 Bugzilla 860. Patch from Wolfgang Breyha.
602 PP/10 GnuTLS: support TLS 1.2 & 1.1.
604 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
607 PP/11 match_* no longer expand right-hand-side by default.
608 New compile-time build option, EXPAND_LISTMATCH_RHS.
609 New expansion conditions, "inlist", "inlisti".
611 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
613 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
615 PP/14 fix log_write() format string regression from TF/03.
616 Bugzilla 1152. Patch from Dmitry Isaikin.
622 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
624 PP/02 Harmonised TLS library version reporting; only show if debugging.
625 Layout now matches that introduced for other libraries in 4.74 PP/03.
627 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
629 PP/04 New "dns_use_edns0" global option.
631 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
634 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
635 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
637 TK/01 Updated PolarSSL code to 0.14.2.
638 Bugzilla 1097. Patch from Andreas Metzler.
640 PP/07 Catch divide-by-zero in ${eval:...}.
643 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
646 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
647 format-string attack -- SECURITY: remote arbitrary code execution.
649 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
650 time unintentionally subject to list matching rules, letting the header
651 cause arbitrary Exim lookups (of items which can occur in lists, *not*
652 arbitrary string expansion). This allowed for information disclosure.
654 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
655 INT_MIN/-1 -- value coerced to INT_MAX.
661 NM/01 Workround for PCRE version dependency in version reporting
664 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
665 This fixes portability to compilers other than gcc, notably
666 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
668 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
669 makefiles for portability to HP-UX and POSIX correctness.
671 PP/01 Permit LOOKUP_foo enabling on the make command-line.
672 Also via indented variable definition in the Makefile.
673 (Debugging by Oliver Heesakkers).
675 PP/02 Restore caching of spamd results with expanded spamd_address.
676 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
678 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
679 Improves build reliability. Fix from: Frank Elsner
681 NM/02 Fix wide character breakage in the rfc2047 coding
682 Fixes bug 1064. Patch from Andrey N. Oktyabrski
684 NM/03 Allow underscore in dnslist lookups
685 Fixes bug 1026. Patch from Graeme Fowler
687 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
688 Code patches from Adam Ciarcinski of NetBSD.
690 NM/04 Fixed exiqgrep to cope with mailq missing size issue
693 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
694 is logged, to avoid truncation. Patch from John Horne.
696 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
697 Patch from Jakob Hirsch.
699 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
700 SQL string expansion failure details.
701 Patch from Andrey Oktyabrski.
703 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
704 Patch from Simon Arlott.
706 PP/09 New lookups functionality failed to compile on old gcc which rejects
707 extern declarations in function scope.
708 Patch from Oliver Fleischmann
710 PP/10 Use sig_atomic_t for flags set from signal handlers.
711 Check getgroups() return and improve debugging.
712 Fixed developed for diagnosis in bug 927 (which turned out to be
715 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
716 Patch from Mark Zealey.
718 PP/12 Bugzilla 1056: Improved spamd server selection.
719 Patch from Mark Zealey.
721 PP/13 Bugzilla 1086: Deal with maildir quota file races.
722 Based on patch from Heiko Schlittermann.
724 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
725 Patch from Uwe Doering, sign-off by Michael Haardt.
727 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
728 variable declaration deep within a block. Bug and patch from
731 PP/15 lookups-Makefile IRIX compatibilty coercion.
733 PP/16 Make DISABLE_DKIM build knob functional.
735 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
736 Patch by Simon Arlott
738 TF/03 Fix valgrind.h portability to C89 compilers that do not support
739 variable argument macros. Our copy now differs from upstream.
745 TF/01 Failure to get a lock on a hints database can have serious
746 consequences so log it to the panic log.
748 TF/02 Log LMTP confirmation messages in the same way as SMTP,
749 controlled using the smtp_confirmation log selector.
751 TF/03 Include the error message when we fail to unlink a spool file.
753 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
754 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
755 for maintaining out-of-tree patches for some time.
757 PP/01 Bugzilla 139: Documentation and portability issues.
758 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
759 Handle per-OS dynamic-module compilation flags.
761 PP/02 Let /dev/null have normal permissions.
762 The 4.73 fixes were a little too stringent and complained about the
763 permissions on /dev/null. Exempt it from some checks.
764 Reported by Andreas M. Kirchwitz.
766 PP/03 Report version information for many libraries, including
767 Exim version information for dynamically loaded libraries. Created
768 version.h, now support a version extension string for distributors
769 who patch heavily. Dynamic module ABI change.
771 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
772 privilege escalation vulnerability whereby the Exim run-time user
773 can cause root to append content of the attacker's choosing to
776 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
779 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
780 If dropping privileges for untrusted macros, we disabled normal logging
781 on the basis that it would fail; for the Exim run-time user, this is not
782 the case, and it resulted in successful deliveries going unlogged.
783 Fixed. Reported by Andreas Metzler.
789 PP/01 Date: & Message-Id: revert to normally being appended to a message,
790 only prepend for the Resent-* case. Fixes regression introduced in
791 Exim 4.70 by NM/22 for Bugzilla 607.
793 PP/02 Include check_rfc2047_length in configure.default because we're seeing
794 increasing numbers of administrators be bitten by this.
796 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
798 PP/03 Bugzilla 994: added openssl_options main configuration option.
800 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
802 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
804 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
806 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
807 without a peer certificate, leading to a segfault because of an
808 assumption that peers always have certificates. Be a little more
809 paranoid. Problem reported by Martin Tscholak.
811 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
812 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
813 NB: ClamAV planning to remove STREAM in "middle of 2010".
814 CL also introduces -bmalware, various -d+acl logging additions and
815 more caution in buffer sizes.
817 PP/09 Implemented reverse_ip expansion operator.
819 PP/10 Bugzilla 937: provide a "debug" ACL control.
821 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
823 PP/12 Bugzilla 973: Implement --version.
825 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
827 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
829 PP/15 Bugzilla 816: support multiple condition rules on Routers.
831 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
832 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
833 ignore trailing whitespace.
835 JJ/02 prevent non-panic DKIM error from being sent to paniclog
837 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
840 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
841 Notification from Dr Andrew Aitchison.
843 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
844 ExtendedDetectionInfo response format.
845 Notification from John Horne.
847 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
850 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
851 XSL and documented dependency on system catalogs, with examples of how
854 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
857 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
858 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
859 configuration file which is writeable by the Exim user or group.
861 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
862 of configuration files to cover files specified with the -C option if
863 they are going to be used with root privileges, not just the default
866 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
867 option (effectively making it always true).
869 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
870 files to be used while preserving root privileges.
872 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
873 that rogue child processes cannot use them.
875 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
876 run-time user, instead of root.
878 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
879 Exim run-time user without dropping privileges.
881 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
882 result string, instead of calling string_vformat() twice with the same
885 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
886 for other users. Others should always drop root privileges if they use
887 -C on the command line, even for a whitelisted configure file.
889 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
891 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
897 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
898 $data_path, and $header_path variables; fixed documentation bugs and
901 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
902 exipick to access non-standard spools, including the "frozen" queue
905 NM/01 Bugzilla 965: Support mysql stored procedures.
906 Patch from Alain Williams
908 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
910 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
911 Patch from Andreas Metzler
913 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
914 Patch from Kirill Miazine
916 NM/05 Bugzilla 671: Added umask to procmail example.
918 JJ/03 installed exipick 20100323.0, fixing doc bug
920 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
921 directory. Notification and patch from Dan Rosenberg.
923 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
925 TK/02 Improve log output when DKIM signing operation fails.
927 MH/01 Treat the transport option dkim_domain as a colon separated
928 list, not as a single string, and sign the message with each element,
929 omitting multiple occurences of the same signer.
931 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
932 Bugzilla 985, 986. Patch by Simon Arlott
934 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
935 Patch by Simon Arlott
937 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
938 MBX locking. Notification from Dan Rosenberg.
944 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
946 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
948 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
950 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
952 NM/04 Bugzilla 915: Flag broken perl installation during build.
958 TK/01 Added patch by Johannes Berg that expands the main option
959 "spamd_address" if it starts with a dollar sign.
961 TK/02 Write list of recipients to X-Envelope-Sender header when building
962 the mbox-format spool file for content scanning (suggested by Jakob
965 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
966 (http://www.dcc-servers.net/) support via dccifd. Activated by
967 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
969 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
970 by Mark Daniel Reidel <mr@df.eu>.
972 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
973 When building exim an external PCRE library is now needed -
974 PCRE is a system library on the majority of modern systems.
975 See entry on PCRE_LIBS in EDITME file.
977 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
978 conversation. Added nologin parameter to request.
979 Patch contributed by Kirill Miazine.
981 TF/01 Do not log submission mode rewrites if they do not change the address.
983 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
985 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
986 log files in place. Contributed by Roberto Lima.
988 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
990 TF/03 Bugzilla 615: When checking the local_parts router precondition
991 after a local_part_suffix or local_part_prefix option, Exim now
992 does not use the address's named list lookup cache, since this
993 contains cached lookups for the whole local part.
995 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
996 Robert Millan. Documentation is in experimental-spec.txt.
998 TF/04 Bugzilla 668: Fix parallel build (make -j).
1000 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1002 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1003 Patch provided by Jan Srzednicki.
1005 TF/05 Leading white space used to be stripped from $spam_report which
1006 wrecked the formatting. Now it is preserved.
1008 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1009 that they are available at delivery time.
1011 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1013 TF/08 TLS error reporting now respects the incoming_interface and
1014 incoming_port log selectors.
1016 TF/09 Produce a more useful error message if an SMTP transport's hosts
1017 setting expands to an empty string.
1019 NM/06 Bugzilla 744: EXPN did not work under TLS.
1020 Patch provided by Phil Pennock.
1022 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1023 Patch provided by Richard Godbee.
1025 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1026 acl_smtp_notquit, added index entry.
1028 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1029 Patch provided by Eugene Bujak.
1031 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1032 accept(). Patch provided by Maxim Dounin.
1034 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1035 Patch provided by Phil Pennock.
1037 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1039 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1040 Patch provided by Brad "anomie" Jorsch.
1042 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1043 Patch provided by Dean Brooks.
1045 TK/05 Add native DKIM support (does not depend on external libraries).
1047 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1048 Patch provided by Graeme Fowler.
1050 NM/16 Bugzilla 851: Documentation example syntax fix.
1052 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1054 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1057 NM/19 Bugzilla 745: TLS version reporting.
1058 Patch provided by Phil Pennock.
1060 NM/20 Bugzilla 167: bool: condition support.
1061 Patch provided by Phil Pennock.
1063 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1064 clients. Patch provided by Phil Pennock.
1066 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1067 Patch provided by Brad "anomie" Jorsch.
1069 NM/23 Bugzilla 687: Fix misparses in eximstats.
1070 Patch provided by Heiko Schlittermann.
1072 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1073 Patch provided by Heiko Schlittermann.
1075 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1076 plus update to original patch.
1078 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1080 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1081 Patch provided by David Brownlee.
1083 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1085 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1087 NM/30 Bugzilla 888: TLS documentation bugfixes.
1089 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1091 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1092 Unlike the original bugzilla I have changed all shell scripts in src tree.
1094 NM/33 Bugzilla 898: Transport filter timeout fix.
1095 Patch by Todd Rinaldo.
1097 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1098 Patch by Serge Demonchaux.
1100 NM/35 Bugzilla 39: Base64 decode bug fixes.
1101 Patch by Jakob Hirsch.
1103 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1105 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1107 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1109 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1115 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1116 ALT-N's libdkim that I have put here:
1117 http://duncanthrax.net/exim-experimental/
1119 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1120 were called 'true' and it seems that C99 defines that as a reserved
1121 keyword to be used with 'bool' variable types. That means you could
1122 not include C99-style headers which use bools without triggering
1123 build errors in sieve.c.
1125 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1126 as mailq or other aliases. Changed the --help handling significantly
1127 to do whats expected. exim_usage() emits usage/help information.
1129 SC/01 Added the -bylocaldomain option to eximstats.
1131 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1133 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1135 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1141 PH/01 Another patch from the Sieve maintainer.
1143 PH/02 When an IPv6 address is converted to a string for single-key lookup
1144 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1145 dots are used instead of colons so that keys in lsearch files need not
1146 contain colons. This was done some time before quoting was made available
1147 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1148 (notated using the quote facility) so as to distinguish them from IPv4
1149 keys. This meant that lookups for IP addresses in host lists did not work
1150 for iplsearch lookups.
1152 This has been fixed by arranging for IPv6 addresses to be expressed with
1153 colons if the lookup type is iplsearch. This is not incompatible, because
1154 previously such lookups could never work.
1156 The situation is now rather anomolous, since one *can* have colons in
1157 ordinary lsearch keys. However, making the change in all cases is
1158 incompatible and would probably break a number of configurations.
1160 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1163 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1164 conversion specification without a maximum field width, thereby enabling
1165 a rogue spamd server to cause a buffer overflow. While nobody in their
1166 right mind would setup Exim to query an untrusted spamd server, an
1167 attacker that gains access to a server running spamd could potentially
1168 exploit this vulnerability to run arbitrary code as the Exim user.
1170 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1171 $primary_hostname instead of what libspf2 thinks the hosts name is.
1173 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1174 a directory entry by the name of the lookup key. Previously, if a
1175 symlink pointed to a non-existing file or a file in a directory that
1176 Exim lacked permissions to read, a lookup for a key matching that
1177 symlink would fail. Now it is enough that a matching directory entry
1178 exists, symlink or not. (Bugzilla 503.)
1180 PH/03 The body_linecount and body_zerocount variables are now exported in the
1183 PH/04 Added the $dnslist_matched variable.
1185 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1186 This means they are set thereafter only if the connection becomes
1189 PH/06 Added the client_condition to authenticators so that some can be skipped
1190 by clients under certain conditions.
1192 PH/07 The error message for a badly-placed control=no_multiline_responses left
1193 "_responses" off the end of the name.
1195 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1197 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1198 (without spaces) instead of just copying the configuration text.
1200 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1202 PH/11 Added $max_received_linelength.
1204 PH/12 Added +ignore_defer and +include_defer to host lists.
1206 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1207 way in which PCRE > 7.0 is built.
1209 PH/14 Implemented queue_only_load_latch.
1211 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1212 MAIL command. The effect was to mangle the value on 64-bit systems.
1214 PH/16 Another patch from the Sieve maintainer.
1216 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1218 PH/18 If a system quota error occurred while trying to create the file for
1219 a maildir delivery, the message "Mailbox is full" was not appended to the
1220 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1221 only to a quota excession during the actual writing of the file.
1223 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1224 characters?) which causes problems in log lines. The DN values are now
1225 passed through string_printing() before being added to log lines.
1227 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1228 and InterBase are left for another time.)
1230 PH/21 Added message_body_newlines option.
1232 PH/22 Guard against possible overflow in moan_check_errorcopy().
1234 PH/23 POSIX allows open() to be a macro; guard against that.
1236 PH/24 If the recipient of an error message contained an @ in the local part
1237 (suitably quoted, of course), incorrect values were put in $domain and
1238 $local_part during the evaluation of errors_copy.
1244 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1245 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1248 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1249 issue a MAIL command.
1251 PH/02 In an ACL statement such as
1253 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1255 if a client was not listed at all, or was listed with a value other than
1256 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1257 the condition was not true (as it should be), so access was not denied.
1258 The bug was that the ! inversion was incorrectly passed on to the second
1259 item. This has been fixed.
1261 PH/03 Added additional dnslists conditions == and =& which are different from
1262 = and & when the dns lookup returns more than one IP address.
1264 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1265 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1267 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1268 FSYNC, which compiles an option called disable_fsync that allows for
1269 bypassing fsync(). The documentation is heavily laced with warnings.
1271 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1273 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1274 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1275 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1276 including adding "make clean"; (3) Added -fPIC when compiling the test
1277 dynamically loaded module, to get rid of a warning.
1279 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1280 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1281 = 0s. The bug is otherwise harmless.
1283 PH/07 There was a bug in the dovecot authenticator such that the value of
1284 $auth1 could be overwritten, and so not correctly preserved, after a
1285 successful authentication. This usually meant that the value preserved by
1286 the server_setid option was incorrect.
1288 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1290 PH/09 Installed PCRE release 7.0.
1292 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1293 run for batched SMTP input. It is now run at the start of every message
1294 in the batch. While fixing this I discovered that the process information
1295 (output by running exiwhat) was not always getting set for -bs and -bS
1296 input. This is fixed, and it now also says "batched" for BSMTP.
1298 PH/11 Added control=no_pipelining.
1300 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1301 patch, slightly modified), and move the expansion of helo_data till after
1302 the connection is made in the smtp transport (so it can use these
1305 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1307 PH/14 Added log_selector = +pid.
1309 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1311 PH/16 Add ${if forany and ${if forall.
1313 PH/17 Added dsn_from option to vary the From: line in DSNs.
1315 PH/18 Flush SMTP output before performing a callout, unless control =
1316 no_callout_flush is set.
1318 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1319 was true (the default) a successful delivery failed to delete the retry
1320 item, thus causing premature timeout of the address. The bug is now
1323 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1325 PH/21 Long custom messages for fakedefer and fakereject are now split up
1326 into multiline reponses in the same way that messages for "deny" and
1327 other ACL rejections are.
1329 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1330 with slight modification.
1332 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1333 draft, changing the syntax and factoring some duplicate code".
1335 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1336 for deliveries of the second and subsequent messages over the same SMTP
1339 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1340 ${reduce, with only minor "tidies".
1342 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1344 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1345 expansion side effects.
1347 PH/27 When a message times out after an over-quota error from an Exim-imposed
1348 quota, the bounce message says "mailbox is full". This message was not
1349 being given when it was a system quota that was exceeded. It now should
1352 MH/03 Made $recipients available in local_scan(). local_scan() already has
1353 better access to the recipient list through recipients_list[], but
1354 $recipients can be useful in postmaster-provided expansion strings.
1356 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1357 in the case of a MAIL command with additional options following the
1358 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1359 were accidentally chopped off.
1361 PH/29 SMTP synchronization checks are implemented when a command is read -
1362 there is a check that no more input is waiting when there shouldn't be
1363 any. However, for some commands, a delay in an ACL can mean that it is
1364 some time before the response is written. In this time, more input might
1365 arrive, invalidly. So now there are extra checks after an ACL has run for
1366 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1367 pipelining has not been advertised.
1369 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1371 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1372 returned in the SMTP response when smtp_return_error_details was false.
1373 This has been fixed.
1375 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1376 instead of the C I/O that was originally supplied, because problems were
1377 reported on Solaris.
1379 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1380 Exim which did not show up earlier: it was assuming that a call to
1381 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1382 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1383 was a macro that became an assignment, so it seemed to work. This has
1384 changed to a proper function call with a void return, hence the compile
1385 error. Exim's code has been fixed.
1387 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1390 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1392 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1394 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1395 because it turns out to be too big or there is a timeout) there is no
1396 "Completed" line in the log. When some messages of this type were
1397 selected by exigrep, they were listed as "not completed". Others were
1398 picked up by some special patterns. I have improved the selection
1399 criteria to be more general.
1401 PH/38 The host_find_failed option in the manualroute router can now be set
1402 to "ignore", to completely ignore a host whose IP address cannot be
1403 found. If all hosts are ignored, the behaviour is controlled by the new
1404 host_all_ignored option.
1406 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1407 homing or because of multiple MX records with /mx) generated more than
1408 one IP address, and the following item turned out to be the local host,
1409 all the secondary addresses of the first item were incorrectly removed
1410 from the list, along with the local host and any following hosts (which
1411 is what is supposed to happen).
1413 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1414 whoever called Exim into the -H file. In the case of the daemon it was
1415 behaving confusingly. When first started, it used values for whoever
1416 started the daemon, but after a SIGHUP it used the Exim user (because it
1417 calls itself on a restart). I have changed the code so that it now always
1420 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1421 message are rejected with the same error (e.g. no authentication or bad
1422 sender address), and a DATA command is nevertheless sent (as can happen
1423 with PIPELINING or a stupid MUA), the error message that was given to the
1424 RCPT commands is included in the rejection of the DATA command. This is
1425 intended to be helpful for MUAs that show only the final error to their
1428 PH/42 Another patch from the Sieve maintainer.
1430 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1431 Eximstats - Fixed some broken HTML links and added missing column headers
1433 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1434 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1436 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1438 PH/43 Yet another patch from the Sieve maintainer.
1440 PH/44 I found a way to check for a TCP/IP connection going away before sending
1441 the response to the final '.' that terminates a message, but only in the
1442 case where the client has not sent further data following the '.'
1443 (unfortunately, this is allowed). However, in many cases there won't be
1444 any further data because there won't be any more messages to send. A call
1445 to select() can be used: if it shows that the input is "ready", there is
1446 either input waiting, or the socket has been closed. An attempt to read
1447 the next input character can distinguish the two cases. Previously, Exim
1448 would have sent an OK response which the client would never have see.
1449 This could lead to message repetition. This fix should cure that, at
1450 least in a lot of common cases.
1452 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1453 advertised in response to EHLO.
1459 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1460 fixed by 4.65/MH/01 (is this a record?) are fixed:
1462 (i) An empty string was always treated as zero by the numeric comparison
1463 operators. This behaviour has been restored.
1465 (ii) It is documented that the numeric comparison operators always treat
1466 their arguments as decimal numbers. This was broken in that numbers
1467 starting with 0 were being interpreted as octal.
1469 While fixing these problems I realized that there was another issue that
1470 hadn't been noticed. Values of message_size_limit (both the global option
1471 and the transport option) were treated as octal if they started with 0.
1472 The documentation was vague. These values are now always treated as
1473 decimal, and I will make that clear in the documentation.
1479 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1480 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1483 MH/01 Don't check that the operands of numeric comparison operators are
1484 integers when their expansion is in "skipping" mode (fixes bug
1485 introduced by 4.64-PH/07).
1487 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1488 child addresses, Exim now panics and dies. Previously, because the count
1489 is held in a short int, deliveries were likely to be lost. As such a
1490 large number of recipients for a single message is ridiculous
1491 (performance will be very, very poor), I have chosen to impose a limit
1492 rather than extend the field.
1498 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1499 leftover -K file (the existence of which was triggered by #402).
1500 While we were at it, introduced process PID as part of the -K
1501 filename. This should rule out race conditions when creating
1504 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1505 processing considerably. Previous code took too long for large mails,
1506 triggering a timeout which in turn triggers #401.
1508 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1509 in the DK code in transports.c. sendfile() is not really portable,
1510 hence the _LINUX specificness.
1512 TF/01 In the add_headers option to the mail command in an Exim filter,
1513 there was a bug that Exim would claim a syntax error in any
1514 header after the first one which had an odd number of characters
1517 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1518 callout verification, Exim cached a "reject" for the entire domain. This
1519 is correct for most verifications, but it is not correct for a recipient
1520 verification with use_sender or use_postmaster set, because in that case
1521 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1522 case of MAIL FROM:<> rejection from other early rejections (e.g.
1523 rejection of HELO). When verifying a recipient using a non-null MAIL
1524 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1525 Whatever the result of the callout, the value of the domain cache is
1526 left unchanged (for any other kind of callout, getting as far as trying
1527 RCPT means that the domain itself is ok).
1529 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1532 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1533 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1534 session, the daemon ignores ECONNECT errors and logs others; it now
1535 ignores EPIPE as well.
1537 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1538 (quoted-printable decoding).
1540 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1541 later the small subsequent patch to fix an introduced bug.
1543 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1545 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1547 PH/08 An error is now given if message_size_limit is specified negative.
1549 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1550 to be given (somewhat) arbitrary names.
1552 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1555 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1556 miscellaneous code fixes
1558 PH/10 Added the log_reject_target ACL modifier to specify where to log
1561 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1562 hostname. This is wrong, because it relates to the incoming message (and
1563 probably the interface on which it is arriving) and not to the outgoing
1564 callout (which could be using a different interface). This has been
1565 changed to use the value of the helo_data option from the smtp transport
1566 instead - this is what is used when a message is actually being sent. If
1567 there is no remote transport (possible with a router that sets up host
1568 addresses), $smtp_active_hostname is used.
1570 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1571 tweaks were necessary in order to get it to work (see also 21 below):
1572 (a) The code assumed that strncpy() returns a negative number on buffer
1573 overflow, which isn't the case. Replaced with Exim's string_format()
1575 (b) There were several signed/unsigned issues. I just did the minimum
1576 hacking in of casts. There is scope for a larger refactoring.
1577 (c) The code used strcasecmp() which is not a standard C function.
1578 Replaced with Exim's strcmpic() function.
1579 (d) The code set only $1; it now sets $auth1 as well.
1580 (e) A simple test gave the error "authentication client didn't specify
1581 service in request". It would seem that Dovecot has changed its
1582 interface. Fortunately there's a specification; I followed it and
1583 changed what the client sends and it appears to be working now.
1585 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1588 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1589 address A is aliased to B and C, where B exists and C does not. Without
1590 -v the output is "A verified" because verification stops after a
1591 successful redirection if more than one address is generated. However,
1592 with -v the child addresses are also verified. Exim was outputting "A
1593 failed to verify" and then showing the successful verification for C,
1594 with its parentage. It now outputs "B failed to verify", showing B's
1595 parentage before showing the successful verification of C.
1597 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1598 look up a TXT record in a specific list after matching in a combined
1601 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1602 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1603 they consult the DNS. I had assumed they would set it the way they
1604 wanted; and indeed my experiments on Linux seem to show that in some
1605 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1606 To be on the safe side, however, I have now made the interface to
1607 host_find_byname() similar to host_find_bydns(), with an argument
1608 containing the DNS resolver options. The host_find_byname() function now
1609 sets these options at its start, just as host_find_bydns() does. The smtp
1610 transport options dns_qualify_single and dns_search_parents are passed to
1611 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
1612 of host_find_byname() use the default settings of RES_DEFNAMES
1613 (qualify_single) but not RES_DNSRCH (search_parents).
1615 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
1616 spool_read_header() do less string testing, by means of a preliminary
1617 switch on the second character of optional "-foo" lines. (This is
1618 overdue, caused by the large number of possibilities that now exist.
1619 Originally there were few.) While I was there, I also converted the
1620 str(n)cmp tests so they don't re-test the leading "-" and the first
1621 character, in the hope this might squeeze out yet more improvement.
1623 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
1624 flag allowing group syntax was set by the header_syntax check but not
1625 turned off, possible causing trouble later; (2) The flag was not being
1626 set at all for the header_verify test, causing "group"-style headers to
1627 be rejected. I have now set it in this case, and also caused header_
1628 verify to ignore an empty address taken from a group. While doing this, I
1629 came across some other cases where the code for allowing group syntax
1630 while scanning a header line wasn't quite right (mostly, not resetting
1631 the flag correctly in the right place). These bugs could have caused
1632 trouble for malformed header lines. I hope it is now all correct.
1634 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
1635 with the "reply" argument non-NULL. The code, however (which originally
1636 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
1637 but it didn't always do it. This confused somebody who was copying the
1638 code for some other use. I have removed all the tests.
1640 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
1641 feature that was used to support insecure browsers during the U.S. crypto
1642 embargo. It requires special client support, and Exim is probably the
1643 only MTA that supported it -- and would never use it because real RSA is
1644 always available. This code has been removed, because it had the bad
1645 effect of slowing Exim down by computing (never used) parameters for the
1646 RSA_EXPORT functionality.
1648 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
1649 authenticator to fail if there's a tab character in the incoming data
1650 (there should never be unless someone is messing about, as it's supposed
1651 to be base64-encoded). Also added, on Timo's advice, the "secured" option
1652 if the connection is using TLS or if the remote IP is the same as the
1653 local IP, and the "valid-client-cert option" if a client certificate has
1656 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
1657 authenticators. This can be used for authorization after authentication
1658 succeeds. (In the case of plaintext, it servers for both authentication
1661 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1662 if any retry times were supplied.
1664 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1665 connection, where there is no client IP address to check. In this
1666 situation, the verify now always succeeds.
1668 PH/25 Applied John Jetmore's -Mset patch.
1670 PH/26 Added -bem to be like -Mset, but loading a message from a file.
1672 PH/27 In a string expansion for a processed (not raw) header when multiple
1673 headers of the same name were present, leading whitespace was being
1674 removed from all of them, but trailing whitespace was being removed only
1675 from the last one. Now trailing whitespace is removed from each header
1676 before concatenation. Completely empty headers in a concatenation (as
1677 before) are ignored.
1679 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1680 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1682 PH/29 [Removed. This was a change that I later backed out, and forgot to
1683 correct the ChangeLog entry (that I had efficiently created) before
1684 committing the later change.]
1686 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1687 address errors (4xx response to RCPT) over the same connection as other
1688 messages routed to the same hosts. Such deliveries are always "forced",
1689 so retry times are not inspected. This resulted in far too many retries
1690 for the affected addresses. The effect occurred only when there were more
1691 hosts than the hosts_max_try setting in the smtp transport when it had
1692 the 4xx errors. Those hosts that it had tried were not added to the list
1693 of hosts for which the message was waiting, so if all were tried, there
1694 was no problem. Two fixes have been applied:
1696 (i) If there are any address or message errors in an SMTP delivery, none
1697 of the hosts (tried or untried) are now added to the list of hosts
1698 for which the message is waiting, so the message should not be a
1699 candidate for sending over the same connection that was used for a
1700 successful delivery of some other message. This seems entirely
1701 reasonable: after all the message is NOT "waiting for some host".
1702 This is so "obvious" that I'm not sure why it wasn't done
1703 previously. Hope I haven't missed anything, but it can't do any
1704 harm, as the worst effect is to miss an optimization.
1706 (ii) If, despite (i), such a delivery is accidentally attempted, the
1707 routing retry time is respected, so at least it doesn't keep
1708 hammering the server.
1710 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1711 in ${readsocket because some servers need this prod.
1713 PH/32 Added some extra debug output when updating a wait-xxx database.
1715 PH/33 The hint "could be header name not terminated by colon", which has been
1716 given for certain expansion errors for a long time, was not being given
1717 for the ${if def:h_colon_omitted{... case.
1719 PH/34 The spec says: "With one important exception, whenever a domain list is
1720 being scanned, $domain contains the subject domain." There was at least
1721 one case where this was not true.
1723 PH/35 The error "getsockname() failed: connection reset by peer" was being
1724 written to the panic log as well as the main log, but it isn't really
1725 panic-worthy as it just means the connection died rather early on. I have
1726 removed the panic log writing for the ECONNRESET error when getsockname()
1729 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1730 runs only) independently of the message's sender address. This meant
1731 that, if the 4xx error was in fact related to the sender, a different
1732 message to the same recipient with a different sender could confuse
1733 things. In particualar, this can happen when sending to a greylisting
1734 server, but other circumstances could also provoke similar problems.
1735 I have changed the default so that the retry time for these errors is now
1736 based a combination of the sender and recipient addresses. This change
1737 can be overridden by setting address_retry_include_sender=false in the
1740 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1741 remote server are returned as part of bounce messages. This was not
1742 happening for LMTP over a pipe (the lmtp transport), but now it is the
1743 same for both kinds of LMTP.
1745 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1746 in header lines that were in fact CNAMEs. This is no longer the case.
1748 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1749 and queue runs started by the daemon processed all messages. This has
1750 been fixed so that -R and -S can now usefully be given with -q<time>.
1752 PH/40 Import PCRE release 6.7 (fixes some bugs).
1754 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1756 PH/42 Give an error if -q is specified more than once.
1758 PH/43 Renamed the variables $interface_address and $interface_port as
1759 $received_ip_address and $received_port, to make it clear that these
1760 values apply to message reception, and not to the outgoing interface when
1761 a message is delivered. (The old names remain recognized, of course.)
1763 PH/44 There was no timeout on the connect() call when using a Unix domain
1764 socket in the ${readsocket expansion. There now is.
1766 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1767 be meaningful with "accept".
1769 SC/01 Eximstats V1.43
1770 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1772 SC/02 Eximstats V1.44
1773 Use a glob alias rather than an array ref in the generated
1774 parser. This improves both readability and performance.
1776 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1777 Collect SpamAssassin and rejection statistics.
1778 Don't display local sender or destination tables unless
1779 there is data to show.
1780 Added average volumes into the top table text output.
1782 SC/04 Eximstats V1.46
1783 Collect data on the number of addresses (recipients)
1784 as well as the number of messages.
1786 SC/05 Eximstats V1.47
1787 Added 'Message too big' to the list of mail rejection
1788 reasons (thanks to Marco Gaiarin).
1790 SC/06 Eximstats V1.48
1791 Mainlog lines which have GMT offsets and are too short to
1792 have a flag are now skipped.
1794 SC/07 Eximstats V1.49 (Alain Williams)
1795 Added the -emptyok flag.
1797 SC/08 Eximstats V1.50
1798 Fixes for obtaining the IP address from reject messages.
1800 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1801 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1802 whitesspace changes from 4.64-PH/27
1804 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1807 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1808 are found, allow negative numbers in numeric criteria)
1810 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1812 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1813 to match changes made in 4.64-PH/43
1815 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1817 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1818 "domain" is actually an IP address, and force a failure. This locks out
1819 those revolvers/nameservers that support "A-for-A" lookups, in
1820 contravention of the specifications.
1822 PH/48 When a host name was looked up from an IP address, and the subsequent
1823 forward lookup of the name timed out, the host name was left in
1824 $sender_host_name, contrary to the specification.
1826 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1827 restricted to single-key lookups, Exim was not diagnosing an error if
1828 * or *@ was used with a query-style lookup.
1830 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1832 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1833 long ago, but noone interested enough thought of it. Let's just say that
1834 the "1.1" means that there are some new functions that weren't there at
1835 some point in the past.
1837 PH/51 Error processing for expansion failure of helo_data from an smtp
1838 transport during callout processing was broken.
1840 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1841 tested/used via the -bh/-bhc/-bs options.
1843 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1844 bug, fixed in subsequent PCRE releases).
1846 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1847 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1849 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1855 SC/01 Use a glob alias rather than an array ref in eximstats generated
1856 parser. This improves both readability and performance.
1858 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1859 Don't display local sender or destination tables in eximstats unless
1860 there is data to show.
1861 Added average volumes into the eximstats top table text output.
1863 SC/03 Collect data on the number of addresses (recipients) as well
1864 as the number of messages in eximstats.
1866 TF/01 Correct an error in the documentation for the redirect router. Exim
1867 does (usually) call initgroups() when daemonizing.
1869 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1870 with consistent privilege compared to when running as a daemon.
1872 TF/03 Note in the spec that $authenticated_id is not set for local
1873 submissions from trusted users.
1875 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1876 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1878 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1879 by adding some example configuration directives to the default
1880 configuration file. A little bit of work is required to uncomment the
1881 directives and define how usernames and passwords are checked, but
1882 there is now a framework to start from.
1884 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1885 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1886 without this. I don't know how relevant this is to other LDAP libraries.
1888 PH/02 Add the verb name to the "unknown ACL verb" error.
1890 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1892 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1894 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1895 directory not expanded when it should be if an expanded home directory
1896 was set for the address (which is overridden by the transport).
1898 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1901 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1902 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1903 because it is too late at that time, and has no effect.
1905 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1906 security issue with \' (bugzilla #107). I could not use the
1907 PQescapeStringConn() function, because it needs a PGconn value as one of
1910 PH/08 When testing addresses using -bt, indicate those final addresses that
1911 are duplicates that would not cause an additional delivery. At least one
1912 person was confused, thinking that -bt output corresponded to deliveries.
1913 (Suppressing duplicates isn't a good idea as you lose the information
1914 about possibly different redirections that led to the duplicates.)
1916 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1917 systems where poll() doesn't work, in particular OS X.
1919 PH/10 Added more information to debugging output for retry time not reached.
1921 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1922 operations in malware.c.
1924 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1927 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1928 syslog_duplication set false, log lines that would normally be written
1929 both the the main log and to the reject log were not written to syslog at
1932 PH/14 In the default configuration, change the use of "message" in ACL warn
1933 statements to "add_header".
1935 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1936 not followed by a command (e.g. "seen endif").
1938 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1939 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1942 PH/17 Added extra conditions to the default value of delay_warning_condition
1946 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1947 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1948 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1951 The Auto-Submitted: and various List- headers are standardised, whereas I
1952 don't think Precedence: ever was.
1954 PH/18 Refactored debugging code in route_finduser() to show more information,
1955 in particular, the error code if getpwnam() issues one.
1957 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1958 This is apparently needed in addition to the PH/07 change above to avoid
1959 any possible encoding problems.
1961 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1962 but not after initializing Perl.
1964 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1965 output them only if debugging. By default they are written stderr,
1966 apparently, which is not desirable.
1968 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1971 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1974 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1976 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1977 authenticated or an ident call has been made. Suppress the default
1978 values for $authenticated_id and $authenticated_sender (but permit -oMai
1979 and -oMas) when testing with -bh.
1981 PH/24 Re-jigged the order of the tests in the default configuration so that the
1982 tests for valid domains and recipients precede the DNS black list and CSA
1983 tests, on the grounds that those ones are more expensive.
1985 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1986 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1987 This bug exists in every version of Exim that I still have, right back to
1990 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1991 However, an attempt to turn on case-sensitivity in a regex key by
1992 including (?-i) didn't work because the subject string was already
1993 lowercased, and the effects were non-intuitive. It turns out that a
1994 one-line patch can be used to allow (?-i) to work as expected.
2000 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2001 other effects) broke the use of negated acl sub-conditions.
2003 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2006 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2007 "Deny" causes Exim to reject the incoming connection with a 554 error.
2008 Unfortunately, if there is a major crisis, such as a disk failure,
2009 tcp-wrappers gives "deny", whereas what one would like would be some
2010 kind of temporary error. A kludge has been added to help with this.
2011 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2012 554 error is used if errno is still zero or contains ENOENT (which occurs
2013 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2016 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2018 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2019 errors. Otherwise a message that provokes a temporary error (when other
2020 messages do not) can cause a whole host to time out.
2022 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2023 addresses were routed directly to files or pipes from a redirect router.
2024 File deliveries just didn't batch; pipe deliveries might have suffered
2027 PH/06 A failure to get a lock for a hints database would erroneously always say
2028 "Failed to get write lock", even when it was really a read lock.
2030 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2031 of 0600. This has been changed to use the value of the lockfile_mode
2032 option (which defaults to 0600).
2034 PH/08 Applied small patch from the Sieve maintainer.
2036 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2037 folder from quota calculations, a direct delivery into this folder messed
2038 up the contents of the maildirsize file. This was because the regex was
2039 used only to exclude .Trash (or whatever) when the size of the mailbox
2040 was calculated. There was no check that a delivery was happening into an
2041 excluded directory. This bug has been fixed by ignoring all quota
2042 processing for deliveries into excluded directories.
2044 PH/10 Added the maildirfolder_create_regex option to appendfile.
2050 PH/01 The code for finding all the local interface addresses on a FreeBSD
2051 system running IPv6 was broken. This may well have applied to all BSD
2052 systems, as well as to others that have similar system calls. The broken
2053 code found IPv4 interfaces correctly, but gave incorrect values for the
2054 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2055 that it would not match correctly against @[] and not recognize the IPv6
2058 PH/02 The ipliteral router was not recognizing addresses of the form user@
2059 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2061 PH/03 Added disable_ipv6.
2063 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2064 decoded form, because it is most often used to construct To: headers
2065 lines in autoreplies, and the decoded form may well be syntactically
2066 invalid. However, $reply_address has leading white space removed, and all
2067 newlines turned into spaces so that the autoreply transport does not
2070 PH/05 If group was specified without a user on a router, and no group or user
2071 was specified on a transport, the group from the router was ignored.
2073 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2074 for visible compile-time settings that can be used to change these
2075 numbers, for those that want even more. Backwards compatibility with old
2076 spool files has been maintained. However, going back to a previous Exim
2077 release will lost any variables that are in spool files.
2079 PH/07 Two small changes when running in the test harness: increase delay when
2080 passing a TCP/IP connection to a new process, in case the original
2081 process has to generate a bounce, and remove special handling of
2082 127.0.0.2 (sic), which is no longer necessary.
2084 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2085 be the same on different OS.
2087 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2090 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2091 whether --show-vars was specified or not
2093 JJ/02 exipick: Added support for new ACL variable spool format introduced
2096 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2097 syntactically invalid From: or Reply-to: line, and a filter used this to
2098 generate an autoreply, and therefore failed to obtain an address for the
2099 autoreply, Exim could try to deliver to a non-existent relative file
2100 name, causing unrelated and misleading errors. What now happens is that
2101 it logs this as a hard delivery error, but does not attempt to create a
2104 PH/11 The exinext utility has a -C option for testing purposes, but although
2105 the given file was scanned by exinext itself; it wasn't being passed on
2106 when Exim was called.
2108 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2109 an end-of-file indication when reading a command response.
2111 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2112 compiled. In many other places in Exim, IPv6 addresses are always
2113 recognized, so I have changed this. It also means that IPv4 domain
2114 literals of the form [IPV4:n.n.n.n] are now always recognized.
2116 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2117 used if the router is not running as root, for example, when verifying at
2118 ACL time, or when using -bh. The debugging output from this situation was
2119 non-existent - all you got was a failure to exec. I have made two
2122 (a) Failures to set uid/gid, the current directory, or a process leader
2123 in a subprocess such as that created by queryprogram now generate
2124 suitable debugging ouput when -d is set.
2126 (b) The queryprogram router detects when it is not running as root,
2127 outputs suitable debugging information if -d is set, and then runs
2128 the subprocess without attempting to change uid/gid.
2130 PH/15 Minor change to Makefile for building test_host (undocumented testing
2133 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2134 additional section of a DNS packet that returns MX or SRV records.
2135 Instead, it always explicitly searches for A/AAAA records. This avoids
2136 major problems that occur when a DNS server includes only records of one
2137 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2138 fixed another bug: if SRV records were looked up and the corresponding
2139 address records were *not* found in the additional section, the port
2140 values from the SRV records were lost.
2142 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2143 using the correct key (the original address) when searching the retry
2144 rules in order to find which one to use for generating the retry hint.
2146 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2147 adding the default one. Similarly, if it contains a Reply-To: header, the
2148 errors_reply_to option, if set, is not used.
2150 PH/19 When calculating a retry time, Exim used to measure the "time since
2151 failure" by looking at the "first failed" field in the retry record. Now
2152 it does not use this if it is later than than the arrival time of the
2153 message. Instead it uses the arrival time. This makes for better
2154 behaviour in cases where some deliveries succeed, thus re-setting the
2155 "first failed" field. An example is a quota failure for a huge message
2156 when small messages continue to be delivered. Without this change, the
2157 "time since failure" will always be short, possible causing more frequent
2158 delivery attempts for the huge message than are intended.
2159 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2161 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2162 $1, $2, $3) because the numerical variables can be reset during some
2163 expansion items (e.g. "match"), thereby losing the authentication data.
2165 PH/21 Make -bV show the size of off_t variables so that the test suite can
2166 decide whether to run tests for quotas > 2G.
2168 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2169 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2170 filecount value is greater than 2G or if a quota value is greater than 2G
2171 on a system where the size of off_t is not greater than 4, a panic error
2174 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2175 never match. The debug and -bh output now contains an explicit error
2176 message indicating a malformed IPv4 address or mask.
2178 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2179 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2180 PH/23 above applies.
2182 PH/25 Do not write to syslog when running in the test harness. The only
2183 occasion when this arises is a failure to open the main or panic logs
2184 (for which there is an explicit test).
2186 PH/26 Added the /no_tell option to "control=freeze".
2188 PH/27 If a host name lookup failed very early in a connection, for example, if
2189 the IP address matched host_lookup and the reverse lookup yielded a name
2190 that did not have a forward lookup, an error message of the form "no IP
2191 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2192 could be logged. Now it outputs the IP address instead of "NULL".
2194 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2195 allows the sender and the authenticated sender to be set when
2196 submitting a message from within Exim. Since child_open_exim() is
2197 documented for local_scan(), the new function should be too.
2199 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2200 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2201 results in an empty string is now treated as unset.
2203 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2205 PH/31 Added sender_verify_fail logging option.
2207 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2208 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2209 tidied the source and removed it altogether.
2211 PH/33 When a queue run was abandoned because the load average was too high, a
2212 log line was always written; now it is written only if the queue_run log
2213 selector is set. In addition, the log line for abandonment now contains
2214 information about the queue run such as the pid. This is always present
2215 in "start" and "stop" lines but was omitted from the "abandon" line.
2217 PH/34 Omit spaces between a header name and the colon in the error message that
2218 is given when verify = headers_syntax fails (if there are lots of them,
2219 the message gets confusing).
2221 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2222 names, as there are now some PTR records that contain slashes. This check
2223 is only to protect against broken name servers that fall over on strange
2224 characters, so the fact that it applies to all lookups doesn't matter.
2226 PH/36 Now that the new test suite is complete, we can remove some of the
2227 special code in Exim that was needed for the old test suite. For example,
2228 sorting DNS records because real resolvers return them in an arbitrary
2229 order. The new test suite's fake resolver always returns records in the
2232 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2233 bounces) except when queue_only is set, to avoid logging races between
2234 the different processes.
2236 PH/38 Panic-die if .include specifies a non-absolute path.
2238 PH/39 A tweak to the "H" retry rule from its user.
2240 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2241 a label. They prevented compilation on older perls.
2243 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2244 a warning to be raised on newish perls.
2246 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2247 on queue. Changes to match documented behaviour of showing count of
2248 messages matching specified criteria.
2250 PH/40 Changed the default ident timeout from 30s to 5s.
2252 PH/41 Added support for the use of login_cap features, on those BSD systems
2253 that have them, for controlling the resources used by pipe deliveries.
2255 PH/42 The content-scanning code uses fopen() to create files in which to put
2256 message data. Previously it was not paying any attention to the mode of
2257 the files. Exim runs with umask(0) because the rest of the code creates
2258 files with open(), and sets the required mode explicitly. Thus, these
2259 files were ending up world-writeable. This was not a big issue, because,
2260 being within the spool directory, they were not world-accessible. I have
2261 created a function called modefopen, which takes an additional mode
2262 argument. It sets umask(777), creates the file, chmods it to the required
2263 mode, then resets the umask. All the relevant calls to fopen() in the
2264 content scanning code have been changed to use this function.
2266 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2267 to 24 hours. This avoids potential overflow problems when processing G
2268 and H retry rules. I suspect nobody ever tinkers with this value.
2270 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2272 PH/45 When the plaintext authenticator is running as a client, the server's
2273 challenges are checked to ensure they are valid base64 strings. By
2274 default, the authentication attempt is cancelled if an invalid string is
2275 received. Setting client_ignore_invalid_base64 true ignores these errors.
2276 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2277 they are received. Thus, the responses can be made to depend on the
2278 challenges. If an invalid string is ignored, an empty string is placed in
2281 PH/46 Messages that are created by the autoreply transport now contains a
2282 References: header, in accordance with RFCs 2822 and 3834.
2284 PH/47 Added authenticated_sender_force to the smtp transport.
2286 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2288 PH/49 Installed latest patch from the Sieve maintainer.
2290 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2291 was also set, the appendfile transport was unnecessarily scanning a
2292 directory of message files (e.g. for maildir delivery) to find the count
2293 of files (along with the size), even though it did not need this
2294 information. It now does the scan only if it needs to find either the
2295 size of the count of files.
2297 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2299 PH/52 Two bugs concerned with error handling when the smtp transport is
2302 (i) Exim was not creating retry information for temporary errors given
2303 for individual recipients after the DATA command when the smtp transport
2304 was used in LMTP mode. This meant that they could be retried too
2305 frequently, and not timed out correctly.
2307 (ii) Exim was setting the flag that allows error details to be returned
2308 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2309 recipients that were returned after the DATA command.
2311 PH/53 This is related to PH/52, but is more general: for any failing address,
2312 when detailed error information was permitted to be returned to the
2313 sender, but the error was temporary, then after the final timeout, only
2314 "retry timeout exceeded" was returned. Now it returns the full error as
2315 well as "retry timeout exceeded".
2317 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2318 do this, and (what is worse) MTAs that accept it.
2320 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2321 will now be deprecated.
2323 PH/56 New os.c-cygwin from the Cygwin maintainer.
2325 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2326 formats (previously only available in exim formats via -bpr, -bpru,
2327 and -bpra. Now also available in native and exiqgrep formats)
2329 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2330 with very large, slow to parse queues
2332 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2334 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2336 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2337 responses to authentication challenges, though it was showing the
2338 challenges; (ii) I've removed the CR characters from the debug output for
2341 PH/58 Allow for the insertion of a newline as well as a space when a string
2342 is turned into more than one encoded-word during RFC 2047 encoding. The
2343 Sieve code now uses this.
2345 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2346 data_4xx, lost_connection, tls_required.
2348 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2349 message was being sent as an SMTP response.
2351 PH/61 Add -l and -k options to exicyclog.
2353 PH/62 When verifying, if an address was redirected to one new address, so that
2354 verification continued, and the new address failed or deferred after
2355 having set something in $address_data, the value of $address_data was not
2356 passed back to the ACL. This was different to the case when no
2357 redirection occurred. The value is now passed back in both cases.
2359 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2360 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2361 use login_cap.h, so on its own it isn't the distinguishing feature. The
2362 new name refers directly to the setclassresources() function.
2364 PH/65 Added configuration files for NetBSD3.
2366 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2368 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2369 is preferred over IPv4.
2371 PH/68 The bounce_return_message and bounce_return_body options were not being
2372 honoured for bounces generated during the reception of non-SMTP messages.
2373 In particular, this applied to messages rejected by the ACL. This bug has
2374 been fixed. However, if bounce_return_message is true and bounce_return_
2375 body is false, the headers that are returned for a non-SMTP message
2376 include only those that have been read before the error was detected.
2377 (In the case of an ACL rejection, they have all been read.)
2379 PH/69 The HTML version of the specification is now built in a directory called
2380 spec_html instead of spec.html, because the latter looks like a path with
2381 a MIME-type, and this confuses some software.
2383 PH/70 Catch two compiler warnings in sieve.c.
2385 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2386 function verify_get_ident() calls ip_connect() to connect a socket, but
2387 if the "connect()" function timed out, ip_connect() used to close the
2388 socket. However, verify_get_ident() also closes the socket later, and in
2389 between Exim writes to the log, which may get opened at this point. When
2390 the socket was closed in ip_connect(), the log could get the same file
2391 descriptor number as the socket. This naturally causes chaos. The fix is
2392 not to close the socket in ip_connect(); the socket should be closed by
2393 the function that creates it. There was only one place in the code where
2394 this was missing, in the iplookup router, which I don't think anybody now
2395 uses, but I've fixed it anyway.
2397 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2398 well as to direct DNS lookups. Otherwise the handling of names in host
2399 lists is inconsistent and therefore confusing.
2405 PH/01 Two changes to the default runtime configuration:
2407 (1) Move the checks for relay_from_hosts and authenticated clients from
2408 after to before the (commented out) DNS black list checks.
2410 (2) Add control=submission to the relay_from_hosts and authenticated
2411 clients checks, on the grounds that messages accepted by these
2412 statements are most likely to be submissions.
2414 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2416 (1) Generate an error if the third argument for the ${prvs expansion is
2419 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2422 (3) Reset the variables that are obtained from the first argument of
2423 ${prvscheck and used in the second argument before leaving the code,
2424 because their memory is reclaimed, so using them afterwards may do
2427 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2428 one (it's much easier than Tom thought :-).
2430 (5) Because of (4), we can now allow for the use of $prvscheck_result
2431 inside the third argument.
2433 PH/03 For some reason, the default setting of PATH when running a command from
2434 a pipe transport was just "/usr/bin". I have changed it to
2437 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2438 anything to be listed in the output from -bV.
2440 PH/05 When a filter generated an autoreply, the entire To: header line was
2441 quoted in the delivery log line, like this:
2443 => >A.N.Other <ano@some.domain> <original@ddress> ...
2445 This has been changed so that it extracts the operative address. There
2446 may be more than one such address. If so, they are comma-separated, like
2449 => >ano@some.domain,ona@other.domain <original@ddress> ...
2451 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2452 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2453 not being looked up in the rDNS to get a host name, Exim was showing the
2454 IP address twice in Received: lines, even though the IP addresses were
2455 identical. For example:
2457 Received: from [1.2.3.4] (helo=[1.2.3.4])
2459 However, if the real host name was known, it was omitting the HELO data
2460 if it matched the actual IP address. This has been tidied up so that it
2461 doesn't show the same IP address twice.
2463 PH/07 When both +timestamp and +memory debugging was on, the value given by
2464 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2465 called by the debug printing, thereby overwriting the timestamp buffer.
2466 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2468 PH/08 When the original message was included in an autoreply transport, it
2469 always said "this is a copy of the message, including all the headers",
2470 even if body_only or headers_only was set. It now gives an appropriate
2473 PH/09 Applied a patch from the Sieve maintainer which:
2475 o fixes some comments
2476 o adds the (disabled) notify extension core
2477 o adds some debug output for the result of if/elsif tests
2478 o points to the current vacation draft in the documentation
2479 and documents the missing references header update
2483 o fixes a bug in processing the envelope test (when testing
2484 multiple envelope elements, the last element determinted the
2487 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2488 Electronic Mail") by including:
2490 Auto-submitted: auto-generated
2492 in the messages that it generates (bounce messages and others, such as
2493 warnings). In the case of bounce messages for non-SMTP mesages, there was
2494 also a typo: it was using "Auto_submitted" (underscore instead of
2495 hyphen). Since every message generated by Exim is necessarily in response
2496 to another message, thes have all been changed to:
2498 Auto-Submitted: auto-replied
2500 in accordance with these statements in the RFC:
2502 The auto-replied keyword:
2504 - SHOULD be used on messages sent in direct response to another
2505 message by an automatic process,
2507 - MUST NOT be used on manually-generated messages,
2509 - MAY be used on Delivery Status Notifications (DSNs) and Message
2510 Disposition Notifications (MDNs),
2512 - MUST NOT be used on messages generated by automatic or periodic
2513 processes, except for messages which are automatic responses to
2516 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2517 to the default Received: header definition.
2519 PH/12 Added log selector acl_warn_skipped (default on).
2521 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2522 variables because (a) they are in the wrong storage pool and (b) even if
2523 they were copied, it wouldn't work properly because of the caching.
2525 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2526 checking when decoding. Apparently there are clients that generate
2527 overlong encoded strings. Why am I not surprised?
2529 PH/15 If the first argument of "${if match_address" was not empty, but did not
2530 contain an "@" character, Exim crashed. Now it writes a panic log message
2531 and treats the condition as false.
2533 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2535 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2536 extension "envelope-auth". The code is finished and in agreement with
2537 other implementations, but there is no documentation so far and in fact,
2538 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2539 not changing the active code.
2541 Print executed "if" and "elsif" statements when debugging is used. This
2542 helps a great deal to understand what a filter does.
2544 Document more things not specified clearly in RFC3028. I had all this
2545 sorted out, when out of a sudden new issues came to my mind. Oops."
2547 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2550 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2551 canonical form (as documented). However, after a host name lookup from
2552 the IP address, check_host() was doing a simple string comparison with
2553 addresses acquired from the DNS when checking that the found name did
2554 have the original IP as one of its addresses. Since any found IPv6
2555 addresses are likely to be in abbreviated form, the comparison could
2556 fail. Luckily, there already exists a function for doing the comparison
2557 by converting both addresses to binary, so now that is used instead of
2558 the text comparison.
2560 PH/20 There was another similar case to PH/19, when a complete host name was
2561 given in a host list; looking up its IP address could give an abbreviated
2562 form, whereas the current host's name might or might not be abbreviated.
2563 The same fix has been applied.
2569 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2570 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2573 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2574 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2576 PH/03 Typo: missing ".o" in src/pcre/Makefile.
2578 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2579 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2580 Also, for "Auto-Submitted", treat anything other than "no" as
2581 non-personal, in accordance with RFC 3834. (Previously it treated
2582 anything starting "auto-" as non-personal.)
2584 TF/01 The control=submission/name=... option had a problem with syntax
2585 errors if the name included a slash character. The /name= option
2586 now slurps the rest of the string, so it can include any characters
2587 but it must come last in the list of options (after /sender_retain
2590 PH/05 Some modifications to the interface to the fake nameserver for the new
2598 TK/01 Added the "success_on_redirect" address verification option. See
2599 NewStuff for rationale and an example.
2601 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2603 PH/02 Patch to exigrep to allow it to work on syslog lines.
2605 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2606 fread() to skip over the body file's header line, because in Cygwin the
2607 header line is locked and is inaccessible.
2609 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
2610 co-exist for some time) to make it clear that it is the Exim ID that is
2611 referenced, not the Message-ID: header line.
2613 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
2614 string_format() function, because snprintf() does not exist on all
2617 PH/06 The use of forbid_filter_existstest now also locks out the use of the
2618 ${stat: expansion item.
2620 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
2621 protocol synchronization error", to keep the pedants happy.
2623 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
2624 well as for IRIX systems, when gcc is being used. See the host.c source
2627 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
2629 PH/10 Named domain lists were not working if used in a queue_smtp_domains
2632 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
2633 transport and to the smtp transport in LMTP mode.
2635 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
2637 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
2638 run a filter in a subprocess. This could lead to confusion in subsequent
2639 lookups in the parent process. There should also be a search_tidyup() at
2640 the end of the subprocess.
2642 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
2643 only if the host matched helo_try_verify_hosts, which caused the
2644 verification to occur when the EHLO/HELO command was issued. The ACL just
2645 tested the remembered result. Now, if a previous verification attempt has
2646 not happened, "verify = helo" does it there and then.
2648 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
2650 TK/03 Fix log output including CR from clamd.
2652 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
2653 exist provoked a memory error which could cause a segfault.
2655 PH/15 Installed PCRE 6.2
2657 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2659 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2660 of the problem. Specifically, suggested +O2 rather than +O1 for the
2663 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2665 PH/20 If a delivery was routed to a non-standard port by means of an SRV
2666 record, the port was not correctly logged when the outgoing_port log
2667 selector was set (it logged the transort's default port).
2669 PH/21 Added support for host-specific ports to manualroute, queryprogram,
2670 fallback_hosts, and "hosts" in the smtp transport.
2672 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2673 host errors such as "Connection refused".
2675 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2676 authentication with radiusclient 0.4.9:
2678 - Error returned from rc_read_config was caught wrongly
2679 - Username/password not passed on to radius server due to wrong length.
2681 The presumption is that some radiusclient API changes for 4.51/PH/17
2682 were not taken care of correctly. The code is still untested by me (my
2683 Linux distribution still has 0.3.2 of radiusclient), but it was
2684 contributed by a Radius user.
2686 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2687 expanding the "port" option of the smtp transport.
2689 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2690 while reading a MIME header. Thanks to Tom Hughes for a patch.
2692 PH/24 Include config.h inside local_scan.h so that configuration settings are
2695 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2696 that in an ACL for RCPT (for example), you can examine exactly what was
2699 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2700 commands, but it was not correctly comparing the address with the actual
2701 client host address. Thus, it would show the EHLO address in Received:
2702 header lines when this was not necessary.
2704 PH/27 Added the % operator to ${eval:}.
2706 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2707 it should be ignoring failures (because with -C, for example, it has lost
2708 privilege). It wasn't ignoring creation failures other than "already
2711 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2714 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2715 because an input line was too long, either on its own, or by virtue of
2716 too many continuations, the temporary file was not being removed, and the
2717 return code was incorrect.
2719 PH/31 Missing "BOOL" in function definition in filtertest.c.
2721 PH/32 Applied Sieve patches from the maintainer.
2723 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2725 PH/33 Added "verify = not_blind".
2727 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2728 Local/Makefile (with some defaults set). These are used in built scripts
2729 such as exicyclog, but they have never been used in the exim_install
2730 script (though there are many overriding facilities there). I have
2731 arranged that the exim_install script now takes note of these two
2734 PH/35 Installed configuration files for Dragonfly.
2736 PH/36 When a locally submitted message by a trusted user did not contain a
2737 From: header, and the sender address was obtained from -f or from an SMTP
2738 MAIL command, and the trusted user did not use -F to supply a sender
2739 name, $originator_name was incorrectly used when constructing a From:
2740 header. Furthermore, $originator_name was used for submission mode
2741 messages from external hosts without From: headers in a similar way,
2742 which is clearly wrong.
2744 PH/37 Added control=suppress_local_fixups.
2746 PH/38 When log_selector = +received_sender was set, and the addition of the
2747 sender made the log line's construction buffer exactly full, or one byte
2748 less than full, an overflow happened when the terminating "\n" was
2751 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2752 when the result of a list match is failure because a DNS lookup failed.
2754 PH/40 RM_COMMAND is now used in the building process.
2756 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2757 the "build-* directories that it finds.
2759 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2760 address in a domain literal was a prefix of an interface address.
2762 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2763 when verifying a sender address, unless rewrite_headers is false.
2765 PH/44 Wrote a long comment about why errors_to addresses are verified as
2766 recipients, not senders.
2768 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2769 the ratelimit ACL was added.
2771 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2773 PH/46 Added extra information about PostgreSQL errors to the error string.
2775 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2776 avoiding the need to install special zones in a real server. This is
2777 backwards compatible; if it can't find the fake resolver, it drops back.
2778 Thus, both old and new test suites can be run.
2780 TF/02 Added util/ratelimit.pl
2782 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2783 clock is set back in time.
2785 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2786 Candler <B.Candler@pobox.com>.
2788 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2789 OK for addresses that are the result of redirections.
2791 PH/48 A number of further additions for the benefit of the new test suite,
2792 including a fake gethostbyname() that interfaces to the fake DNS resolver
2795 TF/06 The fix for widen_domains has also been applied to qualify_single and
2796 search_parents which are the other dnslookup options that can cause
2799 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2802 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2804 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2805 service" if it encountered an MX record with an empty target hostname.
2806 The message is now "an MX or SRV record indicated no SMTP service".
2808 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2809 if the DNS of the sending site is misconfigured. This is quite a
2810 common situation. This change restores the behaviour of treating a
2811 helo verification defer as a failure.
2813 PH/51 If self=fail was set on a router, the bounce message did not include the
2814 actual error message.
2820 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2822 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2823 command itself ended in error, the underlying message about the transport
2824 filter timeout was being overwritten with the pipe command error. Now the
2825 underlying error message should be appended to the second error message.
2827 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2829 PH/02 Reduce the amount of output that "make" produces by default. Full output
2830 can still be requested.
2832 PH/03 The warning log line about a condition test deferring for a "warn" verb
2833 was being output only once per connection, rather than after each
2834 occurrence (because it was using the same function as for successful
2835 "warn" verbs). This seems wrong, so I have changed it.
2837 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2838 it should not have, which might have caused a crash in the right
2839 circumstances, but probably never did.
2841 PH/04 Installed a modified version of Tony Finch's patch to make submission
2842 mode fix the return path as well as the Sender: header line, and to
2843 add a /name= option so that you can make the user's friendly name appear
2846 TF/03 Added the control = fakedefer ACL modifier.
2848 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2849 Mark Lowes for thorough testing.
2851 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2853 TK/03 Merged latest SRS patch from Miles Wilton.
2855 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2856 with the definition in sysexits.h (which is #included earlier).
2857 Fortunately, Exim does not actually use EX_OK. The code used to try to
2858 preserve the sysexits.h value, by assumimg that macro definitions were
2859 scanned for macro replacements. I have been disabused of this notion,
2860 so now the code just undefines EX_OK before #including unistd.h.
2862 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2863 in the smtp transport. When a block could not be written in a single
2864 write() function, the timeout was being re-applied to each part-write.
2865 This seems wrong - if the receiver was accepting one byte at a time it
2866 would take for ever. The timeout is now adjusted when this happens. It
2867 doesn't have to be particularly precise.
2869 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2870 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2872 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2873 without a domain if the check to <postmaster@domain> fails.
2875 SC/01 Eximstats: added -xls and the ability to specify output files
2876 (patch written by Frank Heydlauf).
2878 SC/02 Eximstats: use FileHandles for outputing results.
2880 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2882 SC/04 Eximstats: fixed display of large numbers with -nvr option
2884 SC/05 Eximstats: fixed merging of reports with empty tables.
2886 SC/06 Eximstats: added the -include_original_destination flag
2888 SC/07 Eximstats: removed tabs and trailing whitespace.
2890 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2892 TK/06 MBOX spool code: Add real "From " MBOX separator line
2893 so the .eml file is really in mbox format (even though
2894 most programs do not really care). Patch from Alex Miller.
2896 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2897 The latter is generated from $received_to and is only set if the
2898 message has one envelope recipient. SA can use these headers,
2899 obviously out-of-the-box. Patch from Alex Miller.
2901 PH/08 The ${def test on a variable was returning false if the variable's
2902 value was "0", contrary to what the specification has always said!
2903 The result should be true unless the variable is empty.
2905 PH/09 The syntax error of a character other than { following "${if
2906 def:variable_name" (after optional whitespace) was not being diagnosed.
2907 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2908 accidental colon was present, for example, could give incorrect results.
2910 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2911 result is used (not including appendfile, where other changes are about
2914 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2915 This involved changing a lot of size variables from int to off_t. It
2916 should work with maildirs and everything.
2918 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2919 spamd dying while we are connected to it.
2921 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2924 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2925 function for BDB 4.3.
2927 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2929 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2930 the sources for PCRE have been re-arranged and more files are now
2933 PH/15 The code I had for printing potentially long long variables in PH/11
2934 above was not the best (it lost precision). The length of off_t variables
2935 is now inspected at build time, and an appropriate printing format (%ld
2936 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2937 to be "long long int" or "long int". This is needed for the internal
2938 formatting function string_vformat().
2940 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2941 the configuration file to be ":syslog", then the script "guesses" where
2942 the logs files are, rather than using the compiled in default. In our
2943 case the guess is not the same as the compiled default, so the script
2944 suddenly stopped working when I started to use syslog. The patch checks
2945 to see if log_file_path is "". If so, it attempts to read it from exim
2946 with no configuration file to get the compiled in version, before it
2947 falls back to the previous guessing code."
2949 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2950 implementing BATV in an Exim configuration. See NewStuff for the gory
2953 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2954 Makefile that are specific to HP-UX.
2956 PH/18 If the "use_postmaster" option was set for a recipient callout together
2957 with the "random" option, the postmaster address was used as the MAIL
2958 FROM address for the random test, but not for the subsequent recipient
2959 test. It is now used for both.
2961 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2962 patch removes a few documentation additions to RFC 3028, because the
2963 latest draft now contains them. It adds the new en;ascii-case comparator
2964 and a new error check for 8bit text in MIME parts. Comparator and
2965 require names are now matched exactly. I enabled the subaddress
2966 extension, but it is not well tested yet (read: it works for me)."
2968 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2969 rework some of the code of TK/09 above to avoid the hardwired use of
2970 "%lld" and "long long". Replaced the call to snprintf() with a call to
2973 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2974 records point to non-existent hosts", "retry timeout exceeded", and
2975 "retry time not reached for any host after a long failure period".
2977 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2978 experimental DomainKeys support:
2980 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2981 (2) On an error such as an illegally used "control", the wrong name for
2982 the control was given.
2984 These problems did NOT occur unless DomainKeys support was compiled.
2986 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2988 PH/24 Added ${if match_ip condition.
2990 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2991 databases so that it will be absolutely obvious if a crash occurs in the
2992 DB library. This is a regular occurrence (often caused by mis-matched
2995 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2996 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2997 picked up on a user's system that detects such things. There doesn't seem
2998 to be a gcc warning option for this - only an attribute that has to be
2999 put on the function's prototype. It seems that in Fedora Core 4 they have
3000 set this on a number of new functions. No doubt there will be more in due
3003 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3004 specify a transport. However, if an address that was verified by such a
3005 router was the subject of a callout, Exim crashed because it tried to
3006 read the rcpt_include_affixes from the non-existent transport. Now it
3007 just assumes that the setting of that option is false. This bug was
3008 introduced by 4.51/PH/31.
3010 PH/28 Changed -d+all to exclude +memory, because that information is very
3011 rarely of interest, but it makes the output a lot bigger. People tend to
3012 do -d+all out of habit.
3014 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3015 code in os-type was giving problems when libc.so lives in lib64, like on
3018 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3019 aren't the modern standard, and it seems that some systems' include files
3020 don't always have them. Exim was already checking for some of the newer
3021 ones like T_AAAA, and defining it itself. I've added checks for all the
3022 record types that Exim uses.
3024 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3025 not automatically generating a new one, as it is supposed to. This
3026 prevented TLS from working. If the file did exist, but contained invalid
3027 data, a new version was generated, as expected. It was only the case of a
3028 non-existent file that was broken.
3030 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3031 with a change in libdomainkeys > 0.64.
3033 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3034 from DNS. If the selector record carries the flag, it now has
3035 precedence over the domain-wide flag.
3037 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3039 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3040 the use of a port name that isn't defined in /etc/services) occurred, the
3041 message was deferred as in a normal delivery, and thus remained on the
3042 spool, instead of being failed because of the mua_wrapper setting. This
3043 is now fixed, and I tidied up some of the mua_wrapper messages at the
3046 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3047 the messages in a hash of arrays rather than using individual hashes.
3048 This is a bit cleaner and results in dramatic memory savings, albeit
3049 at a slight CPU cost.
3051 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3052 as requested by Marc Sherman.
3054 SC/10 Eximstats: added histograms for user specified patterns as requested
3057 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3059 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3060 fopen() in the content-scanning modules that did not already have it.
3066 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3067 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3069 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3071 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3073 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3074 to test Sieve filters that use "vacation".
3076 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3077 that changes the way the GnuTLS parameters are stored in the cache file.
3078 The new format can be generated externally. For backward compatibility,
3079 if the data in the cache doesn't make sense, Exim assumes it has read an
3080 old-format file, and it generates new data and writes a new file. This
3081 means that you can't go back to an older release without removing the
3084 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3085 work if there are any delivery delays because "one_time" forces the
3086 parent to be marked "delivered", so its unseen clone is never tried
3087 again. For this reason, Exim now forbids the simultaneous setting of
3090 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3091 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3092 entry if you want to know the details. The fix, however, neglected to
3093 consider the case when local delivery batching is involved. The test for
3094 "previously delivered" was not happening when checking to see if an
3095 address could be batched with a previous (undelivered) one; under
3096 certain circumstances this could lead to multiple deliveries to the same
3099 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3100 in its include files, and this causes problems building Exim.
3102 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3103 header_syntax) but Exim was just ignoring anything given after a slash.
3104 In particular, this caused confusion with an attempt to use "verify =
3105 reverse_host_lookup/defer_ok". An error is now given when options are
3106 supplied for verify items that do not have them. (Maybe reverse_host_
3107 lookup should have a defer_ok option, but that's a different point.)
3109 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3110 defined by RFC 821) to 2048, because there were problems with some AUTH
3111 commands, and RFC 1869 says the size should be increased for extended
3112 SMTP commands that take arguments.
3114 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3117 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3118 "unknown" error; now it says that the functionality isn't in the binary.
3120 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3121 an address' error message when a string expansion fails (syntax or
3122 whatever). Otherwise the password may appear in the log. Following change
3123 PH/42 below, there is no longer a chance of it appearing in a bounce
3126 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3128 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3129 of its addresses was ever tried. (Bugzilla bug #2.)
3131 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3132 the result incorrectly in the debug output. (It correctly added a newline
3133 to what was transported.)
3135 TF/01 Added $received_time.
3137 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3138 commented out examples of how to interface to a virus scanner and to
3139 SpamAssassin. Also added commented examples of av_scanner and
3140 spamd_address settings.
3142 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3143 and controls are allowed in which ACLs. There were a couple of minor
3144 errors. Some of the entries in the conditions table (which is a table of
3145 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3146 negation of where the condition IS allowed.
3148 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3150 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3151 header file does not have a version number, so I've had to invent a new
3152 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3153 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3154 radiusclient), but it was contributed by a Radius user.
3156 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3157 files or directories, for OpenSSL.
3159 PH/19 When an Exim process that is running as root has to create an Exim log
3160 file, it does so in a subprocess that runs as exim:exim so as to get the
3161 ownership right at creation (otherwise, other Exim processes might see
3162 the file with the wrong ownership). There was no test for failure of this
3163 fork() call, which would lead to the process getting stuck as it waited
3164 for a non-existent subprocess. Forks do occasionally fail when resources
3165 run out. I reviewed all the other calls to fork(); they all seem to check
3168 PH/20 When checking for unexpected SMTP input at connect time (before writing
3169 the banner), Exim was not dealing correctly with a non-positive return
3170 from the read() function. If the client had disconnected by this time,
3171 the result was a log entry for a synchronization error with an empty
3172 string after "input=" when read() returned zero. If read() returned -1
3173 (an event I could not check), uninitialized data bytes were printed.
3174 There were reports of junk text (parts of files, etc) appearing after
3177 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3179 PH/22 Added support for macro redefinition, and (re)definition in between
3180 driver and ACL definitions.
3182 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3183 forgetting to use the resulting value; it was using the unexpanded value.
3185 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3186 hadn't been configured. The fix is from Juergen Kreileder, who
3187 understands it better than I do:
3189 "Here's what I see happening with three configured cyrus_sasl
3190 authenticators configured (plain, login, cram-md5):
3192 On startup auth_cyrus_sasl_init() gets called for each of these.
3193 This means three calls to sasl_listmech() without a specified mech_list.
3194 => SASL tests which mechs of all available mechs actually work
3195 => three warnings about OTP not working
3196 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3198 With the patch, sasl_listmech() also gets called three times. But now
3199 SASL's mech_list option is set to the server_mech specified in the the
3200 authenticator. Or in other words, the answer from sasl_listmech()
3201 gets limited to just the mech you're testing for (which is different
3203 => the return list contains just 'plain' or 'login', 'cram-md5' or
3204 nothing depending on the value of ob->server_mech.
3206 I've just tested the patch: Authentication still works fine,
3207 unavailable mechs specified in the exim configuration are still
3208 caught, and the auth.log warnings about OTP are gone."
3210 PH/25 When debugging is enabled, the contents of the command line are added
3211 to the debugging output, even when log_selector=+arguments is not
3214 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3215 answer is "GNU", and only if the return is "GNU/something" is the answer
3218 PH/27 $acl_verify_message is now set immediately after the failure of a
3219 verification in an ACL, and so is available in subsequent modifiers. In
3220 particular, the message can be preserved by coding like this:
3222 warn !verify = sender
3223 set acl_m0 = $acl_verify_message
3225 Previously, $acl_verify_message was set only while expanding "message"
3226 and "log_message" when a very denied access.
3228 PH/28 Modified OS/os.c-Linux with
3230 -#ifndef OS_LOAD_AVERAGE
3231 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3233 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3234 nomenclature these days.)
3236 PH/29 Installed patch from the Sieve maintainer that adds the options
3237 sieve_useraddress and sieve_subaddress to the redirect router.
3239 PH/30 In these circumstances:
3240 . Two addresses routed to the same list of hosts;
3241 . First host does not offer TLS;
3242 . First host accepts first address;
3243 . First host gives temporary error to second address;
3244 . Second host offers TLS and a TLS session is established;
3245 . Second host accepts second address.
3246 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3247 and peerdn, if requested) that were in fact used only for the second
3250 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3251 attention to any local part prefix or suffix that was matched by the
3252 router that accepted the address. It now behaves in the same way as it
3253 does for delivery: the affixes are removed from the local part unless
3254 rcpt_include_affixes is set on the transport.
3256 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3257 timeout during the DATA phase of an incoming message.
3259 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3260 applied a patch sanctioned by the Sieve maintainer.
3262 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3263 the uid or gid is negative. A case of a negative gid caused this to be
3264 noticed. The fix allows for either to be negative.
3266 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3267 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3270 PH/36 The change PH/12 above was broken. Fixed it.
3272 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3273 the grounds that routing the same address twice would always produce the
3274 same answer. This might have been true once, but it is certainly no
3275 longer true now. Routing a child address may depend on the previous
3276 routing that produced that child. Some complicated redirection strategies
3277 went wrong when messages had multiple recipients, and made Exim's
3278 behaviour dependent on the order in which the addresses were given.
3280 I have moved the duplicate checking until after the routing is complete.
3281 Exim scans the addresses that are assigned to local and remote
3282 transports, and removes any duplicates. This means that more work will be
3283 done, as duplicates will always all be routed, but duplicates are
3284 presumably rare, so I don't expect this is of any significance.
3286 For deliveries to pipes, files, and autoreplies, the duplicate checking
3287 still happens during the routing process, since they are not going to be
3290 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3291 It corrects a timeout issue with spamd. This is Ian's comment: "The
3292 background is that sometimes spamd either never reads data from a
3293 connection it has accepted, or it never writes response data. The exiscan
3294 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3295 blindly assumes that writes won't block so it may never time out."
3297 PH/39 Allow G after quota size as well as K and M.
3299 PH/40 The value set for $authenticated_id in an authenticator may not contain
3300 binary zeroes or newlines because the value is written to log lines and
3301 to spool files. There was no check on this. Now the value is run through
3302 the string_printing() function so that such characters are converted to
3303 printable escape sequences.
3305 PH/41 $message_linecount is a new variable that contains the total number of
3306 lines in the message. Compare $body_linecount, which is the count for the
3309 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3310 bounce and delay warning messages, except in certain special cases, which
3313 (a) An SMTP error message from a remote host;
3314 (b) A message specified in a :fail: redirection;
3315 (c) A message specified in a "fail" command in a system filter;
3316 (d) A message specified in a FAIL return from the queryprogram router;
3317 (e) A message specified by the cannot_route_message router option.
3319 In these cases only, Exim does include the error details in bounce and
3320 warning messages. There are also a few cases where bland messages such
3321 as "unrouteable address" or "local delivery error" are given.
3323 PH/43 $value is now also set for the "else" part of a ${run expansion.
3325 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3326 being worked on, but at least Exim now implements the latest version to
3329 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3330 process to complete was treated as a delivery failure, a timeout while
3331 writing the message to the pipe was logged, but erroneously treated as a
3332 successful delivery. Such timeouts include transport filter timeouts. For
3333 consistency with the overall process timeout, these timeouts are now
3334 treated as errors, giving rise to delivery failures by default. However,
3335 there is now a new Boolean option for the pipe transport called
3336 timeout_defer, which, if set TRUE, converts the failures into defers for
3337 both kinds of timeout. A transport filter timeout is now identified in
3340 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3341 systems where "make" and "gmake" are different, calling "gmake" at top
3342 level broke things. I've arranged for the value of $(MAKE) to be passed
3343 from the Makefile to this script so that it can call the same version of
3347 A note about Exim versions 4.44 and 4.50
3348 ----------------------------------------
3350 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3351 changes of various kinds. As a consequence, a big documentation update was
3352 needed. This delayed the release for rather longer than seemed good, especially
3353 in the light of a couple of (minor) security issues. Therefore, the changes
3354 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3355 release. So 4.44 and 4.50 are in effect two different branches that both start
3358 I have left the 4.50 change log unchanged; it contains all the changes since
3359 4.43. The change log for 4.44 is below; many of its items are identical to
3360 those for 4.50. This seems to be the most sensible way to preserve the
3361 historical information.
3367 1. Minor wording change to the doc/README.SIEVE file.
3369 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3370 computation of the current number of files was incorrect.
3372 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3373 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3376 4. Give more explanation in the error message when the command for a transport
3377 filter fails to execute.
3379 5. There are several places where Exim runs a non-Exim command in a
3380 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3381 was being done only for the command run by the queryprogram router. It is
3382 now done for all such subprocesses. The other cases are: ${run, transport
3383 filters, and the commands run by the lmtp and pipe transports.
3385 6. Added CONFIGURE_GROUP build-time option.
3387 7. Some older OS have a limit of 256 on the maximum number of file
3388 descriptors. Exim was using setrlimit() to set 1000 as a large value
3389 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3390 systems. I've change it so that if it can't get 1000, it tries for 256.
3392 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3393 was an oversight, and furthermore, ever since the addition of extra
3394 controls (e.g. 4.43/32), the checks on when to allow different forms of
3395 "control" were broken. There should now be diagnostics for all cases when a
3396 control that does not make sense is encountered.
3398 9. Added the /retain_sender option to "control=submission".
3400 10. $recipients is now available in the predata ACL (oversight).
3402 11. Tidy the search cache before the fork to do a delivery from a message
3403 received from the command line. Otherwise the child will trigger a lookup
3404 failure and thereby defer the delivery if it tries to use (for example) a
3405 cached ldap connection that the parent has called unbind on.
3407 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3408 of $address_data from the recipient verification was clobbered by the
3409 sender verification.
3411 13. The value of address_data from a sender verification is now available in
3412 $sender_address_data in subsequent conditions in the ACL statement.
3414 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3416 15. Added a new option "connect=<time>" to callout options, to set a different
3419 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3420 was its contents. (It was OK if the option was not defined at all.)
3422 17. A "Completed" log line is now written for messages that are removed from
3423 the spool by the -Mrm option.
3425 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3426 information about exactly what failed.
3428 19. Added -dd to debug only the daemon process.
3430 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3431 handles timeouts, both on the server side and network timeouts. Renamed the
3432 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3434 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3435 It is now set to "smtps".
3437 22. $host_address is now set to the target address during the checking of
3438 ignore_target_hosts.
3440 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3441 being passed; this would have caused $sender_host_name to have been used if
3442 matching the list had actually called for a host name (not very likely,
3443 since this list is usually IP addresses). A host name is now passed as
3446 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3447 code that specifies a non-restarting handler (typically sigaction() in
3448 modern systems) in an attempt to fix a rare and obscure crash bug.
3450 25. Narrowed the window for a race in the daemon that could cause it to ignore
3451 SIGCHLD signals. This is not a major problem, because they are used only to
3452 wake it up if nothing else does.
3454 26. A malformed maildirsize file could cause Exim to calculate negative values
3455 for the mailbox size or file count. Odd effects could occur as a result.
3456 The maildirsize information is now recalculated if the size or filecount
3459 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3460 support for a long time. Removed HAVE_SYS_VFS_H.
3462 28. Installed the latest version of exipick from John Jetmore.
3464 29. In an address list, if the pattern was not a regular expression, an empty
3465 subject address (from a bounce message) matched only if the pattern was an
3466 empty string. Non-empty patterns were not even tested. This was the wrong
3467 because it is perfectly reasonable to use an empty address as part of a
3468 database query. An empty address is now tested by patterns that are
3469 lookups. However, all the other forms of pattern expect the subject to
3470 contain a local part and a domain, and therefore, for them, an empty
3471 address still always fails if the pattern is not itself empty.
3473 30. Exim went into a mad DNS loop when attempting to do a callout where the
3474 host was specified on an smtp transport, and looking it up yielded more
3475 than one IP address.
3477 31. Re-factored the code for checking spool and log partition space into a
3478 function that finds that data and another that does the check. The former
3479 is then used to implement four new variables: $spool_space, $log_space,
3480 $spool_inodes, and $log_inodes.
3482 32. The RFC2047 encoding function was originally intended for short strings
3483 such as real names; it was not keeping to the 75-character limit for
3484 encoded words that the RFC imposes. It now respects the limit, and
3485 generates multiple encoded words if necessary. To be on the safe side, I
3486 have increased the buffer size for the ${rfc2047: expansion operator from
3489 33. It is now permitted to omit both strings after an "if" condition; if the
3490 condition is true, the result is "true". As before, when the second string
3491 is omitted, a false condition yields an empty string. This makes it less
3492 cumbersome to write custom ACL and router conditions.
3494 34. Failure to deliver a bounce message always caused it to be frozen, even if
3495 there was an errors_to setting on the router. The errors_to setting is now
3498 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3499 canonical form (fully expanded) before being placed in
3500 $sender_host_address.
3502 36. The table in the code that translates DNS record types into text (T_A to
3503 "A" for instance) was missing entries for NS and CNAME. It is just possible
3504 that this could have caused confusion if both these types were looked up
3505 for the same domain, because the text type is used as part of Exim's
3506 per-process caching. But the chance of anyone hitting this buglet seems
3509 37. The dnsdb lookup has been extended in a number of ways.
3511 (1) There is a new type, "zns", which walks up the domain tree until it
3512 finds some nameserver records. It should be used with care.
3514 (2) There is a new type, "mxh", which is like "mx" except that it returns
3515 just the host names, not the priorities.
3517 (3) It is now possible to give a list of domains (or IP addresses) to be
3518 looked up. The behaviour when one of the lookups defers can be
3519 controlled by a keyword.
3521 (4) It is now possible to specify the separator character for use when
3522 multiple records are returned.
3524 38. The dnslists ACL condition has been extended: it is now possible to supply
3525 a list of IP addresses and/or domains to be looked up in a particular DNS
3528 39. Added log_selector=+queue_time_overall.
3530 40. When running the queue in the test harness, wait just a tad after forking a
3531 delivery process, to get repeatability of debugging output.
3533 41. Include certificate and key file names in error message when GnuTLS fails
3534 to set them up, because the GnuTLS error message doesn't include the name
3535 of the failing file when there is a problem reading it.
3537 42. Allow both -bf and -bF in the same test run.
3539 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3541 44. The "Exiscan patch" is now merged into the mainline Exim source.
3543 45. Sometimes the final signoff response after QUIT could fail to get
3544 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3545 < 0 before doing a fflush(). This bug looks as though it goes back to the
3546 introduction of TLS in release 3.20, but "sometimes" must have been rare
3547 because the tests only now provoked it.
3549 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3550 (this can affect the format of dates).
3552 47. exim_tidydb, when checking for the continued existence of a message for
3553 which it has found a message-specific retry record, was not finding
3554 messages that were in split spool directories. Consequently, it was
3555 deleting retry records that should have stayed in existence.
3557 48. Steve fixed some bugs in eximstats.
3559 49. The SPA authentication driver was not abandoning authentication and moving
3560 on to the next authenticator when an expansion was forced to fail,
3561 contradicting the general specification for all authenticators. Instead it
3562 was generating a temporary error. It now behaves as specified.
3564 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3565 (the order specifies the preference for clients). The order is now AES256,
3566 AES128, 3DES, ARCFOUR128.
3568 51. Small patch to Sieve code - explicitly set From: when generating an
3571 52. Exim crashed if a remote delivery caused a very long error message to be
3572 recorded - for instance if somebody sent an entire SpamAssassin report back
3573 as a large number of 550 error lines. This bug was coincidentally fixed by
3574 increasing the size of one of Exim's internal buffers (big_buffer) that
3575 happened as part of the Exiscan merge. However, to be on the safe side, I
3576 have made the code more robust (and fixed the comments that describe what
3579 53. Now that there can be additional text after "Completed" in log lines (if
3580 the queue_time_overall log selector is set), a one-byte patch to exigrep
3581 was needed to allow it to recognize "Completed" as not the last thing in
3584 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3585 patch that reportedly fixes this has been added. I am not expert enough to
3586 create a test for it. This is what the patch creator wrote:
3588 "I found a little strange behaviour of ldap code when working with
3589 Windows 2003 AD Domain, where users was placed in more than one
3590 Organization Units. When I tried to give exim partial DN, the exit code
3591 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
3592 But simultaneously result of request was absolutely normal ldap result,
3593 so I produce this patch..."
3595 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
3596 REFERENCE, so I have modified the code to exclude the patch when that macro
3599 55. Some experimental protocols are using DNS PTR records for new purposes. The
3600 keys for these records are domain names, not reversed IP addresses. The
3601 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3602 leaves it alone. Component reversal etc. now happens only for IP addresses.
3605 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
3607 57. Double the size of the debug message buffer (to 2048) so that more of very
3608 long debug lines gets shown.
3610 58. The exicyclog utility now does better if the number of log files to keep
3611 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
3613 59. Two changes related to the smtp_active_hostname option:
3615 (1) $smtp_active_hostname is now available as a variable.
3616 (2) The default for smtp_banner uses $smtp_active_hostname instead
3617 of $primary_hostname.
3619 60. The host_aton() function is supposed to be passed a string that is known
3620 to be a valid IP address. However, in the case of IPv6 addresses, it was
3621 not checking this. This is a hostage to fortune. Exim now panics and dies
3622 if the condition is not met. A case was found where this could be provoked
3623 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3624 components; fortuitously, this particular loophole had already been fixed
3625 by change 4.50/55 above.
3627 If there are any other similar loopholes, the new check in host_aton()
3628 itself should stop them being exploited. The report I received stated that
3629 data on the command line could provoke the exploit when Exim was running as
3630 exim, but did not say which command line option was involved. All I could
3631 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3632 running as the user.
3635 61. There was a buffer overflow vulnerability in the SPA authentication code
3636 (which came originally from the Samba project). I have added a test to the
3637 spa_base64_to_bits() function which I hope fixes it.
3640 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
3641 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
3643 63. The daemon start-up calls getloadavg() while still root for those OS that
3644 need the first call to be done as root, but it missed one case: when
3645 deliver_queue_load_max is set with deliver_drop_privilege. This is
3646 necessary for the benefit of the queue runner, because there is no re-exec
3647 when deliver_drop_privilege is set.
3649 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
3650 This has been fixed.
3652 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3653 was in use, was not putting the data itself into the right store pool;
3654 consequently, it could be overwritten for a subsequent message in the same
3655 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3658 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
3660 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
3661 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
3662 the function were treating the return as a boolean value, which happened to
3663 work because 0=false and not-0=true, but is not correct code.
3665 68. The host_aton() function was not handling scoped IPv6 addresses (those
3666 with, for example, "%eth0" on the end) correctly.
3668 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
3669 negated items (that is, ~something) in unsigned ints. Some compilers
3670 apparently mutter when there is no cast.
3672 70. If an address verification called from an ACL failed, and did not produce a
3673 user-specific message (i.e. there was only a "system" message), nothing was
3674 put in $acl_verify_message. In this situation, it now puts the system
3677 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3678 session; change 4.31/43 added the unwanted input to the log line - except
3679 that it did not do this in the start of session case. It now does.
3681 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3682 This is wrong and can cause the other end to generate a synchronization
3683 error if it is another Exim or anything else that does the synchronization
3684 check. A QUIT command is no longer sent after a timeout.
3686 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3687 during host lookups.
3689 74. The defer_ok option of callout verification was not working if it was used
3690 when verifying addresses in header lines, that is, for this case:
3692 verify = header_sender/callout=defer_ok
3694 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3695 those file descriptors could be used for SMTP connections. If anything
3696 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3697 could be sent to the SMTP client, causing chaos. The daemon now opens
3698 stdin, stdout, and stderr to /dev/null when it puts itself into the
3701 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3702 log by default. The user can override this with suitable Perl magic.
3704 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3705 the log message when discard triggers, was not working for the DATA ACL or
3706 for the non-SMTP ACL.
3708 78. Error message wording change in sieve.c.
3710 79. If smtp_accept_max_per_host was set, the number of connections could be
3711 restricted to fewer than expected, because the daemon was trying to set up
3712 a new connection before checking whether the processes handling previous
3713 connections had finished. The check for completed processes is now done
3714 earlier. On busy systems, this bug wouldn't be noticed because something
3715 else would have woken the daemon, and it would have reaped the completed
3718 80. If a message was submitted locally by a user whose login name contained one
3719 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3720 It caused a spool format error. I have fixed the spool reading code. A
3721 related problem was that the "from" clause in the Received: line became
3722 illegal because of the space(s). It is now covered by ${quote_local_part.
3724 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3727 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3729 83. Patch from Sieve maintainer to fix unterminated string problem in
3730 "vacation" handling.
3732 84. Some minor changes to the Linux configuration files to help with other
3733 OS variants using glibc.
3735 85. One more patch for Sieve to update vacation handling to latest spec.
3738 ----------------------------------------------------
3739 See the note above about the 4.44 and 4.50 releases.
3740 ----------------------------------------------------
3746 1. Change 4.43/35 introduced a bug that caused file counts to be
3747 incorrectly computed when quota_filecount was set in an appendfile
3750 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3751 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3754 3. Give more explanation in the error message when the command for a transport
3755 filter fails to execute.
3757 4. There are several places where Exim runs a non-Exim command in a
3758 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3759 was being done only for the command run by the queryprogram router. It is
3760 now done for all such subprocesses. The other cases are: ${run, transport
3761 filters, and the commands run by the lmtp and pipe transports.
3763 5. Some older OS have a limit of 256 on the maximum number of file
3764 descriptors. Exim was using setrlimit() to set 1000 as a large value
3765 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3766 systems. I've change it so that if it can't get 1000, it tries for 256.
3768 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3769 was an oversight, and furthermore, ever since the addition of extra
3770 controls (e.g. 4.43/32), the checks on when to allow different forms of
3771 "control" were broken. There should now be diagnostics for all cases when a
3772 control that does not make sense is encountered.
3774 7. $recipients is now available in the predata ACL (oversight).
3776 8. Tidy the search cache before the fork to do a delivery from a message
3777 received from the command line. Otherwise the child will trigger a lookup
3778 failure and thereby defer the delivery if it tries to use (for example) a
3779 cached ldap connection that the parent has called unbind on.
3781 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3782 of $address_data from the recipient verification was clobbered by the
3783 sender verification.
3785 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3786 was its contents. (It was OK if the option was not defined at all.)
3788 11. A "Completed" log line is now written for messages that are removed from
3789 the spool by the -Mrm option.
3791 12. $host_address is now set to the target address during the checking of
3792 ignore_target_hosts.
3794 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3795 being passed; this would have caused $sender_host_name to have been used if
3796 matching the list had actually called for a host name (not very likely,
3797 since this list is usually IP addresses). A host name is now passed as
3800 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3801 code that specifies a non-restarting handler (typically sigaction() in
3802 modern systems) in an attempt to fix a rare and obscure crash bug.
3804 15. Narrowed the window for a race in the daemon that could cause it to ignore
3805 SIGCHLD signals. This is not a major problem, because they are used only to
3806 wake it up if nothing else does.
3808 16. A malformed maildirsize file could cause Exim to calculate negative values
3809 for the mailbox size or file count. Odd effects could occur as a result.
3810 The maildirsize information is now recalculated if the size or filecount
3813 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3814 support for a long time. Removed HAVE_SYS_VFS_H.
3816 18. Updated exipick to current release from John Jetmore.
3818 19. Allow an empty sender to be matched against a lookup in an address list.
3819 Previously the only cases considered were a regular expression, or an
3822 20. Exim went into a mad DNS lookup loop when doing a callout where the
3823 host was specified on the transport, if the DNS lookup yielded more than
3826 21. The RFC2047 encoding function was originally intended for short strings
3827 such as real names; it was not keeping to the 75-character limit for
3828 encoded words that the RFC imposes. It now respects the limit, and
3829 generates multiple encoded words if necessary. To be on the safe side, I
3830 have increased the buffer size for the ${rfc2047: expansion operator from
3833 22. Failure to deliver a bounce message always caused it to be frozen, even if
3834 there was an errors_to setting on the router. The errors_to setting is now
3837 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3838 canonical form (fully expanded) before being placed in
3839 $sender_host_address.
3841 24. Updated eximstats to version 1.33
3843 25. Include certificate and key file names in error message when GnuTLS fails
3844 to set them up, because the GnuTLS error message doesn't include the name
3845 of the failing file when there is a problem reading it.
3847 26. Expand error message when OpenSSL has problems setting up cert/key files.
3850 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3851 (this can affect the format of dates).
3853 28. exim_tidydb, when checking for the continued existence of a message for
3854 which it has found a message-specific retry record, was not finding
3855 messages that were in split spool directories. Consequently, it was
3856 deleting retry records that should have stayed in existence.
3858 29. eximstats updated to version 1.35
3859 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3860 1.35 - bugfix such that pie charts by volume are generated correctly
3862 30. The SPA authentication driver was not abandoning authentication and moving
3863 on to the next authenticator when an expansion was forced to fail,
3864 contradicting the general specification for all authenticators. Instead it
3865 was generating a temporary error. It now behaves as specified.
3867 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3868 (the order specifies the preference for clients). The order is now AES256,
3869 AES128, 3DES, ARCFOUR128.
3871 31. Small patch to Sieve code - explicitly set From: when generating an
3874 32. Exim crashed if a remote delivery caused a very long error message to be
3875 recorded - for instance if somebody sent an entire SpamAssassin report back
3876 as a large number of 550 error lines. This bug was coincidentally fixed by
3877 increasing the size of one of Exim's internal buffers (big_buffer) that
3878 happened as part of the Exiscan merge. However, to be on the safe side, I
3879 have made the code more robust (and fixed the comments that describe what
3882 33. Some experimental protocols are using DNS PTR records for new purposes. The
3883 keys for these records are domain names, not reversed IP addresses. The
3884 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3885 leaves it alone. Component reversal etc. now happens only for IP addresses.
3888 34. The host_aton() function is supposed to be passed a string that is known
3889 to be a valid IP address. However, in the case of IPv6 addresses, it was
3890 not checking this. This is a hostage to fortune. Exim now panics and dies
3891 if the condition is not met. A case was found where this could be provoked
3892 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3893 components; fortuitously, this particular loophole had already been fixed
3894 by change 4.50/55 or 4.44/33 above.
3896 If there are any other similar loopholes, the new check in host_aton()
3897 itself should stop them being exploited. The report I received stated that
3898 data on the command line could provoke the exploit when Exim was running as
3899 exim, but did not say which command line option was involved. All I could
3900 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3901 running as the user.
3904 35. There was a buffer overflow vulnerability in the SPA authentication code
3905 (which came originally from the Samba project). I have added a test to the
3906 spa_base64_to_bits() function which I hope fixes it.
3909 36. The daemon start-up calls getloadavg() while still root for those OS that
3910 need the first call to be done as root, but it missed one case: when
3911 deliver_queue_load_max is set with deliver_drop_privilege. This is
3912 necessary for the benefit of the queue runner, because there is no re-exec
3913 when deliver_drop_privilege is set.
3915 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3916 was in use, was not putting the data itself into the right store pool;
3917 consequently, it could be overwritten for a subsequent message in the same
3918 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3921 38. Sometimes the final signoff response after QUIT could fail to get
3922 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3923 < 0 before doing a fflush(). This bug looks as though it goes back to the
3924 introduction of TLS in release 3.20, but "sometimes" must have been rare
3925 because the tests only now provoked it.
3931 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3932 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3933 Now it returns an integer. A number of calls were still expecting a T/F
3934 return. Fortuitously, in all cases, the tests worked in OK situations,
3935 which is the norm. However, things would have gone wrong on any write
3936 failures on the smtp file descriptor. This function is used when sending
3937 messages over SMTP and also when doing verify callouts.
3939 2. When Exim is called to do synchronous delivery of a locally submitted
3940 message (the -odf or -odi options), it no longer closes stderr before doing
3943 3. Implemented the mua_wrapper option.
3945 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3947 5. Implemented the functions header_remove(), header_testname(),
3948 header_add_at_position(), and receive_remove_recipient(), and exported them
3951 6. If an ACL "warn" statement specified the addition of headers, Exim already
3952 inserted X-ACL-Warn: at the start if there was no header name. However, it
3953 was not making this test for the second and subsequent header lines if
3954 there were newlines in the string. This meant that an invalid header could
3955 be inserted if Exim was badly configured.
3957 7. Allow an ACL "warn" statement to add header lines at the start or after all
3958 the Received: headers, as well as at the end.
3960 8. Added the rcpt_4xx retry error code.
3962 9. Added postmaster_mailfrom=xxx to callout verification option.
3964 10. Added mailfrom=xxxx to the callout verification option, for verify=
3967 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3968 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3970 12. Inserted some casts to stop certain compilers warning when using pointer
3971 differences as field lengths or precisions in printf-type calls (mostly
3972 affecting debugging statements).
3974 13. Added optional readline() support for -be (dynamically loaded).
3976 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3977 same clock tick as a message's arrival, so that its received time was the
3978 same as the "first fail" time on the retry record, and that message
3979 remained on the queue past the ultimate address timeout, every queue runner
3980 would try a delivery (because it was past the ultimate address timeout) but
3981 after another failure, the ultimate address timeout, which should have then
3982 bounced the address, did not kick in. This was a "< instead of <=" error;
3983 in most cases the first failure would have been in the next clock tick
3984 after the received time, and all would be well.
3986 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3987 being recognized when the domain list was tested by the match_domain
3988 condition in an expansion string.
3990 16. Added the ${str2b64: operator.
3992 17. Exim was always calling setrlimit() to set a large limit for the number of
3993 processes, without checking whether the existing limit was already
3994 adequate. (It did check for the limit on file descriptors.) Furthermore,
3995 errors from getrlimit() and setrlimit() were being ignored. Now they are
3996 logged to the main and panic logs, but Exim does carry on, to try to do its
3997 job under whatever limits there are.
3999 18. Imported PCRE 5.0.
4001 19. Trivial typo in log message " temporarily refused connection" (the leading
4004 20. If the log selector return_path_on_delivery was set and an address was
4005 redirected to /dev/null, the delivery process crashed because it assumed
4006 that a return path would always be set for a "successful" delivery. In this
4007 case, the whole delivery is bypassed as an optimization, and therefore no
4010 21. Internal re-arrangement: the function for sending a challenge and reading
4011 a response while authentication was assuming a zero-terminated challenge
4012 string. It's now changed to take a pointer and a length, to allow for
4013 binary data in such strings.
4015 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4017 23. Exim was not respecting finduser_retries when seeking the login of the
4018 uid under which it was called; it was always trying 10 times. (The default
4019 setting of finduser_retries is zero.) Also, it was sleeping after the final
4020 failure, which is pointless.
4022 24. Implemented tls_on_connect_ports.
4024 25. Implemented acl_smtp_predata.
4026 26. If the domain in control=submission is set empty, Exim assumes that the
4027 authenticated id is a complete email address when it generates From: or
4028 Sender: header lines.
4030 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4031 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4032 chown and chgrp in /bin and hostname in /usr/bin.
4034 28. Exim was keeping the "process log" file open after each use, just as it
4035 does for the main log. This opens the possibility of it remaining open for
4036 long periods when the USR1 signal hits a daemon. Occasional processlog
4037 errors were reported, that could have been caused by this. Anyway, it seems
4038 much more sensible not to leave this file open at all, so that is what now
4041 29. The long-running daemon process does not normally write to the log once it
4042 has entered its main loop, and it closes the log before doing so. This is
4043 so that log files can straightforwardly be renamed and moved. However,
4044 there are a couple of unusual error situations where the daemon does write
4045 log entries, and I had neglected to close the log afterwards.
4047 30. The text of an SMTP error response that was received during a remote
4048 delivery was being truncated at 512 bytes. This is too short for some of
4049 the long messages that one sometimes sees. I've increased the limit to
4052 31. It is now possible to make retry rules that apply only when a message has a
4053 specific sender, in particular, an empty sender.
4055 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4056 it possible to be selective about when SMTP synchronization is enforced.
4058 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4060 32. Implemented hosts_connection_nolog.
4062 33. Added an ACL for QUIT.
4064 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4067 35. Added mailbox_size and mailbox_filecount to appendfile.
4069 36. Added control = no_multiline_responses to ACLs.
4071 37. There was a bug in the logic of the code that waits for the clock to tick
4072 in the case where the clock went backwards by a substantial amount such
4073 that the microsecond fraction of "now" was more than the microsecond
4074 fraction of "then" (but the whole seconds number was less).
4076 38. Added support for the libradius Radius client library this is found on
4077 FreeBSD (previously only the radiusclient library was supported).
4083 1. When certain lookups returned multiple values in the form name=value, the
4084 quoting of the values was not always being done properly. Specifically:
4085 (a) If the value started with a double quote, but contained no whitespace,
4087 (b) If the value contained whitespace other than a space character (i.e.
4088 tabs or newlines or carriage returns) it was not quoted.
4089 This fix has been applied to the mysql and pgsql lookups by writing a
4090 separate quoting function and calling it from the lookup code. The fix
4091 should probably also be applied to nisplus, ibase and oracle lookups, but
4092 since I cannot test any of those, I have not disturbed their existing code.
4094 2. A hit in the callout cache for a specific address caused a log line with no
4095 reason for rejecting RCPT. Now it says "Previous (cached) callout
4096 verification failure".
4098 3. There was an off-by-one bug in the queryprogram router. An over-long
4099 return line was truncated at 256 instead of 255 characters, thereby
4100 overflowing its buffer with the terminating zero. As well as fixing this, I
4101 have increased the buffer size to 1024 (and made a note to document this).
4103 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4104 when Exim is waiting for an SMTP response from a remote server, Exim
4105 restarts its select() call on the socket, thereby resetting its timeout.
4106 This is not a problem when such interrupts are rare. Somebody set up a cron
4107 job to run exiwhat every 2 minutes, which is less than the normal select()
4108 timeout (5 or 10 minutes). This meant that the select() timeout never
4109 kicked in because it was always reset. I have fixed this by comparing the
4110 time when an interrupt arrives with the time at the start of the first call
4111 to select(). If more time than the timeout has elapsed, the interrupt is
4112 treated as a timeout.
4114 5. Some internal re-factoring in preparation for the addition of Sieve
4115 extensions (by MH). In particular, the "personal" test is moved to a
4116 separate function, and given an option for scanning Cc: and Bcc: (which is
4117 not set for Exim filters).
4119 6. When Exim created an email address using the login of the caller as the
4120 local part (e.g. when creating a From: or Sender: header line), it was not
4121 quoting the local part when it contained special characters such as @.
4123 7. Installed new OpenBSD configuration files.
4125 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4126 try to make them clearer.
4128 9. Callout options, other than the timeout value, were being ignored when
4129 verifying sender addresses in header lines. For example, when using
4131 verify = header_sender/callout=no_cache
4133 the cache was (incorrectly) being used.
4135 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4136 only the Cygwin environment.
4138 11. When return_path_on_delivery was set as a log selector, if different remote
4139 addresses in the same message used different return paths and parallel
4140 remote delivery occurred, the wrong values would sometimes be logged.
4141 (Whenever a remote delivery process finished, the return path value from
4142 the most recently started remote delivery process was logged.)
4144 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4145 lines when AUTH and/or TLS are in use. This is the "received protocol"
4146 field. Exim used to use "asmtp" for authenticated SMTP, without any
4147 indication (in the protocol name) for TLS use. Now it follows the RFC and
4148 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4149 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4150 names appear in log lines as well as in Received: header lines.
4152 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4153 extensions, and comparison tests, and to fix some bugs.
4155 14. Changes to the "personal" filter test:
4157 (1) The test was buggy in that it was just doing the equivalent of
4158 "contains" tests on header lines. For example, if a user's address was
4159 anne@some.where, the "personal" test would incorrectly be true for
4161 To: susanne@some.where
4163 This test is now done by extracting each address from the header in turn,
4164 and checking the entire address. Other tests that are part of "personal"
4165 are now done using regular expressions (for example, to check local parts
4166 of addresses in From: header lines).
4168 (2) The list of non-personal local parts in From: addresses has been
4169 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4170 taken from the Sieve specification recommendations.
4172 (3) If the message contains any header line starting with "List-" it is
4173 treated as non-personal.
4175 (4) The test for "circular" in the Subject: header line has been removed
4176 because it now seems ill-conceived.
4178 15. Minor typos in src/EDITME comments corrected.
4180 16. Installed latest exipick from John Jetmore.
4182 17. If headers_add on a router specified a text string that was too long for
4183 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4184 of string_sprintf() is now avoided.
4186 18. $message_body_size was not set (it was always zero) when running the DATA
4187 ACL and the local_scan() function.
4189 19. For the "mail" command in an Exim filter, no default was being set for
4190 the once_repeat time, causing a random time value to be used if "once" was
4191 specified. (If the value happened to be <= 0, no repeat happened.) The
4192 default is now 0s, meaning "never repeat". The "vacation" command was OK
4193 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4194 (I found it when inspecting the code).
4196 20. There is now an overall timeout for performing a callout verification. It
4197 defaults to 4 times the callout timeout, which applies to individual SMTP
4198 commands during the callout. The overall timeout applies when there is more
4199 than one host that can be tried. The timeout is checked before trying the
4200 next host. This prevents very long delays if there are a large number of
4201 hosts and all are timing out (e.g. when the network connections are timing
4202 out). The value of the overall timeout can be changed by specifying an
4203 additional sub-option for "callout", called "maxwait". For example:
4205 verify = sender/callout=5s,maxwait=20s
4207 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4208 to the end before writing, but this should make it even safer).
4210 22. Exim was forgetting that it had advertised PIPELINING for the second and
4211 subsequent messages on an SMTP connection. It was also not resetting its
4212 memory on STARTTLS and an internal HELO.
4214 23. When Exim logs an SMTP synchronization error within a session, it now
4215 records whether PIPELINING has been advertised or not.
4217 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4218 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4219 rather than long int.
4221 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4223 26. Added the never_mail option to autoreply.
4229 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4230 crash if the getsockname() call failed; this can happen if a connection is
4231 closed very soon after it is established. The problem was simply in the
4232 order in which certain operations were done, causing Exim to try to write
4233 to the SMTP stream before it had set up the file descriptor. The bug has
4234 been fixed by making things happen in the correct order.
4240 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4241 before the connection was closed, thus losing the rejection response.
4243 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4244 some early Solaris releases, but causes trouble in current releases where
4245 socklen_t is defined.
4247 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4250 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4253 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4254 that releasing the top of it at the end releases what was used for sub-
4255 expansions (unless the block got too big). However, discard this block if
4256 the first thing is a variable or header, so that we can use its block when
4257 it is dynamic (useful for very large $message_headers, for example).
4259 6. Lookups now cache *every* query, not just the most recent. A new, separate
4260 store pool is used for this. It can be recovered when all lookup caches are
4261 flushed. Lookups now release memory at the end of their result strings.
4262 This has involved some general refactoring of the lookup sources.
4264 7. Some code has been added to the store_xxx() functions to reduce the amount
4265 of flapping under certain conditions.
4267 8. log_incoming_interface used to affect only the <= reception log lines. Now
4268 it causes the local interface and port to be added to several more SMTP log
4269 lines, for example "SMTP connection from", and rejection lines.
4271 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4273 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4275 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4276 could be overwritten at the end of the current message (or the start of a
4277 new message if it was set in a HELO ACL). The value is now preserved for
4278 the duration of the SMTP connection.
4280 12. If a transport had a headers_rewrite setting, and a matching header line
4281 contained an unqualified address, that address was qualified, even if it
4282 did not match any rewriting rules. The underlying bug was that the values
4283 of the flags that permit the existence of unqualified sender and recipient
4284 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4285 non-local messages, and by -bnq for local messages) were not being
4286 preserved with the message after it was received.
4288 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4289 "next input=" as part of the text comprising the host identity instead of
4290 the correct text. The code was using the same buffer for two different
4291 strings. However, depending on which order the printing function evaluated
4292 its arguments, the bug did not always show up. Under Linux, for example, my
4293 test suite worked just fine.
4295 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4296 doesn't work with some older versions of Perl. It has been changed to "my",
4297 which in any case is probably the better facility to use.
4299 15. A really picky compiler found some instances of statements for creating
4300 error messages that either had too many or two few arguments for the format
4303 16. The size of the buffer for calls to the DNS resolver has been increased
4304 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4305 for addresses that have a lot of PTR records. This alleviates a problem; it
4306 does not fully solve it.
4308 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4309 buffer now truncates the list and logs the incident, which is the same
4310 action as happens when Exim is looking up a host name and its aliases.
4311 Previously in this situation something unpredictable would happen;
4312 sometimes it was "internal error: store_reset failed".
4314 18. If a server dropped the connection unexpectedly when an Exim client was
4315 using GnuTLS and trying to read a response, the client delivery process
4316 crashed while trying to generate an error log message.
4318 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4319 string, for example:
4321 warn message = H1: something\nH2: something
4323 the text was added as a single header line from Exim's point of view
4324 though it ended up OK in the delivered message. However, searching for the
4325 second and subsequent header lines using $h_h2: did not work. This has been
4326 fixed. Similarly, if a system filter added multiple headers in this way,
4327 the routers could not see them.
4329 20. Expanded the error message when iplsearch is called with an invalid key to
4330 suggest using net-iplsearch in a host list.
4332 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4333 ACLs are no longer actually imposed (and a message to that effect is
4336 22. If a "gecos" field in a passwd entry contained escaped characters, in
4337 particular, if it contained a \" sequence, Exim got it wrong when building
4338 a From: or a Sender: header from that name. A second bug also caused
4339 incorrect handling when an unquoted " was present following a character
4340 that needed quoting.
4342 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4343 was not being matched caselessly.
4345 24. Arranged for all hyphens in the exim.8 source to be escaped with
4348 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4349 properly. Recipient callout cache records were still being keyed to include
4350 the sender, even when use_sender was set false. This led to far more
4351 callouts that were necessary. The sender is no longer included in the key
4352 when use_sender is false.
4354 26. Added "control = submission" modifier to ACLs.
4356 27. Added the ${base62d: operator to decode base 62 numbers.
4358 28. dnsdb lookups can now access SRV records.
4360 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4361 the configuration file.
4363 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4364 (-v) mode. This makes the output for a verbose queue run more intelligible.
4366 31. Added a use_postmaster feature to recipient callouts.
4368 32. Added the $body_zerocount variable, containing the number of binary zero
4369 bytes in the message body.
4371 33. The time of last modification of the "new" subdirectory is now used as the
4372 "mailbox time last read" when there is a quota error for a maildir
4375 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4377 35. Added +ignore_unknown as a special item in host lists.
4379 36. Code for decoding IPv6 addresses in host lists is now included, even if
4380 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4381 address was recognized as an IP address, but was then not correctly decoded
4382 into binary, causing unexpected and incorrect effects when compared with
4389 1. Very minor rewording of debugging text in manualroute to say "list of
4390 hosts" instead of "hostlist".
4392 2. If verify=header_syntax was set, and a header line with an unqualified
4393 address (no domain) and a large number of spaces between the end of the
4394 name and the colon was received, the reception process suffered a buffer
4395 overflow, and (when I tested it) crashed. This was caused by some obsolete
4396 code that should have been removed. The fix is to remove it!
4398 3. When running in the test harness, delay a bit after writing a bounce
4399 message to get a bit more predictability in the log output.
4401 4. Added a call to search_tidyup() just before forking a reception process. In
4402 theory, someone could use a lookup in the expansion of smtp_accept_max_
4403 per_host which, without the tidyup, could leave open a database connection.
4405 5. Added the variables $recipient_data and $sender_data which get set from a
4406 lookup success in an ACL "recipients" or "senders" condition, or a router
4407 "senders" option, similar to $domain_data and $local_part_data.
4409 6. Moved the writing of debug_print from before to after the "senders" test
4412 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4413 problems for message scanning, either using a data ACL, or using
4414 local_scan() because the Received: header was not generated till after they
4415 were called (in order to set the time as the time of reception completion).
4416 I have revised the way this works. The header is now generated after the
4417 body is received, but before the ACL or local_scan() are called. After they
4418 are run, the timestamp in the header is updated.
4424 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4425 before starting a queue runner without re-exec. This happened only when
4426 deliver_drop_privilege was set or when the Exim user was set to root. The
4427 effect of the bug was that timeouts during subsequent deliveries caused
4428 crashes instead of being properly handled. The handler is now left at its
4429 default (and expected) setting.
4431 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4432 message, again when deliver_drop_privilege is set or Exim is run as root.
4433 The bug described in (1) was not present in this case, but the tidying up
4434 of the other signals was missing. I have made the two cases consistent.
4436 3. The ignore_target_hosts setting on a manualroute router was being ignored
4437 for hosts that were looked up using the /MX notation.
4439 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4442 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4443 operated on the sender address. After changing the $sender_address to <>
4444 for the sender address verify, Exim was re-instated it as the original
4445 (before rewriting) address, but remembering that it had rewritten it, so it
4446 wasn't rewriting it again. This bug also had the effect of breaking the
4447 sender address verification caching when the sender address was rewritten.
4449 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4450 This has been changed so that if the ip literal address matches
4451 ignore_target_hosts, the router declines.
4453 7. Added expansion conditions match_domain, match_address, and match_local_
4454 part (NOT match_host).
4456 8. The placeholder for the Received: header didn't have a length field set.
4458 9. Added code to Exim itself and to exim_lock to test for a specific race
4459 condition that could lead to file corruption when using MBX delivery. The
4460 issue is with the lockfile that is created in /tmp. If this file is removed
4461 after a process has opened it but before that process has acquired a lock,
4462 there is the potential for a second process to recreate the file and also
4463 acquire a lock. This could lead to two Exim processes writing to the file
4464 at the same time. The added code performs the same test as UW imapd; it
4465 checks after acquiring the lock that its file descriptor still refers to
4466 the same named file.
4468 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4469 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4472 11. Added the smtp_active_hostname option. If used, this will typically be made
4473 to depend on the incoming interface address. Because $interface_address is
4474 not set up until the daemon has forked a reception process, error responses
4475 that can happen earlier (such as "too many connections") no longer contain
4478 12. If an expansion in a condition on a "warn" statement fails because a lookup
4479 defers, the "warn" statement is abandoned, and the next ACL statement is
4480 processed. Previously this caused the whole ACL to be aborted.
4482 13. Added the iplsearch lookup type.
4484 14. Added ident_timeout as a log selector.
4486 15. Added tls_certificate_verified as a log selector.
4488 16. Added a global option tls_require_ciphers (compare the smtp transport
4489 option of the same name). This controls incoming TLS connections.
4491 17. I finally figured out how to make tls_require_ciphers do a similar thing
4492 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4493 before starting the TLS session.
4495 18. Tabs are now shown as \t in -bP output.
4497 19. If the log selector return_path_on_delivery was set, Exim crashed when
4498 bouncing a message because it had too many Received: header lines.
4500 20. If two routers both had headers_remove settings, and the first one included
4501 a superfluous trailing colon, the final name in the first list and the
4502 first name in the second list were incorrectly joined into one item (with a
4503 colon in the middle).
4509 1. Added -C and -D options to the exinext utility, mainly to make it easier
4510 to include in the automated testing, but these could be helpful when
4511 multiple configurations are in use.
4513 2. The exinext utility was not formatting the output nicely when there was
4514 an alternate port involved in the retry record key, nor when there was a
4515 message id as well (for retries that were specific to a specific message
4516 and a specific host). It was also confused by IPv6 addresses, because of
4517 the additional colons they contain. I have fixed the IPv4 problem, and
4518 patched it up to do a reasonable job for IPv6.
4520 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4521 delivery, the log line now contains "pipelined" if PIPELINING was used.
4523 4. An SMTP transport process used to panic and die if the bind() call to set
4524 an explicit outgoing interface failed. This has been changed; it is now
4525 treated in the same way as a connect() failure.
4527 5. A reference to $sender_host_name in the part of a conditional expansion
4528 that was being skipped was still causing a DNS lookup. This no longer
4531 6. The def: expansion condition was not recognizing references to header lines
4532 that used bh_ and bheader_.
4534 7. Added the _cache feature to named lists.
4536 8. The code for checking quota_filecount in the appendfile transport was
4537 allowing one more file than it should have been.
4539 9. For compatibility with Sendmail, the command line option
4547 and sets the incoming protocol and host name (for trusted callers). The
4548 host name and its colon can be omitted when only the protocol is to be set.
4549 Note the Exim already has two private options, -pd and -ps, that refer to
4550 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4551 "s", but I don't think that's a major issue.
4553 10. A number of refactoring changes to the code, none of which should affect
4556 (a) The number of logging options was getting close to filling up the
4557 32-bit word that was used as a bit map. I have split them into two classes:
4558 those that are passed in the argument to log_write(), and those that are
4559 only ever tested independently outside of that function. These are now in
4560 separate 32-bit words, so there is plenty of room for expansion again.
4561 There is no change in the user interface or the logging behaviour.
4563 (b) When building, for example, log lines, the code previously used a
4564 macro that called string_cat() twice, in order to add two strings. This is
4565 not really sufficiently general. Furthermore, there was one instance where
4566 it was actually wrong because one of the argument was used twice, and in
4567 one call a function was used. (As it happened, calling the function twice
4568 did not affect the overall behaviour.) The macro has been replaced by a
4569 function that can join an arbitrary number of extra strings onto a growing
4572 (c) The code for expansion conditions now uses a table and a binary chop
4573 instead of a serial search (which was left over from when there were very
4574 few conditions). Also, it now recognizes conditions like "pam" even when
4575 the relevant support is not compiled in: a suitably worded error message is
4576 given if an attempt is made to use such a condition.
4578 11. Added ${time_interval:xxxxx}.
4580 12. A bug was causing one of the ddress fields not to be passed back correctly
4581 from remote delivery subprocesses. The field in question was not being
4582 subsequently used, so this caused to problems in practice.
4584 13. Added new log selectors queue_time and deliver_time.
4586 14. Might have fixed a bug in maildirsizefile handling that threw up
4587 "unexpected character" debug warnings, and recalculated the data
4588 unnecessarily. In any case, I expanded the warning message to give more
4591 15. Added the message "Restricted characters in address" to the statements in
4592 the default ACL that block characters like @ and % in local parts.
4594 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
4595 Three changes have been made:
4597 (a) There was a serious bug; a negative response to MAIL caused the whole
4598 recipient domain to be cached as invalid, thereby blocking all messages
4599 to all local parts at the same domain, from all senders. This bug has
4600 been fixed. The domain is no longer cached after a negative response to
4601 MAIL if the sender used is not empty.
4603 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
4606 (c) A new callout option, "use_sender" has been added for people who want
4607 the modified behaviour.
4613 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
4616 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
4617 indeed breaks things for older releases.
4619 3. Added additional logging to the case where there is a problem reading data
4620 from a filter that is running in a subprocess using a pipe, in order to
4621 try to track down a specific problem.
4623 4. Testing facility fudge: when running in the test harness and attempting
4624 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
4625 getting "No route to host". Convert this to a timeout.
4627 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
4630 6. Some OS don't have socklen_t but use size_t instead. This affects the
4631 fifth argument of getsockopt() amongst other things. This is now
4632 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
4633 can be set for individual OS. I have set it for SunOS5, OSF1, and
4634 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
4635 some earlier ones do not.
4637 7. Change 4.30/15 was not doing the test caselessly.
4639 8. The standard form for an IPv6 address literal was being rejected by address
4640 parsing in, for example, MAIL and RCPT commands. An example of this kind of
4641 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4642 this, as well as the form without the "IPv6" on the front (but only when
4643 address literals are enabled, of course).
4645 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
4647 10. Exim crashed if a message with an empty sender address specified by -f
4648 encountered a router with an errors_to setting. This could be provoked only
4649 by a command such as
4653 where an empty string was supplied; "<>" did not hit this bug.
4655 11. Installed PCRE release 4.5.
4657 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
4658 remained set. It is now erased.
4660 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
4661 times from message ids (which are base 36 rather than the normal 62).
4663 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
4664 were being counted as actual protocol errors, and logged if the log
4665 selector +smtp_protocol_error was set. One cannot be perfect in this test,
4666 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
4667 and DATA following a set of rejected RCPTs do not count as protocol errors.
4668 In other words, Exim assumes they were pipelined, though this may not
4669 actually be the case. Of course, in all cases the client gets an
4670 appropriate error code.
4672 15. If a lookup fails in an ACL condition, a message about the failure may
4673 be available; it is used if testing the ACL cannot continue, because most
4674 such messages specify what the cause of the deferral is. However, some
4675 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
4676 that caused an old message to be retained and used if a later statement
4677 caused a defer, replacing the real cause of the deferral.
4679 16. If an IP address had so many PTR records that the DNS lookup buffer
4680 was not large enough to hold them, Exim could crash while trying to process
4681 the truncated data. It now detects and logs this case.
4683 17. Further to 4.21/58, another change has been made: if (and only if) the
4684 first line of a message (the first header line) ends with CRLF, a bare LF
4685 in a subsequent header line has a space inserted after it, so as not to
4686 terminate the header.
4688 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
4689 unnecessarily, used atoi() instead of strtol(), and didn't check the
4690 termination when getting file sizes from file names by regex.
4692 19. Completely re-implemented the support for maildirsize files, in the light
4693 of a number of problems with the previous contributed implementation
4694 (4.30/29). In particular:
4696 . If the quota is zero, the maildirsize file is maintained, but no quota is
4699 . If the maildir directory does not exist, it is created before any attempt
4700 to write a maildirsize file.
4702 . The quota value in the file is just a cache; if the quota is changed in
4703 the transport, the new value overrides.
4705 . A regular expression is available for excluding directories from the
4708 20. The autoreply transport checks the characters in options that define the
4709 message's headers; it allows continued headers, but it was checking with
4710 isspace() after an embedded newline instead of explicitly looking for a
4713 21. If all the "regular" hosts to which an address was routed had passed their
4714 expiry times, and had not reached their retry times, the address was
4715 bounced, even if fallback hosts were defined. Now Exim should go on to try
4718 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4719 equivalent code in the SMTP transport. Some hosts send humungous responses
4720 to HELO/EHLO, more than 1024 it seems.
4722 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4723 gives compiler warnings in some environments. I've now done it "properly",
4726 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4727 (because of problems with the built-in one) was declared to return uschar *
4728 instead of char *, causing compiler failure.
4730 25. Fixed a file descriptor leak when processing alias/forward files.
4732 26. Fixed a minor format string issue in dbfn.c.
4734 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4736 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4737 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4740 29. When a sender address is verified, it is cached, to save repeating the test
4741 when there is more than one recipient in a message. However, when the
4742 verification involves a callout, it is possible for different callout
4743 options to be set for different recipients. It is too complicated to keep
4744 track of this in the cache, so now Exim always runs a verification when a
4745 callout is required, relying on the callout cache for the optimization.
4746 The overhead is duplication of the address routing, but this should not be
4749 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4750 to be verified with callout=postmaster, and the main callout worked but the
4751 postmaster check failed, the verification correctly failed. However, if a
4752 subsequent RCPT command asked for sender verification *without* the
4753 postmaster check, incorrect caching caused this verification also to fail,
4756 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4757 it was not caching the DNS options (qualify_single, search_parents) that
4758 were used when the lookup failed. A subsequent lookup with different
4759 options therefore always gave the same answer, though there were cases
4760 where it should not have. (Example: a "domains = !$mx_any" option on a
4761 dnslookup router: the "domains" option is always processed without any
4762 widening, but the router might have qualify_single set.) Now Exim uses the
4763 cached value only when the same options are set.
4765 32. Added John Jetmore's "exipick" utility to the distribution.
4767 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4768 than a timeout (e.g. a certificate is required, and is not provided), an
4769 Exim server now closes the connection immediately. Previously it waited for
4770 the client to close - but if the client is SSL, it seems that they each
4771 wait for each other, leading to a delay before one of them times out.
4773 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4774 maintained 0.8.x compatibility because I don't think many are using it, and
4775 it is clearly obsolete.
4777 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4780 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4781 client certificate was expired. A simple patch fixes this, though I don't
4782 understand the full logic of why the verify callback is called multiple
4785 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4786 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4787 which causes problems with some clients (such as the Certicom SSL Plus
4788 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4789 disables the coutermeasure allowing Eudora to connect."
4791 38. Exim was not checking that a write() to a log file succeeded. This could
4792 lead to Bad Things if a log got too big, in particular if it hit a file
4793 size limit. Exim now panics and dies if it cannot write to a log file, just
4794 as it does if it cannot open a log file.
4796 39. Modified OS/Makefile-Linux so that it now contains
4798 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4800 The two -D definitions ensure that Exim is compiled with large file
4801 support, which makes it possible to handle log files that are bigger than
4804 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4805 instance) a domain was checked against a named list that involved a lookup,
4806 causing $domain_data to be set, then another domain was checked against the
4807 same list, then the first domain was re-checked, the value of $domain_data
4808 after the final check could be wrong. In particular, if the second check
4809 failed, it could be set empty. This bug probably also applied to
4812 41. The strip_trailing_dot option was not being applied to the address given
4813 with the -f command-line option.
4815 42. The code for reading a message's header from the spool was incrementing
4816 $received_count, but never initializing it. This meant that the value was
4817 incorrect (doubled) while delivering a message in the same process in which
4818 it was received. In the most common configuration of Exim, this never
4819 happens - a fresh exec is done - but it can happen when
4820 deliver_drop_privilege is set.
4822 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4823 it now includes up to 150 characters of the unexpected data in the log
4826 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4827 and building data strings. The size of both of these buffers was 10 000
4828 bytes - far larger than anybody would *ever* want, thought I. Needless to
4829 say, somebody hit the limit. I have increased the maximum line length to
4830 20 000 and the maximum data length of concatenated lines to 100 000. I have
4831 also fixed two bugs, because there was no checking on these buffers. Tsk,
4832 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4833 buffer is too small.
4835 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4836 lsearch lookups. Now it does.
4838 46. When parsing a route_list item in a manualroute router, a fixed-length
4839 buffer was used for the list of hosts. I made this 1024 bytes long,
4840 thinking that nobody would ever have a list of hosts that long. Wrong.
4841 Somebody had a whole pile of complicated expansion conditions, and the
4842 string was silently truncated, leading to an expansion error. It turns out
4843 that it is easier to change to an unlimited length (owing to other changes
4844 that have happened since this code was originally written) than to build
4845 structure for giving a limitation error. The length of the item that
4846 expands into the list of hosts is now unlimited.
4848 47. The lsearch lookup could not handle data where the length of text line was
4849 more than 4095 characters. Such lines were truncated, leading to shortened
4850 data being returned. It should now handle lines of any length.
4852 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4853 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4856 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4857 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4858 no longer happens in comment lines. A list of replacements is now placed
4859 at the head of all of the source files, except those whose only change is
4860 to replace PERL_COMMAND in the very first #! line.
4862 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4863 messages on the queue, with a bottom-up merge sort, using code contributed
4864 by Michael Haardt. This should make operations like -bp somewhat faster on
4865 large queues. It won't affect queue runners, except when queue_run_in_order
4868 51. Installed eximstats 1.31 in the distribution.
4870 52. Added support for SRV lookups to the dnslookup router.
4872 53. If an ACL referred to $message_body or $message_body_end, the value was not
4873 reset for any messages that followed in the same SMTP session.
4875 54. The store-handling optimization for building very long strings was not
4876 differentiating between the different store pools. I don't think this
4877 actually made any difference in practice, but I've tidied it.
4879 55. While running the routers to verify a sender address, $sender_address
4880 was still set to the sender address. This is wrong, because when routing to
4881 send a bounce to the sender, it would be empty. Therefore, I have changed
4882 it so that, while verifying a sender address, $sender_address is set to <>.
4883 (There is no change to what happens when verifying a recipient address.)
4885 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4886 target A or AAAA records (if not already returned) without resetting the
4887 qualify_single or search_parents options of the DNS resolver. These are
4888 inappropriate in this case because the targets of MX and SRV records must
4889 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4890 target that, when qualified, matched something in the local domain. These
4891 two options are now turned off when doing these lookups.
4893 57. It seems that at least some releases of Reiserfs (which does not have the
4894 concept of a fixed number of inodes) returns zero and not -1 for the
4895 number of available inodes. This interacted badly with check_spool_inodes,
4896 which assumed that -1 was the "no such thing" setting. What I have done is
4897 to check that the total number of inodes is greater than zero before doing
4898 the test of how many are available.
4900 58. When a "warn" ACL statement has a log_message modifier, the message is
4901 remembered, and not repeated. This is to avoid a lot of repetition when a
4902 message has many recipients that cause the same warning to be written.
4903 Howewer, Exim was preserving the list of already written lines for an
4904 entire SMTP session, which doesn't seem right. The memory is now reset if a
4905 new message is started.
4907 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4908 debugging output unless log_rewrite was also set.
4910 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4911 of (int)(handle) when we know that handle contains (void *)(-1).
4913 61. The Exim daemon panic-logs an error return when it closes the incoming
4914 connection. However "connection reset by peer" seems to be common, and
4915 isn't really an error worthy of noting specially, so that particular error
4918 62. When Exim is trying to find all the local interfaces, it used to panic and
4919 die if the ioctl to get the interface flags failed. However, it seems that
4920 on at least one OS (Solaris 9) it is possible to have an interface that is
4921 included in the list of interfaces, but for which you get a failure error
4922 for this call. This happens when the interface is not "plumbed" into a
4923 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4924 failure of the "get flags" call assumes that the interface is down.
4926 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4927 makes life easier for people who are doing arithmetic on fields extracted
4928 from dates, where you often get leading zeros that should not be
4929 interpreted as octal.
4931 64. Added qualify_domain to the redirect router, to override the global
4934 65. If a pathologically long header line contained very many addresses (the
4935 report of this problem mentioned 10 000) and each of them was rewritten,
4936 Exim could use up a very large amount of memory. (It kept on making new
4937 copies of the header line as it rewrote, and never released the old ones.)
4938 At the expense of a bit more processing, the header rewriting function has
4939 been changed so that it no longer eats memory in this way.
4941 66. The generation of the Received: header has been moved from the time that a
4942 message starts to be received, to the time that it finishes. The timestamp
4943 in the Received: header should now be very close to that of the <= log
4944 line. There are two side-effects of this change:
4946 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4947 logged header lines no longer include the local Received: line, because
4948 it has not yet been created. The same applies to a copy of the message
4949 that is returned to a non-SMTP sender when a message is rejected.
4951 (b) When a filter file is tested using -bf, no additional Received: header
4952 is added to the test message. After some thought, I decided that this
4955 This change does not affect the value of $received_for. It is still set
4956 after address rewriting, but before local_scan() is called.
4958 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4960 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4961 gave an unhelpful panic error message, and a defer error. I have managed to
4962 change this behaviour so that it now rejects any supplied certificate,
4963 which seems right, as the list of acceptable certificates is empty.
4965 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4966 gave an unhelpful defer error. I have not managed to make this reject any
4967 supplied certificates, but the error message it gives is "no certificate
4968 supplied", which is not helpful.
4970 70. exigrep's output now also includes lines that are not associated with any
4971 message, but which match the given pattern. Implemented by a patch from
4972 Martin Sluka, which also tidied up the Perl a bit.
4974 71. Recipient callout verification, like sender verification, was using <> in
4975 the MAIL FROM command. This isn't really the right thing, since the actual
4976 sender may affect whether the remote host accepts the recipient or not. I
4977 have changed it to use the actual sender in the callout; this means that
4978 the cache record is now keyed on a recipient/sender pair, not just the
4979 recipient address. There doesn't seem to be a real danger of callout loops,
4980 since a callout by the remote host to check the sender would use <>.
4981 [SEE ABOVE: changed after hitting problems.]
4983 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4984 temporary errors. However, in the case of such a code being given after
4985 the end of a data transmission (i.e. after ".") Exim was failing to write
4986 a retry record for the message. (Yes, there was some broken host that was
4987 actually sending 8xx at this point.)
4989 73. An unknown lookup type in a host list could cause Exim to panic-die when
4990 the list was checked. (An example that provoked this was putting <; in the
4991 middle of a list instead of at the start.) If this happened during a DATA
4992 ACL check, a -D file could be left lying around. This kind of configuration
4993 error no longer causes Exim to die; instead it causes a defer errror. The
4994 incident is still logged to the main and panic logs.
4996 74. Buglet left over from Exim 3 conversion. The message "too many messages
4997 in one connection" was written to the rejectlog but not the mainlog, except
4998 when address rewriting (yes!) was being logged.
5000 75. Added write_rejectlog option.
5002 76. When a system filter was run not as root (that is, when system_filter_user
5003 was set), the values of the $n variables were not being returned to the
5004 main process; thus, they were not subsequently available in the $sn
5007 77. Added +return_path_on_delivery log selector.
5009 78. A connection timeout was being treated differently from recipients deferred
5010 when testing hosts_max_try with a message that was older than the host's
5011 retry timeout. (The host should not be counted, thus allowing all hosts to
5012 be tried at least once before bouncing.) This may have been the cause of an
5013 occasionally reported bug whereby a message would remain on the queue
5014 longer than the retry timeout, but would be bounced if a delivery was
5015 forced. I say "may" because I never totally pinned down the problem;
5016 setting up timeout/retry tests is difficult. See also the next item.
5018 79. The ultimate address timeout was not being applied to errors that involved
5019 a combination of host plus message (for example, a timeout on a MAIL
5020 command). When an address resolved to a number of possible hosts, and they
5021 were not all tried for each delivery (e.g. because of hosts_max_try), a
5022 message could remain on the queue longer than the retry timeout.
5024 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5027 81. Fixed an obscure SMTP outgoing bug which required at least the following
5028 conditions: (a) there was another message waiting for the same server;
5029 (b) the server returned 5xx to all RCPT commands in the first message so
5030 that the message was not completed; (c) the server dropped the connection
5031 or gave a negative response to the RSET that Exim sends to abort the
5032 transaction. The observed case was a dropped connection after DATA that had
5033 been sent in pipelining mode. That is, the server had advertised PIPELINING
5034 but was not implementing it correctly. The effect of the bug was incorrect
5035 behaviour, such as trying another host, and this could lead to a crash.
5041 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5042 and daemon.c were passed as pointers to ints; they should have been
5043 pointers to socklen_t variables (which are typically unsigned ints).
5045 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5048 3. Fixed a really odd bug that affected only the testing scheme; patching a
5049 certain fixed string in the binary changed the value of another string that
5050 happened to be identical to the end of the original first string.
5052 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5053 name", it returns that address as the IP address. On some operating
5054 systems (e.g. Solaris), it also passes back the IP address string as the
5055 "host name". However, on others (e.g. Linux), it passes back an empty
5056 string. Exim wasn't checking for this, and was changing the host name to an
5057 empty string, assuming it had been canonicized.
5059 5. Although rare, it is permitted to have more than one PTR record for a given
5060 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5061 all the names associated with an address, because they do in Solaris.
5062 However, it seems that they do not in Linux for data that comes from the
5063 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5064 I found this out when I moved to a new Linux workstation and tried to run
5065 the Exim test suite.
5067 To get round this problem I have changed the code so that it now does its
5068 own call to the DNS to look up PTR records when searching for a host name.
5069 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5070 addresses that are only in /etc/hosts are still found.
5072 This behaviour is, however, controlled by an option called host_lookup_
5073 order, which defaults to "bydns:byaddr". If people want to use the other
5074 order, or indeed, just use one or the other means of lookup, they can
5075 specify it in this variable.
5077 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5078 some operating systems, this comes back from gethostbyaddr() as an empty
5079 string, and this is what Exim used to test for. However, it seems that in
5080 other systems, "." is yielded. Exim now tests for this case too.
5082 7. The values of check_spool_space and check_log_space are now held internally
5083 as a number of kilobytes instead of an absolute number of bytes. If a
5084 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5085 kilobyte. This means that much larger values can be stored.
5087 8. Exim monitor: an attempt to get the action menu when not actually pointing
5088 at a message produces an empty menu entitled "No message selected". This
5089 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5090 no entries in it ("Shell widget menu has zero width and/or height"). So I
5091 have added a single, blank menu entry in this case.
5093 9. Added ${quote_local_part.
5095 10. MIME decoding is now applied to the contents of Subject: header lines when
5098 11. Now that a reference to $sender_host_address automatically causes a reverse
5099 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5100 host lookup before query-style lookups in lists that might use this
5101 variable. This has therefore been abolished, and the "net-" prefix is no
5102 longer necessary for query-style lookups.
5104 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5105 have been a typo for LFLAGS, so it has been changed.
5107 13. The install script calls Exim with "-C /dev/null" in order to find the
5108 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5109 to be output. Howeve, since Exim outputs its version number before the
5110 error, it didn't break the script. It just looked ugly. I fixed this by
5111 always allowing "-C /dev/null" if the caller is root.
5113 14. Ignore overlarge ACL variable number when reading spool file - insurance
5114 against a later release with more variables having written the file.
5116 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5117 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5118 this, as well as the form without the "IPv6" on the front.
5120 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5121 OS/Makefile-Darwin file.
5123 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5124 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5126 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5127 variable. However, some OS use ints for inodes. I've added cast to long int
5128 to get rid of the compiler warning.
5130 19. I had forgotten to lock out "/../" in configuration file names when
5131 ALT_CONFIG_PREFIX was set.
5133 20. Routers used for verification do not need to specify transports. However,
5134 if such a router generated a host list, and callout was configured, Exim
5135 crashed, because it could not find a port number from the (non-existent)
5136 transport. It now assumes port 25 in this circumstance.
5138 21. Added the -t option to exigrep.
5140 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5141 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5142 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5143 them is set, LOOKUP_LSEARCH is forced.
5145 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5147 24. Added sender and host information to the "rejected by local_scan()" log
5148 line; previously there was no indication of these.
5150 25. Added .include_if_exists.
5152 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5153 Linux. It turns out that not all file systems support this. Apparently some
5154 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5155 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5156 not supported on the file descriptor, is now ignored when Exim is trying to
5157 sync a directory. This applies only to Linux.
5159 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5161 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5162 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5165 29. Added support for maildirsize files from supplied patch (modified a bit).
5167 30. The use of :fail: followed by an empty string could lead Exim to respond to
5168 sender verification failures with (e.g.):
5170 550 Verification failed for <xxx>
5171 550 Sender verify failed
5173 where the first response line was missing the '-' that indicates it is not
5174 the final line of the response.
5176 31. The loop for finding the name of the user that called Exim had a hardwired
5177 limit of 10; it now uses the value of finduser_retries, which is used for
5178 all other user lookups.
5180 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5183 33. Exim was neglecting to zero errno before one call of strtol() when
5184 expanding a string and expecting an integer value. On some systems this
5185 resulted in spurious "integer overflow" errors. Also, it was casting the
5186 result into an int without checking.
5188 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5189 did not work. The code looks as if it has *never* worked, though it appears
5190 to have been documented since at least releast 1.62. I have made it work.
5192 35. The "timeout_DNS" error in retry rules, also documented since at least
5193 1.62, also never worked. As it isn't clear exactly what this means, and
5194 clearly it isn't a major issue, I have abolished the feature by treating it
5195 as "timeout", and writing a warning to the main and panic logs.
5197 36. The display of retry rules for -brt wasn't always showing the error code
5200 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5201 timeout_connect_A, timeout_connect_MX.
5203 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5204 to the empty sender.
5206 39. The daemon was not analysing the content of -oX till after it had closed
5207 stderr and disconnected from the controlling terminal. This meant that any
5208 syntax errors were only noted on the panic log, and the return code from
5209 the command was 0. By re-arranging the code a little, I've made the
5210 decoding happen first, so such errors now appear on stderr, and the return
5211 code is 1. However, the actual setting up of the sockets still happens in
5212 the disconnected process, so errors there are still only recorded on the
5215 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5216 connections (as happens on some IP stacks) was logged at start up time as
5217 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5218 it from "IPv6 and IPv4", which means that two separate sockets are being
5221 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5222 says whether AF_INET or AF_INET6 was passed as an argument.
5224 42. Exiwhat output was messed up when time zones were included in log
5227 43. Exiwhat now gives more information about the daemon's listening ports,
5228 and whether -tls-on-connect was used.
5230 44. The "port" option of the smtp transport is now expanded.
5232 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5233 silently ignored. Now an error message is written to the main and panic
5236 46. There's a new ACL modifier called "logwrite" which writes to a log file
5237 as soon as it is encountered.
5239 47. Added $local_user_uid and $local_user_gid at routing time.
5241 48. Exim crashed when trying to verify a sender address that was being
5244 49. Exim was recognizing only a space character after ".include". It now also
5245 recognizes a tab character.
5247 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5248 extracting the relevant information from the specification. The man page no
5249 longer contains scrambled data for the -d option, and I've added a section
5250 at the front about calling Exim under different names.
5252 51. Added "extra_headers" argument to the "mail" command in filter files.
5254 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5257 53. Installed eximstats 1.29.
5259 54. Added transport_filter_timeout as a generic transport option.
5261 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5262 Cc: header lines. This was required by RFC 822, but it not required by RFC
5265 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5266 incoming messages that did not have them. Now it does so only if the
5267 message originates locally, that is, if there is no associated remote host
5268 address. When Resent- header lines are present, this applies to the Resent-
5269 lines rather than the non-Resent- lines.
5271 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5272 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5274 58. Messages for configuration errors now include the name of the main
5275 configuration file - useful now that there may be more than one file in a
5276 list (.included file names were always shown).
5278 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5279 for those rare installations that do not start the daemon as root or run it
5280 setuid root. I've cut out the call to initgroups() if the daemon is not
5283 60. The Exim user and group can now be bound into the binary as text strings
5284 that are looked up at the start of Exim's processing.
5286 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5288 62. Added $mailstore_basename variable.
5290 63. Installed patch to sieve.c from Michael Haardt.
5292 64. When Exim failed to open the panic log after failing to open the main log,
5293 the original message it was trying to log was written to stderr and debug
5294 output, but if they were not available (the usual case in production), it
5295 was lost. Now it is written to syslog before the two lines that record the
5296 failures to open the logs.
5298 65. Users' Exim filters run in subprocesses under the user's uid. It is
5299 possible for a "deliver" command or an alias in a "personal" command to
5300 provoke an address rewrite. If logging of address rewriting is configured,
5301 this fails because the process is not running as root or exim. There may be
5302 a better way of dealing with this, but for the moment (because 4.30 needs
5303 to be released), I have disabled address rewrite logging when running a
5304 filter in a non-root, non-exim process.
5310 1. The buildconfig auxiliary program wasn't quoting the value set for
5311 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5312 not defined. This bug was masked in 4.22 by the effect that was fixed in
5315 2. Some messages that were rejected after a message id was allocated were
5316 shown as "incomplete" by exigrep. It no longer does this for messages that
5317 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5319 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5320 have allow_domain_literals set, the ID did not get logged in the <= line.
5321 Domain literals are now always recognized in Message-ID: header lines.
5323 4. The first argument for a ${extract expansion item is the key name or field
5324 number. Leading and trailing spaces in this item were not being ignored,
5325 causing some misleading effects.
5327 5. When deliver_drop_privilege was set, single queue runner processes started
5328 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5329 same command in the process it spins off) were not dropping privilege.
5331 6. When the daemon running as "exim" started a queue runner, it always
5332 re-executed Exim in the spun-off process. This is a waste of effort when
5333 deliver_drop_privilege is set. The new process now just calls the
5334 queue-runner function directly.
5340 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5343 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5344 ignored. Though the use of -oP was forcing the writing of a pid file, it
5345 was always written to the default place.
5347 3. If the message "no IP address found for host xxxx" is generated during
5348 incoming verification, it is now followed by identification of the incoming
5349 connection (so you can more easily find what provoked it).
5351 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5353 5. Added some features to "harden" Exim a bit more against certain attacks:
5355 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5356 be put in Local/Makefile. This is like the never_users runtime option,
5357 but it cannot be overridden. The default setting is "root".
5359 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5360 prefix string with which any file named in a -C command line option
5363 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5364 is retained for -C and -D only if the caller of Exim is root. Without
5365 it, the exim user may also use -C and -D and retain privilege.
5367 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5368 command line option is disabled.
5370 6. Macro names set by the -D option must start with an upper case letter, just
5371 like macro names defined in the configuration file.
5373 7. Added "dereference=" facility to LDAP.
5375 8. Two instances of the typo "uknown" in the source files are fixed.
5377 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5378 the Configure-Makefile script screwed up while processing it.
5380 10. Incorporated PCRE 4.4.
5382 11. The SMTP synchronization check was not operating right at the start of an
5383 SMTP session. For example, it could not catch a HELO sent before the client
5384 waited for the greeting. There is now a check for outstanding input at the
5385 point when the greeting is written. Because of the duplex, asynchronous
5386 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5387 way, but not yet received, when the check is performed.
5389 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5390 on TCP/IP sockets, because this apparently causes some broken clients to
5393 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5394 unchanged) from the Cygwin maintainer.
5396 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5397 was supported instead of testing for mbx. Effectively a typo.
5399 15. The spa authenticator server code was not checking that the input it
5400 received was valid base64.
5402 16. The debug output line for the "set" modifier in ACLs was not showing the
5403 name of the variable that was being set.
5405 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5407 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5408 lookup on its own. Something else was needed to trigger the lookup. For
5409 example, a match in host_lookup or the need for a host name in a host list.
5410 Now, if $sender_host_name is referenced and the host name has not yet been
5411 looked up, a lookup is performed. If the lookup fails, the variable remains
5412 empty, and $host_lookup_failed is set to "1".
5414 19. Added "eqi" as a case-independent comparison operator.
5416 20. The saslauthd authentication condition could segfault if neither service
5417 nor realm was specified.
5419 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5420 error message that was logged was misleading, and incoming SMTP
5421 connections were dropped. The message is now more accurate, and temporary
5422 errors are given to SMTP connections.
5424 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5425 (except RSET) with a 421 error, until QUIT is received. However, it was
5426 failing to send a response to QUIT.
5428 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5429 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5430 ACL is now run after the helo_try_verify_hosts code.
5432 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5433 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5434 case-independent, so other case variants are also recognized). Apparently
5435 some systems use these upper case variants.
5437 25. If more than two messages were waiting for the same host, and a transport
5438 filter was specified for the transport, Exim sent two messages over the
5439 same TCP/IP connection, and then failed with "socket operation on non-
5440 socket" when it tried to send the third.
5442 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5444 27. The extern definition of crypt16() in expand.c was not being excluded when
5445 the OS had its own crypt16() function.
5447 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5448 as a preferred synonym for return_size_limit, both as an option and as an
5451 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5453 30. Changed the default configuration ACL to relax the local part checking rule
5454 for addresses that are not in any local domains. For these addresses,
5455 slashes and pipe symbols are allowed within local parts, but the sequence
5456 /../ is explicitly forbidden.
5458 31. SPA server authentication was not clearing the challenge buffer before
5461 32. log_message in a "warn" ACL statement was writing to the reject log as
5462 well as to the main log, which contradicts the documentation and doesn't
5463 seem right (because no rejection is happening). So I have stopped it.
5465 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5466 However, I am unable to do any testing of this.
5468 34. Fixed an infelicity in the appendfile transport. When checking directories
5469 for a mailbox, to see if any needed to be created, it was accidentally
5470 using path names with one or more superfluous leading slashes; tracing
5471 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5473 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5474 contents are added to the log line that is written for every discarded
5475 recipient. (Previously a log_message setting was ignored.)
5477 36. The ${quote: operator now quotes the string if it is empty.
5479 37. The install script runs exim in order to find its version number. If for
5480 some reason other than non-existence or emptiness, which it checks, it
5481 could not run './exim', it was installing it with an empty version number,
5482 i.e. as "exim-". This error state is now caught, and the installation is
5485 38. An argument was missing from the function that creates an error message
5486 when Exim fails to connect to the socket for saslauthd authentication.
5487 This could cause Exim to crash, or give a corrupted message.
5489 39. Added isip, isip4, and isip6 to ${if conditions.
5491 40. The ACL variables $acl_xx are now saved with the message, and can be
5492 accessed later in routers, transports, and filters.
5494 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5495 strings in the file are not string-expanded.
5497 42. If a MAIL command specified a SIZE value that was too large to fit into an
5498 int variable, the check against message_size_limit failed. Such values are
5499 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5500 day this will have to be increased, but I don't think I want to be around
5501 when emails are that large.
5508 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5509 iconv() is not standard in FreeBSD.
5511 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5512 IPv6 enabled. The observed symptom was a segmentation fault on return from
5513 the function os_common_find_running_interfaces() in src/os.c.
5515 3. In the check_special_case() function in daemon.c I had used "errno" as an
5516 argument name, which causes warnings on some systems. This was basically a
5517 typo, since it was named "eno" in the comments!
5519 4. The code that waits for the clock to tick (at a resolution of some fraction
5520 of a second) so as to ensure message-id uniqueness was always waiting for
5521 at least one whole tick, when it could have waited for less. [This is
5522 almost certainly not relevant at current processor speeds, where it is
5523 unlikely to ever wait at all. But we try to future-proof.]
5525 5. The function that sleeps for a time interval that includes fractions of a
5526 second contained a race. It did not block SIGALRM between setting the
5527 timer, and suspending (a couple of lines later). If the interval was short
5528 and the sigsuspend() was delayed until after it had expired, the suspension
5529 never ended. On busy systems this could lead to processes getting stuck for
5532 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5533 process, before it forks any delivery processes. The open lookup caching
5534 mechanism meant that the open file or database connection was passed into
5535 the delivery process. The problem was that delivery processes always tidy
5536 up cached lookup data. This could cause a problem for the next delivery
5537 process started by the queue runner, because the external queue runner
5538 process does not know about the closure. So the next delivery process
5539 still has data in the lookup cache. In the case of a file lookup, there was
5540 no problem because closing a file descriptor in a subprocess doesn't affect
5541 the parent. However, if the lookup was caching a connection to a database,
5542 the connection was closed, and the second delivery process was likely to
5543 see errors such as "PGSQL: query failed: server closed the connection
5544 unexpectedly". The problem has been fixed by closing all cached lookups
5545 in a queue runner before running a delivery process.
5547 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5548 seem to have the "const" qualifier which it has on other OS. I've
5551 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5552 *of the same type* (client or server) with the same public name that an
5553 error should be diagnosed.
5555 9. When Exim looked up a host name for an IP address, but failed to find the
5556 original IP address when looking up the host name (a safety check), it
5557 output the message "<ip address> does not match any IP for NULL", which was
5558 confusing, to say the least. The bug was that the host name should have
5559 appeared instead of "NULL".
5561 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5562 user that is built into the binary, and the -C or -D options is used, root
5563 privilege is dropped before the configuration file is read. In addition,
5564 logging is switched to stderr instead of the normal log files. If the
5565 configuration then re-defines the Exim user, the unprivileged environment
5566 is probably not what is expected, so Exim logs a panic warning message (but
5569 However, if deliver_drop_privilege is set, the unprivileged state may well
5570 be exactly what is intended, so the warning has been cut out in that case,
5571 and Exim is allowed to try to write to its normal log files.
5577 1. smtp_return_error_details was not giving details for temporary sender
5578 or receiver verification errors.
5580 2. Diagnose a configuration error if two authenticators have the same public
5583 3. Exim used not to create the message log file for a message until the first
5584 delivery attempt. This could be confusing when incoming messages were held
5585 for policy or load reasons. The message log file is now created at the time
5586 the message is received, and an initial "Received" line is written to it.
5588 4. The automatically generated man page for command line options had a minor
5589 bug that caused no ill effects; however, a more serious problem was that
5590 the procedure for building the man page automatically didn't always
5591 operate. Consequently, release 4.20 contains an out-of-date version. This
5592 shouldn't happen again.
5594 5. When building Exim with embedded Perl support, the script that builds the
5595 Makefile was calling 'perl' to find its compile-time parameters, ignoring
5596 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
5598 6. The freeze_tell option was not being used for messages that were frozen on
5599 arrival, either by an ACL or by local_scan().
5601 7. Added the smtp_incomplete_transaction log selector.
5603 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
5604 was accepting AUTH without a new EHLO.
5606 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
5607 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
5610 10. Logging of TCP/IP connections (when configured) now happens in the main
5611 daemon process instead of the child process, so that the TCP/IP connection
5612 count is more accurate (but it can never be perfect).
5614 11. The use of "drop" in a nested ACL was not being handled correctly in the
5615 outer ACL. Now, if condition failure induced by the nested "drop" causes
5616 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
5617 or "require"), the connection is dropped.
5619 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
5620 that yield "discard" can now be used with an "accept" or a "discard" verb,
5621 but an error is generated for any others (because I can't see a useful way
5622 to define what should happen).
5624 13. When an ACL is read dynamically from a file (or anywhere else), the lines
5625 are now processed in the same way as lines in the Exim configuration file.
5626 In particular, continuation lines are supported.
5628 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
5630 15. Added -ti meaning -t -i.
5632 16. Check for letters, digits, hyphens, and dots in the names of dnslist
5633 domains, and warn by logging if others are found.
5635 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
5636 returned from GIFCONF when Exim is trying to find the list of interfaces on
5637 a host. The code in os.c has been modified to copy each ifreq to an aligned
5638 structure in all cases.
5640 Also, in some cases, the returned ifreq's were being copied to a 'struct
5641 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
5642 means the last couple of bytes of an IPv6 address could be chopped if the
5643 ifreq contained only a normal sockaddr (14 bytes storage).
5645 18. Named domain lists were not supported in the hosts_treat_as_local option.
5646 An entry such as +xxxx was not recognized, and was treated as a literal
5649 19. Ensure that header lines added by a DATA ACL are included in the reject log
5650 if the ACL subsequently rejects the message.
5652 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
5653 MD5 (which is deprecated).
5655 21. When testing a filter file using -bf, Exim was writing a message when it
5656 took the sender from a "From " line in the message, but it was not doing so
5657 when it took $return_path from a Return-Path: header line. It now does.
5659 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
5660 with a valid header line field name (a series of printing characters
5661 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
5663 23. Changed "disc" in the source to "disk" to conform to the documentation and
5664 the book and for uniformity.
5666 24. Ignore Sendmail's -Ooption=value command line item.
5668 25. When execve() failed while trying to run a command in a pipe transport,
5669 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
5670 could be confused with a return value of 69 from the command itself. This
5671 has been changed to 127, the value the shell returns if it is asked to run
5672 a non-existent command. The wording for the related log line suggests a
5673 non-existent command as the problem.
5675 26. If received_header_text expands to an empty string, do not add a Received:
5676 header line to the message. (Well, it adds a token one on the spool, but
5677 marks it "old" so that it doesn't get used or transmitted.)
5679 27. Installed eximstats 1.28 (addition of -nt option).
5681 28. There was no check for failure on the call to getsockname() in the daemon
5682 code. This can fail if there is a shortage of resources on the system, with
5683 ENOMEM, for example. A temporary error is now given on failure.
5685 29. Contrary to the C standard, it seems that in some environments, the
5686 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
5687 program. Exim now does this explicitly; it affects the formatting of
5688 timestamps using strftime().
5690 30. If exiqsumm was given junk data, it threw up some uninitialized variable
5691 complaints. I've now initialized all the variables, to avoid this.
5693 32. Header lines added by a system filter were not being "seen" during
5694 transport-time rewrites.
5696 33. The info_callback() function passed to OpenSSL is set up with type void
5697 (*)(SSL *, int, int), as described somewhere. However, when calling the
5698 function (actually a macro) that sets it up, the type void(*)() is
5699 expected. I've put in a cast to prevent warnings from picky compilers.
5701 34. If a DNS black list lookup found a CNAME record, but there were no A
5702 records associated with the domain it pointed at, Exim crashed.
5704 35. If a DNS black list lookup returned more than one A record, Exim ignored
5705 all but the first. It now scans all returned addresses if a particular IP
5706 value is being sought. In this situation, the contents of the
5707 $dnslist_value variable are a list of all the addresses, separated by a
5710 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5711 to accept a host name and all its aliases if the forward lookup for any of
5712 them yielded the IP address of the incoming connection. Now it accepts only
5713 those names whose forward lookup yields the correct IP address. Any other
5714 names are discarded. This closes a loophole whereby a rogue DNS
5715 administrator could create reverse DNS records to break through a
5716 wildcarded host restriction in an ACL.
5718 37. If a user filter or a system filter that ran in a subprocess used any of
5719 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5720 the wrong values were passed to the pipe command ($thisaddress had the
5721 value of $0, $0 had the value of $1, etc). This bug was introduced by
5722 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5724 38. Improved the line breaking for long SMTP error messages from ACLs.
5725 Previously, if there was no break point between 40 and 75 characters, Exim
5726 left the rest of the message alone. Two changes have been made: (a) I've
5727 reduced the minimum length to 35 characters; (b) if it can't find a break
5728 point between 35 and 75 characters, it looks ahead and uses the first one
5729 that it finds. This may give the occasional overlong line, but at least the
5730 remaining text gets split now.
5732 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5733 file descriptors might be low, and that setting 1000 would always raise it.
5734 It turns out that in some environments, the limit is already over 1000 and
5735 that lowering it causes trouble. So now Exim takes care not to decrease it.
5737 40. When delivering a message, the value of $return_path is set to $sender_
5738 address at the start of routing (routers may change the value). By an
5739 oversight, this default was not being set up when an address was tested by
5740 -bt or -bv, which affected the outcome if any router or filter referred to
5743 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5744 log only when "message" or "log_message" are set. However, if one of the
5745 conditions was an address verification, or a call to a nested ACL, the
5746 messages generated by the underlying test were being passed through. This
5747 no longer happens. The underlying message is available in $acl_verify_
5748 message for both "message" and "log_message" expansions, so it can be
5749 passed through if needed.
5751 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5752 new expansion $bh_ to give the encoded byte string without charset
5753 translation. Translation happens only if iconv() is available; HAVE_ICONV
5754 indicates this at build time. HEADERS_CHARSET gives the charset to
5755 translate to; headers_charset can change it in the configuration, and
5756 "headers charset" can change it in an individual filter file.
5758 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5759 that creates RFC 2047 encoded "words" labels them as that charset instead
5760 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5761 expansion operator; (ii) when Exim creates a From: line for a local
5762 message; (iii) when a header line is rewritten to include a "phrase" part.
5764 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5765 buggy, causing it to skip the first lines of messages whose message ID
5766 ended in 'D'. This would not have bitten before Exim release 4.14, because
5767 message IDs were unlikely to end in 'D' before then. The effect was to have
5768 incorrect size information for certain domains.
5770 45. #include "config.h" was missing at the start of the crypt16.c module. This
5771 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5774 46. If there was a timeout during a "random" callout check, Exim treated it as
5775 a failure of the random address, and carried on sending RSET and the real
5776 address. If the delay was just some slowness somewhere, the response to the
5777 original RCPT would be taken as a response to RSET and so on, causing
5778 mayhem of various kinds.
5780 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5781 when I implemented it. It didn't allow for the fact that some option values
5782 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5783 the right test for positive values.
5785 48. Domain names in DNS records are case-independent. Exim always looks them up
5786 in lower case. Some resolvers return domain names in exactly the case they
5787 appear in the zone file, that is, they may contain uppercase letters. Not
5788 all resolvers do this - some return always lower case. Exim was treating a
5789 change of case by a resolver as a change of domain, similar to a widening
5790 of a domain abbreviation. This triggered its re-routing code and so it was
5791 trying to route what was effectively the same domain again. This normally
5792 caused routing to fail (because the router wouldn't handle the domain
5793 twice). Now Exim checks for this case specially, and just changes the
5794 casing of the domain that it ultimately uses when it transmits the message
5797 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5800 50. If a filter generated a file delivery with a non-absolute name (possible if
5801 no home directory exists for the router), the forbid_file option was not
5804 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5805 the existing equality matching.
5807 52. Exim was using ints instead of ino_t variables in some places where it was
5808 dealing with inode numbers.
5810 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5811 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5812 called TMPDIR, and if it finds it is different, it changes its value.
5814 54. The smtp_printf() function is now made available to local_scan() so
5815 additional output lines can be written before returning. There is also an
5816 smtp_fflush() function to enable the detection of a dropped connection.
5817 The variables smtp_input and smtp_batched_input are exported to
5820 55. Changed the default runtime configuration: the message "Unknown user"
5821 has been removed from the ACL, and instead placed on the localuser router,
5822 using the cannot_route_message feature. This means that any verification
5823 failures that generate their own messages won't get overridden. Similarly,
5824 the "Unrouteable address" message that was in the ACL for unverifiable
5825 relay addresses has also been removed.
5827 56. Added hosts_avoid_esmtp to the smtp transport.
5829 57. The exicyclog script was not checking for the esoteric option
5830 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5831 will work only if exicyclog is run under the appropriate euid.
5833 58. Following a discussion on the list, the rules by which Exim recognises line
5834 endings on incoming messages have been changed. The -dropcr and drop_cr
5835 options are now no-ops, retained only for backwards compatibility. The
5836 following line terminators are recognized: LF CRLF CR. However, special
5837 processing applies to CR:
5839 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5840 nor a local message in the state where . is a terminator.
5842 (ii) If a bare CR is encountered in a header line, an extra space is added
5843 after the line terminator so as not to end the header. The reasoning
5844 behind this is that bare CRs in header lines are most likely either
5845 to be mistakes, or people trying to play silly games.
5847 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5848 was being incorrectly given as 18 bytes larger than it should have been.
5849 This is a VOB (very old bug).
5851 60. This may never have affected anything current, but just in case it has:
5852 When the local host is found other than at the start of a list of hosts,
5853 the local host, those with the same MX, and any that follow, are discarded.
5854 When the list in question was part of a longer list of hosts, the following
5855 hosts (not currently being processed) were also being discarded. This no
5856 longer happens. I'm not sure if this situation could ever has previously
5859 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5862 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5863 header. This is something that is recommended in a new Internet Draft, and
5864 is something that is documented as being done by Sendmail. There are two
5865 possible values. For messages generated by the autoreply transport, Exim
5868 Auto-Submitted: auto-replied
5870 whereas for all other generated messages (e.g. bounces) it adds
5872 Auto-Submitted: auto-generated
5874 63. The "personal" condition in filters now includes a test for the
5875 Auto-Submitted: header. If it contains the string "auto-" the message it
5876 not considered personal.
5878 64. Added rcpt_include_affixes as a generic transport option.
5880 65. Added queue_only_override (default true).
5882 66. Added the syslog_duplication option.
5884 67. If what should have been the first header line of a message consisted of
5885 a space followed by a colon, Exim was mis-interpreting it as a header line.
5886 It isn't of course - it is syntactically invalid and should therefore be
5887 treated as the start of the message body. The misbehaviour could have
5888 caused a number of strange effects, including loss of data in subsequent
5889 header lines, and spool format errors.
5891 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5892 client host had authenticated. This control can now be exercised by an ACL
5893 for more flexibility.
5895 69. By default, callouts do not happen when testing with -bh. There is now a
5896 variant, -bhc, which does actually run the callout code, including
5897 consulting and updating the callout cache.
5899 70. Added support for saslauthd authentication, courtesy of Alexander
5902 71. If statvfs() failed on the spool or log directories while checking their
5903 size for availability, Exim confusingly gave the error "space shortage".
5904 Furthermore, in debugging mode it crashed with a floating point exception.
5905 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5906 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5907 serious problem, Exim now writes to the main and panic logs when this
5908 happens, with details of the failure. It then refuses to accept the
5909 incoming message, giving the message "spool directory problem" or "log
5910 directory problem" with a 421 code for SMTP messages.
5912 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5913 0, 1, and 2 exist, because some OS complain for execs without them (see
5914 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5915 descriptors. However, the code omitted to check that the open succeeded,
5916 causing mysterious errors if for some reason the permissions on /dev/null
5917 got screwed. Now Exim writes a message to the main and panic logs, and
5918 bombs out if it can't open /dev/null.
5920 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5921 interact so that it is all more flexible. It is supposed to remain
5922 backwards compatible. Also added extra_local_interfaces.
5924 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5925 to bomb out with an assertion failure - to the client this appears as a
5926 connection drop. This problem occurs in the part of the code that was taken
5927 from the Samba project. Fortunately, the assertion is in a very simple
5928 function, so I have fixed this by reproducing the function inline in the
5929 one place where it is called, and arranging for authentication to fail
5930 instead of killing the process with assert().
5932 75. The SPA client code was not working when the server requested OEM rather
5933 than Unicode encoding.
5935 76. Added code to make require_files with a specific uid setting more usable in
5936 the case where statting the file as root fails - usually a non-root-mounted
5937 NFS file system. When this happens and the failure is EACCES, Exim now
5938 forks a subprocess and does the per-uid checking as the relevant uid.
5940 77. Added process_log_path.
5942 78. If log_file_path was not explicitly set, a setting of check_log_space or
5943 check_log_inodes was ignored.
5945 79. If a space check for the spool or log partitions fails, the incident is now
5946 logged. Of course, in the latter case the data may get lost...
5948 80. Added the %p formatting code to string_format() so that it can be used to
5949 print addresses in debug_print(). Adjusted all the address printing in the
5950 debugging in store.c to use %p rather than %d.
5952 81. There was a concern that a line of code in smtp_in.c could overflow a
5953 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5954 initially expressed, the concern was not well-founded, because trailing
5955 spaces are removed early. However, if the trailing spaces were followed by
5956 a NULL, they did not get removed, so the overflow was possible. Two fixes
5959 (a) I re-wrote the offending code in a cleaner fashion.
5960 (b) If an incoming SMTP command contains a NULL character, it is rejected
5963 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5964 runs initgroups(), so that if the Exim user is in any additional groups,
5965 they will be used during message reception.
5971 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob