1 Change log file for Exim from version 4.21
2 -------------------------------------------
7 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
10 JH/02 The smtp transport option "multi_domain" is now expanded.
12 JH/03 The smtp transport now requests PRDR by default, if the server offers
15 JH/04 Certificate name checking on server certificates, when exim is a client,
16 is now done by default. The transport option tls_verify_cert_hostname
17 can be used to disable this per-host. The build option
18 EXPERIMENTAL_CERTNAMES is withdrawn.
20 JH/05 The value of the tls_verify_certificates smtp transport and main options
21 default to the word "system" to access the system default CA bundle.
22 For GnuTLS, only version 3.0.20 or later.
24 JH/06 Verification of the server certificate for a TLS connection is now tried
25 (but not required) by default. The verification status is now logged by
26 default, for both outbound TLS and client-certificate supplying inbound
29 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
32 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
33 Status Notification (bounce) messages are now MIME format per RFC 3464.
34 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
35 under the control of the dsn_advertise_hosts option, and routers may
36 have a dsn_lasthop option.
38 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
39 default, modifiable by a malware= option. The list separator for
40 the options can now be changed in the usual way.
46 TL/01 When running the test suite, the README says that variables such as
47 no_msglog_check are global and can be placed anywhere in a specific
48 test's script, however it was observed that placement needed to be near
49 the beginning for it to behave that way. Changed the runtest perl
50 script to read through the entire script once to detect and set these
51 variables, reset to the beginning of the script, and then run through
52 the script parsing/test process like normal.
54 TL/02 The BSD's have an arc4random API. One of the functions to induce
55 adding randomness was arc4random_stir(), but it has been removed in
56 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
57 function when detected.
59 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
60 cause callback expansion.
62 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
63 syntax errors in an expansion can be treated as a string instead of
64 logging or causing an error, due to the internal use of bool_lax
65 instead of bool when processing it.
67 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
68 server certificates when making smtp deliveries.
70 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
72 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
74 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
76 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
77 Merged patch from Sebastian Wiedenroth.
79 JH/05 Fix results-pipe from transport process. Several recipients, combined
80 with certificate use, exposed issues where response data items split
81 over buffer boundaries were not parsed properly. This eventually
82 resulted in duplicates being sent. This issue only became common enough
83 to notice due to the introduction of conection certificate information,
84 the item size being so much larger. Found and fixed by Wolfgang Breyha.
86 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
87 size buffer was used, resulting in syntax errors when an expansion
90 JH/07 Add support for directories of certificates when compiled with a GnuTLS
91 version 3.3.6 or later.
93 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
94 is EXPERIMENTAL_EVENT, the main-configuration and transport options
95 both become "event_action", the variables become $event_name, $event_data
96 and $event_defer_errno. There is a new variable $verify_mode, usable in
97 routers, transports and related events. The tls:cert event is now also
98 raised for inbound connections, if the main configuration event_action
101 TL/06 In test suite, disable OCSP for old versions of openssl which contained
102 early OCSP support, but no stapling (appears to be less than 1.0.0).
104 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
105 server certificate names available under the smtp transport option
106 "tls_verify_cert_hostname" now do not permit multi-component wildcard
109 JH/10 Time-related extraction expansions from certificates now use the main
110 option "timezone" setting for output formatting, and are consistent
111 between OpenSSL and GnuTLS compilations. Bug 1541.
113 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
114 encoded parameter in the incoming message. Bug 1558.
116 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
117 include certificate info, eximon was claiming there were spoolfile
120 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
122 JH/14 Log delivery-related information more consistently, using the sequence
123 "H=<name> [<ip>]" wherever possible.
125 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
126 are problematic for Debian distribution, omit them from the release
129 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
131 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
134 JH/17 Fix a null-indirection in certextract expansions when a nondefault
135 output list separator was used.
140 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
141 checkers that were complaining about end of non-void function with no
144 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
145 This was a regression intruduced in 4.83 by another bugfix.
147 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
149 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
150 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
156 TF/01 Correctly close the server side of TLS when forking for delivery.
158 When a message was received over SMTP with TLS, Exim failed to clear up
159 the incoming connection properly after forking off the child process to
160 deliver the message. In some situations the subsequent outgoing
161 delivery connection happened to have the same fd number as the incoming
162 connection previously had. Exim would try to use TLS and fail, logging
163 a "Bad file descriptor" error.
165 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
166 utilities have not been installed.
168 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
169 temporary space as the ACL may create new global variables.
171 TL/01 LDAP support uses per connection or global context settings, depending
172 upon the detected version of the libraries at build time.
174 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
175 to extract and use the src ip:port in logging and expansions as if it
176 were a direct connection from the outside internet. PPv2 support was
177 updated based on HAProxy spec change in May 2014.
179 JH/02 Add ${listextract {number}{list}{success}{fail}}.
181 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
182 Properly escape header and check for NULL return.
184 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
187 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
189 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
190 characters in header names, implemented as a verify condition.
191 Contributed by Michael Fischer v. Mollard.
193 TL/05 Rename SPF condition results err_perm and err_temp to standardized
194 results permerror and temperror. Previous values are deprecated but
195 still accepted. In a future release, err_perm and err_temp will be
196 completely removed, which will be a backward incompatibility if the
197 ACL tests for either of these two old results. Patch contributed by
198 user bes-internal on the mailing list.
200 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
202 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
203 selectors, in both main and reject logs.
205 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
208 JH/07 Add malware type "sock" for talking to simple daemon.
210 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
212 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
213 routers/transports under cutthrough routing.
215 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
216 numbers. Touch up "bool" conditional to keep the same definition.
218 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
220 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
222 JH/12 Expand items in router/transport headers_add or headers_remove lists
223 individually rather than the list as a whole. Bug 1452.
225 Required for reasonable handling of multiple headers_ options when
226 they may be empty; requires that headers_remove items with embedded
227 colons must have them doubled (or the list-separator changed).
229 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
230 view the policy declared in the DMARC record. Currently, $dmarc_status
231 is a combined value of both the record presence and the result of the
234 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
236 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
237 dnslookup router and the smtp transport (applying to the forward
240 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
241 of ldap servers used for a specific lookup. Patch provided by Heiko
244 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
245 New variable $lookup_dnssec_authenticated for observability.
247 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
248 Patch submitted by Lars Timman.
250 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
252 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
253 Requires trusted mode and valid format message id, aborts otherwise.
254 Patch contributed by Heiko Schlichting.
256 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
257 certextract with support for various fields. Bug 1358.
259 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
260 is requested by default, modifiable by smtp transport option
263 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
264 operate on certificate variables to give certificate fingerprints
265 Also new ${sha256:cert_variable}.
267 JH/23 The PRDR feature is moved from being Experimental into the mainline.
269 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
270 Christian Aistleitner.
272 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
274 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
275 file. Patch from Wolfgang Breyha.
277 JH/25 Expand the coverage of the delivery $host and $host_address to
278 client authenticators run in verify callout. Bug 1476.
280 JH/26 Port service names are now accepted for tls_on_connect_ports, to
281 align with daemon_smtp_ports. Bug 72.
283 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
284 support and error reporting did not work properly.
286 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
287 and is readable. Patch from Andrew Colin Kissa.
289 TL/14 Enhance documentation of ${run expansion and how it parses the
290 commandline after expansion, particularly in the case when an
291 unquoted variable expansion results in an empty value.
293 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
295 PP/02 Fix internal collision of T_APL on systems which support RFC3123
296 by renaming away from it. Addresses GH issue 15, reported by
299 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
301 TL/15 SECURITY: prevent double expansion in math comparison functions
302 (can expand unsanitized data). Not remotely exploitable.
309 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
311 PP/02 Make -n do something, by making it not do something.
312 When combined with -bP, the name of an option is not output.
314 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
317 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
318 $sender_host_name and config options to manage this, and basic check
321 PP/05 DSCP support for outbound connections and control modifier for inbound.
323 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
324 (Only plugin which currently uses this is kerberos4, which nobody should
325 be using, but we should make it available and other future plugins might
326 conceivably use it, even though it would break NAT; stuff *should* be
327 using channel bindings instead).
329 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
330 name; added for Sendmail compatibility; requires admin caller.
331 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
332 just ignore it); requires trusted caller.
333 Also parse but ignore: -Ac -Am -X<logfile>
336 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
338 TL/02 Add +smtp_confirmation as a default logging option.
340 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
341 Patch by Magnus Holmgren from 2007-02-20.
343 TL/04 Bugzilla 1281 - Spec typo.
344 Bugzilla 1283 - Spec typo.
345 Bugzilla 1290 - Spec grammar fixes.
347 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
349 TL/06 Add Experimental DMARC support using libopendmarc libraries.
351 TL/07 Fix an out of order global option causing a segfault. Reported to dev
352 mailing list by by Dmitry Isaikin.
354 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
356 JH/02 Support "G" suffix to numbers in ${if comparisons.
358 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
360 NM/01 Bugzilla 1197 - Spec typo
361 Bugzilla 1196 - Spec examples corrections
363 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
365 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
366 gnutls_enable_pkcs11, but renamed to more accurately indicate its
369 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
370 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
372 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
373 "acl {{name}{arg}...}", and optional args on acl condition
376 JH/05 Permit multiple router/transport headers_add/remove lines.
378 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
380 JH/07 Avoid using a waiting database for a single-message-only transport.
381 Performance patch from Paul Fisher. Bugzilla 1262.
383 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
386 JH/09 Add $headers_added variable, with content from use of ACL modifier
387 add_header (but not yet added to the message). Bugzilla 199.
389 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
390 Pulled from Bugzilla 817 by Wolfgang Breyha.
392 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
394 (nb: this is the same fix as in Exim 4.80.1)
396 JH/11 Add A= logging on delivery lines, and a client_set_id option on
399 JH/12 Add optional authenticated_sender logging to A= and a log_selector
402 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
404 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
405 advertise SMTP AUTH mechanism to us, instead of a generic
406 protocol violation error. Also, make Exim more robust to bad
407 data from the Dovecot auth socket.
409 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
411 When a queue runner is handling a message, Exim first routes the
412 recipient addresses, during which it prunes them based on the retry
413 hints database. After that it attempts to deliver the message to
414 any remaining recipients. It then updates the hints database using
417 So if a recipient address works intermittently, it can get repeatedly
418 deferred at routing time. The retry hints record remains fresh so the
419 address never reaches the final cutoff time.
421 This is a fairly common occurrence when a user is bumping up against
422 their storage quota. Exim had some logic in its local delivery code
423 to deal with this. However it did not apply to per-recipient defers
424 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
426 This change adds a proper retry rule check during routing so that the
427 final cutoff time is checked against the message's age. We only do
428 this check if there is an address retry record and there is not a
429 domain retry record; this implies that previous attempts to handle
430 the address had the retry_use_local_parts option turned on. We use
431 this as an approximation for the destination being like a local
432 delivery, as in LMTP.
434 I suspect this new check makes the old local delivery cutoff check
435 redundant, but I have not verified this so I left the code in place.
437 TF/02 Correct gecos expansion when From: is a prefix of the username.
439 Test 0254 submits a message to Exim with the header
443 When I ran the test suite under the user fanf2, Exim expanded
444 the header to contain my full name, whereas it should have added
445 a Resent-Sender: header. It erroneously treats any prefix of the
446 username as equal to the username.
448 This change corrects that bug.
450 GF/01 DCC debug and logging tidyup
451 Error conditions log to paniclog rather than rejectlog.
452 Debug lines prefixed by "DCC: " to remove any ambiguity.
454 TF/03 Avoid unnecessary rebuilds of lookup-related code.
456 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
457 Bug spotted by Jeremy Harris; was flawed since initial commit.
458 Would have resulted in OCSP responses post-SNI triggering an Exim
459 NULL dereference and crash.
461 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
463 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
464 Bug detection, analysis and fix by Samuel Thibault.
465 Bugzilla 1331, Debian bug #698092.
467 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
469 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
470 Server implementation by Todd Lyons, client by JH.
471 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
472 config variable "prdr_enable" controls whether the server
473 advertises the facility. If the client requests PRDR a new
474 acl_data_smtp_prdr ACL is called once for each recipient, after
475 the body content is received and before the acl_smtp_data ACL.
476 The client is controlled by bolth of: a hosts_try_prdr option
477 on the smtp transport, and the server advertisement.
478 Default client logging of deliveries and rejections involving
479 PRDR are flagged with the string "PRDR".
481 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
482 fclose(). Diagnosis by Todd Lyons.
484 PP/17 Update configure.default to handle IPv6 localhost better.
485 Patch by Alain Williams (plus minor tweaks).
488 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
489 This is now consistent with GnuTLS, and is now documented: the
490 previous undocumented portable approach to treating the option as
491 unset was to force an expansion failure. That still works, and
492 an empty string is now equivalent.
494 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
495 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
496 not performing validation itself.
498 PP/20 Added force_command boolean option to pipe transport.
499 Patch from Nick Koston, of cPanel Inc.
501 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
504 TF/04 Added udpsend ACL modifer and hexquote expansion operator
506 PP/21 Fix eximon continuous updating with timestamped log-files.
507 Broken in a format-string cleanup in 4.80, missed when I repaired the
508 other false fix of the same issue.
509 Report and fix from Heiko Schlichting.
512 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
513 Report from Prashanth Katuri.
515 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
516 It's SecureTransport, so affects any MacOS clients which use the
517 system-integrated TLS libraries, including email clients.
519 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
520 using a MIME ACL for non-SMTP local injection.
521 Report and assistance in diagnosis by Warren Baker.
523 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
525 JH/16 Fix comparisons for 64b. Bugzilla 1385.
527 TL/09 Add expansion variable $authenticated_fail_id to keep track of
528 last id that failed so it may be referenced in subsequent ACL's.
530 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
533 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
534 ldap library initialization, allowing self-signed CA's to be
535 used. Also properly sets require_cert option later in code by
536 using NULL (global ldap config) instead of ldap handle (per
537 session). Bug diagnosis and testing by alxgomz.
539 TL/12 Enhanced documentation in the ratelimit.pl script provided in
540 the src/util/ subdirectory.
542 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
543 renamed to Transport Post Delivery Action by Jeremy Harris, as
546 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
547 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
548 redis_servers = needs to be configured which will be used by the redis
549 lookup. Patch from Warren Baker, of The Packet Hub.
551 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
553 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
554 hostname or reverse DNS when processing a host list. Used suggestions
555 from multiple comments on this bug.
557 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
559 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
560 Missed a few lines, added it to make the runtest require no keyboard
563 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
564 contains upper case chars. Make router use caseful_local_part.
566 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
567 support when GnuTLS has been built with p11-kit.
573 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
575 This, or similar/improved, will also be change PP/11 of 4.82.
581 PP/01 Handle short writes when writing local log-files.
582 In practice, only affects FreeBSD (8 onwards).
583 Bugzilla 1053, with thanks to Dmitry Isaikin.
585 NM/01 Bugzilla 949 - Documentation tweak
587 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
590 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
592 PP/02 Implemented gsasl authenticator.
594 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
596 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
597 `pkg-config foo` for cflags/libs.
599 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
600 with rest of GSASL and with heimdal_gssapi.
602 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
603 `pkg-config foo` for cflags/libs for the TLS implementation.
605 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
606 properties get this fed in as external SSF. A number of robustness
607 and debugging improvements to the cyrus_sasl authenticator.
609 PP/08 cyrus_sasl server now expands the server_realm option.
611 PP/09 Bugzilla 1214 - Log authentication information in reject log.
612 Patch by Jeremy Harris.
614 PP/10 Added dbmjz lookup type.
616 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
618 PP/12 MAIL args handles TAB as well as SP, for better interop with
619 non-compliant senders.
620 Analysis and variant patch by Todd Lyons.
622 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
623 Bug report from Lars Müller <lars@samba.org> (via SUSE),
624 Patch from Dirk Mueller <dmueller@suse.com>
626 PP/13 tls_peerdn now print-escaped for spool files.
627 Observed some $tls_peerdn in wild which contained \n, which resulted
628 in spool file corruption.
630 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
631 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
632 or write after TLS renegotiation, which otherwise led to messages
635 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
636 as a tracking header (ie: a signed header comes before the signature).
637 Patch from Wolfgang Breyha.
639 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
640 comma-sep list; embedded commas doubled.
642 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
644 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
646 Report and patch from Dmitry Banschikov.
648 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
649 Removed SSL_clear() after SSL_new() which led to protocol negotiation
650 failures. We appear to now support TLS1.1+ with Exim.
652 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
653 lets Exim select keys and certificates based upon TLS SNI from client.
654 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
655 before an outbound SMTP session. New log_selector, +tls_sni.
657 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
658 NULL dereference. Report and patch from Alun Jones.
660 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
661 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
662 Not seeing resolver debug output on NetBSD, but suspect this is a
663 resolver implementation change.
665 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
666 Left warnings. Added "eximon gdb" invocation mode.
668 PP/21 Defaulting "accept_8bitmime" to true, not false.
670 PP/22 Added -bw for inetd wait mode support.
672 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
673 locate the relevant includes and libraries. Made this the default.
675 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
676 Bugzilla 1246, report and most of solution from Tomasz Kusy.
678 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
679 This may cause build issues on older platforms.
681 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
682 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
683 gnutls_require_mac & gnutls_require_protocols (no longer supported).
684 Added SNI support via GnuTLS too.
685 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
687 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
689 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
690 Applied second patch from Janne, implementing suggestion to default
691 multiple-strings-in-record handling to match SPF spec.
693 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
695 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
696 read-only, out of scope).
697 Patch from Wolfgang Breyha, report from Stuart Northfield.
699 PP/29 Fix three issues highlighted by clang analyser static analysis.
700 Only crash-plausible issue would require the Cambridge-specific
701 iplookup router and a misconfiguration.
702 Report from Marcin Mirosław.
704 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
706 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
707 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
708 As part of this, removing so much warning spew let me fix some minor
709 real issues in debug logging.
711 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
712 assignment on my part. Fixed.
714 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
715 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
716 Janne Snabb (who went above and beyond: thank you).
718 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
719 string otherwise requires a connection and a bunch more work and it's
720 relatively easy to get wrong. Should also expose TLS library linkage
723 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
724 64-bit ${eval} (JH/03).
726 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
727 GNU libc to support some of the 64-bit stuff, should not lead to
728 conflicts. Defined before os.h is pulled in, so if a given platform
729 needs to override this, it can.
731 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
732 protection layer was required, which is not implemented.
733 Bugzilla 1254, patch from Wolfgang Breyha.
735 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
736 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
737 tls_dhparam take prime identifiers. Also unbreak combination of
738 OpenSSL+DH_params+TLSSNI.
740 PP/39 Disable SSLv2 by default in OpenSSL support.
746 PP/01 Solaris build fix for Oracle's LDAP libraries.
747 Bugzilla 1109, patch from Stephen Usher.
749 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
751 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
754 TF/02 Fix a couple more cases where we did not log the error message
755 when unlink() failed. See also change 4.74-TF/03.
757 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
758 lock up or crash if it happened to be inside a call to libc when it
759 got a SIGUSR1 from exiwhat.
761 The SIGUSR1 handler appends the current process status to the process
762 log which is later printed by exiwhat. It used to use the general
763 purpose logging code to do this, but several functions it calls are
764 not safe for signals.
766 The new output code in the SIGUSR1 handler is specific to the process
767 log, and simple enough that it's easy to inspect for signal safety.
768 Removing some special cases also simplifies the general logging code.
769 Removing the spurious timestamps from the process log simplifies
772 TF/04 Improved ratelimit ACL condition.
774 The /noupdate option has been deprecated in favour of /readonly which
775 has clearer semantics. The /leaky, /strict, and /readonly update modes
776 are mutually exclusive. The update mode is no longer included in the
777 database key; it just determines when the database is updated. (This
778 means that when you upgrde Exim will forget old rate measurements.)
780 Exim now checks that the per_* options are used with an update mode that
781 makes sense for the current ACL. For example, when Exim is processing a
782 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
783 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
784 must specify per_mail/readonly. If you omit the update mode it defaults to
785 /leaky where that makes sense (as before) or /readonly where required.
787 The /noupdate option is now undocumented but still supported for
788 backwards compatibility. It is equivalent to /readonly except that in
789 ACLs where /readonly is required you may specify /leaky/noupdate or
790 /strict/noupdate which are treated the same as /readonly.
792 A useful new feature is the /count= option. This is a generalization
793 of the per_byte option, so that you can measure the throughput of other
794 aggregate values. For example, the per_byte option is now equivalent
795 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
797 The per_rcpt option has been generalized using the /count= mechanism
798 (though it's more complicated than the per_byte equivalence). When it is
799 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
800 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
801 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
802 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
803 Note that using per_rcpt with a non-readonly update mode in more than
804 one ACL will cause the recipients to be double-counted. (The per_mail
805 and per_byte options don't have this problem.)
807 The handling of very low rates has changed slightly. If the computed rate
808 is less than the event's count (usually one) then this event is the first
809 after a long gap. In this case the rate is set to the same as this event's
810 count, so that the first message of a spam run is counted properly.
812 The major new feature is a mechanism for counting the rate of unique
813 events. The new per_addr option counts the number of different
814 recipients that someone has sent messages to in the last time period. It
815 behaves like per_rcpt if all the recipient addresses are different, but
816 duplicate recipient addresses do not increase the measured rate. Like
817 the /count= option this is a general mechanism, so the per_addr option
818 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
819 example, measure the rate that a client uses different sender addresses
820 with the options per_mail/unique=$sender_address. There are further
821 details in the main documentation.
823 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
825 TF/06 Removed a few PCRE remnants.
827 TF/07 Automatically extract Exim's version number from tags in the git
828 repository when doing development or release builds.
830 PP/02 Raise smtp_cmd_buffer_size to 16kB.
831 Bugzilla 879. Patch from Paul Fisher.
833 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
834 Heavily based on revision 40f9a89a from Simon Arlott's tree.
837 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
839 PP/05 Variable $av_failed, true if the AV scanner deferred.
840 Bugzilla 1078. Patch from John Horne.
842 PP/06 Stop make process more reliably on build failure.
843 Bugzilla 1087. Patch from Heiko Schlittermann.
845 PP/07 Make maildir_use_size_file an _expandable_ boolean.
846 Bugzilla 1089. Patch from Heiko Schlittermann.
848 PP/08 Handle ${run} returning more data than OS pipe buffer size.
849 Bugzilla 1131. Patch from Holger Weiß.
851 PP/09 Handle IPv6 addresses with SPF.
852 Bugzilla 860. Patch from Wolfgang Breyha.
854 PP/10 GnuTLS: support TLS 1.2 & 1.1.
856 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
859 PP/11 match_* no longer expand right-hand-side by default.
860 New compile-time build option, EXPAND_LISTMATCH_RHS.
861 New expansion conditions, "inlist", "inlisti".
863 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
865 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
867 PP/14 fix log_write() format string regression from TF/03.
868 Bugzilla 1152. Patch from Dmitry Isaikin.
874 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
876 PP/02 Harmonised TLS library version reporting; only show if debugging.
877 Layout now matches that introduced for other libraries in 4.74 PP/03.
879 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
881 PP/04 New "dns_use_edns0" global option.
883 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
886 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
887 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
889 TK/01 Updated PolarSSL code to 0.14.2.
890 Bugzilla 1097. Patch from Andreas Metzler.
892 PP/07 Catch divide-by-zero in ${eval:...}.
895 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
898 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
899 format-string attack -- SECURITY: remote arbitrary code execution.
901 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
902 time unintentionally subject to list matching rules, letting the header
903 cause arbitrary Exim lookups (of items which can occur in lists, *not*
904 arbitrary string expansion). This allowed for information disclosure.
906 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
907 INT_MIN/-1 -- value coerced to INT_MAX.
913 NM/01 Workround for PCRE version dependency in version reporting
916 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
917 This fixes portability to compilers other than gcc, notably
918 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
920 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
921 makefiles for portability to HP-UX and POSIX correctness.
923 PP/01 Permit LOOKUP_foo enabling on the make command-line.
924 Also via indented variable definition in the Makefile.
925 (Debugging by Oliver Heesakkers).
927 PP/02 Restore caching of spamd results with expanded spamd_address.
928 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
930 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
931 Improves build reliability. Fix from: Frank Elsner
933 NM/02 Fix wide character breakage in the rfc2047 coding
934 Fixes bug 1064. Patch from Andrey N. Oktyabrski
936 NM/03 Allow underscore in dnslist lookups
937 Fixes bug 1026. Patch from Graeme Fowler
939 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
940 Code patches from Adam Ciarcinski of NetBSD.
942 NM/04 Fixed exiqgrep to cope with mailq missing size issue
945 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
946 is logged, to avoid truncation. Patch from John Horne.
948 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
949 Patch from Jakob Hirsch.
951 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
952 SQL string expansion failure details.
953 Patch from Andrey Oktyabrski.
955 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
956 Patch from Simon Arlott.
958 PP/09 New lookups functionality failed to compile on old gcc which rejects
959 extern declarations in function scope.
960 Patch from Oliver Fleischmann
962 PP/10 Use sig_atomic_t for flags set from signal handlers.
963 Check getgroups() return and improve debugging.
964 Fixed developed for diagnosis in bug 927 (which turned out to be
967 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
968 Patch from Mark Zealey.
970 PP/12 Bugzilla 1056: Improved spamd server selection.
971 Patch from Mark Zealey.
973 PP/13 Bugzilla 1086: Deal with maildir quota file races.
974 Based on patch from Heiko Schlittermann.
976 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
977 Patch from Uwe Doering, sign-off by Michael Haardt.
979 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
980 variable declaration deep within a block. Bug and patch from
983 PP/15 lookups-Makefile IRIX compatibilty coercion.
985 PP/16 Make DISABLE_DKIM build knob functional.
987 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
988 Patch by Simon Arlott
990 TF/03 Fix valgrind.h portability to C89 compilers that do not support
991 variable argument macros. Our copy now differs from upstream.
997 TF/01 Failure to get a lock on a hints database can have serious
998 consequences so log it to the panic log.
1000 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1001 controlled using the smtp_confirmation log selector.
1003 TF/03 Include the error message when we fail to unlink a spool file.
1005 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1006 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1007 for maintaining out-of-tree patches for some time.
1009 PP/01 Bugzilla 139: Documentation and portability issues.
1010 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1011 Handle per-OS dynamic-module compilation flags.
1013 PP/02 Let /dev/null have normal permissions.
1014 The 4.73 fixes were a little too stringent and complained about the
1015 permissions on /dev/null. Exempt it from some checks.
1016 Reported by Andreas M. Kirchwitz.
1018 PP/03 Report version information for many libraries, including
1019 Exim version information for dynamically loaded libraries. Created
1020 version.h, now support a version extension string for distributors
1021 who patch heavily. Dynamic module ABI change.
1023 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1024 privilege escalation vulnerability whereby the Exim run-time user
1025 can cause root to append content of the attacker's choosing to
1028 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1031 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1032 If dropping privileges for untrusted macros, we disabled normal logging
1033 on the basis that it would fail; for the Exim run-time user, this is not
1034 the case, and it resulted in successful deliveries going unlogged.
1035 Fixed. Reported by Andreas Metzler.
1041 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1042 only prepend for the Resent-* case. Fixes regression introduced in
1043 Exim 4.70 by NM/22 for Bugzilla 607.
1045 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1046 increasing numbers of administrators be bitten by this.
1048 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1050 PP/03 Bugzilla 994: added openssl_options main configuration option.
1052 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1054 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1056 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1058 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1059 without a peer certificate, leading to a segfault because of an
1060 assumption that peers always have certificates. Be a little more
1061 paranoid. Problem reported by Martin Tscholak.
1063 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1064 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1065 NB: ClamAV planning to remove STREAM in "middle of 2010".
1066 CL also introduces -bmalware, various -d+acl logging additions and
1067 more caution in buffer sizes.
1069 PP/09 Implemented reverse_ip expansion operator.
1071 PP/10 Bugzilla 937: provide a "debug" ACL control.
1073 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1075 PP/12 Bugzilla 973: Implement --version.
1077 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1079 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1081 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1083 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1084 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1085 ignore trailing whitespace.
1087 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1089 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1092 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1093 Notification from Dr Andrew Aitchison.
1095 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1096 ExtendedDetectionInfo response format.
1097 Notification from John Horne.
1099 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1102 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1103 XSL and documented dependency on system catalogs, with examples of how
1106 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1109 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1110 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1111 configuration file which is writeable by the Exim user or group.
1113 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1114 of configuration files to cover files specified with the -C option if
1115 they are going to be used with root privileges, not just the default
1118 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1119 option (effectively making it always true).
1121 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1122 files to be used while preserving root privileges.
1124 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1125 that rogue child processes cannot use them.
1127 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1128 run-time user, instead of root.
1130 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1131 Exim run-time user without dropping privileges.
1133 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1134 result string, instead of calling string_vformat() twice with the same
1137 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1138 for other users. Others should always drop root privileges if they use
1139 -C on the command line, even for a whitelisted configure file.
1141 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1143 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1149 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1150 $data_path, and $header_path variables; fixed documentation bugs and
1153 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1154 exipick to access non-standard spools, including the "frozen" queue
1157 NM/01 Bugzilla 965: Support mysql stored procedures.
1158 Patch from Alain Williams
1160 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1162 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1163 Patch from Andreas Metzler
1165 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1166 Patch from Kirill Miazine
1168 NM/05 Bugzilla 671: Added umask to procmail example.
1170 JJ/03 installed exipick 20100323.0, fixing doc bug
1172 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1173 directory. Notification and patch from Dan Rosenberg.
1175 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1177 TK/02 Improve log output when DKIM signing operation fails.
1179 MH/01 Treat the transport option dkim_domain as a colon separated
1180 list, not as a single string, and sign the message with each element,
1181 omitting multiple occurences of the same signer.
1183 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1184 Bugzilla 985, 986. Patch by Simon Arlott
1186 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1187 Patch by Simon Arlott
1189 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1190 MBX locking. Notification from Dan Rosenberg.
1196 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1198 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1200 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1202 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1204 NM/04 Bugzilla 915: Flag broken perl installation during build.
1210 TK/01 Added patch by Johannes Berg that expands the main option
1211 "spamd_address" if it starts with a dollar sign.
1213 TK/02 Write list of recipients to X-Envelope-Sender header when building
1214 the mbox-format spool file for content scanning (suggested by Jakob
1217 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1218 (http://www.dcc-servers.net/) support via dccifd. Activated by
1219 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1221 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1222 by Mark Daniel Reidel <mr@df.eu>.
1224 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1225 When building exim an external PCRE library is now needed -
1226 PCRE is a system library on the majority of modern systems.
1227 See entry on PCRE_LIBS in EDITME file.
1229 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1230 conversation. Added nologin parameter to request.
1231 Patch contributed by Kirill Miazine.
1233 TF/01 Do not log submission mode rewrites if they do not change the address.
1235 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1237 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1238 log files in place. Contributed by Roberto Lima.
1240 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1242 TF/03 Bugzilla 615: When checking the local_parts router precondition
1243 after a local_part_suffix or local_part_prefix option, Exim now
1244 does not use the address's named list lookup cache, since this
1245 contains cached lookups for the whole local part.
1247 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1248 Robert Millan. Documentation is in experimental-spec.txt.
1250 TF/04 Bugzilla 668: Fix parallel build (make -j).
1252 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1254 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1255 Patch provided by Jan Srzednicki.
1257 TF/05 Leading white space used to be stripped from $spam_report which
1258 wrecked the formatting. Now it is preserved.
1260 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1261 that they are available at delivery time.
1263 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1265 TF/08 TLS error reporting now respects the incoming_interface and
1266 incoming_port log selectors.
1268 TF/09 Produce a more useful error message if an SMTP transport's hosts
1269 setting expands to an empty string.
1271 NM/06 Bugzilla 744: EXPN did not work under TLS.
1272 Patch provided by Phil Pennock.
1274 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1275 Patch provided by Richard Godbee.
1277 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1278 acl_smtp_notquit, added index entry.
1280 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1281 Patch provided by Eugene Bujak.
1283 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1284 accept(). Patch provided by Maxim Dounin.
1286 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1287 Patch provided by Phil Pennock.
1289 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1291 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1292 Patch provided by Brad "anomie" Jorsch.
1294 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1295 Patch provided by Dean Brooks.
1297 TK/05 Add native DKIM support (does not depend on external libraries).
1299 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1300 Patch provided by Graeme Fowler.
1302 NM/16 Bugzilla 851: Documentation example syntax fix.
1304 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1306 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1309 NM/19 Bugzilla 745: TLS version reporting.
1310 Patch provided by Phil Pennock.
1312 NM/20 Bugzilla 167: bool: condition support.
1313 Patch provided by Phil Pennock.
1315 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1316 clients. Patch provided by Phil Pennock.
1318 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1319 Patch provided by Brad "anomie" Jorsch.
1321 NM/23 Bugzilla 687: Fix misparses in eximstats.
1322 Patch provided by Heiko Schlittermann.
1324 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1325 Patch provided by Heiko Schlittermann.
1327 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1328 plus update to original patch.
1330 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1332 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1333 Patch provided by David Brownlee.
1335 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1337 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1339 NM/30 Bugzilla 888: TLS documentation bugfixes.
1341 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1343 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1344 Unlike the original bugzilla I have changed all shell scripts in src tree.
1346 NM/33 Bugzilla 898: Transport filter timeout fix.
1347 Patch by Todd Rinaldo.
1349 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1350 Patch by Serge Demonchaux.
1352 NM/35 Bugzilla 39: Base64 decode bug fixes.
1353 Patch by Jakob Hirsch.
1355 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1357 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1359 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1361 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1367 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1368 ALT-N's libdkim that I have put here:
1369 http://duncanthrax.net/exim-experimental/
1371 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1372 were called 'true' and it seems that C99 defines that as a reserved
1373 keyword to be used with 'bool' variable types. That means you could
1374 not include C99-style headers which use bools without triggering
1375 build errors in sieve.c.
1377 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1378 as mailq or other aliases. Changed the --help handling significantly
1379 to do whats expected. exim_usage() emits usage/help information.
1381 SC/01 Added the -bylocaldomain option to eximstats.
1383 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1385 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1387 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1393 PH/01 Another patch from the Sieve maintainer.
1395 PH/02 When an IPv6 address is converted to a string for single-key lookup
1396 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1397 dots are used instead of colons so that keys in lsearch files need not
1398 contain colons. This was done some time before quoting was made available
1399 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1400 (notated using the quote facility) so as to distinguish them from IPv4
1401 keys. This meant that lookups for IP addresses in host lists did not work
1402 for iplsearch lookups.
1404 This has been fixed by arranging for IPv6 addresses to be expressed with
1405 colons if the lookup type is iplsearch. This is not incompatible, because
1406 previously such lookups could never work.
1408 The situation is now rather anomolous, since one *can* have colons in
1409 ordinary lsearch keys. However, making the change in all cases is
1410 incompatible and would probably break a number of configurations.
1412 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1415 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1416 conversion specification without a maximum field width, thereby enabling
1417 a rogue spamd server to cause a buffer overflow. While nobody in their
1418 right mind would setup Exim to query an untrusted spamd server, an
1419 attacker that gains access to a server running spamd could potentially
1420 exploit this vulnerability to run arbitrary code as the Exim user.
1422 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1423 $primary_hostname instead of what libspf2 thinks the hosts name is.
1425 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1426 a directory entry by the name of the lookup key. Previously, if a
1427 symlink pointed to a non-existing file or a file in a directory that
1428 Exim lacked permissions to read, a lookup for a key matching that
1429 symlink would fail. Now it is enough that a matching directory entry
1430 exists, symlink or not. (Bugzilla 503.)
1432 PH/03 The body_linecount and body_zerocount variables are now exported in the
1435 PH/04 Added the $dnslist_matched variable.
1437 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1438 This means they are set thereafter only if the connection becomes
1441 PH/06 Added the client_condition to authenticators so that some can be skipped
1442 by clients under certain conditions.
1444 PH/07 The error message for a badly-placed control=no_multiline_responses left
1445 "_responses" off the end of the name.
1447 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1449 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1450 (without spaces) instead of just copying the configuration text.
1452 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1454 PH/11 Added $max_received_linelength.
1456 PH/12 Added +ignore_defer and +include_defer to host lists.
1458 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1459 way in which PCRE > 7.0 is built.
1461 PH/14 Implemented queue_only_load_latch.
1463 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1464 MAIL command. The effect was to mangle the value on 64-bit systems.
1466 PH/16 Another patch from the Sieve maintainer.
1468 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1470 PH/18 If a system quota error occurred while trying to create the file for
1471 a maildir delivery, the message "Mailbox is full" was not appended to the
1472 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1473 only to a quota excession during the actual writing of the file.
1475 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1476 characters?) which causes problems in log lines. The DN values are now
1477 passed through string_printing() before being added to log lines.
1479 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1480 and InterBase are left for another time.)
1482 PH/21 Added message_body_newlines option.
1484 PH/22 Guard against possible overflow in moan_check_errorcopy().
1486 PH/23 POSIX allows open() to be a macro; guard against that.
1488 PH/24 If the recipient of an error message contained an @ in the local part
1489 (suitably quoted, of course), incorrect values were put in $domain and
1490 $local_part during the evaluation of errors_copy.
1496 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1497 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1500 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1501 issue a MAIL command.
1503 PH/02 In an ACL statement such as
1505 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1507 if a client was not listed at all, or was listed with a value other than
1508 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1509 the condition was not true (as it should be), so access was not denied.
1510 The bug was that the ! inversion was incorrectly passed on to the second
1511 item. This has been fixed.
1513 PH/03 Added additional dnslists conditions == and =& which are different from
1514 = and & when the dns lookup returns more than one IP address.
1516 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1517 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1519 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1520 FSYNC, which compiles an option called disable_fsync that allows for
1521 bypassing fsync(). The documentation is heavily laced with warnings.
1523 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1525 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1526 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1527 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1528 including adding "make clean"; (3) Added -fPIC when compiling the test
1529 dynamically loaded module, to get rid of a warning.
1531 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1532 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1533 = 0s. The bug is otherwise harmless.
1535 PH/07 There was a bug in the dovecot authenticator such that the value of
1536 $auth1 could be overwritten, and so not correctly preserved, after a
1537 successful authentication. This usually meant that the value preserved by
1538 the server_setid option was incorrect.
1540 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1542 PH/09 Installed PCRE release 7.0.
1544 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1545 run for batched SMTP input. It is now run at the start of every message
1546 in the batch. While fixing this I discovered that the process information
1547 (output by running exiwhat) was not always getting set for -bs and -bS
1548 input. This is fixed, and it now also says "batched" for BSMTP.
1550 PH/11 Added control=no_pipelining.
1552 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1553 patch, slightly modified), and move the expansion of helo_data till after
1554 the connection is made in the smtp transport (so it can use these
1557 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1559 PH/14 Added log_selector = +pid.
1561 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1563 PH/16 Add ${if forany and ${if forall.
1565 PH/17 Added dsn_from option to vary the From: line in DSNs.
1567 PH/18 Flush SMTP output before performing a callout, unless control =
1568 no_callout_flush is set.
1570 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1571 was true (the default) a successful delivery failed to delete the retry
1572 item, thus causing premature timeout of the address. The bug is now
1575 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1577 PH/21 Long custom messages for fakedefer and fakereject are now split up
1578 into multiline reponses in the same way that messages for "deny" and
1579 other ACL rejections are.
1581 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1582 with slight modification.
1584 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1585 draft, changing the syntax and factoring some duplicate code".
1587 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1588 for deliveries of the second and subsequent messages over the same SMTP
1591 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1592 ${reduce, with only minor "tidies".
1594 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1596 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1597 expansion side effects.
1599 PH/27 When a message times out after an over-quota error from an Exim-imposed
1600 quota, the bounce message says "mailbox is full". This message was not
1601 being given when it was a system quota that was exceeded. It now should
1604 MH/03 Made $recipients available in local_scan(). local_scan() already has
1605 better access to the recipient list through recipients_list[], but
1606 $recipients can be useful in postmaster-provided expansion strings.
1608 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1609 in the case of a MAIL command with additional options following the
1610 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1611 were accidentally chopped off.
1613 PH/29 SMTP synchronization checks are implemented when a command is read -
1614 there is a check that no more input is waiting when there shouldn't be
1615 any. However, for some commands, a delay in an ACL can mean that it is
1616 some time before the response is written. In this time, more input might
1617 arrive, invalidly. So now there are extra checks after an ACL has run for
1618 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1619 pipelining has not been advertised.
1621 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1623 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1624 returned in the SMTP response when smtp_return_error_details was false.
1625 This has been fixed.
1627 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1628 instead of the C I/O that was originally supplied, because problems were
1629 reported on Solaris.
1631 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1632 Exim which did not show up earlier: it was assuming that a call to
1633 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1634 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1635 was a macro that became an assignment, so it seemed to work. This has
1636 changed to a proper function call with a void return, hence the compile
1637 error. Exim's code has been fixed.
1639 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1642 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1644 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1646 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1647 because it turns out to be too big or there is a timeout) there is no
1648 "Completed" line in the log. When some messages of this type were
1649 selected by exigrep, they were listed as "not completed". Others were
1650 picked up by some special patterns. I have improved the selection
1651 criteria to be more general.
1653 PH/38 The host_find_failed option in the manualroute router can now be set
1654 to "ignore", to completely ignore a host whose IP address cannot be
1655 found. If all hosts are ignored, the behaviour is controlled by the new
1656 host_all_ignored option.
1658 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1659 homing or because of multiple MX records with /mx) generated more than
1660 one IP address, and the following item turned out to be the local host,
1661 all the secondary addresses of the first item were incorrectly removed
1662 from the list, along with the local host and any following hosts (which
1663 is what is supposed to happen).
1665 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1666 whoever called Exim into the -H file. In the case of the daemon it was
1667 behaving confusingly. When first started, it used values for whoever
1668 started the daemon, but after a SIGHUP it used the Exim user (because it
1669 calls itself on a restart). I have changed the code so that it now always
1672 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1673 message are rejected with the same error (e.g. no authentication or bad
1674 sender address), and a DATA command is nevertheless sent (as can happen
1675 with PIPELINING or a stupid MUA), the error message that was given to the
1676 RCPT commands is included in the rejection of the DATA command. This is
1677 intended to be helpful for MUAs that show only the final error to their
1680 PH/42 Another patch from the Sieve maintainer.
1682 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1683 Eximstats - Fixed some broken HTML links and added missing column headers
1685 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1686 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1688 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1690 PH/43 Yet another patch from the Sieve maintainer.
1692 PH/44 I found a way to check for a TCP/IP connection going away before sending
1693 the response to the final '.' that terminates a message, but only in the
1694 case where the client has not sent further data following the '.'
1695 (unfortunately, this is allowed). However, in many cases there won't be
1696 any further data because there won't be any more messages to send. A call
1697 to select() can be used: if it shows that the input is "ready", there is
1698 either input waiting, or the socket has been closed. An attempt to read
1699 the next input character can distinguish the two cases. Previously, Exim
1700 would have sent an OK response which the client would never have see.
1701 This could lead to message repetition. This fix should cure that, at
1702 least in a lot of common cases.
1704 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1705 advertised in response to EHLO.
1711 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1712 fixed by 4.65/MH/01 (is this a record?) are fixed:
1714 (i) An empty string was always treated as zero by the numeric comparison
1715 operators. This behaviour has been restored.
1717 (ii) It is documented that the numeric comparison operators always treat
1718 their arguments as decimal numbers. This was broken in that numbers
1719 starting with 0 were being interpreted as octal.
1721 While fixing these problems I realized that there was another issue that
1722 hadn't been noticed. Values of message_size_limit (both the global option
1723 and the transport option) were treated as octal if they started with 0.
1724 The documentation was vague. These values are now always treated as
1725 decimal, and I will make that clear in the documentation.
1731 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1732 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1735 MH/01 Don't check that the operands of numeric comparison operators are
1736 integers when their expansion is in "skipping" mode (fixes bug
1737 introduced by 4.64-PH/07).
1739 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1740 child addresses, Exim now panics and dies. Previously, because the count
1741 is held in a short int, deliveries were likely to be lost. As such a
1742 large number of recipients for a single message is ridiculous
1743 (performance will be very, very poor), I have chosen to impose a limit
1744 rather than extend the field.
1750 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1751 leftover -K file (the existence of which was triggered by #402).
1752 While we were at it, introduced process PID as part of the -K
1753 filename. This should rule out race conditions when creating
1756 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1757 processing considerably. Previous code took too long for large mails,
1758 triggering a timeout which in turn triggers #401.
1760 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1761 in the DK code in transports.c. sendfile() is not really portable,
1762 hence the _LINUX specificness.
1764 TF/01 In the add_headers option to the mail command in an Exim filter,
1765 there was a bug that Exim would claim a syntax error in any
1766 header after the first one which had an odd number of characters
1769 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1770 callout verification, Exim cached a "reject" for the entire domain. This
1771 is correct for most verifications, but it is not correct for a recipient
1772 verification with use_sender or use_postmaster set, because in that case
1773 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1774 case of MAIL FROM:<> rejection from other early rejections (e.g.
1775 rejection of HELO). When verifying a recipient using a non-null MAIL
1776 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1777 Whatever the result of the callout, the value of the domain cache is
1778 left unchanged (for any other kind of callout, getting as far as trying
1779 RCPT means that the domain itself is ok).
1781 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1784 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1785 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1786 session, the daemon ignores ECONNECT errors and logs others; it now
1787 ignores EPIPE as well.
1789 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1790 (quoted-printable decoding).
1792 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1793 later the small subsequent patch to fix an introduced bug.
1795 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1797 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1799 PH/08 An error is now given if message_size_limit is specified negative.
1801 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1802 to be given (somewhat) arbitrary names.
1804 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1807 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1808 miscellaneous code fixes
1810 PH/10 Added the log_reject_target ACL modifier to specify where to log
1813 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1814 hostname. This is wrong, because it relates to the incoming message (and
1815 probably the interface on which it is arriving) and not to the outgoing
1816 callout (which could be using a different interface). This has been
1817 changed to use the value of the helo_data option from the smtp transport
1818 instead - this is what is used when a message is actually being sent. If
1819 there is no remote transport (possible with a router that sets up host
1820 addresses), $smtp_active_hostname is used.
1822 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1823 tweaks were necessary in order to get it to work (see also 21 below):
1824 (a) The code assumed that strncpy() returns a negative number on buffer
1825 overflow, which isn't the case. Replaced with Exim's string_format()
1827 (b) There were several signed/unsigned issues. I just did the minimum
1828 hacking in of casts. There is scope for a larger refactoring.
1829 (c) The code used strcasecmp() which is not a standard C function.
1830 Replaced with Exim's strcmpic() function.
1831 (d) The code set only $1; it now sets $auth1 as well.
1832 (e) A simple test gave the error "authentication client didn't specify
1833 service in request". It would seem that Dovecot has changed its
1834 interface. Fortunately there's a specification; I followed it and
1835 changed what the client sends and it appears to be working now.
1837 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1840 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1841 address A is aliased to B and C, where B exists and C does not. Without
1842 -v the output is "A verified" because verification stops after a
1843 successful redirection if more than one address is generated. However,
1844 with -v the child addresses are also verified. Exim was outputting "A
1845 failed to verify" and then showing the successful verification for C,
1846 with its parentage. It now outputs "B failed to verify", showing B's
1847 parentage before showing the successful verification of C.
1849 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1850 look up a TXT record in a specific list after matching in a combined
1853 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1854 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1855 they consult the DNS. I had assumed they would set it the way they
1856 wanted; and indeed my experiments on Linux seem to show that in some
1857 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1858 To be on the safe side, however, I have now made the interface to
1859 host_find_byname() similar to host_find_bydns(), with an argument
1860 containing the DNS resolver options. The host_find_byname() function now
1861 sets these options at its start, just as host_find_bydns() does. The smtp
1862 transport options dns_qualify_single and dns_search_parents are passed to
1863 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
1864 of host_find_byname() use the default settings of RES_DEFNAMES
1865 (qualify_single) but not RES_DNSRCH (search_parents).
1867 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
1868 spool_read_header() do less string testing, by means of a preliminary
1869 switch on the second character of optional "-foo" lines. (This is
1870 overdue, caused by the large number of possibilities that now exist.
1871 Originally there were few.) While I was there, I also converted the
1872 str(n)cmp tests so they don't re-test the leading "-" and the first
1873 character, in the hope this might squeeze out yet more improvement.
1875 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
1876 flag allowing group syntax was set by the header_syntax check but not
1877 turned off, possible causing trouble later; (2) The flag was not being
1878 set at all for the header_verify test, causing "group"-style headers to
1879 be rejected. I have now set it in this case, and also caused header_
1880 verify to ignore an empty address taken from a group. While doing this, I
1881 came across some other cases where the code for allowing group syntax
1882 while scanning a header line wasn't quite right (mostly, not resetting
1883 the flag correctly in the right place). These bugs could have caused
1884 trouble for malformed header lines. I hope it is now all correct.
1886 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
1887 with the "reply" argument non-NULL. The code, however (which originally
1888 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
1889 but it didn't always do it. This confused somebody who was copying the
1890 code for some other use. I have removed all the tests.
1892 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
1893 feature that was used to support insecure browsers during the U.S. crypto
1894 embargo. It requires special client support, and Exim is probably the
1895 only MTA that supported it -- and would never use it because real RSA is
1896 always available. This code has been removed, because it had the bad
1897 effect of slowing Exim down by computing (never used) parameters for the
1898 RSA_EXPORT functionality.
1900 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
1901 authenticator to fail if there's a tab character in the incoming data
1902 (there should never be unless someone is messing about, as it's supposed
1903 to be base64-encoded). Also added, on Timo's advice, the "secured" option
1904 if the connection is using TLS or if the remote IP is the same as the
1905 local IP, and the "valid-client-cert option" if a client certificate has
1908 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
1909 authenticators. This can be used for authorization after authentication
1910 succeeds. (In the case of plaintext, it servers for both authentication
1913 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1914 if any retry times were supplied.
1916 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1917 connection, where there is no client IP address to check. In this
1918 situation, the verify now always succeeds.
1920 PH/25 Applied John Jetmore's -Mset patch.
1922 PH/26 Added -bem to be like -Mset, but loading a message from a file.
1924 PH/27 In a string expansion for a processed (not raw) header when multiple
1925 headers of the same name were present, leading whitespace was being
1926 removed from all of them, but trailing whitespace was being removed only
1927 from the last one. Now trailing whitespace is removed from each header
1928 before concatenation. Completely empty headers in a concatenation (as
1929 before) are ignored.
1931 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1932 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1934 PH/29 [Removed. This was a change that I later backed out, and forgot to
1935 correct the ChangeLog entry (that I had efficiently created) before
1936 committing the later change.]
1938 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1939 address errors (4xx response to RCPT) over the same connection as other
1940 messages routed to the same hosts. Such deliveries are always "forced",
1941 so retry times are not inspected. This resulted in far too many retries
1942 for the affected addresses. The effect occurred only when there were more
1943 hosts than the hosts_max_try setting in the smtp transport when it had
1944 the 4xx errors. Those hosts that it had tried were not added to the list
1945 of hosts for which the message was waiting, so if all were tried, there
1946 was no problem. Two fixes have been applied:
1948 (i) If there are any address or message errors in an SMTP delivery, none
1949 of the hosts (tried or untried) are now added to the list of hosts
1950 for which the message is waiting, so the message should not be a
1951 candidate for sending over the same connection that was used for a
1952 successful delivery of some other message. This seems entirely
1953 reasonable: after all the message is NOT "waiting for some host".
1954 This is so "obvious" that I'm not sure why it wasn't done
1955 previously. Hope I haven't missed anything, but it can't do any
1956 harm, as the worst effect is to miss an optimization.
1958 (ii) If, despite (i), such a delivery is accidentally attempted, the
1959 routing retry time is respected, so at least it doesn't keep
1960 hammering the server.
1962 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1963 in ${readsocket because some servers need this prod.
1965 PH/32 Added some extra debug output when updating a wait-xxx database.
1967 PH/33 The hint "could be header name not terminated by colon", which has been
1968 given for certain expansion errors for a long time, was not being given
1969 for the ${if def:h_colon_omitted{... case.
1971 PH/34 The spec says: "With one important exception, whenever a domain list is
1972 being scanned, $domain contains the subject domain." There was at least
1973 one case where this was not true.
1975 PH/35 The error "getsockname() failed: connection reset by peer" was being
1976 written to the panic log as well as the main log, but it isn't really
1977 panic-worthy as it just means the connection died rather early on. I have
1978 removed the panic log writing for the ECONNRESET error when getsockname()
1981 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1982 runs only) independently of the message's sender address. This meant
1983 that, if the 4xx error was in fact related to the sender, a different
1984 message to the same recipient with a different sender could confuse
1985 things. In particualar, this can happen when sending to a greylisting
1986 server, but other circumstances could also provoke similar problems.
1987 I have changed the default so that the retry time for these errors is now
1988 based a combination of the sender and recipient addresses. This change
1989 can be overridden by setting address_retry_include_sender=false in the
1992 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1993 remote server are returned as part of bounce messages. This was not
1994 happening for LMTP over a pipe (the lmtp transport), but now it is the
1995 same for both kinds of LMTP.
1997 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1998 in header lines that were in fact CNAMEs. This is no longer the case.
2000 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2001 and queue runs started by the daemon processed all messages. This has
2002 been fixed so that -R and -S can now usefully be given with -q<time>.
2004 PH/40 Import PCRE release 6.7 (fixes some bugs).
2006 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2008 PH/42 Give an error if -q is specified more than once.
2010 PH/43 Renamed the variables $interface_address and $interface_port as
2011 $received_ip_address and $received_port, to make it clear that these
2012 values apply to message reception, and not to the outgoing interface when
2013 a message is delivered. (The old names remain recognized, of course.)
2015 PH/44 There was no timeout on the connect() call when using a Unix domain
2016 socket in the ${readsocket expansion. There now is.
2018 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2019 be meaningful with "accept".
2021 SC/01 Eximstats V1.43
2022 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2024 SC/02 Eximstats V1.44
2025 Use a glob alias rather than an array ref in the generated
2026 parser. This improves both readability and performance.
2028 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2029 Collect SpamAssassin and rejection statistics.
2030 Don't display local sender or destination tables unless
2031 there is data to show.
2032 Added average volumes into the top table text output.
2034 SC/04 Eximstats V1.46
2035 Collect data on the number of addresses (recipients)
2036 as well as the number of messages.
2038 SC/05 Eximstats V1.47
2039 Added 'Message too big' to the list of mail rejection
2040 reasons (thanks to Marco Gaiarin).
2042 SC/06 Eximstats V1.48
2043 Mainlog lines which have GMT offsets and are too short to
2044 have a flag are now skipped.
2046 SC/07 Eximstats V1.49 (Alain Williams)
2047 Added the -emptyok flag.
2049 SC/08 Eximstats V1.50
2050 Fixes for obtaining the IP address from reject messages.
2052 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2053 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2054 whitesspace changes from 4.64-PH/27
2056 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2059 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2060 are found, allow negative numbers in numeric criteria)
2062 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2064 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2065 to match changes made in 4.64-PH/43
2067 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2069 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2070 "domain" is actually an IP address, and force a failure. This locks out
2071 those revolvers/nameservers that support "A-for-A" lookups, in
2072 contravention of the specifications.
2074 PH/48 When a host name was looked up from an IP address, and the subsequent
2075 forward lookup of the name timed out, the host name was left in
2076 $sender_host_name, contrary to the specification.
2078 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2079 restricted to single-key lookups, Exim was not diagnosing an error if
2080 * or *@ was used with a query-style lookup.
2082 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2084 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2085 long ago, but noone interested enough thought of it. Let's just say that
2086 the "1.1" means that there are some new functions that weren't there at
2087 some point in the past.
2089 PH/51 Error processing for expansion failure of helo_data from an smtp
2090 transport during callout processing was broken.
2092 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2093 tested/used via the -bh/-bhc/-bs options.
2095 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2096 bug, fixed in subsequent PCRE releases).
2098 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2099 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2101 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2107 SC/01 Use a glob alias rather than an array ref in eximstats generated
2108 parser. This improves both readability and performance.
2110 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2111 Don't display local sender or destination tables in eximstats unless
2112 there is data to show.
2113 Added average volumes into the eximstats top table text output.
2115 SC/03 Collect data on the number of addresses (recipients) as well
2116 as the number of messages in eximstats.
2118 TF/01 Correct an error in the documentation for the redirect router. Exim
2119 does (usually) call initgroups() when daemonizing.
2121 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2122 with consistent privilege compared to when running as a daemon.
2124 TF/03 Note in the spec that $authenticated_id is not set for local
2125 submissions from trusted users.
2127 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2128 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2130 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2131 by adding some example configuration directives to the default
2132 configuration file. A little bit of work is required to uncomment the
2133 directives and define how usernames and passwords are checked, but
2134 there is now a framework to start from.
2136 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2137 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2138 without this. I don't know how relevant this is to other LDAP libraries.
2140 PH/02 Add the verb name to the "unknown ACL verb" error.
2142 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2144 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2146 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2147 directory not expanded when it should be if an expanded home directory
2148 was set for the address (which is overridden by the transport).
2150 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2153 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2154 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2155 because it is too late at that time, and has no effect.
2157 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2158 security issue with \' (bugzilla #107). I could not use the
2159 PQescapeStringConn() function, because it needs a PGconn value as one of
2162 PH/08 When testing addresses using -bt, indicate those final addresses that
2163 are duplicates that would not cause an additional delivery. At least one
2164 person was confused, thinking that -bt output corresponded to deliveries.
2165 (Suppressing duplicates isn't a good idea as you lose the information
2166 about possibly different redirections that led to the duplicates.)
2168 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2169 systems where poll() doesn't work, in particular OS X.
2171 PH/10 Added more information to debugging output for retry time not reached.
2173 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2174 operations in malware.c.
2176 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2179 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2180 syslog_duplication set false, log lines that would normally be written
2181 both the the main log and to the reject log were not written to syslog at
2184 PH/14 In the default configuration, change the use of "message" in ACL warn
2185 statements to "add_header".
2187 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2188 not followed by a command (e.g. "seen endif").
2190 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2191 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2194 PH/17 Added extra conditions to the default value of delay_warning_condition
2198 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2199 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2200 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2203 The Auto-Submitted: and various List- headers are standardised, whereas I
2204 don't think Precedence: ever was.
2206 PH/18 Refactored debugging code in route_finduser() to show more information,
2207 in particular, the error code if getpwnam() issues one.
2209 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2210 This is apparently needed in addition to the PH/07 change above to avoid
2211 any possible encoding problems.
2213 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2214 but not after initializing Perl.
2216 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2217 output them only if debugging. By default they are written stderr,
2218 apparently, which is not desirable.
2220 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2223 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2226 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2228 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2229 authenticated or an ident call has been made. Suppress the default
2230 values for $authenticated_id and $authenticated_sender (but permit -oMai
2231 and -oMas) when testing with -bh.
2233 PH/24 Re-jigged the order of the tests in the default configuration so that the
2234 tests for valid domains and recipients precede the DNS black list and CSA
2235 tests, on the grounds that those ones are more expensive.
2237 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2238 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2239 This bug exists in every version of Exim that I still have, right back to
2242 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2243 However, an attempt to turn on case-sensitivity in a regex key by
2244 including (?-i) didn't work because the subject string was already
2245 lowercased, and the effects were non-intuitive. It turns out that a
2246 one-line patch can be used to allow (?-i) to work as expected.
2252 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2253 other effects) broke the use of negated acl sub-conditions.
2255 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2258 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2259 "Deny" causes Exim to reject the incoming connection with a 554 error.
2260 Unfortunately, if there is a major crisis, such as a disk failure,
2261 tcp-wrappers gives "deny", whereas what one would like would be some
2262 kind of temporary error. A kludge has been added to help with this.
2263 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2264 554 error is used if errno is still zero or contains ENOENT (which occurs
2265 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2268 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2270 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2271 errors. Otherwise a message that provokes a temporary error (when other
2272 messages do not) can cause a whole host to time out.
2274 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2275 addresses were routed directly to files or pipes from a redirect router.
2276 File deliveries just didn't batch; pipe deliveries might have suffered
2279 PH/06 A failure to get a lock for a hints database would erroneously always say
2280 "Failed to get write lock", even when it was really a read lock.
2282 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2283 of 0600. This has been changed to use the value of the lockfile_mode
2284 option (which defaults to 0600).
2286 PH/08 Applied small patch from the Sieve maintainer.
2288 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2289 folder from quota calculations, a direct delivery into this folder messed
2290 up the contents of the maildirsize file. This was because the regex was
2291 used only to exclude .Trash (or whatever) when the size of the mailbox
2292 was calculated. There was no check that a delivery was happening into an
2293 excluded directory. This bug has been fixed by ignoring all quota
2294 processing for deliveries into excluded directories.
2296 PH/10 Added the maildirfolder_create_regex option to appendfile.
2302 PH/01 The code for finding all the local interface addresses on a FreeBSD
2303 system running IPv6 was broken. This may well have applied to all BSD
2304 systems, as well as to others that have similar system calls. The broken
2305 code found IPv4 interfaces correctly, but gave incorrect values for the
2306 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2307 that it would not match correctly against @[] and not recognize the IPv6
2310 PH/02 The ipliteral router was not recognizing addresses of the form user@
2311 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2313 PH/03 Added disable_ipv6.
2315 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2316 decoded form, because it is most often used to construct To: headers
2317 lines in autoreplies, and the decoded form may well be syntactically
2318 invalid. However, $reply_address has leading white space removed, and all
2319 newlines turned into spaces so that the autoreply transport does not
2322 PH/05 If group was specified without a user on a router, and no group or user
2323 was specified on a transport, the group from the router was ignored.
2325 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2326 for visible compile-time settings that can be used to change these
2327 numbers, for those that want even more. Backwards compatibility with old
2328 spool files has been maintained. However, going back to a previous Exim
2329 release will lost any variables that are in spool files.
2331 PH/07 Two small changes when running in the test harness: increase delay when
2332 passing a TCP/IP connection to a new process, in case the original
2333 process has to generate a bounce, and remove special handling of
2334 127.0.0.2 (sic), which is no longer necessary.
2336 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2337 be the same on different OS.
2339 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2342 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2343 whether --show-vars was specified or not
2345 JJ/02 exipick: Added support for new ACL variable spool format introduced
2348 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2349 syntactically invalid From: or Reply-to: line, and a filter used this to
2350 generate an autoreply, and therefore failed to obtain an address for the
2351 autoreply, Exim could try to deliver to a non-existent relative file
2352 name, causing unrelated and misleading errors. What now happens is that
2353 it logs this as a hard delivery error, but does not attempt to create a
2356 PH/11 The exinext utility has a -C option for testing purposes, but although
2357 the given file was scanned by exinext itself; it wasn't being passed on
2358 when Exim was called.
2360 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2361 an end-of-file indication when reading a command response.
2363 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2364 compiled. In many other places in Exim, IPv6 addresses are always
2365 recognized, so I have changed this. It also means that IPv4 domain
2366 literals of the form [IPV4:n.n.n.n] are now always recognized.
2368 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2369 used if the router is not running as root, for example, when verifying at
2370 ACL time, or when using -bh. The debugging output from this situation was
2371 non-existent - all you got was a failure to exec. I have made two
2374 (a) Failures to set uid/gid, the current directory, or a process leader
2375 in a subprocess such as that created by queryprogram now generate
2376 suitable debugging ouput when -d is set.
2378 (b) The queryprogram router detects when it is not running as root,
2379 outputs suitable debugging information if -d is set, and then runs
2380 the subprocess without attempting to change uid/gid.
2382 PH/15 Minor change to Makefile for building test_host (undocumented testing
2385 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2386 additional section of a DNS packet that returns MX or SRV records.
2387 Instead, it always explicitly searches for A/AAAA records. This avoids
2388 major problems that occur when a DNS server includes only records of one
2389 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2390 fixed another bug: if SRV records were looked up and the corresponding
2391 address records were *not* found in the additional section, the port
2392 values from the SRV records were lost.
2394 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2395 using the correct key (the original address) when searching the retry
2396 rules in order to find which one to use for generating the retry hint.
2398 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2399 adding the default one. Similarly, if it contains a Reply-To: header, the
2400 errors_reply_to option, if set, is not used.
2402 PH/19 When calculating a retry time, Exim used to measure the "time since
2403 failure" by looking at the "first failed" field in the retry record. Now
2404 it does not use this if it is later than than the arrival time of the
2405 message. Instead it uses the arrival time. This makes for better
2406 behaviour in cases where some deliveries succeed, thus re-setting the
2407 "first failed" field. An example is a quota failure for a huge message
2408 when small messages continue to be delivered. Without this change, the
2409 "time since failure" will always be short, possible causing more frequent
2410 delivery attempts for the huge message than are intended.
2411 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2413 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2414 $1, $2, $3) because the numerical variables can be reset during some
2415 expansion items (e.g. "match"), thereby losing the authentication data.
2417 PH/21 Make -bV show the size of off_t variables so that the test suite can
2418 decide whether to run tests for quotas > 2G.
2420 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2421 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2422 filecount value is greater than 2G or if a quota value is greater than 2G
2423 on a system where the size of off_t is not greater than 4, a panic error
2426 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2427 never match. The debug and -bh output now contains an explicit error
2428 message indicating a malformed IPv4 address or mask.
2430 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2431 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2432 PH/23 above applies.
2434 PH/25 Do not write to syslog when running in the test harness. The only
2435 occasion when this arises is a failure to open the main or panic logs
2436 (for which there is an explicit test).
2438 PH/26 Added the /no_tell option to "control=freeze".
2440 PH/27 If a host name lookup failed very early in a connection, for example, if
2441 the IP address matched host_lookup and the reverse lookup yielded a name
2442 that did not have a forward lookup, an error message of the form "no IP
2443 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2444 could be logged. Now it outputs the IP address instead of "NULL".
2446 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2447 allows the sender and the authenticated sender to be set when
2448 submitting a message from within Exim. Since child_open_exim() is
2449 documented for local_scan(), the new function should be too.
2451 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2452 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2453 results in an empty string is now treated as unset.
2455 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2457 PH/31 Added sender_verify_fail logging option.
2459 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2460 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2461 tidied the source and removed it altogether.
2463 PH/33 When a queue run was abandoned because the load average was too high, a
2464 log line was always written; now it is written only if the queue_run log
2465 selector is set. In addition, the log line for abandonment now contains
2466 information about the queue run such as the pid. This is always present
2467 in "start" and "stop" lines but was omitted from the "abandon" line.
2469 PH/34 Omit spaces between a header name and the colon in the error message that
2470 is given when verify = headers_syntax fails (if there are lots of them,
2471 the message gets confusing).
2473 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2474 names, as there are now some PTR records that contain slashes. This check
2475 is only to protect against broken name servers that fall over on strange
2476 characters, so the fact that it applies to all lookups doesn't matter.
2478 PH/36 Now that the new test suite is complete, we can remove some of the
2479 special code in Exim that was needed for the old test suite. For example,
2480 sorting DNS records because real resolvers return them in an arbitrary
2481 order. The new test suite's fake resolver always returns records in the
2484 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2485 bounces) except when queue_only is set, to avoid logging races between
2486 the different processes.
2488 PH/38 Panic-die if .include specifies a non-absolute path.
2490 PH/39 A tweak to the "H" retry rule from its user.
2492 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2493 a label. They prevented compilation on older perls.
2495 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2496 a warning to be raised on newish perls.
2498 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2499 on queue. Changes to match documented behaviour of showing count of
2500 messages matching specified criteria.
2502 PH/40 Changed the default ident timeout from 30s to 5s.
2504 PH/41 Added support for the use of login_cap features, on those BSD systems
2505 that have them, for controlling the resources used by pipe deliveries.
2507 PH/42 The content-scanning code uses fopen() to create files in which to put
2508 message data. Previously it was not paying any attention to the mode of
2509 the files. Exim runs with umask(0) because the rest of the code creates
2510 files with open(), and sets the required mode explicitly. Thus, these
2511 files were ending up world-writeable. This was not a big issue, because,
2512 being within the spool directory, they were not world-accessible. I have
2513 created a function called modefopen, which takes an additional mode
2514 argument. It sets umask(777), creates the file, chmods it to the required
2515 mode, then resets the umask. All the relevant calls to fopen() in the
2516 content scanning code have been changed to use this function.
2518 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2519 to 24 hours. This avoids potential overflow problems when processing G
2520 and H retry rules. I suspect nobody ever tinkers with this value.
2522 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2524 PH/45 When the plaintext authenticator is running as a client, the server's
2525 challenges are checked to ensure they are valid base64 strings. By
2526 default, the authentication attempt is cancelled if an invalid string is
2527 received. Setting client_ignore_invalid_base64 true ignores these errors.
2528 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2529 they are received. Thus, the responses can be made to depend on the
2530 challenges. If an invalid string is ignored, an empty string is placed in
2533 PH/46 Messages that are created by the autoreply transport now contains a
2534 References: header, in accordance with RFCs 2822 and 3834.
2536 PH/47 Added authenticated_sender_force to the smtp transport.
2538 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2540 PH/49 Installed latest patch from the Sieve maintainer.
2542 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2543 was also set, the appendfile transport was unnecessarily scanning a
2544 directory of message files (e.g. for maildir delivery) to find the count
2545 of files (along with the size), even though it did not need this
2546 information. It now does the scan only if it needs to find either the
2547 size of the count of files.
2549 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2551 PH/52 Two bugs concerned with error handling when the smtp transport is
2554 (i) Exim was not creating retry information for temporary errors given
2555 for individual recipients after the DATA command when the smtp transport
2556 was used in LMTP mode. This meant that they could be retried too
2557 frequently, and not timed out correctly.
2559 (ii) Exim was setting the flag that allows error details to be returned
2560 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2561 recipients that were returned after the DATA command.
2563 PH/53 This is related to PH/52, but is more general: for any failing address,
2564 when detailed error information was permitted to be returned to the
2565 sender, but the error was temporary, then after the final timeout, only
2566 "retry timeout exceeded" was returned. Now it returns the full error as
2567 well as "retry timeout exceeded".
2569 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2570 do this, and (what is worse) MTAs that accept it.
2572 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2573 will now be deprecated.
2575 PH/56 New os.c-cygwin from the Cygwin maintainer.
2577 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2578 formats (previously only available in exim formats via -bpr, -bpru,
2579 and -bpra. Now also available in native and exiqgrep formats)
2581 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2582 with very large, slow to parse queues
2584 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2586 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2588 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2589 responses to authentication challenges, though it was showing the
2590 challenges; (ii) I've removed the CR characters from the debug output for
2593 PH/58 Allow for the insertion of a newline as well as a space when a string
2594 is turned into more than one encoded-word during RFC 2047 encoding. The
2595 Sieve code now uses this.
2597 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2598 data_4xx, lost_connection, tls_required.
2600 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2601 message was being sent as an SMTP response.
2603 PH/61 Add -l and -k options to exicyclog.
2605 PH/62 When verifying, if an address was redirected to one new address, so that
2606 verification continued, and the new address failed or deferred after
2607 having set something in $address_data, the value of $address_data was not
2608 passed back to the ACL. This was different to the case when no
2609 redirection occurred. The value is now passed back in both cases.
2611 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2612 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2613 use login_cap.h, so on its own it isn't the distinguishing feature. The
2614 new name refers directly to the setclassresources() function.
2616 PH/65 Added configuration files for NetBSD3.
2618 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2620 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2621 is preferred over IPv4.
2623 PH/68 The bounce_return_message and bounce_return_body options were not being
2624 honoured for bounces generated during the reception of non-SMTP messages.
2625 In particular, this applied to messages rejected by the ACL. This bug has
2626 been fixed. However, if bounce_return_message is true and bounce_return_
2627 body is false, the headers that are returned for a non-SMTP message
2628 include only those that have been read before the error was detected.
2629 (In the case of an ACL rejection, they have all been read.)
2631 PH/69 The HTML version of the specification is now built in a directory called
2632 spec_html instead of spec.html, because the latter looks like a path with
2633 a MIME-type, and this confuses some software.
2635 PH/70 Catch two compiler warnings in sieve.c.
2637 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2638 function verify_get_ident() calls ip_connect() to connect a socket, but
2639 if the "connect()" function timed out, ip_connect() used to close the
2640 socket. However, verify_get_ident() also closes the socket later, and in
2641 between Exim writes to the log, which may get opened at this point. When
2642 the socket was closed in ip_connect(), the log could get the same file
2643 descriptor number as the socket. This naturally causes chaos. The fix is
2644 not to close the socket in ip_connect(); the socket should be closed by
2645 the function that creates it. There was only one place in the code where
2646 this was missing, in the iplookup router, which I don't think anybody now
2647 uses, but I've fixed it anyway.
2649 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2650 well as to direct DNS lookups. Otherwise the handling of names in host
2651 lists is inconsistent and therefore confusing.
2657 PH/01 Two changes to the default runtime configuration:
2659 (1) Move the checks for relay_from_hosts and authenticated clients from
2660 after to before the (commented out) DNS black list checks.
2662 (2) Add control=submission to the relay_from_hosts and authenticated
2663 clients checks, on the grounds that messages accepted by these
2664 statements are most likely to be submissions.
2666 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2668 (1) Generate an error if the third argument for the ${prvs expansion is
2671 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2674 (3) Reset the variables that are obtained from the first argument of
2675 ${prvscheck and used in the second argument before leaving the code,
2676 because their memory is reclaimed, so using them afterwards may do
2679 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2680 one (it's much easier than Tom thought :-).
2682 (5) Because of (4), we can now allow for the use of $prvscheck_result
2683 inside the third argument.
2685 PH/03 For some reason, the default setting of PATH when running a command from
2686 a pipe transport was just "/usr/bin". I have changed it to
2689 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2690 anything to be listed in the output from -bV.
2692 PH/05 When a filter generated an autoreply, the entire To: header line was
2693 quoted in the delivery log line, like this:
2695 => >A.N.Other <ano@some.domain> <original@ddress> ...
2697 This has been changed so that it extracts the operative address. There
2698 may be more than one such address. If so, they are comma-separated, like
2701 => >ano@some.domain,ona@other.domain <original@ddress> ...
2703 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2704 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2705 not being looked up in the rDNS to get a host name, Exim was showing the
2706 IP address twice in Received: lines, even though the IP addresses were
2707 identical. For example:
2709 Received: from [1.2.3.4] (helo=[1.2.3.4])
2711 However, if the real host name was known, it was omitting the HELO data
2712 if it matched the actual IP address. This has been tidied up so that it
2713 doesn't show the same IP address twice.
2715 PH/07 When both +timestamp and +memory debugging was on, the value given by
2716 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2717 called by the debug printing, thereby overwriting the timestamp buffer.
2718 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2720 PH/08 When the original message was included in an autoreply transport, it
2721 always said "this is a copy of the message, including all the headers",
2722 even if body_only or headers_only was set. It now gives an appropriate
2725 PH/09 Applied a patch from the Sieve maintainer which:
2727 o fixes some comments
2728 o adds the (disabled) notify extension core
2729 o adds some debug output for the result of if/elsif tests
2730 o points to the current vacation draft in the documentation
2731 and documents the missing references header update
2735 o fixes a bug in processing the envelope test (when testing
2736 multiple envelope elements, the last element determinted the
2739 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2740 Electronic Mail") by including:
2742 Auto-submitted: auto-generated
2744 in the messages that it generates (bounce messages and others, such as
2745 warnings). In the case of bounce messages for non-SMTP mesages, there was
2746 also a typo: it was using "Auto_submitted" (underscore instead of
2747 hyphen). Since every message generated by Exim is necessarily in response
2748 to another message, thes have all been changed to:
2750 Auto-Submitted: auto-replied
2752 in accordance with these statements in the RFC:
2754 The auto-replied keyword:
2756 - SHOULD be used on messages sent in direct response to another
2757 message by an automatic process,
2759 - MUST NOT be used on manually-generated messages,
2761 - MAY be used on Delivery Status Notifications (DSNs) and Message
2762 Disposition Notifications (MDNs),
2764 - MUST NOT be used on messages generated by automatic or periodic
2765 processes, except for messages which are automatic responses to
2768 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2769 to the default Received: header definition.
2771 PH/12 Added log selector acl_warn_skipped (default on).
2773 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2774 variables because (a) they are in the wrong storage pool and (b) even if
2775 they were copied, it wouldn't work properly because of the caching.
2777 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2778 checking when decoding. Apparently there are clients that generate
2779 overlong encoded strings. Why am I not surprised?
2781 PH/15 If the first argument of "${if match_address" was not empty, but did not
2782 contain an "@" character, Exim crashed. Now it writes a panic log message
2783 and treats the condition as false.
2785 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2787 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2788 extension "envelope-auth". The code is finished and in agreement with
2789 other implementations, but there is no documentation so far and in fact,
2790 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2791 not changing the active code.
2793 Print executed "if" and "elsif" statements when debugging is used. This
2794 helps a great deal to understand what a filter does.
2796 Document more things not specified clearly in RFC3028. I had all this
2797 sorted out, when out of a sudden new issues came to my mind. Oops."
2799 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2802 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2803 canonical form (as documented). However, after a host name lookup from
2804 the IP address, check_host() was doing a simple string comparison with
2805 addresses acquired from the DNS when checking that the found name did
2806 have the original IP as one of its addresses. Since any found IPv6
2807 addresses are likely to be in abbreviated form, the comparison could
2808 fail. Luckily, there already exists a function for doing the comparison
2809 by converting both addresses to binary, so now that is used instead of
2810 the text comparison.
2812 PH/20 There was another similar case to PH/19, when a complete host name was
2813 given in a host list; looking up its IP address could give an abbreviated
2814 form, whereas the current host's name might or might not be abbreviated.
2815 The same fix has been applied.
2821 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2822 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2825 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2826 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2828 PH/03 Typo: missing ".o" in src/pcre/Makefile.
2830 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2831 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2832 Also, for "Auto-Submitted", treat anything other than "no" as
2833 non-personal, in accordance with RFC 3834. (Previously it treated
2834 anything starting "auto-" as non-personal.)
2836 TF/01 The control=submission/name=... option had a problem with syntax
2837 errors if the name included a slash character. The /name= option
2838 now slurps the rest of the string, so it can include any characters
2839 but it must come last in the list of options (after /sender_retain
2842 PH/05 Some modifications to the interface to the fake nameserver for the new
2850 TK/01 Added the "success_on_redirect" address verification option. See
2851 NewStuff for rationale and an example.
2853 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2855 PH/02 Patch to exigrep to allow it to work on syslog lines.
2857 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2858 fread() to skip over the body file's header line, because in Cygwin the
2859 header line is locked and is inaccessible.
2861 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
2862 co-exist for some time) to make it clear that it is the Exim ID that is
2863 referenced, not the Message-ID: header line.
2865 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
2866 string_format() function, because snprintf() does not exist on all
2869 PH/06 The use of forbid_filter_existstest now also locks out the use of the
2870 ${stat: expansion item.
2872 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
2873 protocol synchronization error", to keep the pedants happy.
2875 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
2876 well as for IRIX systems, when gcc is being used. See the host.c source
2879 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
2881 PH/10 Named domain lists were not working if used in a queue_smtp_domains
2884 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
2885 transport and to the smtp transport in LMTP mode.
2887 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
2889 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
2890 run a filter in a subprocess. This could lead to confusion in subsequent
2891 lookups in the parent process. There should also be a search_tidyup() at
2892 the end of the subprocess.
2894 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
2895 only if the host matched helo_try_verify_hosts, which caused the
2896 verification to occur when the EHLO/HELO command was issued. The ACL just
2897 tested the remembered result. Now, if a previous verification attempt has
2898 not happened, "verify = helo" does it there and then.
2900 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
2902 TK/03 Fix log output including CR from clamd.
2904 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
2905 exist provoked a memory error which could cause a segfault.
2907 PH/15 Installed PCRE 6.2
2909 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2911 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2912 of the problem. Specifically, suggested +O2 rather than +O1 for the
2915 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2917 PH/20 If a delivery was routed to a non-standard port by means of an SRV
2918 record, the port was not correctly logged when the outgoing_port log
2919 selector was set (it logged the transort's default port).
2921 PH/21 Added support for host-specific ports to manualroute, queryprogram,
2922 fallback_hosts, and "hosts" in the smtp transport.
2924 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2925 host errors such as "Connection refused".
2927 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2928 authentication with radiusclient 0.4.9:
2930 - Error returned from rc_read_config was caught wrongly
2931 - Username/password not passed on to radius server due to wrong length.
2933 The presumption is that some radiusclient API changes for 4.51/PH/17
2934 were not taken care of correctly. The code is still untested by me (my
2935 Linux distribution still has 0.3.2 of radiusclient), but it was
2936 contributed by a Radius user.
2938 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2939 expanding the "port" option of the smtp transport.
2941 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2942 while reading a MIME header. Thanks to Tom Hughes for a patch.
2944 PH/24 Include config.h inside local_scan.h so that configuration settings are
2947 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2948 that in an ACL for RCPT (for example), you can examine exactly what was
2951 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2952 commands, but it was not correctly comparing the address with the actual
2953 client host address. Thus, it would show the EHLO address in Received:
2954 header lines when this was not necessary.
2956 PH/27 Added the % operator to ${eval:}.
2958 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2959 it should be ignoring failures (because with -C, for example, it has lost
2960 privilege). It wasn't ignoring creation failures other than "already
2963 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2966 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2967 because an input line was too long, either on its own, or by virtue of
2968 too many continuations, the temporary file was not being removed, and the
2969 return code was incorrect.
2971 PH/31 Missing "BOOL" in function definition in filtertest.c.
2973 PH/32 Applied Sieve patches from the maintainer.
2975 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2977 PH/33 Added "verify = not_blind".
2979 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2980 Local/Makefile (with some defaults set). These are used in built scripts
2981 such as exicyclog, but they have never been used in the exim_install
2982 script (though there are many overriding facilities there). I have
2983 arranged that the exim_install script now takes note of these two
2986 PH/35 Installed configuration files for Dragonfly.
2988 PH/36 When a locally submitted message by a trusted user did not contain a
2989 From: header, and the sender address was obtained from -f or from an SMTP
2990 MAIL command, and the trusted user did not use -F to supply a sender
2991 name, $originator_name was incorrectly used when constructing a From:
2992 header. Furthermore, $originator_name was used for submission mode
2993 messages from external hosts without From: headers in a similar way,
2994 which is clearly wrong.
2996 PH/37 Added control=suppress_local_fixups.
2998 PH/38 When log_selector = +received_sender was set, and the addition of the
2999 sender made the log line's construction buffer exactly full, or one byte
3000 less than full, an overflow happened when the terminating "\n" was
3003 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3004 when the result of a list match is failure because a DNS lookup failed.
3006 PH/40 RM_COMMAND is now used in the building process.
3008 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3009 the "build-* directories that it finds.
3011 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3012 address in a domain literal was a prefix of an interface address.
3014 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3015 when verifying a sender address, unless rewrite_headers is false.
3017 PH/44 Wrote a long comment about why errors_to addresses are verified as
3018 recipients, not senders.
3020 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3021 the ratelimit ACL was added.
3023 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3025 PH/46 Added extra information about PostgreSQL errors to the error string.
3027 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3028 avoiding the need to install special zones in a real server. This is
3029 backwards compatible; if it can't find the fake resolver, it drops back.
3030 Thus, both old and new test suites can be run.
3032 TF/02 Added util/ratelimit.pl
3034 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3035 clock is set back in time.
3037 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3038 Candler <B.Candler@pobox.com>.
3040 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3041 OK for addresses that are the result of redirections.
3043 PH/48 A number of further additions for the benefit of the new test suite,
3044 including a fake gethostbyname() that interfaces to the fake DNS resolver
3047 TF/06 The fix for widen_domains has also been applied to qualify_single and
3048 search_parents which are the other dnslookup options that can cause
3051 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3054 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3056 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3057 service" if it encountered an MX record with an empty target hostname.
3058 The message is now "an MX or SRV record indicated no SMTP service".
3060 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3061 if the DNS of the sending site is misconfigured. This is quite a
3062 common situation. This change restores the behaviour of treating a
3063 helo verification defer as a failure.
3065 PH/51 If self=fail was set on a router, the bounce message did not include the
3066 actual error message.
3072 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3074 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3075 command itself ended in error, the underlying message about the transport
3076 filter timeout was being overwritten with the pipe command error. Now the
3077 underlying error message should be appended to the second error message.
3079 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3081 PH/02 Reduce the amount of output that "make" produces by default. Full output
3082 can still be requested.
3084 PH/03 The warning log line about a condition test deferring for a "warn" verb
3085 was being output only once per connection, rather than after each
3086 occurrence (because it was using the same function as for successful
3087 "warn" verbs). This seems wrong, so I have changed it.
3089 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3090 it should not have, which might have caused a crash in the right
3091 circumstances, but probably never did.
3093 PH/04 Installed a modified version of Tony Finch's patch to make submission
3094 mode fix the return path as well as the Sender: header line, and to
3095 add a /name= option so that you can make the user's friendly name appear
3098 TF/03 Added the control = fakedefer ACL modifier.
3100 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3101 Mark Lowes for thorough testing.
3103 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3105 TK/03 Merged latest SRS patch from Miles Wilton.
3107 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3108 with the definition in sysexits.h (which is #included earlier).
3109 Fortunately, Exim does not actually use EX_OK. The code used to try to
3110 preserve the sysexits.h value, by assumimg that macro definitions were
3111 scanned for macro replacements. I have been disabused of this notion,
3112 so now the code just undefines EX_OK before #including unistd.h.
3114 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3115 in the smtp transport. When a block could not be written in a single
3116 write() function, the timeout was being re-applied to each part-write.
3117 This seems wrong - if the receiver was accepting one byte at a time it
3118 would take for ever. The timeout is now adjusted when this happens. It
3119 doesn't have to be particularly precise.
3121 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3122 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3124 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3125 without a domain if the check to <postmaster@domain> fails.
3127 SC/01 Eximstats: added -xls and the ability to specify output files
3128 (patch written by Frank Heydlauf).
3130 SC/02 Eximstats: use FileHandles for outputing results.
3132 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3134 SC/04 Eximstats: fixed display of large numbers with -nvr option
3136 SC/05 Eximstats: fixed merging of reports with empty tables.
3138 SC/06 Eximstats: added the -include_original_destination flag
3140 SC/07 Eximstats: removed tabs and trailing whitespace.
3142 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3144 TK/06 MBOX spool code: Add real "From " MBOX separator line
3145 so the .eml file is really in mbox format (even though
3146 most programs do not really care). Patch from Alex Miller.
3148 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3149 The latter is generated from $received_to and is only set if the
3150 message has one envelope recipient. SA can use these headers,
3151 obviously out-of-the-box. Patch from Alex Miller.
3153 PH/08 The ${def test on a variable was returning false if the variable's
3154 value was "0", contrary to what the specification has always said!
3155 The result should be true unless the variable is empty.
3157 PH/09 The syntax error of a character other than { following "${if
3158 def:variable_name" (after optional whitespace) was not being diagnosed.
3159 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3160 accidental colon was present, for example, could give incorrect results.
3162 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3163 result is used (not including appendfile, where other changes are about
3166 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3167 This involved changing a lot of size variables from int to off_t. It
3168 should work with maildirs and everything.
3170 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3171 spamd dying while we are connected to it.
3173 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3176 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3177 function for BDB 4.3.
3179 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3181 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3182 the sources for PCRE have been re-arranged and more files are now
3185 PH/15 The code I had for printing potentially long long variables in PH/11
3186 above was not the best (it lost precision). The length of off_t variables
3187 is now inspected at build time, and an appropriate printing format (%ld
3188 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3189 to be "long long int" or "long int". This is needed for the internal
3190 formatting function string_vformat().
3192 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3193 the configuration file to be ":syslog", then the script "guesses" where
3194 the logs files are, rather than using the compiled in default. In our
3195 case the guess is not the same as the compiled default, so the script
3196 suddenly stopped working when I started to use syslog. The patch checks
3197 to see if log_file_path is "". If so, it attempts to read it from exim
3198 with no configuration file to get the compiled in version, before it
3199 falls back to the previous guessing code."
3201 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3202 implementing BATV in an Exim configuration. See NewStuff for the gory
3205 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3206 Makefile that are specific to HP-UX.
3208 PH/18 If the "use_postmaster" option was set for a recipient callout together
3209 with the "random" option, the postmaster address was used as the MAIL
3210 FROM address for the random test, but not for the subsequent recipient
3211 test. It is now used for both.
3213 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3214 patch removes a few documentation additions to RFC 3028, because the
3215 latest draft now contains them. It adds the new en;ascii-case comparator
3216 and a new error check for 8bit text in MIME parts. Comparator and
3217 require names are now matched exactly. I enabled the subaddress
3218 extension, but it is not well tested yet (read: it works for me)."
3220 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3221 rework some of the code of TK/09 above to avoid the hardwired use of
3222 "%lld" and "long long". Replaced the call to snprintf() with a call to
3225 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3226 records point to non-existent hosts", "retry timeout exceeded", and
3227 "retry time not reached for any host after a long failure period".
3229 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3230 experimental DomainKeys support:
3232 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3233 (2) On an error such as an illegally used "control", the wrong name for
3234 the control was given.
3236 These problems did NOT occur unless DomainKeys support was compiled.
3238 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3240 PH/24 Added ${if match_ip condition.
3242 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3243 databases so that it will be absolutely obvious if a crash occurs in the
3244 DB library. This is a regular occurrence (often caused by mis-matched
3247 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3248 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3249 picked up on a user's system that detects such things. There doesn't seem
3250 to be a gcc warning option for this - only an attribute that has to be
3251 put on the function's prototype. It seems that in Fedora Core 4 they have
3252 set this on a number of new functions. No doubt there will be more in due
3255 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3256 specify a transport. However, if an address that was verified by such a
3257 router was the subject of a callout, Exim crashed because it tried to
3258 read the rcpt_include_affixes from the non-existent transport. Now it
3259 just assumes that the setting of that option is false. This bug was
3260 introduced by 4.51/PH/31.
3262 PH/28 Changed -d+all to exclude +memory, because that information is very
3263 rarely of interest, but it makes the output a lot bigger. People tend to
3264 do -d+all out of habit.
3266 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3267 code in os-type was giving problems when libc.so lives in lib64, like on
3270 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3271 aren't the modern standard, and it seems that some systems' include files
3272 don't always have them. Exim was already checking for some of the newer
3273 ones like T_AAAA, and defining it itself. I've added checks for all the
3274 record types that Exim uses.
3276 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3277 not automatically generating a new one, as it is supposed to. This
3278 prevented TLS from working. If the file did exist, but contained invalid
3279 data, a new version was generated, as expected. It was only the case of a
3280 non-existent file that was broken.
3282 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3283 with a change in libdomainkeys > 0.64.
3285 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3286 from DNS. If the selector record carries the flag, it now has
3287 precedence over the domain-wide flag.
3289 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3291 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3292 the use of a port name that isn't defined in /etc/services) occurred, the
3293 message was deferred as in a normal delivery, and thus remained on the
3294 spool, instead of being failed because of the mua_wrapper setting. This
3295 is now fixed, and I tidied up some of the mua_wrapper messages at the
3298 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3299 the messages in a hash of arrays rather than using individual hashes.
3300 This is a bit cleaner and results in dramatic memory savings, albeit
3301 at a slight CPU cost.
3303 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3304 as requested by Marc Sherman.
3306 SC/10 Eximstats: added histograms for user specified patterns as requested
3309 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3311 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3312 fopen() in the content-scanning modules that did not already have it.
3318 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3319 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3321 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3323 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3325 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3326 to test Sieve filters that use "vacation".
3328 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3329 that changes the way the GnuTLS parameters are stored in the cache file.
3330 The new format can be generated externally. For backward compatibility,
3331 if the data in the cache doesn't make sense, Exim assumes it has read an
3332 old-format file, and it generates new data and writes a new file. This
3333 means that you can't go back to an older release without removing the
3336 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3337 work if there are any delivery delays because "one_time" forces the
3338 parent to be marked "delivered", so its unseen clone is never tried
3339 again. For this reason, Exim now forbids the simultaneous setting of
3342 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3343 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3344 entry if you want to know the details. The fix, however, neglected to
3345 consider the case when local delivery batching is involved. The test for
3346 "previously delivered" was not happening when checking to see if an
3347 address could be batched with a previous (undelivered) one; under
3348 certain circumstances this could lead to multiple deliveries to the same
3351 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3352 in its include files, and this causes problems building Exim.
3354 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3355 header_syntax) but Exim was just ignoring anything given after a slash.
3356 In particular, this caused confusion with an attempt to use "verify =
3357 reverse_host_lookup/defer_ok". An error is now given when options are
3358 supplied for verify items that do not have them. (Maybe reverse_host_
3359 lookup should have a defer_ok option, but that's a different point.)
3361 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3362 defined by RFC 821) to 2048, because there were problems with some AUTH
3363 commands, and RFC 1869 says the size should be increased for extended
3364 SMTP commands that take arguments.
3366 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3369 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3370 "unknown" error; now it says that the functionality isn't in the binary.
3372 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3373 an address' error message when a string expansion fails (syntax or
3374 whatever). Otherwise the password may appear in the log. Following change
3375 PH/42 below, there is no longer a chance of it appearing in a bounce
3378 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3380 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3381 of its addresses was ever tried. (Bugzilla bug #2.)
3383 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3384 the result incorrectly in the debug output. (It correctly added a newline
3385 to what was transported.)
3387 TF/01 Added $received_time.
3389 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3390 commented out examples of how to interface to a virus scanner and to
3391 SpamAssassin. Also added commented examples of av_scanner and
3392 spamd_address settings.
3394 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3395 and controls are allowed in which ACLs. There were a couple of minor
3396 errors. Some of the entries in the conditions table (which is a table of
3397 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3398 negation of where the condition IS allowed.
3400 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3402 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3403 header file does not have a version number, so I've had to invent a new
3404 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3405 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3406 radiusclient), but it was contributed by a Radius user.
3408 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3409 files or directories, for OpenSSL.
3411 PH/19 When an Exim process that is running as root has to create an Exim log
3412 file, it does so in a subprocess that runs as exim:exim so as to get the
3413 ownership right at creation (otherwise, other Exim processes might see
3414 the file with the wrong ownership). There was no test for failure of this
3415 fork() call, which would lead to the process getting stuck as it waited
3416 for a non-existent subprocess. Forks do occasionally fail when resources
3417 run out. I reviewed all the other calls to fork(); they all seem to check
3420 PH/20 When checking for unexpected SMTP input at connect time (before writing
3421 the banner), Exim was not dealing correctly with a non-positive return
3422 from the read() function. If the client had disconnected by this time,
3423 the result was a log entry for a synchronization error with an empty
3424 string after "input=" when read() returned zero. If read() returned -1
3425 (an event I could not check), uninitialized data bytes were printed.
3426 There were reports of junk text (parts of files, etc) appearing after
3429 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3431 PH/22 Added support for macro redefinition, and (re)definition in between
3432 driver and ACL definitions.
3434 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3435 forgetting to use the resulting value; it was using the unexpanded value.
3437 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3438 hadn't been configured. The fix is from Juergen Kreileder, who
3439 understands it better than I do:
3441 "Here's what I see happening with three configured cyrus_sasl
3442 authenticators configured (plain, login, cram-md5):
3444 On startup auth_cyrus_sasl_init() gets called for each of these.
3445 This means three calls to sasl_listmech() without a specified mech_list.
3446 => SASL tests which mechs of all available mechs actually work
3447 => three warnings about OTP not working
3448 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3450 With the patch, sasl_listmech() also gets called three times. But now
3451 SASL's mech_list option is set to the server_mech specified in the the
3452 authenticator. Or in other words, the answer from sasl_listmech()
3453 gets limited to just the mech you're testing for (which is different
3455 => the return list contains just 'plain' or 'login', 'cram-md5' or
3456 nothing depending on the value of ob->server_mech.
3458 I've just tested the patch: Authentication still works fine,
3459 unavailable mechs specified in the exim configuration are still
3460 caught, and the auth.log warnings about OTP are gone."
3462 PH/25 When debugging is enabled, the contents of the command line are added
3463 to the debugging output, even when log_selector=+arguments is not
3466 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3467 answer is "GNU", and only if the return is "GNU/something" is the answer
3470 PH/27 $acl_verify_message is now set immediately after the failure of a
3471 verification in an ACL, and so is available in subsequent modifiers. In
3472 particular, the message can be preserved by coding like this:
3474 warn !verify = sender
3475 set acl_m0 = $acl_verify_message
3477 Previously, $acl_verify_message was set only while expanding "message"
3478 and "log_message" when a very denied access.
3480 PH/28 Modified OS/os.c-Linux with
3482 -#ifndef OS_LOAD_AVERAGE
3483 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3485 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3486 nomenclature these days.)
3488 PH/29 Installed patch from the Sieve maintainer that adds the options
3489 sieve_useraddress and sieve_subaddress to the redirect router.
3491 PH/30 In these circumstances:
3492 . Two addresses routed to the same list of hosts;
3493 . First host does not offer TLS;
3494 . First host accepts first address;
3495 . First host gives temporary error to second address;
3496 . Second host offers TLS and a TLS session is established;
3497 . Second host accepts second address.
3498 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3499 and peerdn, if requested) that were in fact used only for the second
3502 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3503 attention to any local part prefix or suffix that was matched by the
3504 router that accepted the address. It now behaves in the same way as it
3505 does for delivery: the affixes are removed from the local part unless
3506 rcpt_include_affixes is set on the transport.
3508 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3509 timeout during the DATA phase of an incoming message.
3511 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3512 applied a patch sanctioned by the Sieve maintainer.
3514 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3515 the uid or gid is negative. A case of a negative gid caused this to be
3516 noticed. The fix allows for either to be negative.
3518 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3519 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3522 PH/36 The change PH/12 above was broken. Fixed it.
3524 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3525 the grounds that routing the same address twice would always produce the
3526 same answer. This might have been true once, but it is certainly no
3527 longer true now. Routing a child address may depend on the previous
3528 routing that produced that child. Some complicated redirection strategies
3529 went wrong when messages had multiple recipients, and made Exim's
3530 behaviour dependent on the order in which the addresses were given.
3532 I have moved the duplicate checking until after the routing is complete.
3533 Exim scans the addresses that are assigned to local and remote
3534 transports, and removes any duplicates. This means that more work will be
3535 done, as duplicates will always all be routed, but duplicates are
3536 presumably rare, so I don't expect this is of any significance.
3538 For deliveries to pipes, files, and autoreplies, the duplicate checking
3539 still happens during the routing process, since they are not going to be
3542 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3543 It corrects a timeout issue with spamd. This is Ian's comment: "The
3544 background is that sometimes spamd either never reads data from a
3545 connection it has accepted, or it never writes response data. The exiscan
3546 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3547 blindly assumes that writes won't block so it may never time out."
3549 PH/39 Allow G after quota size as well as K and M.
3551 PH/40 The value set for $authenticated_id in an authenticator may not contain
3552 binary zeroes or newlines because the value is written to log lines and
3553 to spool files. There was no check on this. Now the value is run through
3554 the string_printing() function so that such characters are converted to
3555 printable escape sequences.
3557 PH/41 $message_linecount is a new variable that contains the total number of
3558 lines in the message. Compare $body_linecount, which is the count for the
3561 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3562 bounce and delay warning messages, except in certain special cases, which
3565 (a) An SMTP error message from a remote host;
3566 (b) A message specified in a :fail: redirection;
3567 (c) A message specified in a "fail" command in a system filter;
3568 (d) A message specified in a FAIL return from the queryprogram router;
3569 (e) A message specified by the cannot_route_message router option.
3571 In these cases only, Exim does include the error details in bounce and
3572 warning messages. There are also a few cases where bland messages such
3573 as "unrouteable address" or "local delivery error" are given.
3575 PH/43 $value is now also set for the "else" part of a ${run expansion.
3577 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3578 being worked on, but at least Exim now implements the latest version to
3581 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3582 process to complete was treated as a delivery failure, a timeout while
3583 writing the message to the pipe was logged, but erroneously treated as a
3584 successful delivery. Such timeouts include transport filter timeouts. For
3585 consistency with the overall process timeout, these timeouts are now
3586 treated as errors, giving rise to delivery failures by default. However,
3587 there is now a new Boolean option for the pipe transport called
3588 timeout_defer, which, if set TRUE, converts the failures into defers for
3589 both kinds of timeout. A transport filter timeout is now identified in
3592 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3593 systems where "make" and "gmake" are different, calling "gmake" at top
3594 level broke things. I've arranged for the value of $(MAKE) to be passed
3595 from the Makefile to this script so that it can call the same version of
3599 A note about Exim versions 4.44 and 4.50
3600 ----------------------------------------
3602 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3603 changes of various kinds. As a consequence, a big documentation update was
3604 needed. This delayed the release for rather longer than seemed good, especially
3605 in the light of a couple of (minor) security issues. Therefore, the changes
3606 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3607 release. So 4.44 and 4.50 are in effect two different branches that both start
3610 I have left the 4.50 change log unchanged; it contains all the changes since
3611 4.43. The change log for 4.44 is below; many of its items are identical to
3612 those for 4.50. This seems to be the most sensible way to preserve the
3613 historical information.
3619 1. Minor wording change to the doc/README.SIEVE file.
3621 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3622 computation of the current number of files was incorrect.
3624 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3625 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3628 4. Give more explanation in the error message when the command for a transport
3629 filter fails to execute.
3631 5. There are several places where Exim runs a non-Exim command in a
3632 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3633 was being done only for the command run by the queryprogram router. It is
3634 now done for all such subprocesses. The other cases are: ${run, transport
3635 filters, and the commands run by the lmtp and pipe transports.
3637 6. Added CONFIGURE_GROUP build-time option.
3639 7. Some older OS have a limit of 256 on the maximum number of file
3640 descriptors. Exim was using setrlimit() to set 1000 as a large value
3641 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3642 systems. I've change it so that if it can't get 1000, it tries for 256.
3644 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3645 was an oversight, and furthermore, ever since the addition of extra
3646 controls (e.g. 4.43/32), the checks on when to allow different forms of
3647 "control" were broken. There should now be diagnostics for all cases when a
3648 control that does not make sense is encountered.
3650 9. Added the /retain_sender option to "control=submission".
3652 10. $recipients is now available in the predata ACL (oversight).
3654 11. Tidy the search cache before the fork to do a delivery from a message
3655 received from the command line. Otherwise the child will trigger a lookup
3656 failure and thereby defer the delivery if it tries to use (for example) a
3657 cached ldap connection that the parent has called unbind on.
3659 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3660 of $address_data from the recipient verification was clobbered by the
3661 sender verification.
3663 13. The value of address_data from a sender verification is now available in
3664 $sender_address_data in subsequent conditions in the ACL statement.
3666 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3668 15. Added a new option "connect=<time>" to callout options, to set a different
3671 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3672 was its contents. (It was OK if the option was not defined at all.)
3674 17. A "Completed" log line is now written for messages that are removed from
3675 the spool by the -Mrm option.
3677 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3678 information about exactly what failed.
3680 19. Added -dd to debug only the daemon process.
3682 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3683 handles timeouts, both on the server side and network timeouts. Renamed the
3684 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3686 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3687 It is now set to "smtps".
3689 22. $host_address is now set to the target address during the checking of
3690 ignore_target_hosts.
3692 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3693 being passed; this would have caused $sender_host_name to have been used if
3694 matching the list had actually called for a host name (not very likely,
3695 since this list is usually IP addresses). A host name is now passed as
3698 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3699 code that specifies a non-restarting handler (typically sigaction() in
3700 modern systems) in an attempt to fix a rare and obscure crash bug.
3702 25. Narrowed the window for a race in the daemon that could cause it to ignore
3703 SIGCHLD signals. This is not a major problem, because they are used only to
3704 wake it up if nothing else does.
3706 26. A malformed maildirsize file could cause Exim to calculate negative values
3707 for the mailbox size or file count. Odd effects could occur as a result.
3708 The maildirsize information is now recalculated if the size or filecount
3711 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3712 support for a long time. Removed HAVE_SYS_VFS_H.
3714 28. Installed the latest version of exipick from John Jetmore.
3716 29. In an address list, if the pattern was not a regular expression, an empty
3717 subject address (from a bounce message) matched only if the pattern was an
3718 empty string. Non-empty patterns were not even tested. This was the wrong
3719 because it is perfectly reasonable to use an empty address as part of a
3720 database query. An empty address is now tested by patterns that are
3721 lookups. However, all the other forms of pattern expect the subject to
3722 contain a local part and a domain, and therefore, for them, an empty
3723 address still always fails if the pattern is not itself empty.
3725 30. Exim went into a mad DNS loop when attempting to do a callout where the
3726 host was specified on an smtp transport, and looking it up yielded more
3727 than one IP address.
3729 31. Re-factored the code for checking spool and log partition space into a
3730 function that finds that data and another that does the check. The former
3731 is then used to implement four new variables: $spool_space, $log_space,
3732 $spool_inodes, and $log_inodes.
3734 32. The RFC2047 encoding function was originally intended for short strings
3735 such as real names; it was not keeping to the 75-character limit for
3736 encoded words that the RFC imposes. It now respects the limit, and
3737 generates multiple encoded words if necessary. To be on the safe side, I
3738 have increased the buffer size for the ${rfc2047: expansion operator from
3741 33. It is now permitted to omit both strings after an "if" condition; if the
3742 condition is true, the result is "true". As before, when the second string
3743 is omitted, a false condition yields an empty string. This makes it less
3744 cumbersome to write custom ACL and router conditions.
3746 34. Failure to deliver a bounce message always caused it to be frozen, even if
3747 there was an errors_to setting on the router. The errors_to setting is now
3750 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3751 canonical form (fully expanded) before being placed in
3752 $sender_host_address.
3754 36. The table in the code that translates DNS record types into text (T_A to
3755 "A" for instance) was missing entries for NS and CNAME. It is just possible
3756 that this could have caused confusion if both these types were looked up
3757 for the same domain, because the text type is used as part of Exim's
3758 per-process caching. But the chance of anyone hitting this buglet seems
3761 37. The dnsdb lookup has been extended in a number of ways.
3763 (1) There is a new type, "zns", which walks up the domain tree until it
3764 finds some nameserver records. It should be used with care.
3766 (2) There is a new type, "mxh", which is like "mx" except that it returns
3767 just the host names, not the priorities.
3769 (3) It is now possible to give a list of domains (or IP addresses) to be
3770 looked up. The behaviour when one of the lookups defers can be
3771 controlled by a keyword.
3773 (4) It is now possible to specify the separator character for use when
3774 multiple records are returned.
3776 38. The dnslists ACL condition has been extended: it is now possible to supply
3777 a list of IP addresses and/or domains to be looked up in a particular DNS
3780 39. Added log_selector=+queue_time_overall.
3782 40. When running the queue in the test harness, wait just a tad after forking a
3783 delivery process, to get repeatability of debugging output.
3785 41. Include certificate and key file names in error message when GnuTLS fails
3786 to set them up, because the GnuTLS error message doesn't include the name
3787 of the failing file when there is a problem reading it.
3789 42. Allow both -bf and -bF in the same test run.
3791 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3793 44. The "Exiscan patch" is now merged into the mainline Exim source.
3795 45. Sometimes the final signoff response after QUIT could fail to get
3796 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3797 < 0 before doing a fflush(). This bug looks as though it goes back to the
3798 introduction of TLS in release 3.20, but "sometimes" must have been rare
3799 because the tests only now provoked it.
3801 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3802 (this can affect the format of dates).
3804 47. exim_tidydb, when checking for the continued existence of a message for
3805 which it has found a message-specific retry record, was not finding
3806 messages that were in split spool directories. Consequently, it was
3807 deleting retry records that should have stayed in existence.
3809 48. Steve fixed some bugs in eximstats.
3811 49. The SPA authentication driver was not abandoning authentication and moving
3812 on to the next authenticator when an expansion was forced to fail,
3813 contradicting the general specification for all authenticators. Instead it
3814 was generating a temporary error. It now behaves as specified.
3816 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3817 (the order specifies the preference for clients). The order is now AES256,
3818 AES128, 3DES, ARCFOUR128.
3820 51. Small patch to Sieve code - explicitly set From: when generating an
3823 52. Exim crashed if a remote delivery caused a very long error message to be
3824 recorded - for instance if somebody sent an entire SpamAssassin report back
3825 as a large number of 550 error lines. This bug was coincidentally fixed by
3826 increasing the size of one of Exim's internal buffers (big_buffer) that
3827 happened as part of the Exiscan merge. However, to be on the safe side, I
3828 have made the code more robust (and fixed the comments that describe what
3831 53. Now that there can be additional text after "Completed" in log lines (if
3832 the queue_time_overall log selector is set), a one-byte patch to exigrep
3833 was needed to allow it to recognize "Completed" as not the last thing in
3836 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3837 patch that reportedly fixes this has been added. I am not expert enough to
3838 create a test for it. This is what the patch creator wrote:
3840 "I found a little strange behaviour of ldap code when working with
3841 Windows 2003 AD Domain, where users was placed in more than one
3842 Organization Units. When I tried to give exim partial DN, the exit code
3843 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
3844 But simultaneously result of request was absolutely normal ldap result,
3845 so I produce this patch..."
3847 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
3848 REFERENCE, so I have modified the code to exclude the patch when that macro
3851 55. Some experimental protocols are using DNS PTR records for new purposes. The
3852 keys for these records are domain names, not reversed IP addresses. The
3853 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3854 leaves it alone. Component reversal etc. now happens only for IP addresses.
3857 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
3859 57. Double the size of the debug message buffer (to 2048) so that more of very
3860 long debug lines gets shown.
3862 58. The exicyclog utility now does better if the number of log files to keep
3863 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
3865 59. Two changes related to the smtp_active_hostname option:
3867 (1) $smtp_active_hostname is now available as a variable.
3868 (2) The default for smtp_banner uses $smtp_active_hostname instead
3869 of $primary_hostname.
3871 60. The host_aton() function is supposed to be passed a string that is known
3872 to be a valid IP address. However, in the case of IPv6 addresses, it was
3873 not checking this. This is a hostage to fortune. Exim now panics and dies
3874 if the condition is not met. A case was found where this could be provoked
3875 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3876 components; fortuitously, this particular loophole had already been fixed
3877 by change 4.50/55 above.
3879 If there are any other similar loopholes, the new check in host_aton()
3880 itself should stop them being exploited. The report I received stated that
3881 data on the command line could provoke the exploit when Exim was running as
3882 exim, but did not say which command line option was involved. All I could
3883 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3884 running as the user.
3887 61. There was a buffer overflow vulnerability in the SPA authentication code
3888 (which came originally from the Samba project). I have added a test to the
3889 spa_base64_to_bits() function which I hope fixes it.
3892 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
3893 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
3895 63. The daemon start-up calls getloadavg() while still root for those OS that
3896 need the first call to be done as root, but it missed one case: when
3897 deliver_queue_load_max is set with deliver_drop_privilege. This is
3898 necessary for the benefit of the queue runner, because there is no re-exec
3899 when deliver_drop_privilege is set.
3901 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
3902 This has been fixed.
3904 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3905 was in use, was not putting the data itself into the right store pool;
3906 consequently, it could be overwritten for a subsequent message in the same
3907 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3910 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
3912 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
3913 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
3914 the function were treating the return as a boolean value, which happened to
3915 work because 0=false and not-0=true, but is not correct code.
3917 68. The host_aton() function was not handling scoped IPv6 addresses (those
3918 with, for example, "%eth0" on the end) correctly.
3920 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
3921 negated items (that is, ~something) in unsigned ints. Some compilers
3922 apparently mutter when there is no cast.
3924 70. If an address verification called from an ACL failed, and did not produce a
3925 user-specific message (i.e. there was only a "system" message), nothing was
3926 put in $acl_verify_message. In this situation, it now puts the system
3929 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3930 session; change 4.31/43 added the unwanted input to the log line - except
3931 that it did not do this in the start of session case. It now does.
3933 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3934 This is wrong and can cause the other end to generate a synchronization
3935 error if it is another Exim or anything else that does the synchronization
3936 check. A QUIT command is no longer sent after a timeout.
3938 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3939 during host lookups.
3941 74. The defer_ok option of callout verification was not working if it was used
3942 when verifying addresses in header lines, that is, for this case:
3944 verify = header_sender/callout=defer_ok
3946 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3947 those file descriptors could be used for SMTP connections. If anything
3948 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3949 could be sent to the SMTP client, causing chaos. The daemon now opens
3950 stdin, stdout, and stderr to /dev/null when it puts itself into the
3953 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3954 log by default. The user can override this with suitable Perl magic.
3956 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3957 the log message when discard triggers, was not working for the DATA ACL or
3958 for the non-SMTP ACL.
3960 78. Error message wording change in sieve.c.
3962 79. If smtp_accept_max_per_host was set, the number of connections could be
3963 restricted to fewer than expected, because the daemon was trying to set up
3964 a new connection before checking whether the processes handling previous
3965 connections had finished. The check for completed processes is now done
3966 earlier. On busy systems, this bug wouldn't be noticed because something
3967 else would have woken the daemon, and it would have reaped the completed
3970 80. If a message was submitted locally by a user whose login name contained one
3971 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3972 It caused a spool format error. I have fixed the spool reading code. A
3973 related problem was that the "from" clause in the Received: line became
3974 illegal because of the space(s). It is now covered by ${quote_local_part.
3976 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3979 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3981 83. Patch from Sieve maintainer to fix unterminated string problem in
3982 "vacation" handling.
3984 84. Some minor changes to the Linux configuration files to help with other
3985 OS variants using glibc.
3987 85. One more patch for Sieve to update vacation handling to latest spec.
3990 ----------------------------------------------------
3991 See the note above about the 4.44 and 4.50 releases.
3992 ----------------------------------------------------
3998 1. Change 4.43/35 introduced a bug that caused file counts to be
3999 incorrectly computed when quota_filecount was set in an appendfile
4002 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4003 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4006 3. Give more explanation in the error message when the command for a transport
4007 filter fails to execute.
4009 4. There are several places where Exim runs a non-Exim command in a
4010 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4011 was being done only for the command run by the queryprogram router. It is
4012 now done for all such subprocesses. The other cases are: ${run, transport
4013 filters, and the commands run by the lmtp and pipe transports.
4015 5. Some older OS have a limit of 256 on the maximum number of file
4016 descriptors. Exim was using setrlimit() to set 1000 as a large value
4017 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4018 systems. I've change it so that if it can't get 1000, it tries for 256.
4020 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4021 was an oversight, and furthermore, ever since the addition of extra
4022 controls (e.g. 4.43/32), the checks on when to allow different forms of
4023 "control" were broken. There should now be diagnostics for all cases when a
4024 control that does not make sense is encountered.
4026 7. $recipients is now available in the predata ACL (oversight).
4028 8. Tidy the search cache before the fork to do a delivery from a message
4029 received from the command line. Otherwise the child will trigger a lookup
4030 failure and thereby defer the delivery if it tries to use (for example) a
4031 cached ldap connection that the parent has called unbind on.
4033 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4034 of $address_data from the recipient verification was clobbered by the
4035 sender verification.
4037 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4038 was its contents. (It was OK if the option was not defined at all.)
4040 11. A "Completed" log line is now written for messages that are removed from
4041 the spool by the -Mrm option.
4043 12. $host_address is now set to the target address during the checking of
4044 ignore_target_hosts.
4046 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4047 being passed; this would have caused $sender_host_name to have been used if
4048 matching the list had actually called for a host name (not very likely,
4049 since this list is usually IP addresses). A host name is now passed as
4052 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4053 code that specifies a non-restarting handler (typically sigaction() in
4054 modern systems) in an attempt to fix a rare and obscure crash bug.
4056 15. Narrowed the window for a race in the daemon that could cause it to ignore
4057 SIGCHLD signals. This is not a major problem, because they are used only to
4058 wake it up if nothing else does.
4060 16. A malformed maildirsize file could cause Exim to calculate negative values
4061 for the mailbox size or file count. Odd effects could occur as a result.
4062 The maildirsize information is now recalculated if the size or filecount
4065 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4066 support for a long time. Removed HAVE_SYS_VFS_H.
4068 18. Updated exipick to current release from John Jetmore.
4070 19. Allow an empty sender to be matched against a lookup in an address list.
4071 Previously the only cases considered were a regular expression, or an
4074 20. Exim went into a mad DNS lookup loop when doing a callout where the
4075 host was specified on the transport, if the DNS lookup yielded more than
4078 21. The RFC2047 encoding function was originally intended for short strings
4079 such as real names; it was not keeping to the 75-character limit for
4080 encoded words that the RFC imposes. It now respects the limit, and
4081 generates multiple encoded words if necessary. To be on the safe side, I
4082 have increased the buffer size for the ${rfc2047: expansion operator from
4085 22. Failure to deliver a bounce message always caused it to be frozen, even if
4086 there was an errors_to setting on the router. The errors_to setting is now
4089 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4090 canonical form (fully expanded) before being placed in
4091 $sender_host_address.
4093 24. Updated eximstats to version 1.33
4095 25. Include certificate and key file names in error message when GnuTLS fails
4096 to set them up, because the GnuTLS error message doesn't include the name
4097 of the failing file when there is a problem reading it.
4099 26. Expand error message when OpenSSL has problems setting up cert/key files.
4102 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4103 (this can affect the format of dates).
4105 28. exim_tidydb, when checking for the continued existence of a message for
4106 which it has found a message-specific retry record, was not finding
4107 messages that were in split spool directories. Consequently, it was
4108 deleting retry records that should have stayed in existence.
4110 29. eximstats updated to version 1.35
4111 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4112 1.35 - bugfix such that pie charts by volume are generated correctly
4114 30. The SPA authentication driver was not abandoning authentication and moving
4115 on to the next authenticator when an expansion was forced to fail,
4116 contradicting the general specification for all authenticators. Instead it
4117 was generating a temporary error. It now behaves as specified.
4119 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4120 (the order specifies the preference for clients). The order is now AES256,
4121 AES128, 3DES, ARCFOUR128.
4123 31. Small patch to Sieve code - explicitly set From: when generating an
4126 32. Exim crashed if a remote delivery caused a very long error message to be
4127 recorded - for instance if somebody sent an entire SpamAssassin report back
4128 as a large number of 550 error lines. This bug was coincidentally fixed by
4129 increasing the size of one of Exim's internal buffers (big_buffer) that
4130 happened as part of the Exiscan merge. However, to be on the safe side, I
4131 have made the code more robust (and fixed the comments that describe what
4134 33. Some experimental protocols are using DNS PTR records for new purposes. The
4135 keys for these records are domain names, not reversed IP addresses. The
4136 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4137 leaves it alone. Component reversal etc. now happens only for IP addresses.
4140 34. The host_aton() function is supposed to be passed a string that is known
4141 to be a valid IP address. However, in the case of IPv6 addresses, it was
4142 not checking this. This is a hostage to fortune. Exim now panics and dies
4143 if the condition is not met. A case was found where this could be provoked
4144 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4145 components; fortuitously, this particular loophole had already been fixed
4146 by change 4.50/55 or 4.44/33 above.
4148 If there are any other similar loopholes, the new check in host_aton()
4149 itself should stop them being exploited. The report I received stated that
4150 data on the command line could provoke the exploit when Exim was running as
4151 exim, but did not say which command line option was involved. All I could
4152 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4153 running as the user.
4156 35. There was a buffer overflow vulnerability in the SPA authentication code
4157 (which came originally from the Samba project). I have added a test to the
4158 spa_base64_to_bits() function which I hope fixes it.
4161 36. The daemon start-up calls getloadavg() while still root for those OS that
4162 need the first call to be done as root, but it missed one case: when
4163 deliver_queue_load_max is set with deliver_drop_privilege. This is
4164 necessary for the benefit of the queue runner, because there is no re-exec
4165 when deliver_drop_privilege is set.
4167 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4168 was in use, was not putting the data itself into the right store pool;
4169 consequently, it could be overwritten for a subsequent message in the same
4170 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4173 38. Sometimes the final signoff response after QUIT could fail to get
4174 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4175 < 0 before doing a fflush(). This bug looks as though it goes back to the
4176 introduction of TLS in release 3.20, but "sometimes" must have been rare
4177 because the tests only now provoked it.
4183 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4184 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4185 Now it returns an integer. A number of calls were still expecting a T/F
4186 return. Fortuitously, in all cases, the tests worked in OK situations,
4187 which is the norm. However, things would have gone wrong on any write
4188 failures on the smtp file descriptor. This function is used when sending
4189 messages over SMTP and also when doing verify callouts.
4191 2. When Exim is called to do synchronous delivery of a locally submitted
4192 message (the -odf or -odi options), it no longer closes stderr before doing
4195 3. Implemented the mua_wrapper option.
4197 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4199 5. Implemented the functions header_remove(), header_testname(),
4200 header_add_at_position(), and receive_remove_recipient(), and exported them
4203 6. If an ACL "warn" statement specified the addition of headers, Exim already
4204 inserted X-ACL-Warn: at the start if there was no header name. However, it
4205 was not making this test for the second and subsequent header lines if
4206 there were newlines in the string. This meant that an invalid header could
4207 be inserted if Exim was badly configured.
4209 7. Allow an ACL "warn" statement to add header lines at the start or after all
4210 the Received: headers, as well as at the end.
4212 8. Added the rcpt_4xx retry error code.
4214 9. Added postmaster_mailfrom=xxx to callout verification option.
4216 10. Added mailfrom=xxxx to the callout verification option, for verify=
4219 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4220 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4222 12. Inserted some casts to stop certain compilers warning when using pointer
4223 differences as field lengths or precisions in printf-type calls (mostly
4224 affecting debugging statements).
4226 13. Added optional readline() support for -be (dynamically loaded).
4228 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4229 same clock tick as a message's arrival, so that its received time was the
4230 same as the "first fail" time on the retry record, and that message
4231 remained on the queue past the ultimate address timeout, every queue runner
4232 would try a delivery (because it was past the ultimate address timeout) but
4233 after another failure, the ultimate address timeout, which should have then
4234 bounced the address, did not kick in. This was a "< instead of <=" error;
4235 in most cases the first failure would have been in the next clock tick
4236 after the received time, and all would be well.
4238 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4239 being recognized when the domain list was tested by the match_domain
4240 condition in an expansion string.
4242 16. Added the ${str2b64: operator.
4244 17. Exim was always calling setrlimit() to set a large limit for the number of
4245 processes, without checking whether the existing limit was already
4246 adequate. (It did check for the limit on file descriptors.) Furthermore,
4247 errors from getrlimit() and setrlimit() were being ignored. Now they are
4248 logged to the main and panic logs, but Exim does carry on, to try to do its
4249 job under whatever limits there are.
4251 18. Imported PCRE 5.0.
4253 19. Trivial typo in log message " temporarily refused connection" (the leading
4256 20. If the log selector return_path_on_delivery was set and an address was
4257 redirected to /dev/null, the delivery process crashed because it assumed
4258 that a return path would always be set for a "successful" delivery. In this
4259 case, the whole delivery is bypassed as an optimization, and therefore no
4262 21. Internal re-arrangement: the function for sending a challenge and reading
4263 a response while authentication was assuming a zero-terminated challenge
4264 string. It's now changed to take a pointer and a length, to allow for
4265 binary data in such strings.
4267 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4269 23. Exim was not respecting finduser_retries when seeking the login of the
4270 uid under which it was called; it was always trying 10 times. (The default
4271 setting of finduser_retries is zero.) Also, it was sleeping after the final
4272 failure, which is pointless.
4274 24. Implemented tls_on_connect_ports.
4276 25. Implemented acl_smtp_predata.
4278 26. If the domain in control=submission is set empty, Exim assumes that the
4279 authenticated id is a complete email address when it generates From: or
4280 Sender: header lines.
4282 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4283 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4284 chown and chgrp in /bin and hostname in /usr/bin.
4286 28. Exim was keeping the "process log" file open after each use, just as it
4287 does for the main log. This opens the possibility of it remaining open for
4288 long periods when the USR1 signal hits a daemon. Occasional processlog
4289 errors were reported, that could have been caused by this. Anyway, it seems
4290 much more sensible not to leave this file open at all, so that is what now
4293 29. The long-running daemon process does not normally write to the log once it
4294 has entered its main loop, and it closes the log before doing so. This is
4295 so that log files can straightforwardly be renamed and moved. However,
4296 there are a couple of unusual error situations where the daemon does write
4297 log entries, and I had neglected to close the log afterwards.
4299 30. The text of an SMTP error response that was received during a remote
4300 delivery was being truncated at 512 bytes. This is too short for some of
4301 the long messages that one sometimes sees. I've increased the limit to
4304 31. It is now possible to make retry rules that apply only when a message has a
4305 specific sender, in particular, an empty sender.
4307 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4308 it possible to be selective about when SMTP synchronization is enforced.
4310 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4312 32. Implemented hosts_connection_nolog.
4314 33. Added an ACL for QUIT.
4316 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4319 35. Added mailbox_size and mailbox_filecount to appendfile.
4321 36. Added control = no_multiline_responses to ACLs.
4323 37. There was a bug in the logic of the code that waits for the clock to tick
4324 in the case where the clock went backwards by a substantial amount such
4325 that the microsecond fraction of "now" was more than the microsecond
4326 fraction of "then" (but the whole seconds number was less).
4328 38. Added support for the libradius Radius client library this is found on
4329 FreeBSD (previously only the radiusclient library was supported).
4335 1. When certain lookups returned multiple values in the form name=value, the
4336 quoting of the values was not always being done properly. Specifically:
4337 (a) If the value started with a double quote, but contained no whitespace,
4339 (b) If the value contained whitespace other than a space character (i.e.
4340 tabs or newlines or carriage returns) it was not quoted.
4341 This fix has been applied to the mysql and pgsql lookups by writing a
4342 separate quoting function and calling it from the lookup code. The fix
4343 should probably also be applied to nisplus, ibase and oracle lookups, but
4344 since I cannot test any of those, I have not disturbed their existing code.
4346 2. A hit in the callout cache for a specific address caused a log line with no
4347 reason for rejecting RCPT. Now it says "Previous (cached) callout
4348 verification failure".
4350 3. There was an off-by-one bug in the queryprogram router. An over-long
4351 return line was truncated at 256 instead of 255 characters, thereby
4352 overflowing its buffer with the terminating zero. As well as fixing this, I
4353 have increased the buffer size to 1024 (and made a note to document this).
4355 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4356 when Exim is waiting for an SMTP response from a remote server, Exim
4357 restarts its select() call on the socket, thereby resetting its timeout.
4358 This is not a problem when such interrupts are rare. Somebody set up a cron
4359 job to run exiwhat every 2 minutes, which is less than the normal select()
4360 timeout (5 or 10 minutes). This meant that the select() timeout never
4361 kicked in because it was always reset. I have fixed this by comparing the
4362 time when an interrupt arrives with the time at the start of the first call
4363 to select(). If more time than the timeout has elapsed, the interrupt is
4364 treated as a timeout.
4366 5. Some internal re-factoring in preparation for the addition of Sieve
4367 extensions (by MH). In particular, the "personal" test is moved to a
4368 separate function, and given an option for scanning Cc: and Bcc: (which is
4369 not set for Exim filters).
4371 6. When Exim created an email address using the login of the caller as the
4372 local part (e.g. when creating a From: or Sender: header line), it was not
4373 quoting the local part when it contained special characters such as @.
4375 7. Installed new OpenBSD configuration files.
4377 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4378 try to make them clearer.
4380 9. Callout options, other than the timeout value, were being ignored when
4381 verifying sender addresses in header lines. For example, when using
4383 verify = header_sender/callout=no_cache
4385 the cache was (incorrectly) being used.
4387 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4388 only the Cygwin environment.
4390 11. When return_path_on_delivery was set as a log selector, if different remote
4391 addresses in the same message used different return paths and parallel
4392 remote delivery occurred, the wrong values would sometimes be logged.
4393 (Whenever a remote delivery process finished, the return path value from
4394 the most recently started remote delivery process was logged.)
4396 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4397 lines when AUTH and/or TLS are in use. This is the "received protocol"
4398 field. Exim used to use "asmtp" for authenticated SMTP, without any
4399 indication (in the protocol name) for TLS use. Now it follows the RFC and
4400 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4401 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4402 names appear in log lines as well as in Received: header lines.
4404 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4405 extensions, and comparison tests, and to fix some bugs.
4407 14. Changes to the "personal" filter test:
4409 (1) The test was buggy in that it was just doing the equivalent of
4410 "contains" tests on header lines. For example, if a user's address was
4411 anne@some.where, the "personal" test would incorrectly be true for
4413 To: susanne@some.where
4415 This test is now done by extracting each address from the header in turn,
4416 and checking the entire address. Other tests that are part of "personal"
4417 are now done using regular expressions (for example, to check local parts
4418 of addresses in From: header lines).
4420 (2) The list of non-personal local parts in From: addresses has been
4421 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4422 taken from the Sieve specification recommendations.
4424 (3) If the message contains any header line starting with "List-" it is
4425 treated as non-personal.
4427 (4) The test for "circular" in the Subject: header line has been removed
4428 because it now seems ill-conceived.
4430 15. Minor typos in src/EDITME comments corrected.
4432 16. Installed latest exipick from John Jetmore.
4434 17. If headers_add on a router specified a text string that was too long for
4435 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4436 of string_sprintf() is now avoided.
4438 18. $message_body_size was not set (it was always zero) when running the DATA
4439 ACL and the local_scan() function.
4441 19. For the "mail" command in an Exim filter, no default was being set for
4442 the once_repeat time, causing a random time value to be used if "once" was
4443 specified. (If the value happened to be <= 0, no repeat happened.) The
4444 default is now 0s, meaning "never repeat". The "vacation" command was OK
4445 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4446 (I found it when inspecting the code).
4448 20. There is now an overall timeout for performing a callout verification. It
4449 defaults to 4 times the callout timeout, which applies to individual SMTP
4450 commands during the callout. The overall timeout applies when there is more
4451 than one host that can be tried. The timeout is checked before trying the
4452 next host. This prevents very long delays if there are a large number of
4453 hosts and all are timing out (e.g. when the network connections are timing
4454 out). The value of the overall timeout can be changed by specifying an
4455 additional sub-option for "callout", called "maxwait". For example:
4457 verify = sender/callout=5s,maxwait=20s
4459 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4460 to the end before writing, but this should make it even safer).
4462 22. Exim was forgetting that it had advertised PIPELINING for the second and
4463 subsequent messages on an SMTP connection. It was also not resetting its
4464 memory on STARTTLS and an internal HELO.
4466 23. When Exim logs an SMTP synchronization error within a session, it now
4467 records whether PIPELINING has been advertised or not.
4469 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4470 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4471 rather than long int.
4473 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4475 26. Added the never_mail option to autoreply.
4481 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4482 crash if the getsockname() call failed; this can happen if a connection is
4483 closed very soon after it is established. The problem was simply in the
4484 order in which certain operations were done, causing Exim to try to write
4485 to the SMTP stream before it had set up the file descriptor. The bug has
4486 been fixed by making things happen in the correct order.
4492 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4493 before the connection was closed, thus losing the rejection response.
4495 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4496 some early Solaris releases, but causes trouble in current releases where
4497 socklen_t is defined.
4499 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4502 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4505 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4506 that releasing the top of it at the end releases what was used for sub-
4507 expansions (unless the block got too big). However, discard this block if
4508 the first thing is a variable or header, so that we can use its block when
4509 it is dynamic (useful for very large $message_headers, for example).
4511 6. Lookups now cache *every* query, not just the most recent. A new, separate
4512 store pool is used for this. It can be recovered when all lookup caches are
4513 flushed. Lookups now release memory at the end of their result strings.
4514 This has involved some general refactoring of the lookup sources.
4516 7. Some code has been added to the store_xxx() functions to reduce the amount
4517 of flapping under certain conditions.
4519 8. log_incoming_interface used to affect only the <= reception log lines. Now
4520 it causes the local interface and port to be added to several more SMTP log
4521 lines, for example "SMTP connection from", and rejection lines.
4523 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4525 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4527 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4528 could be overwritten at the end of the current message (or the start of a
4529 new message if it was set in a HELO ACL). The value is now preserved for
4530 the duration of the SMTP connection.
4532 12. If a transport had a headers_rewrite setting, and a matching header line
4533 contained an unqualified address, that address was qualified, even if it
4534 did not match any rewriting rules. The underlying bug was that the values
4535 of the flags that permit the existence of unqualified sender and recipient
4536 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4537 non-local messages, and by -bnq for local messages) were not being
4538 preserved with the message after it was received.
4540 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4541 "next input=" as part of the text comprising the host identity instead of
4542 the correct text. The code was using the same buffer for two different
4543 strings. However, depending on which order the printing function evaluated
4544 its arguments, the bug did not always show up. Under Linux, for example, my
4545 test suite worked just fine.
4547 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4548 doesn't work with some older versions of Perl. It has been changed to "my",
4549 which in any case is probably the better facility to use.
4551 15. A really picky compiler found some instances of statements for creating
4552 error messages that either had too many or two few arguments for the format
4555 16. The size of the buffer for calls to the DNS resolver has been increased
4556 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4557 for addresses that have a lot of PTR records. This alleviates a problem; it
4558 does not fully solve it.
4560 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4561 buffer now truncates the list and logs the incident, which is the same
4562 action as happens when Exim is looking up a host name and its aliases.
4563 Previously in this situation something unpredictable would happen;
4564 sometimes it was "internal error: store_reset failed".
4566 18. If a server dropped the connection unexpectedly when an Exim client was
4567 using GnuTLS and trying to read a response, the client delivery process
4568 crashed while trying to generate an error log message.
4570 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4571 string, for example:
4573 warn message = H1: something\nH2: something
4575 the text was added as a single header line from Exim's point of view
4576 though it ended up OK in the delivered message. However, searching for the
4577 second and subsequent header lines using $h_h2: did not work. This has been
4578 fixed. Similarly, if a system filter added multiple headers in this way,
4579 the routers could not see them.
4581 20. Expanded the error message when iplsearch is called with an invalid key to
4582 suggest using net-iplsearch in a host list.
4584 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4585 ACLs are no longer actually imposed (and a message to that effect is
4588 22. If a "gecos" field in a passwd entry contained escaped characters, in
4589 particular, if it contained a \" sequence, Exim got it wrong when building
4590 a From: or a Sender: header from that name. A second bug also caused
4591 incorrect handling when an unquoted " was present following a character
4592 that needed quoting.
4594 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4595 was not being matched caselessly.
4597 24. Arranged for all hyphens in the exim.8 source to be escaped with
4600 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4601 properly. Recipient callout cache records were still being keyed to include
4602 the sender, even when use_sender was set false. This led to far more
4603 callouts that were necessary. The sender is no longer included in the key
4604 when use_sender is false.
4606 26. Added "control = submission" modifier to ACLs.
4608 27. Added the ${base62d: operator to decode base 62 numbers.
4610 28. dnsdb lookups can now access SRV records.
4612 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4613 the configuration file.
4615 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4616 (-v) mode. This makes the output for a verbose queue run more intelligible.
4618 31. Added a use_postmaster feature to recipient callouts.
4620 32. Added the $body_zerocount variable, containing the number of binary zero
4621 bytes in the message body.
4623 33. The time of last modification of the "new" subdirectory is now used as the
4624 "mailbox time last read" when there is a quota error for a maildir
4627 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4629 35. Added +ignore_unknown as a special item in host lists.
4631 36. Code for decoding IPv6 addresses in host lists is now included, even if
4632 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4633 address was recognized as an IP address, but was then not correctly decoded
4634 into binary, causing unexpected and incorrect effects when compared with
4641 1. Very minor rewording of debugging text in manualroute to say "list of
4642 hosts" instead of "hostlist".
4644 2. If verify=header_syntax was set, and a header line with an unqualified
4645 address (no domain) and a large number of spaces between the end of the
4646 name and the colon was received, the reception process suffered a buffer
4647 overflow, and (when I tested it) crashed. This was caused by some obsolete
4648 code that should have been removed. The fix is to remove it!
4650 3. When running in the test harness, delay a bit after writing a bounce
4651 message to get a bit more predictability in the log output.
4653 4. Added a call to search_tidyup() just before forking a reception process. In
4654 theory, someone could use a lookup in the expansion of smtp_accept_max_
4655 per_host which, without the tidyup, could leave open a database connection.
4657 5. Added the variables $recipient_data and $sender_data which get set from a
4658 lookup success in an ACL "recipients" or "senders" condition, or a router
4659 "senders" option, similar to $domain_data and $local_part_data.
4661 6. Moved the writing of debug_print from before to after the "senders" test
4664 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4665 problems for message scanning, either using a data ACL, or using
4666 local_scan() because the Received: header was not generated till after they
4667 were called (in order to set the time as the time of reception completion).
4668 I have revised the way this works. The header is now generated after the
4669 body is received, but before the ACL or local_scan() are called. After they
4670 are run, the timestamp in the header is updated.
4676 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4677 before starting a queue runner without re-exec. This happened only when
4678 deliver_drop_privilege was set or when the Exim user was set to root. The
4679 effect of the bug was that timeouts during subsequent deliveries caused
4680 crashes instead of being properly handled. The handler is now left at its
4681 default (and expected) setting.
4683 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4684 message, again when deliver_drop_privilege is set or Exim is run as root.
4685 The bug described in (1) was not present in this case, but the tidying up
4686 of the other signals was missing. I have made the two cases consistent.
4688 3. The ignore_target_hosts setting on a manualroute router was being ignored
4689 for hosts that were looked up using the /MX notation.
4691 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4694 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4695 operated on the sender address. After changing the $sender_address to <>
4696 for the sender address verify, Exim was re-instated it as the original
4697 (before rewriting) address, but remembering that it had rewritten it, so it
4698 wasn't rewriting it again. This bug also had the effect of breaking the
4699 sender address verification caching when the sender address was rewritten.
4701 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4702 This has been changed so that if the ip literal address matches
4703 ignore_target_hosts, the router declines.
4705 7. Added expansion conditions match_domain, match_address, and match_local_
4706 part (NOT match_host).
4708 8. The placeholder for the Received: header didn't have a length field set.
4710 9. Added code to Exim itself and to exim_lock to test for a specific race
4711 condition that could lead to file corruption when using MBX delivery. The
4712 issue is with the lockfile that is created in /tmp. If this file is removed
4713 after a process has opened it but before that process has acquired a lock,
4714 there is the potential for a second process to recreate the file and also
4715 acquire a lock. This could lead to two Exim processes writing to the file
4716 at the same time. The added code performs the same test as UW imapd; it
4717 checks after acquiring the lock that its file descriptor still refers to
4718 the same named file.
4720 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4721 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4724 11. Added the smtp_active_hostname option. If used, this will typically be made
4725 to depend on the incoming interface address. Because $interface_address is
4726 not set up until the daemon has forked a reception process, error responses
4727 that can happen earlier (such as "too many connections") no longer contain
4730 12. If an expansion in a condition on a "warn" statement fails because a lookup
4731 defers, the "warn" statement is abandoned, and the next ACL statement is
4732 processed. Previously this caused the whole ACL to be aborted.
4734 13. Added the iplsearch lookup type.
4736 14. Added ident_timeout as a log selector.
4738 15. Added tls_certificate_verified as a log selector.
4740 16. Added a global option tls_require_ciphers (compare the smtp transport
4741 option of the same name). This controls incoming TLS connections.
4743 17. I finally figured out how to make tls_require_ciphers do a similar thing
4744 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4745 before starting the TLS session.
4747 18. Tabs are now shown as \t in -bP output.
4749 19. If the log selector return_path_on_delivery was set, Exim crashed when
4750 bouncing a message because it had too many Received: header lines.
4752 20. If two routers both had headers_remove settings, and the first one included
4753 a superfluous trailing colon, the final name in the first list and the
4754 first name in the second list were incorrectly joined into one item (with a
4755 colon in the middle).
4761 1. Added -C and -D options to the exinext utility, mainly to make it easier
4762 to include in the automated testing, but these could be helpful when
4763 multiple configurations are in use.
4765 2. The exinext utility was not formatting the output nicely when there was
4766 an alternate port involved in the retry record key, nor when there was a
4767 message id as well (for retries that were specific to a specific message
4768 and a specific host). It was also confused by IPv6 addresses, because of
4769 the additional colons they contain. I have fixed the IPv4 problem, and
4770 patched it up to do a reasonable job for IPv6.
4772 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4773 delivery, the log line now contains "pipelined" if PIPELINING was used.
4775 4. An SMTP transport process used to panic and die if the bind() call to set
4776 an explicit outgoing interface failed. This has been changed; it is now
4777 treated in the same way as a connect() failure.
4779 5. A reference to $sender_host_name in the part of a conditional expansion
4780 that was being skipped was still causing a DNS lookup. This no longer
4783 6. The def: expansion condition was not recognizing references to header lines
4784 that used bh_ and bheader_.
4786 7. Added the _cache feature to named lists.
4788 8. The code for checking quota_filecount in the appendfile transport was
4789 allowing one more file than it should have been.
4791 9. For compatibility with Sendmail, the command line option
4799 and sets the incoming protocol and host name (for trusted callers). The
4800 host name and its colon can be omitted when only the protocol is to be set.
4801 Note the Exim already has two private options, -pd and -ps, that refer to
4802 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4803 "s", but I don't think that's a major issue.
4805 10. A number of refactoring changes to the code, none of which should affect
4808 (a) The number of logging options was getting close to filling up the
4809 32-bit word that was used as a bit map. I have split them into two classes:
4810 those that are passed in the argument to log_write(), and those that are
4811 only ever tested independently outside of that function. These are now in
4812 separate 32-bit words, so there is plenty of room for expansion again.
4813 There is no change in the user interface or the logging behaviour.
4815 (b) When building, for example, log lines, the code previously used a
4816 macro that called string_cat() twice, in order to add two strings. This is
4817 not really sufficiently general. Furthermore, there was one instance where
4818 it was actually wrong because one of the argument was used twice, and in
4819 one call a function was used. (As it happened, calling the function twice
4820 did not affect the overall behaviour.) The macro has been replaced by a
4821 function that can join an arbitrary number of extra strings onto a growing
4824 (c) The code for expansion conditions now uses a table and a binary chop
4825 instead of a serial search (which was left over from when there were very
4826 few conditions). Also, it now recognizes conditions like "pam" even when
4827 the relevant support is not compiled in: a suitably worded error message is
4828 given if an attempt is made to use such a condition.
4830 11. Added ${time_interval:xxxxx}.
4832 12. A bug was causing one of the ddress fields not to be passed back correctly
4833 from remote delivery subprocesses. The field in question was not being
4834 subsequently used, so this caused to problems in practice.
4836 13. Added new log selectors queue_time and deliver_time.
4838 14. Might have fixed a bug in maildirsizefile handling that threw up
4839 "unexpected character" debug warnings, and recalculated the data
4840 unnecessarily. In any case, I expanded the warning message to give more
4843 15. Added the message "Restricted characters in address" to the statements in
4844 the default ACL that block characters like @ and % in local parts.
4846 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
4847 Three changes have been made:
4849 (a) There was a serious bug; a negative response to MAIL caused the whole
4850 recipient domain to be cached as invalid, thereby blocking all messages
4851 to all local parts at the same domain, from all senders. This bug has
4852 been fixed. The domain is no longer cached after a negative response to
4853 MAIL if the sender used is not empty.
4855 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
4858 (c) A new callout option, "use_sender" has been added for people who want
4859 the modified behaviour.
4865 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
4868 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
4869 indeed breaks things for older releases.
4871 3. Added additional logging to the case where there is a problem reading data
4872 from a filter that is running in a subprocess using a pipe, in order to
4873 try to track down a specific problem.
4875 4. Testing facility fudge: when running in the test harness and attempting
4876 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
4877 getting "No route to host". Convert this to a timeout.
4879 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
4882 6. Some OS don't have socklen_t but use size_t instead. This affects the
4883 fifth argument of getsockopt() amongst other things. This is now
4884 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
4885 can be set for individual OS. I have set it for SunOS5, OSF1, and
4886 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
4887 some earlier ones do not.
4889 7. Change 4.30/15 was not doing the test caselessly.
4891 8. The standard form for an IPv6 address literal was being rejected by address
4892 parsing in, for example, MAIL and RCPT commands. An example of this kind of
4893 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4894 this, as well as the form without the "IPv6" on the front (but only when
4895 address literals are enabled, of course).
4897 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
4899 10. Exim crashed if a message with an empty sender address specified by -f
4900 encountered a router with an errors_to setting. This could be provoked only
4901 by a command such as
4905 where an empty string was supplied; "<>" did not hit this bug.
4907 11. Installed PCRE release 4.5.
4909 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
4910 remained set. It is now erased.
4912 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
4913 times from message ids (which are base 36 rather than the normal 62).
4915 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
4916 were being counted as actual protocol errors, and logged if the log
4917 selector +smtp_protocol_error was set. One cannot be perfect in this test,
4918 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
4919 and DATA following a set of rejected RCPTs do not count as protocol errors.
4920 In other words, Exim assumes they were pipelined, though this may not
4921 actually be the case. Of course, in all cases the client gets an
4922 appropriate error code.
4924 15. If a lookup fails in an ACL condition, a message about the failure may
4925 be available; it is used if testing the ACL cannot continue, because most
4926 such messages specify what the cause of the deferral is. However, some
4927 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
4928 that caused an old message to be retained and used if a later statement
4929 caused a defer, replacing the real cause of the deferral.
4931 16. If an IP address had so many PTR records that the DNS lookup buffer
4932 was not large enough to hold them, Exim could crash while trying to process
4933 the truncated data. It now detects and logs this case.
4935 17. Further to 4.21/58, another change has been made: if (and only if) the
4936 first line of a message (the first header line) ends with CRLF, a bare LF
4937 in a subsequent header line has a space inserted after it, so as not to
4938 terminate the header.
4940 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
4941 unnecessarily, used atoi() instead of strtol(), and didn't check the
4942 termination when getting file sizes from file names by regex.
4944 19. Completely re-implemented the support for maildirsize files, in the light
4945 of a number of problems with the previous contributed implementation
4946 (4.30/29). In particular:
4948 . If the quota is zero, the maildirsize file is maintained, but no quota is
4951 . If the maildir directory does not exist, it is created before any attempt
4952 to write a maildirsize file.
4954 . The quota value in the file is just a cache; if the quota is changed in
4955 the transport, the new value overrides.
4957 . A regular expression is available for excluding directories from the
4960 20. The autoreply transport checks the characters in options that define the
4961 message's headers; it allows continued headers, but it was checking with
4962 isspace() after an embedded newline instead of explicitly looking for a
4965 21. If all the "regular" hosts to which an address was routed had passed their
4966 expiry times, and had not reached their retry times, the address was
4967 bounced, even if fallback hosts were defined. Now Exim should go on to try
4970 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4971 equivalent code in the SMTP transport. Some hosts send humungous responses
4972 to HELO/EHLO, more than 1024 it seems.
4974 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4975 gives compiler warnings in some environments. I've now done it "properly",
4978 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4979 (because of problems with the built-in one) was declared to return uschar *
4980 instead of char *, causing compiler failure.
4982 25. Fixed a file descriptor leak when processing alias/forward files.
4984 26. Fixed a minor format string issue in dbfn.c.
4986 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4988 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4989 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4992 29. When a sender address is verified, it is cached, to save repeating the test
4993 when there is more than one recipient in a message. However, when the
4994 verification involves a callout, it is possible for different callout
4995 options to be set for different recipients. It is too complicated to keep
4996 track of this in the cache, so now Exim always runs a verification when a
4997 callout is required, relying on the callout cache for the optimization.
4998 The overhead is duplication of the address routing, but this should not be
5001 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5002 to be verified with callout=postmaster, and the main callout worked but the
5003 postmaster check failed, the verification correctly failed. However, if a
5004 subsequent RCPT command asked for sender verification *without* the
5005 postmaster check, incorrect caching caused this verification also to fail,
5008 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5009 it was not caching the DNS options (qualify_single, search_parents) that
5010 were used when the lookup failed. A subsequent lookup with different
5011 options therefore always gave the same answer, though there were cases
5012 where it should not have. (Example: a "domains = !$mx_any" option on a
5013 dnslookup router: the "domains" option is always processed without any
5014 widening, but the router might have qualify_single set.) Now Exim uses the
5015 cached value only when the same options are set.
5017 32. Added John Jetmore's "exipick" utility to the distribution.
5019 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5020 than a timeout (e.g. a certificate is required, and is not provided), an
5021 Exim server now closes the connection immediately. Previously it waited for
5022 the client to close - but if the client is SSL, it seems that they each
5023 wait for each other, leading to a delay before one of them times out.
5025 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5026 maintained 0.8.x compatibility because I don't think many are using it, and
5027 it is clearly obsolete.
5029 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5032 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5033 client certificate was expired. A simple patch fixes this, though I don't
5034 understand the full logic of why the verify callback is called multiple
5037 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5038 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5039 which causes problems with some clients (such as the Certicom SSL Plus
5040 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5041 disables the coutermeasure allowing Eudora to connect."
5043 38. Exim was not checking that a write() to a log file succeeded. This could
5044 lead to Bad Things if a log got too big, in particular if it hit a file
5045 size limit. Exim now panics and dies if it cannot write to a log file, just
5046 as it does if it cannot open a log file.
5048 39. Modified OS/Makefile-Linux so that it now contains
5050 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5052 The two -D definitions ensure that Exim is compiled with large file
5053 support, which makes it possible to handle log files that are bigger than
5056 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5057 instance) a domain was checked against a named list that involved a lookup,
5058 causing $domain_data to be set, then another domain was checked against the
5059 same list, then the first domain was re-checked, the value of $domain_data
5060 after the final check could be wrong. In particular, if the second check
5061 failed, it could be set empty. This bug probably also applied to
5064 41. The strip_trailing_dot option was not being applied to the address given
5065 with the -f command-line option.
5067 42. The code for reading a message's header from the spool was incrementing
5068 $received_count, but never initializing it. This meant that the value was
5069 incorrect (doubled) while delivering a message in the same process in which
5070 it was received. In the most common configuration of Exim, this never
5071 happens - a fresh exec is done - but it can happen when
5072 deliver_drop_privilege is set.
5074 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5075 it now includes up to 150 characters of the unexpected data in the log
5078 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5079 and building data strings. The size of both of these buffers was 10 000
5080 bytes - far larger than anybody would *ever* want, thought I. Needless to
5081 say, somebody hit the limit. I have increased the maximum line length to
5082 20 000 and the maximum data length of concatenated lines to 100 000. I have
5083 also fixed two bugs, because there was no checking on these buffers. Tsk,
5084 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5085 buffer is too small.
5087 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5088 lsearch lookups. Now it does.
5090 46. When parsing a route_list item in a manualroute router, a fixed-length
5091 buffer was used for the list of hosts. I made this 1024 bytes long,
5092 thinking that nobody would ever have a list of hosts that long. Wrong.
5093 Somebody had a whole pile of complicated expansion conditions, and the
5094 string was silently truncated, leading to an expansion error. It turns out
5095 that it is easier to change to an unlimited length (owing to other changes
5096 that have happened since this code was originally written) than to build
5097 structure for giving a limitation error. The length of the item that
5098 expands into the list of hosts is now unlimited.
5100 47. The lsearch lookup could not handle data where the length of text line was
5101 more than 4095 characters. Such lines were truncated, leading to shortened
5102 data being returned. It should now handle lines of any length.
5104 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5105 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5108 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5109 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5110 no longer happens in comment lines. A list of replacements is now placed
5111 at the head of all of the source files, except those whose only change is
5112 to replace PERL_COMMAND in the very first #! line.
5114 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5115 messages on the queue, with a bottom-up merge sort, using code contributed
5116 by Michael Haardt. This should make operations like -bp somewhat faster on
5117 large queues. It won't affect queue runners, except when queue_run_in_order
5120 51. Installed eximstats 1.31 in the distribution.
5122 52. Added support for SRV lookups to the dnslookup router.
5124 53. If an ACL referred to $message_body or $message_body_end, the value was not
5125 reset for any messages that followed in the same SMTP session.
5127 54. The store-handling optimization for building very long strings was not
5128 differentiating between the different store pools. I don't think this
5129 actually made any difference in practice, but I've tidied it.
5131 55. While running the routers to verify a sender address, $sender_address
5132 was still set to the sender address. This is wrong, because when routing to
5133 send a bounce to the sender, it would be empty. Therefore, I have changed
5134 it so that, while verifying a sender address, $sender_address is set to <>.
5135 (There is no change to what happens when verifying a recipient address.)
5137 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5138 target A or AAAA records (if not already returned) without resetting the
5139 qualify_single or search_parents options of the DNS resolver. These are
5140 inappropriate in this case because the targets of MX and SRV records must
5141 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5142 target that, when qualified, matched something in the local domain. These
5143 two options are now turned off when doing these lookups.
5145 57. It seems that at least some releases of Reiserfs (which does not have the
5146 concept of a fixed number of inodes) returns zero and not -1 for the
5147 number of available inodes. This interacted badly with check_spool_inodes,
5148 which assumed that -1 was the "no such thing" setting. What I have done is
5149 to check that the total number of inodes is greater than zero before doing
5150 the test of how many are available.
5152 58. When a "warn" ACL statement has a log_message modifier, the message is
5153 remembered, and not repeated. This is to avoid a lot of repetition when a
5154 message has many recipients that cause the same warning to be written.
5155 Howewer, Exim was preserving the list of already written lines for an
5156 entire SMTP session, which doesn't seem right. The memory is now reset if a
5157 new message is started.
5159 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5160 debugging output unless log_rewrite was also set.
5162 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5163 of (int)(handle) when we know that handle contains (void *)(-1).
5165 61. The Exim daemon panic-logs an error return when it closes the incoming
5166 connection. However "connection reset by peer" seems to be common, and
5167 isn't really an error worthy of noting specially, so that particular error
5170 62. When Exim is trying to find all the local interfaces, it used to panic and
5171 die if the ioctl to get the interface flags failed. However, it seems that
5172 on at least one OS (Solaris 9) it is possible to have an interface that is
5173 included in the list of interfaces, but for which you get a failure error
5174 for this call. This happens when the interface is not "plumbed" into a
5175 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5176 failure of the "get flags" call assumes that the interface is down.
5178 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5179 makes life easier for people who are doing arithmetic on fields extracted
5180 from dates, where you often get leading zeros that should not be
5181 interpreted as octal.
5183 64. Added qualify_domain to the redirect router, to override the global
5186 65. If a pathologically long header line contained very many addresses (the
5187 report of this problem mentioned 10 000) and each of them was rewritten,
5188 Exim could use up a very large amount of memory. (It kept on making new
5189 copies of the header line as it rewrote, and never released the old ones.)
5190 At the expense of a bit more processing, the header rewriting function has
5191 been changed so that it no longer eats memory in this way.
5193 66. The generation of the Received: header has been moved from the time that a
5194 message starts to be received, to the time that it finishes. The timestamp
5195 in the Received: header should now be very close to that of the <= log
5196 line. There are two side-effects of this change:
5198 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5199 logged header lines no longer include the local Received: line, because
5200 it has not yet been created. The same applies to a copy of the message
5201 that is returned to a non-SMTP sender when a message is rejected.
5203 (b) When a filter file is tested using -bf, no additional Received: header
5204 is added to the test message. After some thought, I decided that this
5207 This change does not affect the value of $received_for. It is still set
5208 after address rewriting, but before local_scan() is called.
5210 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5212 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5213 gave an unhelpful panic error message, and a defer error. I have managed to
5214 change this behaviour so that it now rejects any supplied certificate,
5215 which seems right, as the list of acceptable certificates is empty.
5217 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5218 gave an unhelpful defer error. I have not managed to make this reject any
5219 supplied certificates, but the error message it gives is "no certificate
5220 supplied", which is not helpful.
5222 70. exigrep's output now also includes lines that are not associated with any
5223 message, but which match the given pattern. Implemented by a patch from
5224 Martin Sluka, which also tidied up the Perl a bit.
5226 71. Recipient callout verification, like sender verification, was using <> in
5227 the MAIL FROM command. This isn't really the right thing, since the actual
5228 sender may affect whether the remote host accepts the recipient or not. I
5229 have changed it to use the actual sender in the callout; this means that
5230 the cache record is now keyed on a recipient/sender pair, not just the
5231 recipient address. There doesn't seem to be a real danger of callout loops,
5232 since a callout by the remote host to check the sender would use <>.
5233 [SEE ABOVE: changed after hitting problems.]
5235 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5236 temporary errors. However, in the case of such a code being given after
5237 the end of a data transmission (i.e. after ".") Exim was failing to write
5238 a retry record for the message. (Yes, there was some broken host that was
5239 actually sending 8xx at this point.)
5241 73. An unknown lookup type in a host list could cause Exim to panic-die when
5242 the list was checked. (An example that provoked this was putting <; in the
5243 middle of a list instead of at the start.) If this happened during a DATA
5244 ACL check, a -D file could be left lying around. This kind of configuration
5245 error no longer causes Exim to die; instead it causes a defer errror. The
5246 incident is still logged to the main and panic logs.
5248 74. Buglet left over from Exim 3 conversion. The message "too many messages
5249 in one connection" was written to the rejectlog but not the mainlog, except
5250 when address rewriting (yes!) was being logged.
5252 75. Added write_rejectlog option.
5254 76. When a system filter was run not as root (that is, when system_filter_user
5255 was set), the values of the $n variables were not being returned to the
5256 main process; thus, they were not subsequently available in the $sn
5259 77. Added +return_path_on_delivery log selector.
5261 78. A connection timeout was being treated differently from recipients deferred
5262 when testing hosts_max_try with a message that was older than the host's
5263 retry timeout. (The host should not be counted, thus allowing all hosts to
5264 be tried at least once before bouncing.) This may have been the cause of an
5265 occasionally reported bug whereby a message would remain on the queue
5266 longer than the retry timeout, but would be bounced if a delivery was
5267 forced. I say "may" because I never totally pinned down the problem;
5268 setting up timeout/retry tests is difficult. See also the next item.
5270 79. The ultimate address timeout was not being applied to errors that involved
5271 a combination of host plus message (for example, a timeout on a MAIL
5272 command). When an address resolved to a number of possible hosts, and they
5273 were not all tried for each delivery (e.g. because of hosts_max_try), a
5274 message could remain on the queue longer than the retry timeout.
5276 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5279 81. Fixed an obscure SMTP outgoing bug which required at least the following
5280 conditions: (a) there was another message waiting for the same server;
5281 (b) the server returned 5xx to all RCPT commands in the first message so
5282 that the message was not completed; (c) the server dropped the connection
5283 or gave a negative response to the RSET that Exim sends to abort the
5284 transaction. The observed case was a dropped connection after DATA that had
5285 been sent in pipelining mode. That is, the server had advertised PIPELINING
5286 but was not implementing it correctly. The effect of the bug was incorrect
5287 behaviour, such as trying another host, and this could lead to a crash.
5293 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5294 and daemon.c were passed as pointers to ints; they should have been
5295 pointers to socklen_t variables (which are typically unsigned ints).
5297 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5300 3. Fixed a really odd bug that affected only the testing scheme; patching a
5301 certain fixed string in the binary changed the value of another string that
5302 happened to be identical to the end of the original first string.
5304 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5305 name", it returns that address as the IP address. On some operating
5306 systems (e.g. Solaris), it also passes back the IP address string as the
5307 "host name". However, on others (e.g. Linux), it passes back an empty
5308 string. Exim wasn't checking for this, and was changing the host name to an
5309 empty string, assuming it had been canonicized.
5311 5. Although rare, it is permitted to have more than one PTR record for a given
5312 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5313 all the names associated with an address, because they do in Solaris.
5314 However, it seems that they do not in Linux for data that comes from the
5315 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5316 I found this out when I moved to a new Linux workstation and tried to run
5317 the Exim test suite.
5319 To get round this problem I have changed the code so that it now does its
5320 own call to the DNS to look up PTR records when searching for a host name.
5321 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5322 addresses that are only in /etc/hosts are still found.
5324 This behaviour is, however, controlled by an option called host_lookup_
5325 order, which defaults to "bydns:byaddr". If people want to use the other
5326 order, or indeed, just use one or the other means of lookup, they can
5327 specify it in this variable.
5329 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5330 some operating systems, this comes back from gethostbyaddr() as an empty
5331 string, and this is what Exim used to test for. However, it seems that in
5332 other systems, "." is yielded. Exim now tests for this case too.
5334 7. The values of check_spool_space and check_log_space are now held internally
5335 as a number of kilobytes instead of an absolute number of bytes. If a
5336 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5337 kilobyte. This means that much larger values can be stored.
5339 8. Exim monitor: an attempt to get the action menu when not actually pointing
5340 at a message produces an empty menu entitled "No message selected". This
5341 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5342 no entries in it ("Shell widget menu has zero width and/or height"). So I
5343 have added a single, blank menu entry in this case.
5345 9. Added ${quote_local_part.
5347 10. MIME decoding is now applied to the contents of Subject: header lines when
5350 11. Now that a reference to $sender_host_address automatically causes a reverse
5351 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5352 host lookup before query-style lookups in lists that might use this
5353 variable. This has therefore been abolished, and the "net-" prefix is no
5354 longer necessary for query-style lookups.
5356 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5357 have been a typo for LFLAGS, so it has been changed.
5359 13. The install script calls Exim with "-C /dev/null" in order to find the
5360 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5361 to be output. Howeve, since Exim outputs its version number before the
5362 error, it didn't break the script. It just looked ugly. I fixed this by
5363 always allowing "-C /dev/null" if the caller is root.
5365 14. Ignore overlarge ACL variable number when reading spool file - insurance
5366 against a later release with more variables having written the file.
5368 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5369 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5370 this, as well as the form without the "IPv6" on the front.
5372 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5373 OS/Makefile-Darwin file.
5375 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5376 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5378 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5379 variable. However, some OS use ints for inodes. I've added cast to long int
5380 to get rid of the compiler warning.
5382 19. I had forgotten to lock out "/../" in configuration file names when
5383 ALT_CONFIG_PREFIX was set.
5385 20. Routers used for verification do not need to specify transports. However,
5386 if such a router generated a host list, and callout was configured, Exim
5387 crashed, because it could not find a port number from the (non-existent)
5388 transport. It now assumes port 25 in this circumstance.
5390 21. Added the -t option to exigrep.
5392 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5393 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5394 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5395 them is set, LOOKUP_LSEARCH is forced.
5397 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5399 24. Added sender and host information to the "rejected by local_scan()" log
5400 line; previously there was no indication of these.
5402 25. Added .include_if_exists.
5404 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5405 Linux. It turns out that not all file systems support this. Apparently some
5406 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5407 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5408 not supported on the file descriptor, is now ignored when Exim is trying to
5409 sync a directory. This applies only to Linux.
5411 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5413 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5414 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5417 29. Added support for maildirsize files from supplied patch (modified a bit).
5419 30. The use of :fail: followed by an empty string could lead Exim to respond to
5420 sender verification failures with (e.g.):
5422 550 Verification failed for <xxx>
5423 550 Sender verify failed
5425 where the first response line was missing the '-' that indicates it is not
5426 the final line of the response.
5428 31. The loop for finding the name of the user that called Exim had a hardwired
5429 limit of 10; it now uses the value of finduser_retries, which is used for
5430 all other user lookups.
5432 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5435 33. Exim was neglecting to zero errno before one call of strtol() when
5436 expanding a string and expecting an integer value. On some systems this
5437 resulted in spurious "integer overflow" errors. Also, it was casting the
5438 result into an int without checking.
5440 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5441 did not work. The code looks as if it has *never* worked, though it appears
5442 to have been documented since at least releast 1.62. I have made it work.
5444 35. The "timeout_DNS" error in retry rules, also documented since at least
5445 1.62, also never worked. As it isn't clear exactly what this means, and
5446 clearly it isn't a major issue, I have abolished the feature by treating it
5447 as "timeout", and writing a warning to the main and panic logs.
5449 36. The display of retry rules for -brt wasn't always showing the error code
5452 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5453 timeout_connect_A, timeout_connect_MX.
5455 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5456 to the empty sender.
5458 39. The daemon was not analysing the content of -oX till after it had closed
5459 stderr and disconnected from the controlling terminal. This meant that any
5460 syntax errors were only noted on the panic log, and the return code from
5461 the command was 0. By re-arranging the code a little, I've made the
5462 decoding happen first, so such errors now appear on stderr, and the return
5463 code is 1. However, the actual setting up of the sockets still happens in
5464 the disconnected process, so errors there are still only recorded on the
5467 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5468 connections (as happens on some IP stacks) was logged at start up time as
5469 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5470 it from "IPv6 and IPv4", which means that two separate sockets are being
5473 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5474 says whether AF_INET or AF_INET6 was passed as an argument.
5476 42. Exiwhat output was messed up when time zones were included in log
5479 43. Exiwhat now gives more information about the daemon's listening ports,
5480 and whether -tls-on-connect was used.
5482 44. The "port" option of the smtp transport is now expanded.
5484 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5485 silently ignored. Now an error message is written to the main and panic
5488 46. There's a new ACL modifier called "logwrite" which writes to a log file
5489 as soon as it is encountered.
5491 47. Added $local_user_uid and $local_user_gid at routing time.
5493 48. Exim crashed when trying to verify a sender address that was being
5496 49. Exim was recognizing only a space character after ".include". It now also
5497 recognizes a tab character.
5499 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5500 extracting the relevant information from the specification. The man page no
5501 longer contains scrambled data for the -d option, and I've added a section
5502 at the front about calling Exim under different names.
5504 51. Added "extra_headers" argument to the "mail" command in filter files.
5506 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5509 53. Installed eximstats 1.29.
5511 54. Added transport_filter_timeout as a generic transport option.
5513 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5514 Cc: header lines. This was required by RFC 822, but it not required by RFC
5517 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5518 incoming messages that did not have them. Now it does so only if the
5519 message originates locally, that is, if there is no associated remote host
5520 address. When Resent- header lines are present, this applies to the Resent-
5521 lines rather than the non-Resent- lines.
5523 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5524 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5526 58. Messages for configuration errors now include the name of the main
5527 configuration file - useful now that there may be more than one file in a
5528 list (.included file names were always shown).
5530 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5531 for those rare installations that do not start the daemon as root or run it
5532 setuid root. I've cut out the call to initgroups() if the daemon is not
5535 60. The Exim user and group can now be bound into the binary as text strings
5536 that are looked up at the start of Exim's processing.
5538 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5540 62. Added $mailstore_basename variable.
5542 63. Installed patch to sieve.c from Michael Haardt.
5544 64. When Exim failed to open the panic log after failing to open the main log,
5545 the original message it was trying to log was written to stderr and debug
5546 output, but if they were not available (the usual case in production), it
5547 was lost. Now it is written to syslog before the two lines that record the
5548 failures to open the logs.
5550 65. Users' Exim filters run in subprocesses under the user's uid. It is
5551 possible for a "deliver" command or an alias in a "personal" command to
5552 provoke an address rewrite. If logging of address rewriting is configured,
5553 this fails because the process is not running as root or exim. There may be
5554 a better way of dealing with this, but for the moment (because 4.30 needs
5555 to be released), I have disabled address rewrite logging when running a
5556 filter in a non-root, non-exim process.
5562 1. The buildconfig auxiliary program wasn't quoting the value set for
5563 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5564 not defined. This bug was masked in 4.22 by the effect that was fixed in
5567 2. Some messages that were rejected after a message id was allocated were
5568 shown as "incomplete" by exigrep. It no longer does this for messages that
5569 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5571 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5572 have allow_domain_literals set, the ID did not get logged in the <= line.
5573 Domain literals are now always recognized in Message-ID: header lines.
5575 4. The first argument for a ${extract expansion item is the key name or field
5576 number. Leading and trailing spaces in this item were not being ignored,
5577 causing some misleading effects.
5579 5. When deliver_drop_privilege was set, single queue runner processes started
5580 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5581 same command in the process it spins off) were not dropping privilege.
5583 6. When the daemon running as "exim" started a queue runner, it always
5584 re-executed Exim in the spun-off process. This is a waste of effort when
5585 deliver_drop_privilege is set. The new process now just calls the
5586 queue-runner function directly.
5592 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5595 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5596 ignored. Though the use of -oP was forcing the writing of a pid file, it
5597 was always written to the default place.
5599 3. If the message "no IP address found for host xxxx" is generated during
5600 incoming verification, it is now followed by identification of the incoming
5601 connection (so you can more easily find what provoked it).
5603 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5605 5. Added some features to "harden" Exim a bit more against certain attacks:
5607 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5608 be put in Local/Makefile. This is like the never_users runtime option,
5609 but it cannot be overridden. The default setting is "root".
5611 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5612 prefix string with which any file named in a -C command line option
5615 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5616 is retained for -C and -D only if the caller of Exim is root. Without
5617 it, the exim user may also use -C and -D and retain privilege.
5619 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5620 command line option is disabled.
5622 6. Macro names set by the -D option must start with an upper case letter, just
5623 like macro names defined in the configuration file.
5625 7. Added "dereference=" facility to LDAP.
5627 8. Two instances of the typo "uknown" in the source files are fixed.
5629 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5630 the Configure-Makefile script screwed up while processing it.
5632 10. Incorporated PCRE 4.4.
5634 11. The SMTP synchronization check was not operating right at the start of an
5635 SMTP session. For example, it could not catch a HELO sent before the client
5636 waited for the greeting. There is now a check for outstanding input at the
5637 point when the greeting is written. Because of the duplex, asynchronous
5638 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5639 way, but not yet received, when the check is performed.
5641 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5642 on TCP/IP sockets, because this apparently causes some broken clients to
5645 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5646 unchanged) from the Cygwin maintainer.
5648 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5649 was supported instead of testing for mbx. Effectively a typo.
5651 15. The spa authenticator server code was not checking that the input it
5652 received was valid base64.
5654 16. The debug output line for the "set" modifier in ACLs was not showing the
5655 name of the variable that was being set.
5657 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5659 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5660 lookup on its own. Something else was needed to trigger the lookup. For
5661 example, a match in host_lookup or the need for a host name in a host list.
5662 Now, if $sender_host_name is referenced and the host name has not yet been
5663 looked up, a lookup is performed. If the lookup fails, the variable remains
5664 empty, and $host_lookup_failed is set to "1".
5666 19. Added "eqi" as a case-independent comparison operator.
5668 20. The saslauthd authentication condition could segfault if neither service
5669 nor realm was specified.
5671 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5672 error message that was logged was misleading, and incoming SMTP
5673 connections were dropped. The message is now more accurate, and temporary
5674 errors are given to SMTP connections.
5676 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5677 (except RSET) with a 421 error, until QUIT is received. However, it was
5678 failing to send a response to QUIT.
5680 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5681 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5682 ACL is now run after the helo_try_verify_hosts code.
5684 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5685 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5686 case-independent, so other case variants are also recognized). Apparently
5687 some systems use these upper case variants.
5689 25. If more than two messages were waiting for the same host, and a transport
5690 filter was specified for the transport, Exim sent two messages over the
5691 same TCP/IP connection, and then failed with "socket operation on non-
5692 socket" when it tried to send the third.
5694 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5696 27. The extern definition of crypt16() in expand.c was not being excluded when
5697 the OS had its own crypt16() function.
5699 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5700 as a preferred synonym for return_size_limit, both as an option and as an
5703 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5705 30. Changed the default configuration ACL to relax the local part checking rule
5706 for addresses that are not in any local domains. For these addresses,
5707 slashes and pipe symbols are allowed within local parts, but the sequence
5708 /../ is explicitly forbidden.
5710 31. SPA server authentication was not clearing the challenge buffer before
5713 32. log_message in a "warn" ACL statement was writing to the reject log as
5714 well as to the main log, which contradicts the documentation and doesn't
5715 seem right (because no rejection is happening). So I have stopped it.
5717 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5718 However, I am unable to do any testing of this.
5720 34. Fixed an infelicity in the appendfile transport. When checking directories
5721 for a mailbox, to see if any needed to be created, it was accidentally
5722 using path names with one or more superfluous leading slashes; tracing
5723 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5725 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5726 contents are added to the log line that is written for every discarded
5727 recipient. (Previously a log_message setting was ignored.)
5729 36. The ${quote: operator now quotes the string if it is empty.
5731 37. The install script runs exim in order to find its version number. If for
5732 some reason other than non-existence or emptiness, which it checks, it
5733 could not run './exim', it was installing it with an empty version number,
5734 i.e. as "exim-". This error state is now caught, and the installation is
5737 38. An argument was missing from the function that creates an error message
5738 when Exim fails to connect to the socket for saslauthd authentication.
5739 This could cause Exim to crash, or give a corrupted message.
5741 39. Added isip, isip4, and isip6 to ${if conditions.
5743 40. The ACL variables $acl_xx are now saved with the message, and can be
5744 accessed later in routers, transports, and filters.
5746 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5747 strings in the file are not string-expanded.
5749 42. If a MAIL command specified a SIZE value that was too large to fit into an
5750 int variable, the check against message_size_limit failed. Such values are
5751 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5752 day this will have to be increased, but I don't think I want to be around
5753 when emails are that large.
5760 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5761 iconv() is not standard in FreeBSD.
5763 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5764 IPv6 enabled. The observed symptom was a segmentation fault on return from
5765 the function os_common_find_running_interfaces() in src/os.c.
5767 3. In the check_special_case() function in daemon.c I had used "errno" as an
5768 argument name, which causes warnings on some systems. This was basically a
5769 typo, since it was named "eno" in the comments!
5771 4. The code that waits for the clock to tick (at a resolution of some fraction
5772 of a second) so as to ensure message-id uniqueness was always waiting for
5773 at least one whole tick, when it could have waited for less. [This is
5774 almost certainly not relevant at current processor speeds, where it is
5775 unlikely to ever wait at all. But we try to future-proof.]
5777 5. The function that sleeps for a time interval that includes fractions of a
5778 second contained a race. It did not block SIGALRM between setting the
5779 timer, and suspending (a couple of lines later). If the interval was short
5780 and the sigsuspend() was delayed until after it had expired, the suspension
5781 never ended. On busy systems this could lead to processes getting stuck for
5784 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5785 process, before it forks any delivery processes. The open lookup caching
5786 mechanism meant that the open file or database connection was passed into
5787 the delivery process. The problem was that delivery processes always tidy
5788 up cached lookup data. This could cause a problem for the next delivery
5789 process started by the queue runner, because the external queue runner
5790 process does not know about the closure. So the next delivery process
5791 still has data in the lookup cache. In the case of a file lookup, there was
5792 no problem because closing a file descriptor in a subprocess doesn't affect
5793 the parent. However, if the lookup was caching a connection to a database,
5794 the connection was closed, and the second delivery process was likely to
5795 see errors such as "PGSQL: query failed: server closed the connection
5796 unexpectedly". The problem has been fixed by closing all cached lookups
5797 in a queue runner before running a delivery process.
5799 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5800 seem to have the "const" qualifier which it has on other OS. I've
5803 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5804 *of the same type* (client or server) with the same public name that an
5805 error should be diagnosed.
5807 9. When Exim looked up a host name for an IP address, but failed to find the
5808 original IP address when looking up the host name (a safety check), it
5809 output the message "<ip address> does not match any IP for NULL", which was
5810 confusing, to say the least. The bug was that the host name should have
5811 appeared instead of "NULL".
5813 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5814 user that is built into the binary, and the -C or -D options is used, root
5815 privilege is dropped before the configuration file is read. In addition,
5816 logging is switched to stderr instead of the normal log files. If the
5817 configuration then re-defines the Exim user, the unprivileged environment
5818 is probably not what is expected, so Exim logs a panic warning message (but
5821 However, if deliver_drop_privilege is set, the unprivileged state may well
5822 be exactly what is intended, so the warning has been cut out in that case,
5823 and Exim is allowed to try to write to its normal log files.
5829 1. smtp_return_error_details was not giving details for temporary sender
5830 or receiver verification errors.
5832 2. Diagnose a configuration error if two authenticators have the same public
5835 3. Exim used not to create the message log file for a message until the first
5836 delivery attempt. This could be confusing when incoming messages were held
5837 for policy or load reasons. The message log file is now created at the time
5838 the message is received, and an initial "Received" line is written to it.
5840 4. The automatically generated man page for command line options had a minor
5841 bug that caused no ill effects; however, a more serious problem was that
5842 the procedure for building the man page automatically didn't always
5843 operate. Consequently, release 4.20 contains an out-of-date version. This
5844 shouldn't happen again.
5846 5. When building Exim with embedded Perl support, the script that builds the
5847 Makefile was calling 'perl' to find its compile-time parameters, ignoring
5848 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
5850 6. The freeze_tell option was not being used for messages that were frozen on
5851 arrival, either by an ACL or by local_scan().
5853 7. Added the smtp_incomplete_transaction log selector.
5855 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
5856 was accepting AUTH without a new EHLO.
5858 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
5859 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
5862 10. Logging of TCP/IP connections (when configured) now happens in the main
5863 daemon process instead of the child process, so that the TCP/IP connection
5864 count is more accurate (but it can never be perfect).
5866 11. The use of "drop" in a nested ACL was not being handled correctly in the
5867 outer ACL. Now, if condition failure induced by the nested "drop" causes
5868 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
5869 or "require"), the connection is dropped.
5871 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
5872 that yield "discard" can now be used with an "accept" or a "discard" verb,
5873 but an error is generated for any others (because I can't see a useful way
5874 to define what should happen).
5876 13. When an ACL is read dynamically from a file (or anywhere else), the lines
5877 are now processed in the same way as lines in the Exim configuration file.
5878 In particular, continuation lines are supported.
5880 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
5882 15. Added -ti meaning -t -i.
5884 16. Check for letters, digits, hyphens, and dots in the names of dnslist
5885 domains, and warn by logging if others are found.
5887 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
5888 returned from GIFCONF when Exim is trying to find the list of interfaces on
5889 a host. The code in os.c has been modified to copy each ifreq to an aligned
5890 structure in all cases.
5892 Also, in some cases, the returned ifreq's were being copied to a 'struct
5893 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
5894 means the last couple of bytes of an IPv6 address could be chopped if the
5895 ifreq contained only a normal sockaddr (14 bytes storage).
5897 18. Named domain lists were not supported in the hosts_treat_as_local option.
5898 An entry such as +xxxx was not recognized, and was treated as a literal
5901 19. Ensure that header lines added by a DATA ACL are included in the reject log
5902 if the ACL subsequently rejects the message.
5904 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
5905 MD5 (which is deprecated).
5907 21. When testing a filter file using -bf, Exim was writing a message when it
5908 took the sender from a "From " line in the message, but it was not doing so
5909 when it took $return_path from a Return-Path: header line. It now does.
5911 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
5912 with a valid header line field name (a series of printing characters
5913 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
5915 23. Changed "disc" in the source to "disk" to conform to the documentation and
5916 the book and for uniformity.
5918 24. Ignore Sendmail's -Ooption=value command line item.
5920 25. When execve() failed while trying to run a command in a pipe transport,
5921 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
5922 could be confused with a return value of 69 from the command itself. This
5923 has been changed to 127, the value the shell returns if it is asked to run
5924 a non-existent command. The wording for the related log line suggests a
5925 non-existent command as the problem.
5927 26. If received_header_text expands to an empty string, do not add a Received:
5928 header line to the message. (Well, it adds a token one on the spool, but
5929 marks it "old" so that it doesn't get used or transmitted.)
5931 27. Installed eximstats 1.28 (addition of -nt option).
5933 28. There was no check for failure on the call to getsockname() in the daemon
5934 code. This can fail if there is a shortage of resources on the system, with
5935 ENOMEM, for example. A temporary error is now given on failure.
5937 29. Contrary to the C standard, it seems that in some environments, the
5938 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
5939 program. Exim now does this explicitly; it affects the formatting of
5940 timestamps using strftime().
5942 30. If exiqsumm was given junk data, it threw up some uninitialized variable
5943 complaints. I've now initialized all the variables, to avoid this.
5945 32. Header lines added by a system filter were not being "seen" during
5946 transport-time rewrites.
5948 33. The info_callback() function passed to OpenSSL is set up with type void
5949 (*)(SSL *, int, int), as described somewhere. However, when calling the
5950 function (actually a macro) that sets it up, the type void(*)() is
5951 expected. I've put in a cast to prevent warnings from picky compilers.
5953 34. If a DNS black list lookup found a CNAME record, but there were no A
5954 records associated with the domain it pointed at, Exim crashed.
5956 35. If a DNS black list lookup returned more than one A record, Exim ignored
5957 all but the first. It now scans all returned addresses if a particular IP
5958 value is being sought. In this situation, the contents of the
5959 $dnslist_value variable are a list of all the addresses, separated by a
5962 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5963 to accept a host name and all its aliases if the forward lookup for any of
5964 them yielded the IP address of the incoming connection. Now it accepts only
5965 those names whose forward lookup yields the correct IP address. Any other
5966 names are discarded. This closes a loophole whereby a rogue DNS
5967 administrator could create reverse DNS records to break through a
5968 wildcarded host restriction in an ACL.
5970 37. If a user filter or a system filter that ran in a subprocess used any of
5971 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5972 the wrong values were passed to the pipe command ($thisaddress had the
5973 value of $0, $0 had the value of $1, etc). This bug was introduced by
5974 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5976 38. Improved the line breaking for long SMTP error messages from ACLs.
5977 Previously, if there was no break point between 40 and 75 characters, Exim
5978 left the rest of the message alone. Two changes have been made: (a) I've
5979 reduced the minimum length to 35 characters; (b) if it can't find a break
5980 point between 35 and 75 characters, it looks ahead and uses the first one
5981 that it finds. This may give the occasional overlong line, but at least the
5982 remaining text gets split now.
5984 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5985 file descriptors might be low, and that setting 1000 would always raise it.
5986 It turns out that in some environments, the limit is already over 1000 and
5987 that lowering it causes trouble. So now Exim takes care not to decrease it.
5989 40. When delivering a message, the value of $return_path is set to $sender_
5990 address at the start of routing (routers may change the value). By an
5991 oversight, this default was not being set up when an address was tested by
5992 -bt or -bv, which affected the outcome if any router or filter referred to
5995 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5996 log only when "message" or "log_message" are set. However, if one of the
5997 conditions was an address verification, or a call to a nested ACL, the
5998 messages generated by the underlying test were being passed through. This
5999 no longer happens. The underlying message is available in $acl_verify_
6000 message for both "message" and "log_message" expansions, so it can be
6001 passed through if needed.
6003 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6004 new expansion $bh_ to give the encoded byte string without charset
6005 translation. Translation happens only if iconv() is available; HAVE_ICONV
6006 indicates this at build time. HEADERS_CHARSET gives the charset to
6007 translate to; headers_charset can change it in the configuration, and
6008 "headers charset" can change it in an individual filter file.
6010 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6011 that creates RFC 2047 encoded "words" labels them as that charset instead
6012 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6013 expansion operator; (ii) when Exim creates a From: line for a local
6014 message; (iii) when a header line is rewritten to include a "phrase" part.
6016 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6017 buggy, causing it to skip the first lines of messages whose message ID
6018 ended in 'D'. This would not have bitten before Exim release 4.14, because
6019 message IDs were unlikely to end in 'D' before then. The effect was to have
6020 incorrect size information for certain domains.
6022 45. #include "config.h" was missing at the start of the crypt16.c module. This
6023 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6026 46. If there was a timeout during a "random" callout check, Exim treated it as
6027 a failure of the random address, and carried on sending RSET and the real
6028 address. If the delay was just some slowness somewhere, the response to the
6029 original RCPT would be taken as a response to RSET and so on, causing
6030 mayhem of various kinds.
6032 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6033 when I implemented it. It didn't allow for the fact that some option values
6034 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6035 the right test for positive values.
6037 48. Domain names in DNS records are case-independent. Exim always looks them up
6038 in lower case. Some resolvers return domain names in exactly the case they
6039 appear in the zone file, that is, they may contain uppercase letters. Not
6040 all resolvers do this - some return always lower case. Exim was treating a
6041 change of case by a resolver as a change of domain, similar to a widening
6042 of a domain abbreviation. This triggered its re-routing code and so it was
6043 trying to route what was effectively the same domain again. This normally
6044 caused routing to fail (because the router wouldn't handle the domain
6045 twice). Now Exim checks for this case specially, and just changes the
6046 casing of the domain that it ultimately uses when it transmits the message
6049 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6052 50. If a filter generated a file delivery with a non-absolute name (possible if
6053 no home directory exists for the router), the forbid_file option was not
6056 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6057 the existing equality matching.
6059 52. Exim was using ints instead of ino_t variables in some places where it was
6060 dealing with inode numbers.
6062 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6063 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6064 called TMPDIR, and if it finds it is different, it changes its value.
6066 54. The smtp_printf() function is now made available to local_scan() so
6067 additional output lines can be written before returning. There is also an
6068 smtp_fflush() function to enable the detection of a dropped connection.
6069 The variables smtp_input and smtp_batched_input are exported to
6072 55. Changed the default runtime configuration: the message "Unknown user"
6073 has been removed from the ACL, and instead placed on the localuser router,
6074 using the cannot_route_message feature. This means that any verification
6075 failures that generate their own messages won't get overridden. Similarly,
6076 the "Unrouteable address" message that was in the ACL for unverifiable
6077 relay addresses has also been removed.
6079 56. Added hosts_avoid_esmtp to the smtp transport.
6081 57. The exicyclog script was not checking for the esoteric option
6082 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6083 will work only if exicyclog is run under the appropriate euid.
6085 58. Following a discussion on the list, the rules by which Exim recognises line
6086 endings on incoming messages have been changed. The -dropcr and drop_cr
6087 options are now no-ops, retained only for backwards compatibility. The
6088 following line terminators are recognized: LF CRLF CR. However, special
6089 processing applies to CR:
6091 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6092 nor a local message in the state where . is a terminator.
6094 (ii) If a bare CR is encountered in a header line, an extra space is added
6095 after the line terminator so as not to end the header. The reasoning
6096 behind this is that bare CRs in header lines are most likely either
6097 to be mistakes, or people trying to play silly games.
6099 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6100 was being incorrectly given as 18 bytes larger than it should have been.
6101 This is a VOB (very old bug).
6103 60. This may never have affected anything current, but just in case it has:
6104 When the local host is found other than at the start of a list of hosts,
6105 the local host, those with the same MX, and any that follow, are discarded.
6106 When the list in question was part of a longer list of hosts, the following
6107 hosts (not currently being processed) were also being discarded. This no
6108 longer happens. I'm not sure if this situation could ever has previously
6111 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6114 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6115 header. This is something that is recommended in a new Internet Draft, and
6116 is something that is documented as being done by Sendmail. There are two
6117 possible values. For messages generated by the autoreply transport, Exim
6120 Auto-Submitted: auto-replied
6122 whereas for all other generated messages (e.g. bounces) it adds
6124 Auto-Submitted: auto-generated
6126 63. The "personal" condition in filters now includes a test for the
6127 Auto-Submitted: header. If it contains the string "auto-" the message it
6128 not considered personal.
6130 64. Added rcpt_include_affixes as a generic transport option.
6132 65. Added queue_only_override (default true).
6134 66. Added the syslog_duplication option.
6136 67. If what should have been the first header line of a message consisted of
6137 a space followed by a colon, Exim was mis-interpreting it as a header line.
6138 It isn't of course - it is syntactically invalid and should therefore be
6139 treated as the start of the message body. The misbehaviour could have
6140 caused a number of strange effects, including loss of data in subsequent
6141 header lines, and spool format errors.
6143 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6144 client host had authenticated. This control can now be exercised by an ACL
6145 for more flexibility.
6147 69. By default, callouts do not happen when testing with -bh. There is now a
6148 variant, -bhc, which does actually run the callout code, including
6149 consulting and updating the callout cache.
6151 70. Added support for saslauthd authentication, courtesy of Alexander
6154 71. If statvfs() failed on the spool or log directories while checking their
6155 size for availability, Exim confusingly gave the error "space shortage".
6156 Furthermore, in debugging mode it crashed with a floating point exception.
6157 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6158 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6159 serious problem, Exim now writes to the main and panic logs when this
6160 happens, with details of the failure. It then refuses to accept the
6161 incoming message, giving the message "spool directory problem" or "log
6162 directory problem" with a 421 code for SMTP messages.
6164 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6165 0, 1, and 2 exist, because some OS complain for execs without them (see
6166 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6167 descriptors. However, the code omitted to check that the open succeeded,
6168 causing mysterious errors if for some reason the permissions on /dev/null
6169 got screwed. Now Exim writes a message to the main and panic logs, and
6170 bombs out if it can't open /dev/null.
6172 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6173 interact so that it is all more flexible. It is supposed to remain
6174 backwards compatible. Also added extra_local_interfaces.
6176 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6177 to bomb out with an assertion failure - to the client this appears as a
6178 connection drop. This problem occurs in the part of the code that was taken
6179 from the Samba project. Fortunately, the assertion is in a very simple
6180 function, so I have fixed this by reproducing the function inline in the
6181 one place where it is called, and arranging for authentication to fail
6182 instead of killing the process with assert().
6184 75. The SPA client code was not working when the server requested OEM rather
6185 than Unicode encoding.
6187 76. Added code to make require_files with a specific uid setting more usable in
6188 the case where statting the file as root fails - usually a non-root-mounted
6189 NFS file system. When this happens and the failure is EACCES, Exim now
6190 forks a subprocess and does the per-uid checking as the relevant uid.
6192 77. Added process_log_path.
6194 78. If log_file_path was not explicitly set, a setting of check_log_space or
6195 check_log_inodes was ignored.
6197 79. If a space check for the spool or log partitions fails, the incident is now
6198 logged. Of course, in the latter case the data may get lost...
6200 80. Added the %p formatting code to string_format() so that it can be used to
6201 print addresses in debug_print(). Adjusted all the address printing in the
6202 debugging in store.c to use %p rather than %d.
6204 81. There was a concern that a line of code in smtp_in.c could overflow a
6205 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6206 initially expressed, the concern was not well-founded, because trailing
6207 spaces are removed early. However, if the trailing spaces were followed by
6208 a NULL, they did not get removed, so the overflow was possible. Two fixes
6211 (a) I re-wrote the offending code in a cleaner fashion.
6212 (b) If an incoming SMTP command contains a NULL character, it is rejected
6215 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6216 runs initgroups(), so that if the Exim user is in any additional groups,
6217 they will be used during message reception.
6223 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob