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18 EXPERIMENTAL_CERTNAMES is withdrawn.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. Bug 68.
42 JH/10 The smtp_receive_timeout main option is now expanded before use.
44 JH/11 The incoming_interface log option now also enables logging of the
45 local interface on delivery outgoing connections.
47 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
48 if the interface and destination host and port all match.
50 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
53 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
54 Patch from Andrew Lewis.
56 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
57 now supports optional time-restrictions, weighting, and priority
58 modifiers per server. Patch originally by <rommer@active.by>.
60 JH/16 The spamd_address main option now supports a mixed list of local
61 and remote servers. Remote servers can be IPv6 addresses, and
64 JH/17 Bug 68: The spamd_address main option now supports an optional
65 timeout value per server.
67 JH/18 Bug 1581: Router and transport options headers_add/remove can
68 now have the list separator specified.
70 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
73 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
76 JH/21 Support for the A6 type of dns record is withdrawn.
78 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
79 rather than the verbs used.
81 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
82 from 255 to 1024 chars.
88 TL/01 When running the test suite, the README says that variables such as
89 no_msglog_check are global and can be placed anywhere in a specific
90 test's script, however it was observed that placement needed to be near
91 the beginning for it to behave that way. Changed the runtest perl
92 script to read through the entire script once to detect and set these
93 variables, reset to the beginning of the script, and then run through
94 the script parsing/test process like normal.
96 TL/02 The BSD's have an arc4random API. One of the functions to induce
97 adding randomness was arc4random_stir(), but it has been removed in
98 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
99 function when detected.
101 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
102 cause callback expansion.
104 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
105 syntax errors in an expansion can be treated as a string instead of
106 logging or causing an error, due to the internal use of bool_lax
107 instead of bool when processing it.
109 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
110 server certificates when making smtp deliveries.
112 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
114 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
116 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
118 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
119 Merged patch from Sebastian Wiedenroth.
121 JH/05 Fix results-pipe from transport process. Several recipients, combined
122 with certificate use, exposed issues where response data items split
123 over buffer boundaries were not parsed properly. This eventually
124 resulted in duplicates being sent. This issue only became common enough
125 to notice due to the introduction of conection certificate information,
126 the item size being so much larger. Found and fixed by Wolfgang Breyha.
128 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
129 size buffer was used, resulting in syntax errors when an expansion
132 JH/07 Add support for directories of certificates when compiled with a GnuTLS
133 version 3.3.6 or later.
135 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
136 is EXPERIMENTAL_EVENT, the main-configuration and transport options
137 both become "event_action", the variables become $event_name, $event_data
138 and $event_defer_errno. There is a new variable $verify_mode, usable in
139 routers, transports and related events. The tls:cert event is now also
140 raised for inbound connections, if the main configuration event_action
143 TL/06 In test suite, disable OCSP for old versions of openssl which contained
144 early OCSP support, but no stapling (appears to be less than 1.0.0).
146 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
147 server certificate names available under the smtp transport option
148 "tls_verify_cert_hostname" now do not permit multi-component wildcard
151 JH/10 Time-related extraction expansions from certificates now use the main
152 option "timezone" setting for output formatting, and are consistent
153 between OpenSSL and GnuTLS compilations. Bug 1541.
155 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
156 encoded parameter in the incoming message. Bug 1558.
158 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
159 include certificate info, eximon was claiming there were spoolfile
162 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
164 JH/14 Log delivery-related information more consistently, using the sequence
165 "H=<name> [<ip>]" wherever possible.
167 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
168 are problematic for Debian distribution, omit them from the release
171 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
173 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
176 JH/17 Fix a null-indirection in certextract expansions when a nondefault
177 output list separator was used.
182 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
183 checkers that were complaining about end of non-void function with no
186 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
187 This was a regression intruduced in 4.83 by another bugfix.
189 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
191 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
192 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
198 TF/01 Correctly close the server side of TLS when forking for delivery.
200 When a message was received over SMTP with TLS, Exim failed to clear up
201 the incoming connection properly after forking off the child process to
202 deliver the message. In some situations the subsequent outgoing
203 delivery connection happened to have the same fd number as the incoming
204 connection previously had. Exim would try to use TLS and fail, logging
205 a "Bad file descriptor" error.
207 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
208 utilities have not been installed.
210 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
211 temporary space as the ACL may create new global variables.
213 TL/01 LDAP support uses per connection or global context settings, depending
214 upon the detected version of the libraries at build time.
216 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
217 to extract and use the src ip:port in logging and expansions as if it
218 were a direct connection from the outside internet. PPv2 support was
219 updated based on HAProxy spec change in May 2014.
221 JH/02 Add ${listextract {number}{list}{success}{fail}}.
223 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
224 Properly escape header and check for NULL return.
226 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
229 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
231 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
232 characters in header names, implemented as a verify condition.
233 Contributed by Michael Fischer v. Mollard.
235 TL/05 Rename SPF condition results err_perm and err_temp to standardized
236 results permerror and temperror. Previous values are deprecated but
237 still accepted. In a future release, err_perm and err_temp will be
238 completely removed, which will be a backward incompatibility if the
239 ACL tests for either of these two old results. Patch contributed by
240 user bes-internal on the mailing list.
242 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
244 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
245 selectors, in both main and reject logs.
247 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
250 JH/07 Add malware type "sock" for talking to simple daemon.
252 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
254 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
255 routers/transports under cutthrough routing.
257 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
258 numbers. Touch up "bool" conditional to keep the same definition.
260 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
262 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
264 JH/12 Expand items in router/transport headers_add or headers_remove lists
265 individually rather than the list as a whole. Bug 1452.
267 Required for reasonable handling of multiple headers_ options when
268 they may be empty; requires that headers_remove items with embedded
269 colons must have them doubled (or the list-separator changed).
271 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
272 view the policy declared in the DMARC record. Currently, $dmarc_status
273 is a combined value of both the record presence and the result of the
276 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
278 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
279 dnslookup router and the smtp transport (applying to the forward
282 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
283 of ldap servers used for a specific lookup. Patch provided by Heiko
286 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
287 New variable $lookup_dnssec_authenticated for observability.
289 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
290 Patch submitted by Lars Timman.
292 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
294 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
295 Requires trusted mode and valid format message id, aborts otherwise.
296 Patch contributed by Heiko Schlichting.
298 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
299 certextract with support for various fields. Bug 1358.
301 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
302 is requested by default, modifiable by smtp transport option
305 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
306 operate on certificate variables to give certificate fingerprints
307 Also new ${sha256:cert_variable}.
309 JH/23 The PRDR feature is moved from being Experimental into the mainline.
311 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
312 Christian Aistleitner.
314 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
316 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
317 file. Patch from Wolfgang Breyha.
319 JH/25 Expand the coverage of the delivery $host and $host_address to
320 client authenticators run in verify callout. Bug 1476.
322 JH/26 Port service names are now accepted for tls_on_connect_ports, to
323 align with daemon_smtp_ports. Bug 72.
325 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
326 support and error reporting did not work properly.
328 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
329 and is readable. Patch from Andrew Colin Kissa.
331 TL/14 Enhance documentation of ${run expansion and how it parses the
332 commandline after expansion, particularly in the case when an
333 unquoted variable expansion results in an empty value.
335 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
337 PP/02 Fix internal collision of T_APL on systems which support RFC3123
338 by renaming away from it. Addresses GH issue 15, reported by
341 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
343 TL/15 SECURITY: prevent double expansion in math comparison functions
344 (can expand unsanitized data). Not remotely exploitable.
351 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
353 PP/02 Make -n do something, by making it not do something.
354 When combined with -bP, the name of an option is not output.
356 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
359 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
360 $sender_host_name and config options to manage this, and basic check
363 PP/05 DSCP support for outbound connections and control modifier for inbound.
365 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
366 (Only plugin which currently uses this is kerberos4, which nobody should
367 be using, but we should make it available and other future plugins might
368 conceivably use it, even though it would break NAT; stuff *should* be
369 using channel bindings instead).
371 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
372 name; added for Sendmail compatibility; requires admin caller.
373 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
374 just ignore it); requires trusted caller.
375 Also parse but ignore: -Ac -Am -X<logfile>
378 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
380 TL/02 Add +smtp_confirmation as a default logging option.
382 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
383 Patch by Magnus Holmgren from 2007-02-20.
385 TL/04 Bugzilla 1281 - Spec typo.
386 Bugzilla 1283 - Spec typo.
387 Bugzilla 1290 - Spec grammar fixes.
389 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
391 TL/06 Add Experimental DMARC support using libopendmarc libraries.
393 TL/07 Fix an out of order global option causing a segfault. Reported to dev
394 mailing list by by Dmitry Isaikin.
396 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
398 JH/02 Support "G" suffix to numbers in ${if comparisons.
400 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
402 NM/01 Bugzilla 1197 - Spec typo
403 Bugzilla 1196 - Spec examples corrections
405 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
407 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
408 gnutls_enable_pkcs11, but renamed to more accurately indicate its
411 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
412 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
414 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
415 "acl {{name}{arg}...}", and optional args on acl condition
418 JH/05 Permit multiple router/transport headers_add/remove lines.
420 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
422 JH/07 Avoid using a waiting database for a single-message-only transport.
423 Performance patch from Paul Fisher. Bugzilla 1262.
425 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
428 JH/09 Add $headers_added variable, with content from use of ACL modifier
429 add_header (but not yet added to the message). Bugzilla 199.
431 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
432 Pulled from Bugzilla 817 by Wolfgang Breyha.
434 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
436 (nb: this is the same fix as in Exim 4.80.1)
438 JH/11 Add A= logging on delivery lines, and a client_set_id option on
441 JH/12 Add optional authenticated_sender logging to A= and a log_selector
444 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
446 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
447 advertise SMTP AUTH mechanism to us, instead of a generic
448 protocol violation error. Also, make Exim more robust to bad
449 data from the Dovecot auth socket.
451 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
453 When a queue runner is handling a message, Exim first routes the
454 recipient addresses, during which it prunes them based on the retry
455 hints database. After that it attempts to deliver the message to
456 any remaining recipients. It then updates the hints database using
459 So if a recipient address works intermittently, it can get repeatedly
460 deferred at routing time. The retry hints record remains fresh so the
461 address never reaches the final cutoff time.
463 This is a fairly common occurrence when a user is bumping up against
464 their storage quota. Exim had some logic in its local delivery code
465 to deal with this. However it did not apply to per-recipient defers
466 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
468 This change adds a proper retry rule check during routing so that the
469 final cutoff time is checked against the message's age. We only do
470 this check if there is an address retry record and there is not a
471 domain retry record; this implies that previous attempts to handle
472 the address had the retry_use_local_parts option turned on. We use
473 this as an approximation for the destination being like a local
474 delivery, as in LMTP.
476 I suspect this new check makes the old local delivery cutoff check
477 redundant, but I have not verified this so I left the code in place.
479 TF/02 Correct gecos expansion when From: is a prefix of the username.
481 Test 0254 submits a message to Exim with the header
485 When I ran the test suite under the user fanf2, Exim expanded
486 the header to contain my full name, whereas it should have added
487 a Resent-Sender: header. It erroneously treats any prefix of the
488 username as equal to the username.
490 This change corrects that bug.
492 GF/01 DCC debug and logging tidyup
493 Error conditions log to paniclog rather than rejectlog.
494 Debug lines prefixed by "DCC: " to remove any ambiguity.
496 TF/03 Avoid unnecessary rebuilds of lookup-related code.
498 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
499 Bug spotted by Jeremy Harris; was flawed since initial commit.
500 Would have resulted in OCSP responses post-SNI triggering an Exim
501 NULL dereference and crash.
503 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
505 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
506 Bug detection, analysis and fix by Samuel Thibault.
507 Bugzilla 1331, Debian bug #698092.
509 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
511 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
512 Server implementation by Todd Lyons, client by JH.
513 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
514 config variable "prdr_enable" controls whether the server
515 advertises the facility. If the client requests PRDR a new
516 acl_data_smtp_prdr ACL is called once for each recipient, after
517 the body content is received and before the acl_smtp_data ACL.
518 The client is controlled by bolth of: a hosts_try_prdr option
519 on the smtp transport, and the server advertisement.
520 Default client logging of deliveries and rejections involving
521 PRDR are flagged with the string "PRDR".
523 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
524 fclose(). Diagnosis by Todd Lyons.
526 PP/17 Update configure.default to handle IPv6 localhost better.
527 Patch by Alain Williams (plus minor tweaks).
530 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
531 This is now consistent with GnuTLS, and is now documented: the
532 previous undocumented portable approach to treating the option as
533 unset was to force an expansion failure. That still works, and
534 an empty string is now equivalent.
536 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
537 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
538 not performing validation itself.
540 PP/20 Added force_command boolean option to pipe transport.
541 Patch from Nick Koston, of cPanel Inc.
543 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
546 TF/04 Added udpsend ACL modifer and hexquote expansion operator
548 PP/21 Fix eximon continuous updating with timestamped log-files.
549 Broken in a format-string cleanup in 4.80, missed when I repaired the
550 other false fix of the same issue.
551 Report and fix from Heiko Schlichting.
554 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
555 Report from Prashanth Katuri.
557 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
558 It's SecureTransport, so affects any MacOS clients which use the
559 system-integrated TLS libraries, including email clients.
561 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
562 using a MIME ACL for non-SMTP local injection.
563 Report and assistance in diagnosis by Warren Baker.
565 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
567 JH/16 Fix comparisons for 64b. Bugzilla 1385.
569 TL/09 Add expansion variable $authenticated_fail_id to keep track of
570 last id that failed so it may be referenced in subsequent ACL's.
572 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
575 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
576 ldap library initialization, allowing self-signed CA's to be
577 used. Also properly sets require_cert option later in code by
578 using NULL (global ldap config) instead of ldap handle (per
579 session). Bug diagnosis and testing by alxgomz.
581 TL/12 Enhanced documentation in the ratelimit.pl script provided in
582 the src/util/ subdirectory.
584 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
585 renamed to Transport Post Delivery Action by Jeremy Harris, as
588 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
589 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
590 redis_servers = needs to be configured which will be used by the redis
591 lookup. Patch from Warren Baker, of The Packet Hub.
593 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
595 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
596 hostname or reverse DNS when processing a host list. Used suggestions
597 from multiple comments on this bug.
599 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
601 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
602 Missed a few lines, added it to make the runtest require no keyboard
605 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
606 contains upper case chars. Make router use caseful_local_part.
608 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
609 support when GnuTLS has been built with p11-kit.
615 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
617 This, or similar/improved, will also be change PP/11 of 4.82.
623 PP/01 Handle short writes when writing local log-files.
624 In practice, only affects FreeBSD (8 onwards).
625 Bugzilla 1053, with thanks to Dmitry Isaikin.
627 NM/01 Bugzilla 949 - Documentation tweak
629 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
632 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
634 PP/02 Implemented gsasl authenticator.
636 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
638 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
639 `pkg-config foo` for cflags/libs.
641 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
642 with rest of GSASL and with heimdal_gssapi.
644 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
645 `pkg-config foo` for cflags/libs for the TLS implementation.
647 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
648 properties get this fed in as external SSF. A number of robustness
649 and debugging improvements to the cyrus_sasl authenticator.
651 PP/08 cyrus_sasl server now expands the server_realm option.
653 PP/09 Bugzilla 1214 - Log authentication information in reject log.
654 Patch by Jeremy Harris.
656 PP/10 Added dbmjz lookup type.
658 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
660 PP/12 MAIL args handles TAB as well as SP, for better interop with
661 non-compliant senders.
662 Analysis and variant patch by Todd Lyons.
664 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
665 Bug report from Lars Müller <lars@samba.org> (via SUSE),
666 Patch from Dirk Mueller <dmueller@suse.com>
668 PP/13 tls_peerdn now print-escaped for spool files.
669 Observed some $tls_peerdn in wild which contained \n, which resulted
670 in spool file corruption.
672 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
673 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
674 or write after TLS renegotiation, which otherwise led to messages
677 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
678 as a tracking header (ie: a signed header comes before the signature).
679 Patch from Wolfgang Breyha.
681 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
682 comma-sep list; embedded commas doubled.
684 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
686 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
688 Report and patch from Dmitry Banschikov.
690 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
691 Removed SSL_clear() after SSL_new() which led to protocol negotiation
692 failures. We appear to now support TLS1.1+ with Exim.
694 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
695 lets Exim select keys and certificates based upon TLS SNI from client.
696 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
697 before an outbound SMTP session. New log_selector, +tls_sni.
699 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
700 NULL dereference. Report and patch from Alun Jones.
702 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
703 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
704 Not seeing resolver debug output on NetBSD, but suspect this is a
705 resolver implementation change.
707 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
708 Left warnings. Added "eximon gdb" invocation mode.
710 PP/21 Defaulting "accept_8bitmime" to true, not false.
712 PP/22 Added -bw for inetd wait mode support.
714 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
715 locate the relevant includes and libraries. Made this the default.
717 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
718 Bugzilla 1246, report and most of solution from Tomasz Kusy.
720 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
721 This may cause build issues on older platforms.
723 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
724 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
725 gnutls_require_mac & gnutls_require_protocols (no longer supported).
726 Added SNI support via GnuTLS too.
727 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
729 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
731 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
732 Applied second patch from Janne, implementing suggestion to default
733 multiple-strings-in-record handling to match SPF spec.
735 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
737 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
738 read-only, out of scope).
739 Patch from Wolfgang Breyha, report from Stuart Northfield.
741 PP/29 Fix three issues highlighted by clang analyser static analysis.
742 Only crash-plausible issue would require the Cambridge-specific
743 iplookup router and a misconfiguration.
744 Report from Marcin Mirosław.
746 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
748 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
749 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
750 As part of this, removing so much warning spew let me fix some minor
751 real issues in debug logging.
753 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
754 assignment on my part. Fixed.
756 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
757 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
758 Janne Snabb (who went above and beyond: thank you).
760 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
761 string otherwise requires a connection and a bunch more work and it's
762 relatively easy to get wrong. Should also expose TLS library linkage
765 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
766 64-bit ${eval} (JH/03).
768 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
769 GNU libc to support some of the 64-bit stuff, should not lead to
770 conflicts. Defined before os.h is pulled in, so if a given platform
771 needs to override this, it can.
773 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
774 protection layer was required, which is not implemented.
775 Bugzilla 1254, patch from Wolfgang Breyha.
777 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
778 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
779 tls_dhparam take prime identifiers. Also unbreak combination of
780 OpenSSL+DH_params+TLSSNI.
782 PP/39 Disable SSLv2 by default in OpenSSL support.
788 PP/01 Solaris build fix for Oracle's LDAP libraries.
789 Bugzilla 1109, patch from Stephen Usher.
791 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
793 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
796 TF/02 Fix a couple more cases where we did not log the error message
797 when unlink() failed. See also change 4.74-TF/03.
799 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
800 lock up or crash if it happened to be inside a call to libc when it
801 got a SIGUSR1 from exiwhat.
803 The SIGUSR1 handler appends the current process status to the process
804 log which is later printed by exiwhat. It used to use the general
805 purpose logging code to do this, but several functions it calls are
806 not safe for signals.
808 The new output code in the SIGUSR1 handler is specific to the process
809 log, and simple enough that it's easy to inspect for signal safety.
810 Removing some special cases also simplifies the general logging code.
811 Removing the spurious timestamps from the process log simplifies
814 TF/04 Improved ratelimit ACL condition.
816 The /noupdate option has been deprecated in favour of /readonly which
817 has clearer semantics. The /leaky, /strict, and /readonly update modes
818 are mutually exclusive. The update mode is no longer included in the
819 database key; it just determines when the database is updated. (This
820 means that when you upgrde Exim will forget old rate measurements.)
822 Exim now checks that the per_* options are used with an update mode that
823 makes sense for the current ACL. For example, when Exim is processing a
824 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
825 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
826 must specify per_mail/readonly. If you omit the update mode it defaults to
827 /leaky where that makes sense (as before) or /readonly where required.
829 The /noupdate option is now undocumented but still supported for
830 backwards compatibility. It is equivalent to /readonly except that in
831 ACLs where /readonly is required you may specify /leaky/noupdate or
832 /strict/noupdate which are treated the same as /readonly.
834 A useful new feature is the /count= option. This is a generalization
835 of the per_byte option, so that you can measure the throughput of other
836 aggregate values. For example, the per_byte option is now equivalent
837 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
839 The per_rcpt option has been generalized using the /count= mechanism
840 (though it's more complicated than the per_byte equivalence). When it is
841 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
842 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
843 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
844 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
845 Note that using per_rcpt with a non-readonly update mode in more than
846 one ACL will cause the recipients to be double-counted. (The per_mail
847 and per_byte options don't have this problem.)
849 The handling of very low rates has changed slightly. If the computed rate
850 is less than the event's count (usually one) then this event is the first
851 after a long gap. In this case the rate is set to the same as this event's
852 count, so that the first message of a spam run is counted properly.
854 The major new feature is a mechanism for counting the rate of unique
855 events. The new per_addr option counts the number of different
856 recipients that someone has sent messages to in the last time period. It
857 behaves like per_rcpt if all the recipient addresses are different, but
858 duplicate recipient addresses do not increase the measured rate. Like
859 the /count= option this is a general mechanism, so the per_addr option
860 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
861 example, measure the rate that a client uses different sender addresses
862 with the options per_mail/unique=$sender_address. There are further
863 details in the main documentation.
865 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
867 TF/06 Removed a few PCRE remnants.
869 TF/07 Automatically extract Exim's version number from tags in the git
870 repository when doing development or release builds.
872 PP/02 Raise smtp_cmd_buffer_size to 16kB.
873 Bugzilla 879. Patch from Paul Fisher.
875 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
876 Heavily based on revision 40f9a89a from Simon Arlott's tree.
879 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
881 PP/05 Variable $av_failed, true if the AV scanner deferred.
882 Bugzilla 1078. Patch from John Horne.
884 PP/06 Stop make process more reliably on build failure.
885 Bugzilla 1087. Patch from Heiko Schlittermann.
887 PP/07 Make maildir_use_size_file an _expandable_ boolean.
888 Bugzilla 1089. Patch from Heiko Schlittermann.
890 PP/08 Handle ${run} returning more data than OS pipe buffer size.
891 Bugzilla 1131. Patch from Holger Weiß.
893 PP/09 Handle IPv6 addresses with SPF.
894 Bugzilla 860. Patch from Wolfgang Breyha.
896 PP/10 GnuTLS: support TLS 1.2 & 1.1.
898 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
901 PP/11 match_* no longer expand right-hand-side by default.
902 New compile-time build option, EXPAND_LISTMATCH_RHS.
903 New expansion conditions, "inlist", "inlisti".
905 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
907 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
909 PP/14 fix log_write() format string regression from TF/03.
910 Bugzilla 1152. Patch from Dmitry Isaikin.
916 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
918 PP/02 Harmonised TLS library version reporting; only show if debugging.
919 Layout now matches that introduced for other libraries in 4.74 PP/03.
921 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
923 PP/04 New "dns_use_edns0" global option.
925 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
928 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
929 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
931 TK/01 Updated PolarSSL code to 0.14.2.
932 Bugzilla 1097. Patch from Andreas Metzler.
934 PP/07 Catch divide-by-zero in ${eval:...}.
937 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
940 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
941 format-string attack -- SECURITY: remote arbitrary code execution.
943 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
944 time unintentionally subject to list matching rules, letting the header
945 cause arbitrary Exim lookups (of items which can occur in lists, *not*
946 arbitrary string expansion). This allowed for information disclosure.
948 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
949 INT_MIN/-1 -- value coerced to INT_MAX.
955 NM/01 Workround for PCRE version dependency in version reporting
958 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
959 This fixes portability to compilers other than gcc, notably
960 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
962 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
963 makefiles for portability to HP-UX and POSIX correctness.
965 PP/01 Permit LOOKUP_foo enabling on the make command-line.
966 Also via indented variable definition in the Makefile.
967 (Debugging by Oliver Heesakkers).
969 PP/02 Restore caching of spamd results with expanded spamd_address.
970 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
972 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
973 Improves build reliability. Fix from: Frank Elsner
975 NM/02 Fix wide character breakage in the rfc2047 coding
976 Fixes bug 1064. Patch from Andrey N. Oktyabrski
978 NM/03 Allow underscore in dnslist lookups
979 Fixes bug 1026. Patch from Graeme Fowler
981 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
982 Code patches from Adam Ciarcinski of NetBSD.
984 NM/04 Fixed exiqgrep to cope with mailq missing size issue
987 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
988 is logged, to avoid truncation. Patch from John Horne.
990 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
991 Patch from Jakob Hirsch.
993 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
994 SQL string expansion failure details.
995 Patch from Andrey Oktyabrski.
997 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
998 Patch from Simon Arlott.
1000 PP/09 New lookups functionality failed to compile on old gcc which rejects
1001 extern declarations in function scope.
1002 Patch from Oliver Fleischmann
1004 PP/10 Use sig_atomic_t for flags set from signal handlers.
1005 Check getgroups() return and improve debugging.
1006 Fixed developed for diagnosis in bug 927 (which turned out to be
1009 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1010 Patch from Mark Zealey.
1012 PP/12 Bugzilla 1056: Improved spamd server selection.
1013 Patch from Mark Zealey.
1015 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1016 Based on patch from Heiko Schlittermann.
1018 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1019 Patch from Uwe Doering, sign-off by Michael Haardt.
1021 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1022 variable declaration deep within a block. Bug and patch from
1025 PP/15 lookups-Makefile IRIX compatibilty coercion.
1027 PP/16 Make DISABLE_DKIM build knob functional.
1029 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1030 Patch by Simon Arlott
1032 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1033 variable argument macros. Our copy now differs from upstream.
1039 TF/01 Failure to get a lock on a hints database can have serious
1040 consequences so log it to the panic log.
1042 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1043 controlled using the smtp_confirmation log selector.
1045 TF/03 Include the error message when we fail to unlink a spool file.
1047 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1048 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1049 for maintaining out-of-tree patches for some time.
1051 PP/01 Bugzilla 139: Documentation and portability issues.
1052 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1053 Handle per-OS dynamic-module compilation flags.
1055 PP/02 Let /dev/null have normal permissions.
1056 The 4.73 fixes were a little too stringent and complained about the
1057 permissions on /dev/null. Exempt it from some checks.
1058 Reported by Andreas M. Kirchwitz.
1060 PP/03 Report version information for many libraries, including
1061 Exim version information for dynamically loaded libraries. Created
1062 version.h, now support a version extension string for distributors
1063 who patch heavily. Dynamic module ABI change.
1065 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1066 privilege escalation vulnerability whereby the Exim run-time user
1067 can cause root to append content of the attacker's choosing to
1070 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1073 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1074 If dropping privileges for untrusted macros, we disabled normal logging
1075 on the basis that it would fail; for the Exim run-time user, this is not
1076 the case, and it resulted in successful deliveries going unlogged.
1077 Fixed. Reported by Andreas Metzler.
1083 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1084 only prepend for the Resent-* case. Fixes regression introduced in
1085 Exim 4.70 by NM/22 for Bugzilla 607.
1087 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1088 increasing numbers of administrators be bitten by this.
1090 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1092 PP/03 Bugzilla 994: added openssl_options main configuration option.
1094 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1096 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1098 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1100 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1101 without a peer certificate, leading to a segfault because of an
1102 assumption that peers always have certificates. Be a little more
1103 paranoid. Problem reported by Martin Tscholak.
1105 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1106 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1107 NB: ClamAV planning to remove STREAM in "middle of 2010".
1108 CL also introduces -bmalware, various -d+acl logging additions and
1109 more caution in buffer sizes.
1111 PP/09 Implemented reverse_ip expansion operator.
1113 PP/10 Bugzilla 937: provide a "debug" ACL control.
1115 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1117 PP/12 Bugzilla 973: Implement --version.
1119 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1121 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1123 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1125 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1126 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1127 ignore trailing whitespace.
1129 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1131 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1134 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1135 Notification from Dr Andrew Aitchison.
1137 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1138 ExtendedDetectionInfo response format.
1139 Notification from John Horne.
1141 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1144 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1145 XSL and documented dependency on system catalogs, with examples of how
1148 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1151 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1152 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1153 configuration file which is writeable by the Exim user or group.
1155 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1156 of configuration files to cover files specified with the -C option if
1157 they are going to be used with root privileges, not just the default
1160 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1161 option (effectively making it always true).
1163 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1164 files to be used while preserving root privileges.
1166 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1167 that rogue child processes cannot use them.
1169 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1170 run-time user, instead of root.
1172 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1173 Exim run-time user without dropping privileges.
1175 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1176 result string, instead of calling string_vformat() twice with the same
1179 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1180 for other users. Others should always drop root privileges if they use
1181 -C on the command line, even for a whitelisted configure file.
1183 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1185 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1191 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1192 $data_path, and $header_path variables; fixed documentation bugs and
1195 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1196 exipick to access non-standard spools, including the "frozen" queue
1199 NM/01 Bugzilla 965: Support mysql stored procedures.
1200 Patch from Alain Williams
1202 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1204 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1205 Patch from Andreas Metzler
1207 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1208 Patch from Kirill Miazine
1210 NM/05 Bugzilla 671: Added umask to procmail example.
1212 JJ/03 installed exipick 20100323.0, fixing doc bug
1214 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1215 directory. Notification and patch from Dan Rosenberg.
1217 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1219 TK/02 Improve log output when DKIM signing operation fails.
1221 MH/01 Treat the transport option dkim_domain as a colon separated
1222 list, not as a single string, and sign the message with each element,
1223 omitting multiple occurences of the same signer.
1225 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1226 Bugzilla 985, 986. Patch by Simon Arlott
1228 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1229 Patch by Simon Arlott
1231 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1232 MBX locking. Notification from Dan Rosenberg.
1238 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1240 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1242 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1244 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1246 NM/04 Bugzilla 915: Flag broken perl installation during build.
1252 TK/01 Added patch by Johannes Berg that expands the main option
1253 "spamd_address" if it starts with a dollar sign.
1255 TK/02 Write list of recipients to X-Envelope-Sender header when building
1256 the mbox-format spool file for content scanning (suggested by Jakob
1259 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1260 (http://www.dcc-servers.net/) support via dccifd. Activated by
1261 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1263 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1264 by Mark Daniel Reidel <mr@df.eu>.
1266 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1267 When building exim an external PCRE library is now needed -
1268 PCRE is a system library on the majority of modern systems.
1269 See entry on PCRE_LIBS in EDITME file.
1271 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1272 conversation. Added nologin parameter to request.
1273 Patch contributed by Kirill Miazine.
1275 TF/01 Do not log submission mode rewrites if they do not change the address.
1277 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1279 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1280 log files in place. Contributed by Roberto Lima.
1282 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1284 TF/03 Bugzilla 615: When checking the local_parts router precondition
1285 after a local_part_suffix or local_part_prefix option, Exim now
1286 does not use the address's named list lookup cache, since this
1287 contains cached lookups for the whole local part.
1289 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1290 Robert Millan. Documentation is in experimental-spec.txt.
1292 TF/04 Bugzilla 668: Fix parallel build (make -j).
1294 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1296 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1297 Patch provided by Jan Srzednicki.
1299 TF/05 Leading white space used to be stripped from $spam_report which
1300 wrecked the formatting. Now it is preserved.
1302 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1303 that they are available at delivery time.
1305 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1307 TF/08 TLS error reporting now respects the incoming_interface and
1308 incoming_port log selectors.
1310 TF/09 Produce a more useful error message if an SMTP transport's hosts
1311 setting expands to an empty string.
1313 NM/06 Bugzilla 744: EXPN did not work under TLS.
1314 Patch provided by Phil Pennock.
1316 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1317 Patch provided by Richard Godbee.
1319 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1320 acl_smtp_notquit, added index entry.
1322 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1323 Patch provided by Eugene Bujak.
1325 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1326 accept(). Patch provided by Maxim Dounin.
1328 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1329 Patch provided by Phil Pennock.
1331 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1333 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1334 Patch provided by Brad "anomie" Jorsch.
1336 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1337 Patch provided by Dean Brooks.
1339 TK/05 Add native DKIM support (does not depend on external libraries).
1341 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1342 Patch provided by Graeme Fowler.
1344 NM/16 Bugzilla 851: Documentation example syntax fix.
1346 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1348 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1351 NM/19 Bugzilla 745: TLS version reporting.
1352 Patch provided by Phil Pennock.
1354 NM/20 Bugzilla 167: bool: condition support.
1355 Patch provided by Phil Pennock.
1357 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1358 clients. Patch provided by Phil Pennock.
1360 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1361 Patch provided by Brad "anomie" Jorsch.
1363 NM/23 Bugzilla 687: Fix misparses in eximstats.
1364 Patch provided by Heiko Schlittermann.
1366 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1367 Patch provided by Heiko Schlittermann.
1369 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1370 plus update to original patch.
1372 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1374 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1375 Patch provided by David Brownlee.
1377 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1379 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1381 NM/30 Bugzilla 888: TLS documentation bugfixes.
1383 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1385 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1386 Unlike the original bugzilla I have changed all shell scripts in src tree.
1388 NM/33 Bugzilla 898: Transport filter timeout fix.
1389 Patch by Todd Rinaldo.
1391 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1392 Patch by Serge Demonchaux.
1394 NM/35 Bugzilla 39: Base64 decode bug fixes.
1395 Patch by Jakob Hirsch.
1397 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1399 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1401 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1403 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1409 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1410 ALT-N's libdkim that I have put here:
1411 http://duncanthrax.net/exim-experimental/
1413 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1414 were called 'true' and it seems that C99 defines that as a reserved
1415 keyword to be used with 'bool' variable types. That means you could
1416 not include C99-style headers which use bools without triggering
1417 build errors in sieve.c.
1419 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1420 as mailq or other aliases. Changed the --help handling significantly
1421 to do whats expected. exim_usage() emits usage/help information.
1423 SC/01 Added the -bylocaldomain option to eximstats.
1425 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1427 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1429 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1435 PH/01 Another patch from the Sieve maintainer.
1437 PH/02 When an IPv6 address is converted to a string for single-key lookup
1438 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1439 dots are used instead of colons so that keys in lsearch files need not
1440 contain colons. This was done some time before quoting was made available
1441 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1442 (notated using the quote facility) so as to distinguish them from IPv4
1443 keys. This meant that lookups for IP addresses in host lists did not work
1444 for iplsearch lookups.
1446 This has been fixed by arranging for IPv6 addresses to be expressed with
1447 colons if the lookup type is iplsearch. This is not incompatible, because
1448 previously such lookups could never work.
1450 The situation is now rather anomolous, since one *can* have colons in
1451 ordinary lsearch keys. However, making the change in all cases is
1452 incompatible and would probably break a number of configurations.
1454 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1457 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1458 conversion specification without a maximum field width, thereby enabling
1459 a rogue spamd server to cause a buffer overflow. While nobody in their
1460 right mind would setup Exim to query an untrusted spamd server, an
1461 attacker that gains access to a server running spamd could potentially
1462 exploit this vulnerability to run arbitrary code as the Exim user.
1464 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1465 $primary_hostname instead of what libspf2 thinks the hosts name is.
1467 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1468 a directory entry by the name of the lookup key. Previously, if a
1469 symlink pointed to a non-existing file or a file in a directory that
1470 Exim lacked permissions to read, a lookup for a key matching that
1471 symlink would fail. Now it is enough that a matching directory entry
1472 exists, symlink or not. (Bugzilla 503.)
1474 PH/03 The body_linecount and body_zerocount variables are now exported in the
1477 PH/04 Added the $dnslist_matched variable.
1479 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1480 This means they are set thereafter only if the connection becomes
1483 PH/06 Added the client_condition to authenticators so that some can be skipped
1484 by clients under certain conditions.
1486 PH/07 The error message for a badly-placed control=no_multiline_responses left
1487 "_responses" off the end of the name.
1489 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1491 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1492 (without spaces) instead of just copying the configuration text.
1494 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1496 PH/11 Added $max_received_linelength.
1498 PH/12 Added +ignore_defer and +include_defer to host lists.
1500 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1501 way in which PCRE > 7.0 is built.
1503 PH/14 Implemented queue_only_load_latch.
1505 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1506 MAIL command. The effect was to mangle the value on 64-bit systems.
1508 PH/16 Another patch from the Sieve maintainer.
1510 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1512 PH/18 If a system quota error occurred while trying to create the file for
1513 a maildir delivery, the message "Mailbox is full" was not appended to the
1514 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1515 only to a quota excession during the actual writing of the file.
1517 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1518 characters?) which causes problems in log lines. The DN values are now
1519 passed through string_printing() before being added to log lines.
1521 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1522 and InterBase are left for another time.)
1524 PH/21 Added message_body_newlines option.
1526 PH/22 Guard against possible overflow in moan_check_errorcopy().
1528 PH/23 POSIX allows open() to be a macro; guard against that.
1530 PH/24 If the recipient of an error message contained an @ in the local part
1531 (suitably quoted, of course), incorrect values were put in $domain and
1532 $local_part during the evaluation of errors_copy.
1538 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1539 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1542 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1543 issue a MAIL command.
1545 PH/02 In an ACL statement such as
1547 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1549 if a client was not listed at all, or was listed with a value other than
1550 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1551 the condition was not true (as it should be), so access was not denied.
1552 The bug was that the ! inversion was incorrectly passed on to the second
1553 item. This has been fixed.
1555 PH/03 Added additional dnslists conditions == and =& which are different from
1556 = and & when the dns lookup returns more than one IP address.
1558 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1559 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1561 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1562 FSYNC, which compiles an option called disable_fsync that allows for
1563 bypassing fsync(). The documentation is heavily laced with warnings.
1565 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1567 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1568 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1569 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1570 including adding "make clean"; (3) Added -fPIC when compiling the test
1571 dynamically loaded module, to get rid of a warning.
1573 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1574 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1575 = 0s. The bug is otherwise harmless.
1577 PH/07 There was a bug in the dovecot authenticator such that the value of
1578 $auth1 could be overwritten, and so not correctly preserved, after a
1579 successful authentication. This usually meant that the value preserved by
1580 the server_setid option was incorrect.
1582 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1584 PH/09 Installed PCRE release 7.0.
1586 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1587 run for batched SMTP input. It is now run at the start of every message
1588 in the batch. While fixing this I discovered that the process information
1589 (output by running exiwhat) was not always getting set for -bs and -bS
1590 input. This is fixed, and it now also says "batched" for BSMTP.
1592 PH/11 Added control=no_pipelining.
1594 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1595 patch, slightly modified), and move the expansion of helo_data till after
1596 the connection is made in the smtp transport (so it can use these
1599 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1601 PH/14 Added log_selector = +pid.
1603 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1605 PH/16 Add ${if forany and ${if forall.
1607 PH/17 Added dsn_from option to vary the From: line in DSNs.
1609 PH/18 Flush SMTP output before performing a callout, unless control =
1610 no_callout_flush is set.
1612 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1613 was true (the default) a successful delivery failed to delete the retry
1614 item, thus causing premature timeout of the address. The bug is now
1617 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1619 PH/21 Long custom messages for fakedefer and fakereject are now split up
1620 into multiline reponses in the same way that messages for "deny" and
1621 other ACL rejections are.
1623 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1624 with slight modification.
1626 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1627 draft, changing the syntax and factoring some duplicate code".
1629 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1630 for deliveries of the second and subsequent messages over the same SMTP
1633 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1634 ${reduce, with only minor "tidies".
1636 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1638 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1639 expansion side effects.
1641 PH/27 When a message times out after an over-quota error from an Exim-imposed
1642 quota, the bounce message says "mailbox is full". This message was not
1643 being given when it was a system quota that was exceeded. It now should
1646 MH/03 Made $recipients available in local_scan(). local_scan() already has
1647 better access to the recipient list through recipients_list[], but
1648 $recipients can be useful in postmaster-provided expansion strings.
1650 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1651 in the case of a MAIL command with additional options following the
1652 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1653 were accidentally chopped off.
1655 PH/29 SMTP synchronization checks are implemented when a command is read -
1656 there is a check that no more input is waiting when there shouldn't be
1657 any. However, for some commands, a delay in an ACL can mean that it is
1658 some time before the response is written. In this time, more input might
1659 arrive, invalidly. So now there are extra checks after an ACL has run for
1660 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1661 pipelining has not been advertised.
1663 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1665 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1666 returned in the SMTP response when smtp_return_error_details was false.
1667 This has been fixed.
1669 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1670 instead of the C I/O that was originally supplied, because problems were
1671 reported on Solaris.
1673 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1674 Exim which did not show up earlier: it was assuming that a call to
1675 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1676 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1677 was a macro that became an assignment, so it seemed to work. This has
1678 changed to a proper function call with a void return, hence the compile
1679 error. Exim's code has been fixed.
1681 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1684 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1686 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1688 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1689 because it turns out to be too big or there is a timeout) there is no
1690 "Completed" line in the log. When some messages of this type were
1691 selected by exigrep, they were listed as "not completed". Others were
1692 picked up by some special patterns. I have improved the selection
1693 criteria to be more general.
1695 PH/38 The host_find_failed option in the manualroute router can now be set
1696 to "ignore", to completely ignore a host whose IP address cannot be
1697 found. If all hosts are ignored, the behaviour is controlled by the new
1698 host_all_ignored option.
1700 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1701 homing or because of multiple MX records with /mx) generated more than
1702 one IP address, and the following item turned out to be the local host,
1703 all the secondary addresses of the first item were incorrectly removed
1704 from the list, along with the local host and any following hosts (which
1705 is what is supposed to happen).
1707 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1708 whoever called Exim into the -H file. In the case of the daemon it was
1709 behaving confusingly. When first started, it used values for whoever
1710 started the daemon, but after a SIGHUP it used the Exim user (because it
1711 calls itself on a restart). I have changed the code so that it now always
1714 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1715 message are rejected with the same error (e.g. no authentication or bad
1716 sender address), and a DATA command is nevertheless sent (as can happen
1717 with PIPELINING or a stupid MUA), the error message that was given to the
1718 RCPT commands is included in the rejection of the DATA command. This is
1719 intended to be helpful for MUAs that show only the final error to their
1722 PH/42 Another patch from the Sieve maintainer.
1724 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1725 Eximstats - Fixed some broken HTML links and added missing column headers
1727 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1728 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1730 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1732 PH/43 Yet another patch from the Sieve maintainer.
1734 PH/44 I found a way to check for a TCP/IP connection going away before sending
1735 the response to the final '.' that terminates a message, but only in the
1736 case where the client has not sent further data following the '.'
1737 (unfortunately, this is allowed). However, in many cases there won't be
1738 any further data because there won't be any more messages to send. A call
1739 to select() can be used: if it shows that the input is "ready", there is
1740 either input waiting, or the socket has been closed. An attempt to read
1741 the next input character can distinguish the two cases. Previously, Exim
1742 would have sent an OK response which the client would never have see.
1743 This could lead to message repetition. This fix should cure that, at
1744 least in a lot of common cases.
1746 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1747 advertised in response to EHLO.
1753 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1754 fixed by 4.65/MH/01 (is this a record?) are fixed:
1756 (i) An empty string was always treated as zero by the numeric comparison
1757 operators. This behaviour has been restored.
1759 (ii) It is documented that the numeric comparison operators always treat
1760 their arguments as decimal numbers. This was broken in that numbers
1761 starting with 0 were being interpreted as octal.
1763 While fixing these problems I realized that there was another issue that
1764 hadn't been noticed. Values of message_size_limit (both the global option
1765 and the transport option) were treated as octal if they started with 0.
1766 The documentation was vague. These values are now always treated as
1767 decimal, and I will make that clear in the documentation.
1773 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1774 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1777 MH/01 Don't check that the operands of numeric comparison operators are
1778 integers when their expansion is in "skipping" mode (fixes bug
1779 introduced by 4.64-PH/07).
1781 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1782 child addresses, Exim now panics and dies. Previously, because the count
1783 is held in a short int, deliveries were likely to be lost. As such a
1784 large number of recipients for a single message is ridiculous
1785 (performance will be very, very poor), I have chosen to impose a limit
1786 rather than extend the field.
1792 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1793 leftover -K file (the existence of which was triggered by #402).
1794 While we were at it, introduced process PID as part of the -K
1795 filename. This should rule out race conditions when creating
1798 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1799 processing considerably. Previous code took too long for large mails,
1800 triggering a timeout which in turn triggers #401.
1802 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1803 in the DK code in transports.c. sendfile() is not really portable,
1804 hence the _LINUX specificness.
1806 TF/01 In the add_headers option to the mail command in an Exim filter,
1807 there was a bug that Exim would claim a syntax error in any
1808 header after the first one which had an odd number of characters
1811 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1812 callout verification, Exim cached a "reject" for the entire domain. This
1813 is correct for most verifications, but it is not correct for a recipient
1814 verification with use_sender or use_postmaster set, because in that case
1815 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1816 case of MAIL FROM:<> rejection from other early rejections (e.g.
1817 rejection of HELO). When verifying a recipient using a non-null MAIL
1818 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1819 Whatever the result of the callout, the value of the domain cache is
1820 left unchanged (for any other kind of callout, getting as far as trying
1821 RCPT means that the domain itself is ok).
1823 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1826 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1827 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1828 session, the daemon ignores ECONNECT errors and logs others; it now
1829 ignores EPIPE as well.
1831 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1832 (quoted-printable decoding).
1834 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1835 later the small subsequent patch to fix an introduced bug.
1837 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1839 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1841 PH/08 An error is now given if message_size_limit is specified negative.
1843 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1844 to be given (somewhat) arbitrary names.
1846 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1849 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1850 miscellaneous code fixes
1852 PH/10 Added the log_reject_target ACL modifier to specify where to log
1855 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1856 hostname. This is wrong, because it relates to the incoming message (and
1857 probably the interface on which it is arriving) and not to the outgoing
1858 callout (which could be using a different interface). This has been
1859 changed to use the value of the helo_data option from the smtp transport
1860 instead - this is what is used when a message is actually being sent. If
1861 there is no remote transport (possible with a router that sets up host
1862 addresses), $smtp_active_hostname is used.
1864 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1865 tweaks were necessary in order to get it to work (see also 21 below):
1866 (a) The code assumed that strncpy() returns a negative number on buffer
1867 overflow, which isn't the case. Replaced with Exim's string_format()
1869 (b) There were several signed/unsigned issues. I just did the minimum
1870 hacking in of casts. There is scope for a larger refactoring.
1871 (c) The code used strcasecmp() which is not a standard C function.
1872 Replaced with Exim's strcmpic() function.
1873 (d) The code set only $1; it now sets $auth1 as well.
1874 (e) A simple test gave the error "authentication client didn't specify
1875 service in request". It would seem that Dovecot has changed its
1876 interface. Fortunately there's a specification; I followed it and
1877 changed what the client sends and it appears to be working now.
1879 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1882 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1883 address A is aliased to B and C, where B exists and C does not. Without
1884 -v the output is "A verified" because verification stops after a
1885 successful redirection if more than one address is generated. However,
1886 with -v the child addresses are also verified. Exim was outputting "A
1887 failed to verify" and then showing the successful verification for C,
1888 with its parentage. It now outputs "B failed to verify", showing B's
1889 parentage before showing the successful verification of C.
1891 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1892 look up a TXT record in a specific list after matching in a combined
1895 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1896 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1897 they consult the DNS. I had assumed they would set it the way they
1898 wanted; and indeed my experiments on Linux seem to show that in some
1899 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1900 To be on the safe side, however, I have now made the interface to
1901 host_find_byname() similar to host_find_bydns(), with an argument
1902 containing the DNS resolver options. The host_find_byname() function now
1903 sets these options at its start, just as host_find_bydns() does. The smtp
1904 transport options dns_qualify_single and dns_search_parents are passed to
1905 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
1906 of host_find_byname() use the default settings of RES_DEFNAMES
1907 (qualify_single) but not RES_DNSRCH (search_parents).
1909 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
1910 spool_read_header() do less string testing, by means of a preliminary
1911 switch on the second character of optional "-foo" lines. (This is
1912 overdue, caused by the large number of possibilities that now exist.
1913 Originally there were few.) While I was there, I also converted the
1914 str(n)cmp tests so they don't re-test the leading "-" and the first
1915 character, in the hope this might squeeze out yet more improvement.
1917 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
1918 flag allowing group syntax was set by the header_syntax check but not
1919 turned off, possible causing trouble later; (2) The flag was not being
1920 set at all for the header_verify test, causing "group"-style headers to
1921 be rejected. I have now set it in this case, and also caused header_
1922 verify to ignore an empty address taken from a group. While doing this, I
1923 came across some other cases where the code for allowing group syntax
1924 while scanning a header line wasn't quite right (mostly, not resetting
1925 the flag correctly in the right place). These bugs could have caused
1926 trouble for malformed header lines. I hope it is now all correct.
1928 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
1929 with the "reply" argument non-NULL. The code, however (which originally
1930 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
1931 but it didn't always do it. This confused somebody who was copying the
1932 code for some other use. I have removed all the tests.
1934 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
1935 feature that was used to support insecure browsers during the U.S. crypto
1936 embargo. It requires special client support, and Exim is probably the
1937 only MTA that supported it -- and would never use it because real RSA is
1938 always available. This code has been removed, because it had the bad
1939 effect of slowing Exim down by computing (never used) parameters for the
1940 RSA_EXPORT functionality.
1942 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
1943 authenticator to fail if there's a tab character in the incoming data
1944 (there should never be unless someone is messing about, as it's supposed
1945 to be base64-encoded). Also added, on Timo's advice, the "secured" option
1946 if the connection is using TLS or if the remote IP is the same as the
1947 local IP, and the "valid-client-cert option" if a client certificate has
1950 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
1951 authenticators. This can be used for authorization after authentication
1952 succeeds. (In the case of plaintext, it servers for both authentication
1955 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1956 if any retry times were supplied.
1958 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1959 connection, where there is no client IP address to check. In this
1960 situation, the verify now always succeeds.
1962 PH/25 Applied John Jetmore's -Mset patch.
1964 PH/26 Added -bem to be like -Mset, but loading a message from a file.
1966 PH/27 In a string expansion for a processed (not raw) header when multiple
1967 headers of the same name were present, leading whitespace was being
1968 removed from all of them, but trailing whitespace was being removed only
1969 from the last one. Now trailing whitespace is removed from each header
1970 before concatenation. Completely empty headers in a concatenation (as
1971 before) are ignored.
1973 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1974 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1976 PH/29 [Removed. This was a change that I later backed out, and forgot to
1977 correct the ChangeLog entry (that I had efficiently created) before
1978 committing the later change.]
1980 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1981 address errors (4xx response to RCPT) over the same connection as other
1982 messages routed to the same hosts. Such deliveries are always "forced",
1983 so retry times are not inspected. This resulted in far too many retries
1984 for the affected addresses. The effect occurred only when there were more
1985 hosts than the hosts_max_try setting in the smtp transport when it had
1986 the 4xx errors. Those hosts that it had tried were not added to the list
1987 of hosts for which the message was waiting, so if all were tried, there
1988 was no problem. Two fixes have been applied:
1990 (i) If there are any address or message errors in an SMTP delivery, none
1991 of the hosts (tried or untried) are now added to the list of hosts
1992 for which the message is waiting, so the message should not be a
1993 candidate for sending over the same connection that was used for a
1994 successful delivery of some other message. This seems entirely
1995 reasonable: after all the message is NOT "waiting for some host".
1996 This is so "obvious" that I'm not sure why it wasn't done
1997 previously. Hope I haven't missed anything, but it can't do any
1998 harm, as the worst effect is to miss an optimization.
2000 (ii) If, despite (i), such a delivery is accidentally attempted, the
2001 routing retry time is respected, so at least it doesn't keep
2002 hammering the server.
2004 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2005 in ${readsocket because some servers need this prod.
2007 PH/32 Added some extra debug output when updating a wait-xxx database.
2009 PH/33 The hint "could be header name not terminated by colon", which has been
2010 given for certain expansion errors for a long time, was not being given
2011 for the ${if def:h_colon_omitted{... case.
2013 PH/34 The spec says: "With one important exception, whenever a domain list is
2014 being scanned, $domain contains the subject domain." There was at least
2015 one case where this was not true.
2017 PH/35 The error "getsockname() failed: connection reset by peer" was being
2018 written to the panic log as well as the main log, but it isn't really
2019 panic-worthy as it just means the connection died rather early on. I have
2020 removed the panic log writing for the ECONNRESET error when getsockname()
2023 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2024 runs only) independently of the message's sender address. This meant
2025 that, if the 4xx error was in fact related to the sender, a different
2026 message to the same recipient with a different sender could confuse
2027 things. In particualar, this can happen when sending to a greylisting
2028 server, but other circumstances could also provoke similar problems.
2029 I have changed the default so that the retry time for these errors is now
2030 based a combination of the sender and recipient addresses. This change
2031 can be overridden by setting address_retry_include_sender=false in the
2034 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2035 remote server are returned as part of bounce messages. This was not
2036 happening for LMTP over a pipe (the lmtp transport), but now it is the
2037 same for both kinds of LMTP.
2039 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2040 in header lines that were in fact CNAMEs. This is no longer the case.
2042 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2043 and queue runs started by the daemon processed all messages. This has
2044 been fixed so that -R and -S can now usefully be given with -q<time>.
2046 PH/40 Import PCRE release 6.7 (fixes some bugs).
2048 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2050 PH/42 Give an error if -q is specified more than once.
2052 PH/43 Renamed the variables $interface_address and $interface_port as
2053 $received_ip_address and $received_port, to make it clear that these
2054 values apply to message reception, and not to the outgoing interface when
2055 a message is delivered. (The old names remain recognized, of course.)
2057 PH/44 There was no timeout on the connect() call when using a Unix domain
2058 socket in the ${readsocket expansion. There now is.
2060 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2061 be meaningful with "accept".
2063 SC/01 Eximstats V1.43
2064 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2066 SC/02 Eximstats V1.44
2067 Use a glob alias rather than an array ref in the generated
2068 parser. This improves both readability and performance.
2070 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2071 Collect SpamAssassin and rejection statistics.
2072 Don't display local sender or destination tables unless
2073 there is data to show.
2074 Added average volumes into the top table text output.
2076 SC/04 Eximstats V1.46
2077 Collect data on the number of addresses (recipients)
2078 as well as the number of messages.
2080 SC/05 Eximstats V1.47
2081 Added 'Message too big' to the list of mail rejection
2082 reasons (thanks to Marco Gaiarin).
2084 SC/06 Eximstats V1.48
2085 Mainlog lines which have GMT offsets and are too short to
2086 have a flag are now skipped.
2088 SC/07 Eximstats V1.49 (Alain Williams)
2089 Added the -emptyok flag.
2091 SC/08 Eximstats V1.50
2092 Fixes for obtaining the IP address from reject messages.
2094 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2095 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2096 whitesspace changes from 4.64-PH/27
2098 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2101 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2102 are found, allow negative numbers in numeric criteria)
2104 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2106 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2107 to match changes made in 4.64-PH/43
2109 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2111 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2112 "domain" is actually an IP address, and force a failure. This locks out
2113 those revolvers/nameservers that support "A-for-A" lookups, in
2114 contravention of the specifications.
2116 PH/48 When a host name was looked up from an IP address, and the subsequent
2117 forward lookup of the name timed out, the host name was left in
2118 $sender_host_name, contrary to the specification.
2120 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2121 restricted to single-key lookups, Exim was not diagnosing an error if
2122 * or *@ was used with a query-style lookup.
2124 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2126 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2127 long ago, but noone interested enough thought of it. Let's just say that
2128 the "1.1" means that there are some new functions that weren't there at
2129 some point in the past.
2131 PH/51 Error processing for expansion failure of helo_data from an smtp
2132 transport during callout processing was broken.
2134 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2135 tested/used via the -bh/-bhc/-bs options.
2137 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2138 bug, fixed in subsequent PCRE releases).
2140 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2141 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2143 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2149 SC/01 Use a glob alias rather than an array ref in eximstats generated
2150 parser. This improves both readability and performance.
2152 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2153 Don't display local sender or destination tables in eximstats unless
2154 there is data to show.
2155 Added average volumes into the eximstats top table text output.
2157 SC/03 Collect data on the number of addresses (recipients) as well
2158 as the number of messages in eximstats.
2160 TF/01 Correct an error in the documentation for the redirect router. Exim
2161 does (usually) call initgroups() when daemonizing.
2163 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2164 with consistent privilege compared to when running as a daemon.
2166 TF/03 Note in the spec that $authenticated_id is not set for local
2167 submissions from trusted users.
2169 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2170 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2172 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2173 by adding some example configuration directives to the default
2174 configuration file. A little bit of work is required to uncomment the
2175 directives and define how usernames and passwords are checked, but
2176 there is now a framework to start from.
2178 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2179 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2180 without this. I don't know how relevant this is to other LDAP libraries.
2182 PH/02 Add the verb name to the "unknown ACL verb" error.
2184 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2186 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2188 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2189 directory not expanded when it should be if an expanded home directory
2190 was set for the address (which is overridden by the transport).
2192 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2195 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2196 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2197 because it is too late at that time, and has no effect.
2199 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2200 security issue with \' (bugzilla #107). I could not use the
2201 PQescapeStringConn() function, because it needs a PGconn value as one of
2204 PH/08 When testing addresses using -bt, indicate those final addresses that
2205 are duplicates that would not cause an additional delivery. At least one
2206 person was confused, thinking that -bt output corresponded to deliveries.
2207 (Suppressing duplicates isn't a good idea as you lose the information
2208 about possibly different redirections that led to the duplicates.)
2210 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2211 systems where poll() doesn't work, in particular OS X.
2213 PH/10 Added more information to debugging output for retry time not reached.
2215 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2216 operations in malware.c.
2218 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2221 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2222 syslog_duplication set false, log lines that would normally be written
2223 both the the main log and to the reject log were not written to syslog at
2226 PH/14 In the default configuration, change the use of "message" in ACL warn
2227 statements to "add_header".
2229 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2230 not followed by a command (e.g. "seen endif").
2232 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2233 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2236 PH/17 Added extra conditions to the default value of delay_warning_condition
2240 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2241 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2242 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2245 The Auto-Submitted: and various List- headers are standardised, whereas I
2246 don't think Precedence: ever was.
2248 PH/18 Refactored debugging code in route_finduser() to show more information,
2249 in particular, the error code if getpwnam() issues one.
2251 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2252 This is apparently needed in addition to the PH/07 change above to avoid
2253 any possible encoding problems.
2255 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2256 but not after initializing Perl.
2258 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2259 output them only if debugging. By default they are written stderr,
2260 apparently, which is not desirable.
2262 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2265 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2268 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2270 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2271 authenticated or an ident call has been made. Suppress the default
2272 values for $authenticated_id and $authenticated_sender (but permit -oMai
2273 and -oMas) when testing with -bh.
2275 PH/24 Re-jigged the order of the tests in the default configuration so that the
2276 tests for valid domains and recipients precede the DNS black list and CSA
2277 tests, on the grounds that those ones are more expensive.
2279 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2280 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2281 This bug exists in every version of Exim that I still have, right back to
2284 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2285 However, an attempt to turn on case-sensitivity in a regex key by
2286 including (?-i) didn't work because the subject string was already
2287 lowercased, and the effects were non-intuitive. It turns out that a
2288 one-line patch can be used to allow (?-i) to work as expected.
2294 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2295 other effects) broke the use of negated acl sub-conditions.
2297 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2300 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2301 "Deny" causes Exim to reject the incoming connection with a 554 error.
2302 Unfortunately, if there is a major crisis, such as a disk failure,
2303 tcp-wrappers gives "deny", whereas what one would like would be some
2304 kind of temporary error. A kludge has been added to help with this.
2305 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2306 554 error is used if errno is still zero or contains ENOENT (which occurs
2307 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2310 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2312 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2313 errors. Otherwise a message that provokes a temporary error (when other
2314 messages do not) can cause a whole host to time out.
2316 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2317 addresses were routed directly to files or pipes from a redirect router.
2318 File deliveries just didn't batch; pipe deliveries might have suffered
2321 PH/06 A failure to get a lock for a hints database would erroneously always say
2322 "Failed to get write lock", even when it was really a read lock.
2324 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2325 of 0600. This has been changed to use the value of the lockfile_mode
2326 option (which defaults to 0600).
2328 PH/08 Applied small patch from the Sieve maintainer.
2330 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2331 folder from quota calculations, a direct delivery into this folder messed
2332 up the contents of the maildirsize file. This was because the regex was
2333 used only to exclude .Trash (or whatever) when the size of the mailbox
2334 was calculated. There was no check that a delivery was happening into an
2335 excluded directory. This bug has been fixed by ignoring all quota
2336 processing for deliveries into excluded directories.
2338 PH/10 Added the maildirfolder_create_regex option to appendfile.
2344 PH/01 The code for finding all the local interface addresses on a FreeBSD
2345 system running IPv6 was broken. This may well have applied to all BSD
2346 systems, as well as to others that have similar system calls. The broken
2347 code found IPv4 interfaces correctly, but gave incorrect values for the
2348 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2349 that it would not match correctly against @[] and not recognize the IPv6
2352 PH/02 The ipliteral router was not recognizing addresses of the form user@
2353 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2355 PH/03 Added disable_ipv6.
2357 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2358 decoded form, because it is most often used to construct To: headers
2359 lines in autoreplies, and the decoded form may well be syntactically
2360 invalid. However, $reply_address has leading white space removed, and all
2361 newlines turned into spaces so that the autoreply transport does not
2364 PH/05 If group was specified without a user on a router, and no group or user
2365 was specified on a transport, the group from the router was ignored.
2367 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2368 for visible compile-time settings that can be used to change these
2369 numbers, for those that want even more. Backwards compatibility with old
2370 spool files has been maintained. However, going back to a previous Exim
2371 release will lost any variables that are in spool files.
2373 PH/07 Two small changes when running in the test harness: increase delay when
2374 passing a TCP/IP connection to a new process, in case the original
2375 process has to generate a bounce, and remove special handling of
2376 127.0.0.2 (sic), which is no longer necessary.
2378 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2379 be the same on different OS.
2381 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2384 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2385 whether --show-vars was specified or not
2387 JJ/02 exipick: Added support for new ACL variable spool format introduced
2390 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2391 syntactically invalid From: or Reply-to: line, and a filter used this to
2392 generate an autoreply, and therefore failed to obtain an address for the
2393 autoreply, Exim could try to deliver to a non-existent relative file
2394 name, causing unrelated and misleading errors. What now happens is that
2395 it logs this as a hard delivery error, but does not attempt to create a
2398 PH/11 The exinext utility has a -C option for testing purposes, but although
2399 the given file was scanned by exinext itself; it wasn't being passed on
2400 when Exim was called.
2402 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2403 an end-of-file indication when reading a command response.
2405 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2406 compiled. In many other places in Exim, IPv6 addresses are always
2407 recognized, so I have changed this. It also means that IPv4 domain
2408 literals of the form [IPV4:n.n.n.n] are now always recognized.
2410 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2411 used if the router is not running as root, for example, when verifying at
2412 ACL time, or when using -bh. The debugging output from this situation was
2413 non-existent - all you got was a failure to exec. I have made two
2416 (a) Failures to set uid/gid, the current directory, or a process leader
2417 in a subprocess such as that created by queryprogram now generate
2418 suitable debugging ouput when -d is set.
2420 (b) The queryprogram router detects when it is not running as root,
2421 outputs suitable debugging information if -d is set, and then runs
2422 the subprocess without attempting to change uid/gid.
2424 PH/15 Minor change to Makefile for building test_host (undocumented testing
2427 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2428 additional section of a DNS packet that returns MX or SRV records.
2429 Instead, it always explicitly searches for A/AAAA records. This avoids
2430 major problems that occur when a DNS server includes only records of one
2431 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2432 fixed another bug: if SRV records were looked up and the corresponding
2433 address records were *not* found in the additional section, the port
2434 values from the SRV records were lost.
2436 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2437 using the correct key (the original address) when searching the retry
2438 rules in order to find which one to use for generating the retry hint.
2440 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2441 adding the default one. Similarly, if it contains a Reply-To: header, the
2442 errors_reply_to option, if set, is not used.
2444 PH/19 When calculating a retry time, Exim used to measure the "time since
2445 failure" by looking at the "first failed" field in the retry record. Now
2446 it does not use this if it is later than than the arrival time of the
2447 message. Instead it uses the arrival time. This makes for better
2448 behaviour in cases where some deliveries succeed, thus re-setting the
2449 "first failed" field. An example is a quota failure for a huge message
2450 when small messages continue to be delivered. Without this change, the
2451 "time since failure" will always be short, possible causing more frequent
2452 delivery attempts for the huge message than are intended.
2453 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2455 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2456 $1, $2, $3) because the numerical variables can be reset during some
2457 expansion items (e.g. "match"), thereby losing the authentication data.
2459 PH/21 Make -bV show the size of off_t variables so that the test suite can
2460 decide whether to run tests for quotas > 2G.
2462 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2463 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2464 filecount value is greater than 2G or if a quota value is greater than 2G
2465 on a system where the size of off_t is not greater than 4, a panic error
2468 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2469 never match. The debug and -bh output now contains an explicit error
2470 message indicating a malformed IPv4 address or mask.
2472 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2473 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2474 PH/23 above applies.
2476 PH/25 Do not write to syslog when running in the test harness. The only
2477 occasion when this arises is a failure to open the main or panic logs
2478 (for which there is an explicit test).
2480 PH/26 Added the /no_tell option to "control=freeze".
2482 PH/27 If a host name lookup failed very early in a connection, for example, if
2483 the IP address matched host_lookup and the reverse lookup yielded a name
2484 that did not have a forward lookup, an error message of the form "no IP
2485 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2486 could be logged. Now it outputs the IP address instead of "NULL".
2488 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2489 allows the sender and the authenticated sender to be set when
2490 submitting a message from within Exim. Since child_open_exim() is
2491 documented for local_scan(), the new function should be too.
2493 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2494 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2495 results in an empty string is now treated as unset.
2497 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2499 PH/31 Added sender_verify_fail logging option.
2501 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2502 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2503 tidied the source and removed it altogether.
2505 PH/33 When a queue run was abandoned because the load average was too high, a
2506 log line was always written; now it is written only if the queue_run log
2507 selector is set. In addition, the log line for abandonment now contains
2508 information about the queue run such as the pid. This is always present
2509 in "start" and "stop" lines but was omitted from the "abandon" line.
2511 PH/34 Omit spaces between a header name and the colon in the error message that
2512 is given when verify = headers_syntax fails (if there are lots of them,
2513 the message gets confusing).
2515 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2516 names, as there are now some PTR records that contain slashes. This check
2517 is only to protect against broken name servers that fall over on strange
2518 characters, so the fact that it applies to all lookups doesn't matter.
2520 PH/36 Now that the new test suite is complete, we can remove some of the
2521 special code in Exim that was needed for the old test suite. For example,
2522 sorting DNS records because real resolvers return them in an arbitrary
2523 order. The new test suite's fake resolver always returns records in the
2526 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2527 bounces) except when queue_only is set, to avoid logging races between
2528 the different processes.
2530 PH/38 Panic-die if .include specifies a non-absolute path.
2532 PH/39 A tweak to the "H" retry rule from its user.
2534 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2535 a label. They prevented compilation on older perls.
2537 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2538 a warning to be raised on newish perls.
2540 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2541 on queue. Changes to match documented behaviour of showing count of
2542 messages matching specified criteria.
2544 PH/40 Changed the default ident timeout from 30s to 5s.
2546 PH/41 Added support for the use of login_cap features, on those BSD systems
2547 that have them, for controlling the resources used by pipe deliveries.
2549 PH/42 The content-scanning code uses fopen() to create files in which to put
2550 message data. Previously it was not paying any attention to the mode of
2551 the files. Exim runs with umask(0) because the rest of the code creates
2552 files with open(), and sets the required mode explicitly. Thus, these
2553 files were ending up world-writeable. This was not a big issue, because,
2554 being within the spool directory, they were not world-accessible. I have
2555 created a function called modefopen, which takes an additional mode
2556 argument. It sets umask(777), creates the file, chmods it to the required
2557 mode, then resets the umask. All the relevant calls to fopen() in the
2558 content scanning code have been changed to use this function.
2560 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2561 to 24 hours. This avoids potential overflow problems when processing G
2562 and H retry rules. I suspect nobody ever tinkers with this value.
2564 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2566 PH/45 When the plaintext authenticator is running as a client, the server's
2567 challenges are checked to ensure they are valid base64 strings. By
2568 default, the authentication attempt is cancelled if an invalid string is
2569 received. Setting client_ignore_invalid_base64 true ignores these errors.
2570 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2571 they are received. Thus, the responses can be made to depend on the
2572 challenges. If an invalid string is ignored, an empty string is placed in
2575 PH/46 Messages that are created by the autoreply transport now contains a
2576 References: header, in accordance with RFCs 2822 and 3834.
2578 PH/47 Added authenticated_sender_force to the smtp transport.
2580 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2582 PH/49 Installed latest patch from the Sieve maintainer.
2584 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2585 was also set, the appendfile transport was unnecessarily scanning a
2586 directory of message files (e.g. for maildir delivery) to find the count
2587 of files (along with the size), even though it did not need this
2588 information. It now does the scan only if it needs to find either the
2589 size of the count of files.
2591 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2593 PH/52 Two bugs concerned with error handling when the smtp transport is
2596 (i) Exim was not creating retry information for temporary errors given
2597 for individual recipients after the DATA command when the smtp transport
2598 was used in LMTP mode. This meant that they could be retried too
2599 frequently, and not timed out correctly.
2601 (ii) Exim was setting the flag that allows error details to be returned
2602 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2603 recipients that were returned after the DATA command.
2605 PH/53 This is related to PH/52, but is more general: for any failing address,
2606 when detailed error information was permitted to be returned to the
2607 sender, but the error was temporary, then after the final timeout, only
2608 "retry timeout exceeded" was returned. Now it returns the full error as
2609 well as "retry timeout exceeded".
2611 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2612 do this, and (what is worse) MTAs that accept it.
2614 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2615 will now be deprecated.
2617 PH/56 New os.c-cygwin from the Cygwin maintainer.
2619 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2620 formats (previously only available in exim formats via -bpr, -bpru,
2621 and -bpra. Now also available in native and exiqgrep formats)
2623 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2624 with very large, slow to parse queues
2626 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2628 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2630 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2631 responses to authentication challenges, though it was showing the
2632 challenges; (ii) I've removed the CR characters from the debug output for
2635 PH/58 Allow for the insertion of a newline as well as a space when a string
2636 is turned into more than one encoded-word during RFC 2047 encoding. The
2637 Sieve code now uses this.
2639 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2640 data_4xx, lost_connection, tls_required.
2642 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2643 message was being sent as an SMTP response.
2645 PH/61 Add -l and -k options to exicyclog.
2647 PH/62 When verifying, if an address was redirected to one new address, so that
2648 verification continued, and the new address failed or deferred after
2649 having set something in $address_data, the value of $address_data was not
2650 passed back to the ACL. This was different to the case when no
2651 redirection occurred. The value is now passed back in both cases.
2653 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2654 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2655 use login_cap.h, so on its own it isn't the distinguishing feature. The
2656 new name refers directly to the setclassresources() function.
2658 PH/65 Added configuration files for NetBSD3.
2660 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2662 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2663 is preferred over IPv4.
2665 PH/68 The bounce_return_message and bounce_return_body options were not being
2666 honoured for bounces generated during the reception of non-SMTP messages.
2667 In particular, this applied to messages rejected by the ACL. This bug has
2668 been fixed. However, if bounce_return_message is true and bounce_return_
2669 body is false, the headers that are returned for a non-SMTP message
2670 include only those that have been read before the error was detected.
2671 (In the case of an ACL rejection, they have all been read.)
2673 PH/69 The HTML version of the specification is now built in a directory called
2674 spec_html instead of spec.html, because the latter looks like a path with
2675 a MIME-type, and this confuses some software.
2677 PH/70 Catch two compiler warnings in sieve.c.
2679 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2680 function verify_get_ident() calls ip_connect() to connect a socket, but
2681 if the "connect()" function timed out, ip_connect() used to close the
2682 socket. However, verify_get_ident() also closes the socket later, and in
2683 between Exim writes to the log, which may get opened at this point. When
2684 the socket was closed in ip_connect(), the log could get the same file
2685 descriptor number as the socket. This naturally causes chaos. The fix is
2686 not to close the socket in ip_connect(); the socket should be closed by
2687 the function that creates it. There was only one place in the code where
2688 this was missing, in the iplookup router, which I don't think anybody now
2689 uses, but I've fixed it anyway.
2691 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2692 well as to direct DNS lookups. Otherwise the handling of names in host
2693 lists is inconsistent and therefore confusing.
2699 PH/01 Two changes to the default runtime configuration:
2701 (1) Move the checks for relay_from_hosts and authenticated clients from
2702 after to before the (commented out) DNS black list checks.
2704 (2) Add control=submission to the relay_from_hosts and authenticated
2705 clients checks, on the grounds that messages accepted by these
2706 statements are most likely to be submissions.
2708 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2710 (1) Generate an error if the third argument for the ${prvs expansion is
2713 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2716 (3) Reset the variables that are obtained from the first argument of
2717 ${prvscheck and used in the second argument before leaving the code,
2718 because their memory is reclaimed, so using them afterwards may do
2721 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2722 one (it's much easier than Tom thought :-).
2724 (5) Because of (4), we can now allow for the use of $prvscheck_result
2725 inside the third argument.
2727 PH/03 For some reason, the default setting of PATH when running a command from
2728 a pipe transport was just "/usr/bin". I have changed it to
2731 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2732 anything to be listed in the output from -bV.
2734 PH/05 When a filter generated an autoreply, the entire To: header line was
2735 quoted in the delivery log line, like this:
2737 => >A.N.Other <ano@some.domain> <original@ddress> ...
2739 This has been changed so that it extracts the operative address. There
2740 may be more than one such address. If so, they are comma-separated, like
2743 => >ano@some.domain,ona@other.domain <original@ddress> ...
2745 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2746 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2747 not being looked up in the rDNS to get a host name, Exim was showing the
2748 IP address twice in Received: lines, even though the IP addresses were
2749 identical. For example:
2751 Received: from [1.2.3.4] (helo=[1.2.3.4])
2753 However, if the real host name was known, it was omitting the HELO data
2754 if it matched the actual IP address. This has been tidied up so that it
2755 doesn't show the same IP address twice.
2757 PH/07 When both +timestamp and +memory debugging was on, the value given by
2758 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2759 called by the debug printing, thereby overwriting the timestamp buffer.
2760 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2762 PH/08 When the original message was included in an autoreply transport, it
2763 always said "this is a copy of the message, including all the headers",
2764 even if body_only or headers_only was set. It now gives an appropriate
2767 PH/09 Applied a patch from the Sieve maintainer which:
2769 o fixes some comments
2770 o adds the (disabled) notify extension core
2771 o adds some debug output for the result of if/elsif tests
2772 o points to the current vacation draft in the documentation
2773 and documents the missing references header update
2777 o fixes a bug in processing the envelope test (when testing
2778 multiple envelope elements, the last element determinted the
2781 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2782 Electronic Mail") by including:
2784 Auto-submitted: auto-generated
2786 in the messages that it generates (bounce messages and others, such as
2787 warnings). In the case of bounce messages for non-SMTP mesages, there was
2788 also a typo: it was using "Auto_submitted" (underscore instead of
2789 hyphen). Since every message generated by Exim is necessarily in response
2790 to another message, thes have all been changed to:
2792 Auto-Submitted: auto-replied
2794 in accordance with these statements in the RFC:
2796 The auto-replied keyword:
2798 - SHOULD be used on messages sent in direct response to another
2799 message by an automatic process,
2801 - MUST NOT be used on manually-generated messages,
2803 - MAY be used on Delivery Status Notifications (DSNs) and Message
2804 Disposition Notifications (MDNs),
2806 - MUST NOT be used on messages generated by automatic or periodic
2807 processes, except for messages which are automatic responses to
2810 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2811 to the default Received: header definition.
2813 PH/12 Added log selector acl_warn_skipped (default on).
2815 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2816 variables because (a) they are in the wrong storage pool and (b) even if
2817 they were copied, it wouldn't work properly because of the caching.
2819 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2820 checking when decoding. Apparently there are clients that generate
2821 overlong encoded strings. Why am I not surprised?
2823 PH/15 If the first argument of "${if match_address" was not empty, but did not
2824 contain an "@" character, Exim crashed. Now it writes a panic log message
2825 and treats the condition as false.
2827 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2829 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2830 extension "envelope-auth". The code is finished and in agreement with
2831 other implementations, but there is no documentation so far and in fact,
2832 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2833 not changing the active code.
2835 Print executed "if" and "elsif" statements when debugging is used. This
2836 helps a great deal to understand what a filter does.
2838 Document more things not specified clearly in RFC3028. I had all this
2839 sorted out, when out of a sudden new issues came to my mind. Oops."
2841 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2844 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2845 canonical form (as documented). However, after a host name lookup from
2846 the IP address, check_host() was doing a simple string comparison with
2847 addresses acquired from the DNS when checking that the found name did
2848 have the original IP as one of its addresses. Since any found IPv6
2849 addresses are likely to be in abbreviated form, the comparison could
2850 fail. Luckily, there already exists a function for doing the comparison
2851 by converting both addresses to binary, so now that is used instead of
2852 the text comparison.
2854 PH/20 There was another similar case to PH/19, when a complete host name was
2855 given in a host list; looking up its IP address could give an abbreviated
2856 form, whereas the current host's name might or might not be abbreviated.
2857 The same fix has been applied.
2863 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2864 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2867 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2868 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2870 PH/03 Typo: missing ".o" in src/pcre/Makefile.
2872 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2873 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2874 Also, for "Auto-Submitted", treat anything other than "no" as
2875 non-personal, in accordance with RFC 3834. (Previously it treated
2876 anything starting "auto-" as non-personal.)
2878 TF/01 The control=submission/name=... option had a problem with syntax
2879 errors if the name included a slash character. The /name= option
2880 now slurps the rest of the string, so it can include any characters
2881 but it must come last in the list of options (after /sender_retain
2884 PH/05 Some modifications to the interface to the fake nameserver for the new
2892 TK/01 Added the "success_on_redirect" address verification option. See
2893 NewStuff for rationale and an example.
2895 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2897 PH/02 Patch to exigrep to allow it to work on syslog lines.
2899 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2900 fread() to skip over the body file's header line, because in Cygwin the
2901 header line is locked and is inaccessible.
2903 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
2904 co-exist for some time) to make it clear that it is the Exim ID that is
2905 referenced, not the Message-ID: header line.
2907 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
2908 string_format() function, because snprintf() does not exist on all
2911 PH/06 The use of forbid_filter_existstest now also locks out the use of the
2912 ${stat: expansion item.
2914 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
2915 protocol synchronization error", to keep the pedants happy.
2917 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
2918 well as for IRIX systems, when gcc is being used. See the host.c source
2921 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
2923 PH/10 Named domain lists were not working if used in a queue_smtp_domains
2926 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
2927 transport and to the smtp transport in LMTP mode.
2929 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
2931 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
2932 run a filter in a subprocess. This could lead to confusion in subsequent
2933 lookups in the parent process. There should also be a search_tidyup() at
2934 the end of the subprocess.
2936 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
2937 only if the host matched helo_try_verify_hosts, which caused the
2938 verification to occur when the EHLO/HELO command was issued. The ACL just
2939 tested the remembered result. Now, if a previous verification attempt has
2940 not happened, "verify = helo" does it there and then.
2942 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
2944 TK/03 Fix log output including CR from clamd.
2946 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
2947 exist provoked a memory error which could cause a segfault.
2949 PH/15 Installed PCRE 6.2
2951 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2953 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2954 of the problem. Specifically, suggested +O2 rather than +O1 for the
2957 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2959 PH/20 If a delivery was routed to a non-standard port by means of an SRV
2960 record, the port was not correctly logged when the outgoing_port log
2961 selector was set (it logged the transort's default port).
2963 PH/21 Added support for host-specific ports to manualroute, queryprogram,
2964 fallback_hosts, and "hosts" in the smtp transport.
2966 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2967 host errors such as "Connection refused".
2969 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2970 authentication with radiusclient 0.4.9:
2972 - Error returned from rc_read_config was caught wrongly
2973 - Username/password not passed on to radius server due to wrong length.
2975 The presumption is that some radiusclient API changes for 4.51/PH/17
2976 were not taken care of correctly. The code is still untested by me (my
2977 Linux distribution still has 0.3.2 of radiusclient), but it was
2978 contributed by a Radius user.
2980 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2981 expanding the "port" option of the smtp transport.
2983 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2984 while reading a MIME header. Thanks to Tom Hughes for a patch.
2986 PH/24 Include config.h inside local_scan.h so that configuration settings are
2989 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2990 that in an ACL for RCPT (for example), you can examine exactly what was
2993 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2994 commands, but it was not correctly comparing the address with the actual
2995 client host address. Thus, it would show the EHLO address in Received:
2996 header lines when this was not necessary.
2998 PH/27 Added the % operator to ${eval:}.
3000 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3001 it should be ignoring failures (because with -C, for example, it has lost
3002 privilege). It wasn't ignoring creation failures other than "already
3005 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3008 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3009 because an input line was too long, either on its own, or by virtue of
3010 too many continuations, the temporary file was not being removed, and the
3011 return code was incorrect.
3013 PH/31 Missing "BOOL" in function definition in filtertest.c.
3015 PH/32 Applied Sieve patches from the maintainer.
3017 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3019 PH/33 Added "verify = not_blind".
3021 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3022 Local/Makefile (with some defaults set). These are used in built scripts
3023 such as exicyclog, but they have never been used in the exim_install
3024 script (though there are many overriding facilities there). I have
3025 arranged that the exim_install script now takes note of these two
3028 PH/35 Installed configuration files for Dragonfly.
3030 PH/36 When a locally submitted message by a trusted user did not contain a
3031 From: header, and the sender address was obtained from -f or from an SMTP
3032 MAIL command, and the trusted user did not use -F to supply a sender
3033 name, $originator_name was incorrectly used when constructing a From:
3034 header. Furthermore, $originator_name was used for submission mode
3035 messages from external hosts without From: headers in a similar way,
3036 which is clearly wrong.
3038 PH/37 Added control=suppress_local_fixups.
3040 PH/38 When log_selector = +received_sender was set, and the addition of the
3041 sender made the log line's construction buffer exactly full, or one byte
3042 less than full, an overflow happened when the terminating "\n" was
3045 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3046 when the result of a list match is failure because a DNS lookup failed.
3048 PH/40 RM_COMMAND is now used in the building process.
3050 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3051 the "build-* directories that it finds.
3053 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3054 address in a domain literal was a prefix of an interface address.
3056 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3057 when verifying a sender address, unless rewrite_headers is false.
3059 PH/44 Wrote a long comment about why errors_to addresses are verified as
3060 recipients, not senders.
3062 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3063 the ratelimit ACL was added.
3065 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3067 PH/46 Added extra information about PostgreSQL errors to the error string.
3069 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3070 avoiding the need to install special zones in a real server. This is
3071 backwards compatible; if it can't find the fake resolver, it drops back.
3072 Thus, both old and new test suites can be run.
3074 TF/02 Added util/ratelimit.pl
3076 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3077 clock is set back in time.
3079 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3080 Candler <B.Candler@pobox.com>.
3082 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3083 OK for addresses that are the result of redirections.
3085 PH/48 A number of further additions for the benefit of the new test suite,
3086 including a fake gethostbyname() that interfaces to the fake DNS resolver
3089 TF/06 The fix for widen_domains has also been applied to qualify_single and
3090 search_parents which are the other dnslookup options that can cause
3093 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3096 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3098 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3099 service" if it encountered an MX record with an empty target hostname.
3100 The message is now "an MX or SRV record indicated no SMTP service".
3102 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3103 if the DNS of the sending site is misconfigured. This is quite a
3104 common situation. This change restores the behaviour of treating a
3105 helo verification defer as a failure.
3107 PH/51 If self=fail was set on a router, the bounce message did not include the
3108 actual error message.
3114 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3116 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3117 command itself ended in error, the underlying message about the transport
3118 filter timeout was being overwritten with the pipe command error. Now the
3119 underlying error message should be appended to the second error message.
3121 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3123 PH/02 Reduce the amount of output that "make" produces by default. Full output
3124 can still be requested.
3126 PH/03 The warning log line about a condition test deferring for a "warn" verb
3127 was being output only once per connection, rather than after each
3128 occurrence (because it was using the same function as for successful
3129 "warn" verbs). This seems wrong, so I have changed it.
3131 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3132 it should not have, which might have caused a crash in the right
3133 circumstances, but probably never did.
3135 PH/04 Installed a modified version of Tony Finch's patch to make submission
3136 mode fix the return path as well as the Sender: header line, and to
3137 add a /name= option so that you can make the user's friendly name appear
3140 TF/03 Added the control = fakedefer ACL modifier.
3142 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3143 Mark Lowes for thorough testing.
3145 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3147 TK/03 Merged latest SRS patch from Miles Wilton.
3149 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3150 with the definition in sysexits.h (which is #included earlier).
3151 Fortunately, Exim does not actually use EX_OK. The code used to try to
3152 preserve the sysexits.h value, by assumimg that macro definitions were
3153 scanned for macro replacements. I have been disabused of this notion,
3154 so now the code just undefines EX_OK before #including unistd.h.
3156 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3157 in the smtp transport. When a block could not be written in a single
3158 write() function, the timeout was being re-applied to each part-write.
3159 This seems wrong - if the receiver was accepting one byte at a time it
3160 would take for ever. The timeout is now adjusted when this happens. It
3161 doesn't have to be particularly precise.
3163 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3164 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3166 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3167 without a domain if the check to <postmaster@domain> fails.
3169 SC/01 Eximstats: added -xls and the ability to specify output files
3170 (patch written by Frank Heydlauf).
3172 SC/02 Eximstats: use FileHandles for outputing results.
3174 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3176 SC/04 Eximstats: fixed display of large numbers with -nvr option
3178 SC/05 Eximstats: fixed merging of reports with empty tables.
3180 SC/06 Eximstats: added the -include_original_destination flag
3182 SC/07 Eximstats: removed tabs and trailing whitespace.
3184 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3186 TK/06 MBOX spool code: Add real "From " MBOX separator line
3187 so the .eml file is really in mbox format (even though
3188 most programs do not really care). Patch from Alex Miller.
3190 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3191 The latter is generated from $received_to and is only set if the
3192 message has one envelope recipient. SA can use these headers,
3193 obviously out-of-the-box. Patch from Alex Miller.
3195 PH/08 The ${def test on a variable was returning false if the variable's
3196 value was "0", contrary to what the specification has always said!
3197 The result should be true unless the variable is empty.
3199 PH/09 The syntax error of a character other than { following "${if
3200 def:variable_name" (after optional whitespace) was not being diagnosed.
3201 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3202 accidental colon was present, for example, could give incorrect results.
3204 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3205 result is used (not including appendfile, where other changes are about
3208 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3209 This involved changing a lot of size variables from int to off_t. It
3210 should work with maildirs and everything.
3212 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3213 spamd dying while we are connected to it.
3215 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3218 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3219 function for BDB 4.3.
3221 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3223 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3224 the sources for PCRE have been re-arranged and more files are now
3227 PH/15 The code I had for printing potentially long long variables in PH/11
3228 above was not the best (it lost precision). The length of off_t variables
3229 is now inspected at build time, and an appropriate printing format (%ld
3230 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3231 to be "long long int" or "long int". This is needed for the internal
3232 formatting function string_vformat().
3234 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3235 the configuration file to be ":syslog", then the script "guesses" where
3236 the logs files are, rather than using the compiled in default. In our
3237 case the guess is not the same as the compiled default, so the script
3238 suddenly stopped working when I started to use syslog. The patch checks
3239 to see if log_file_path is "". If so, it attempts to read it from exim
3240 with no configuration file to get the compiled in version, before it
3241 falls back to the previous guessing code."
3243 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3244 implementing BATV in an Exim configuration. See NewStuff for the gory
3247 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3248 Makefile that are specific to HP-UX.
3250 PH/18 If the "use_postmaster" option was set for a recipient callout together
3251 with the "random" option, the postmaster address was used as the MAIL
3252 FROM address for the random test, but not for the subsequent recipient
3253 test. It is now used for both.
3255 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3256 patch removes a few documentation additions to RFC 3028, because the
3257 latest draft now contains them. It adds the new en;ascii-case comparator
3258 and a new error check for 8bit text in MIME parts. Comparator and
3259 require names are now matched exactly. I enabled the subaddress
3260 extension, but it is not well tested yet (read: it works for me)."
3262 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3263 rework some of the code of TK/09 above to avoid the hardwired use of
3264 "%lld" and "long long". Replaced the call to snprintf() with a call to
3267 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3268 records point to non-existent hosts", "retry timeout exceeded", and
3269 "retry time not reached for any host after a long failure period".
3271 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3272 experimental DomainKeys support:
3274 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3275 (2) On an error such as an illegally used "control", the wrong name for
3276 the control was given.
3278 These problems did NOT occur unless DomainKeys support was compiled.
3280 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3282 PH/24 Added ${if match_ip condition.
3284 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3285 databases so that it will be absolutely obvious if a crash occurs in the
3286 DB library. This is a regular occurrence (often caused by mis-matched
3289 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3290 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3291 picked up on a user's system that detects such things. There doesn't seem
3292 to be a gcc warning option for this - only an attribute that has to be
3293 put on the function's prototype. It seems that in Fedora Core 4 they have
3294 set this on a number of new functions. No doubt there will be more in due
3297 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3298 specify a transport. However, if an address that was verified by such a
3299 router was the subject of a callout, Exim crashed because it tried to
3300 read the rcpt_include_affixes from the non-existent transport. Now it
3301 just assumes that the setting of that option is false. This bug was
3302 introduced by 4.51/PH/31.
3304 PH/28 Changed -d+all to exclude +memory, because that information is very
3305 rarely of interest, but it makes the output a lot bigger. People tend to
3306 do -d+all out of habit.
3308 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3309 code in os-type was giving problems when libc.so lives in lib64, like on
3312 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3313 aren't the modern standard, and it seems that some systems' include files
3314 don't always have them. Exim was already checking for some of the newer
3315 ones like T_AAAA, and defining it itself. I've added checks for all the
3316 record types that Exim uses.
3318 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3319 not automatically generating a new one, as it is supposed to. This
3320 prevented TLS from working. If the file did exist, but contained invalid
3321 data, a new version was generated, as expected. It was only the case of a
3322 non-existent file that was broken.
3324 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3325 with a change in libdomainkeys > 0.64.
3327 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3328 from DNS. If the selector record carries the flag, it now has
3329 precedence over the domain-wide flag.
3331 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3333 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3334 the use of a port name that isn't defined in /etc/services) occurred, the
3335 message was deferred as in a normal delivery, and thus remained on the
3336 spool, instead of being failed because of the mua_wrapper setting. This
3337 is now fixed, and I tidied up some of the mua_wrapper messages at the
3340 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3341 the messages in a hash of arrays rather than using individual hashes.
3342 This is a bit cleaner and results in dramatic memory savings, albeit
3343 at a slight CPU cost.
3345 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3346 as requested by Marc Sherman.
3348 SC/10 Eximstats: added histograms for user specified patterns as requested
3351 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3353 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3354 fopen() in the content-scanning modules that did not already have it.
3360 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3361 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3363 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3365 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3367 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3368 to test Sieve filters that use "vacation".
3370 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3371 that changes the way the GnuTLS parameters are stored in the cache file.
3372 The new format can be generated externally. For backward compatibility,
3373 if the data in the cache doesn't make sense, Exim assumes it has read an
3374 old-format file, and it generates new data and writes a new file. This
3375 means that you can't go back to an older release without removing the
3378 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3379 work if there are any delivery delays because "one_time" forces the
3380 parent to be marked "delivered", so its unseen clone is never tried
3381 again. For this reason, Exim now forbids the simultaneous setting of
3384 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3385 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3386 entry if you want to know the details. The fix, however, neglected to
3387 consider the case when local delivery batching is involved. The test for
3388 "previously delivered" was not happening when checking to see if an
3389 address could be batched with a previous (undelivered) one; under
3390 certain circumstances this could lead to multiple deliveries to the same
3393 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3394 in its include files, and this causes problems building Exim.
3396 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3397 header_syntax) but Exim was just ignoring anything given after a slash.
3398 In particular, this caused confusion with an attempt to use "verify =
3399 reverse_host_lookup/defer_ok". An error is now given when options are
3400 supplied for verify items that do not have them. (Maybe reverse_host_
3401 lookup should have a defer_ok option, but that's a different point.)
3403 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3404 defined by RFC 821) to 2048, because there were problems with some AUTH
3405 commands, and RFC 1869 says the size should be increased for extended
3406 SMTP commands that take arguments.
3408 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3411 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3412 "unknown" error; now it says that the functionality isn't in the binary.
3414 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3415 an address' error message when a string expansion fails (syntax or
3416 whatever). Otherwise the password may appear in the log. Following change
3417 PH/42 below, there is no longer a chance of it appearing in a bounce
3420 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3422 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3423 of its addresses was ever tried. (Bugzilla bug #2.)
3425 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3426 the result incorrectly in the debug output. (It correctly added a newline
3427 to what was transported.)
3429 TF/01 Added $received_time.
3431 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3432 commented out examples of how to interface to a virus scanner and to
3433 SpamAssassin. Also added commented examples of av_scanner and
3434 spamd_address settings.
3436 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3437 and controls are allowed in which ACLs. There were a couple of minor
3438 errors. Some of the entries in the conditions table (which is a table of
3439 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3440 negation of where the condition IS allowed.
3442 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3444 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3445 header file does not have a version number, so I've had to invent a new
3446 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3447 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3448 radiusclient), but it was contributed by a Radius user.
3450 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3451 files or directories, for OpenSSL.
3453 PH/19 When an Exim process that is running as root has to create an Exim log
3454 file, it does so in a subprocess that runs as exim:exim so as to get the
3455 ownership right at creation (otherwise, other Exim processes might see
3456 the file with the wrong ownership). There was no test for failure of this
3457 fork() call, which would lead to the process getting stuck as it waited
3458 for a non-existent subprocess. Forks do occasionally fail when resources
3459 run out. I reviewed all the other calls to fork(); they all seem to check
3462 PH/20 When checking for unexpected SMTP input at connect time (before writing
3463 the banner), Exim was not dealing correctly with a non-positive return
3464 from the read() function. If the client had disconnected by this time,
3465 the result was a log entry for a synchronization error with an empty
3466 string after "input=" when read() returned zero. If read() returned -1
3467 (an event I could not check), uninitialized data bytes were printed.
3468 There were reports of junk text (parts of files, etc) appearing after
3471 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3473 PH/22 Added support for macro redefinition, and (re)definition in between
3474 driver and ACL definitions.
3476 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3477 forgetting to use the resulting value; it was using the unexpanded value.
3479 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3480 hadn't been configured. The fix is from Juergen Kreileder, who
3481 understands it better than I do:
3483 "Here's what I see happening with three configured cyrus_sasl
3484 authenticators configured (plain, login, cram-md5):
3486 On startup auth_cyrus_sasl_init() gets called for each of these.
3487 This means three calls to sasl_listmech() without a specified mech_list.
3488 => SASL tests which mechs of all available mechs actually work
3489 => three warnings about OTP not working
3490 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3492 With the patch, sasl_listmech() also gets called three times. But now
3493 SASL's mech_list option is set to the server_mech specified in the the
3494 authenticator. Or in other words, the answer from sasl_listmech()
3495 gets limited to just the mech you're testing for (which is different
3497 => the return list contains just 'plain' or 'login', 'cram-md5' or
3498 nothing depending on the value of ob->server_mech.
3500 I've just tested the patch: Authentication still works fine,
3501 unavailable mechs specified in the exim configuration are still
3502 caught, and the auth.log warnings about OTP are gone."
3504 PH/25 When debugging is enabled, the contents of the command line are added
3505 to the debugging output, even when log_selector=+arguments is not
3508 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3509 answer is "GNU", and only if the return is "GNU/something" is the answer
3512 PH/27 $acl_verify_message is now set immediately after the failure of a
3513 verification in an ACL, and so is available in subsequent modifiers. In
3514 particular, the message can be preserved by coding like this:
3516 warn !verify = sender
3517 set acl_m0 = $acl_verify_message
3519 Previously, $acl_verify_message was set only while expanding "message"
3520 and "log_message" when a very denied access.
3522 PH/28 Modified OS/os.c-Linux with
3524 -#ifndef OS_LOAD_AVERAGE
3525 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3527 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3528 nomenclature these days.)
3530 PH/29 Installed patch from the Sieve maintainer that adds the options
3531 sieve_useraddress and sieve_subaddress to the redirect router.
3533 PH/30 In these circumstances:
3534 . Two addresses routed to the same list of hosts;
3535 . First host does not offer TLS;
3536 . First host accepts first address;
3537 . First host gives temporary error to second address;
3538 . Second host offers TLS and a TLS session is established;
3539 . Second host accepts second address.
3540 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3541 and peerdn, if requested) that were in fact used only for the second
3544 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3545 attention to any local part prefix or suffix that was matched by the
3546 router that accepted the address. It now behaves in the same way as it
3547 does for delivery: the affixes are removed from the local part unless
3548 rcpt_include_affixes is set on the transport.
3550 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3551 timeout during the DATA phase of an incoming message.
3553 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3554 applied a patch sanctioned by the Sieve maintainer.
3556 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3557 the uid or gid is negative. A case of a negative gid caused this to be
3558 noticed. The fix allows for either to be negative.
3560 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3561 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3564 PH/36 The change PH/12 above was broken. Fixed it.
3566 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3567 the grounds that routing the same address twice would always produce the
3568 same answer. This might have been true once, but it is certainly no
3569 longer true now. Routing a child address may depend on the previous
3570 routing that produced that child. Some complicated redirection strategies
3571 went wrong when messages had multiple recipients, and made Exim's
3572 behaviour dependent on the order in which the addresses were given.
3574 I have moved the duplicate checking until after the routing is complete.
3575 Exim scans the addresses that are assigned to local and remote
3576 transports, and removes any duplicates. This means that more work will be
3577 done, as duplicates will always all be routed, but duplicates are
3578 presumably rare, so I don't expect this is of any significance.
3580 For deliveries to pipes, files, and autoreplies, the duplicate checking
3581 still happens during the routing process, since they are not going to be
3584 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3585 It corrects a timeout issue with spamd. This is Ian's comment: "The
3586 background is that sometimes spamd either never reads data from a
3587 connection it has accepted, or it never writes response data. The exiscan
3588 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3589 blindly assumes that writes won't block so it may never time out."
3591 PH/39 Allow G after quota size as well as K and M.
3593 PH/40 The value set for $authenticated_id in an authenticator may not contain
3594 binary zeroes or newlines because the value is written to log lines and
3595 to spool files. There was no check on this. Now the value is run through
3596 the string_printing() function so that such characters are converted to
3597 printable escape sequences.
3599 PH/41 $message_linecount is a new variable that contains the total number of
3600 lines in the message. Compare $body_linecount, which is the count for the
3603 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3604 bounce and delay warning messages, except in certain special cases, which
3607 (a) An SMTP error message from a remote host;
3608 (b) A message specified in a :fail: redirection;
3609 (c) A message specified in a "fail" command in a system filter;
3610 (d) A message specified in a FAIL return from the queryprogram router;
3611 (e) A message specified by the cannot_route_message router option.
3613 In these cases only, Exim does include the error details in bounce and
3614 warning messages. There are also a few cases where bland messages such
3615 as "unrouteable address" or "local delivery error" are given.
3617 PH/43 $value is now also set for the "else" part of a ${run expansion.
3619 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3620 being worked on, but at least Exim now implements the latest version to
3623 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3624 process to complete was treated as a delivery failure, a timeout while
3625 writing the message to the pipe was logged, but erroneously treated as a
3626 successful delivery. Such timeouts include transport filter timeouts. For
3627 consistency with the overall process timeout, these timeouts are now
3628 treated as errors, giving rise to delivery failures by default. However,
3629 there is now a new Boolean option for the pipe transport called
3630 timeout_defer, which, if set TRUE, converts the failures into defers for
3631 both kinds of timeout. A transport filter timeout is now identified in
3634 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3635 systems where "make" and "gmake" are different, calling "gmake" at top
3636 level broke things. I've arranged for the value of $(MAKE) to be passed
3637 from the Makefile to this script so that it can call the same version of
3641 A note about Exim versions 4.44 and 4.50
3642 ----------------------------------------
3644 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3645 changes of various kinds. As a consequence, a big documentation update was
3646 needed. This delayed the release for rather longer than seemed good, especially
3647 in the light of a couple of (minor) security issues. Therefore, the changes
3648 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3649 release. So 4.44 and 4.50 are in effect two different branches that both start
3652 I have left the 4.50 change log unchanged; it contains all the changes since
3653 4.43. The change log for 4.44 is below; many of its items are identical to
3654 those for 4.50. This seems to be the most sensible way to preserve the
3655 historical information.
3661 1. Minor wording change to the doc/README.SIEVE file.
3663 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3664 computation of the current number of files was incorrect.
3666 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3667 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3670 4. Give more explanation in the error message when the command for a transport
3671 filter fails to execute.
3673 5. There are several places where Exim runs a non-Exim command in a
3674 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3675 was being done only for the command run by the queryprogram router. It is
3676 now done for all such subprocesses. The other cases are: ${run, transport
3677 filters, and the commands run by the lmtp and pipe transports.
3679 6. Added CONFIGURE_GROUP build-time option.
3681 7. Some older OS have a limit of 256 on the maximum number of file
3682 descriptors. Exim was using setrlimit() to set 1000 as a large value
3683 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3684 systems. I've change it so that if it can't get 1000, it tries for 256.
3686 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3687 was an oversight, and furthermore, ever since the addition of extra
3688 controls (e.g. 4.43/32), the checks on when to allow different forms of
3689 "control" were broken. There should now be diagnostics for all cases when a
3690 control that does not make sense is encountered.
3692 9. Added the /retain_sender option to "control=submission".
3694 10. $recipients is now available in the predata ACL (oversight).
3696 11. Tidy the search cache before the fork to do a delivery from a message
3697 received from the command line. Otherwise the child will trigger a lookup
3698 failure and thereby defer the delivery if it tries to use (for example) a
3699 cached ldap connection that the parent has called unbind on.
3701 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3702 of $address_data from the recipient verification was clobbered by the
3703 sender verification.
3705 13. The value of address_data from a sender verification is now available in
3706 $sender_address_data in subsequent conditions in the ACL statement.
3708 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3710 15. Added a new option "connect=<time>" to callout options, to set a different
3713 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3714 was its contents. (It was OK if the option was not defined at all.)
3716 17. A "Completed" log line is now written for messages that are removed from
3717 the spool by the -Mrm option.
3719 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3720 information about exactly what failed.
3722 19. Added -dd to debug only the daemon process.
3724 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3725 handles timeouts, both on the server side and network timeouts. Renamed the
3726 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3728 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3729 It is now set to "smtps".
3731 22. $host_address is now set to the target address during the checking of
3732 ignore_target_hosts.
3734 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3735 being passed; this would have caused $sender_host_name to have been used if
3736 matching the list had actually called for a host name (not very likely,
3737 since this list is usually IP addresses). A host name is now passed as
3740 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3741 code that specifies a non-restarting handler (typically sigaction() in
3742 modern systems) in an attempt to fix a rare and obscure crash bug.
3744 25. Narrowed the window for a race in the daemon that could cause it to ignore
3745 SIGCHLD signals. This is not a major problem, because they are used only to
3746 wake it up if nothing else does.
3748 26. A malformed maildirsize file could cause Exim to calculate negative values
3749 for the mailbox size or file count. Odd effects could occur as a result.
3750 The maildirsize information is now recalculated if the size or filecount
3753 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3754 support for a long time. Removed HAVE_SYS_VFS_H.
3756 28. Installed the latest version of exipick from John Jetmore.
3758 29. In an address list, if the pattern was not a regular expression, an empty
3759 subject address (from a bounce message) matched only if the pattern was an
3760 empty string. Non-empty patterns were not even tested. This was the wrong
3761 because it is perfectly reasonable to use an empty address as part of a
3762 database query. An empty address is now tested by patterns that are
3763 lookups. However, all the other forms of pattern expect the subject to
3764 contain a local part and a domain, and therefore, for them, an empty
3765 address still always fails if the pattern is not itself empty.
3767 30. Exim went into a mad DNS loop when attempting to do a callout where the
3768 host was specified on an smtp transport, and looking it up yielded more
3769 than one IP address.
3771 31. Re-factored the code for checking spool and log partition space into a
3772 function that finds that data and another that does the check. The former
3773 is then used to implement four new variables: $spool_space, $log_space,
3774 $spool_inodes, and $log_inodes.
3776 32. The RFC2047 encoding function was originally intended for short strings
3777 such as real names; it was not keeping to the 75-character limit for
3778 encoded words that the RFC imposes. It now respects the limit, and
3779 generates multiple encoded words if necessary. To be on the safe side, I
3780 have increased the buffer size for the ${rfc2047: expansion operator from
3783 33. It is now permitted to omit both strings after an "if" condition; if the
3784 condition is true, the result is "true". As before, when the second string
3785 is omitted, a false condition yields an empty string. This makes it less
3786 cumbersome to write custom ACL and router conditions.
3788 34. Failure to deliver a bounce message always caused it to be frozen, even if
3789 there was an errors_to setting on the router. The errors_to setting is now
3792 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3793 canonical form (fully expanded) before being placed in
3794 $sender_host_address.
3796 36. The table in the code that translates DNS record types into text (T_A to
3797 "A" for instance) was missing entries for NS and CNAME. It is just possible
3798 that this could have caused confusion if both these types were looked up
3799 for the same domain, because the text type is used as part of Exim's
3800 per-process caching. But the chance of anyone hitting this buglet seems
3803 37. The dnsdb lookup has been extended in a number of ways.
3805 (1) There is a new type, "zns", which walks up the domain tree until it
3806 finds some nameserver records. It should be used with care.
3808 (2) There is a new type, "mxh", which is like "mx" except that it returns
3809 just the host names, not the priorities.
3811 (3) It is now possible to give a list of domains (or IP addresses) to be
3812 looked up. The behaviour when one of the lookups defers can be
3813 controlled by a keyword.
3815 (4) It is now possible to specify the separator character for use when
3816 multiple records are returned.
3818 38. The dnslists ACL condition has been extended: it is now possible to supply
3819 a list of IP addresses and/or domains to be looked up in a particular DNS
3822 39. Added log_selector=+queue_time_overall.
3824 40. When running the queue in the test harness, wait just a tad after forking a
3825 delivery process, to get repeatability of debugging output.
3827 41. Include certificate and key file names in error message when GnuTLS fails
3828 to set them up, because the GnuTLS error message doesn't include the name
3829 of the failing file when there is a problem reading it.
3831 42. Allow both -bf and -bF in the same test run.
3833 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3835 44. The "Exiscan patch" is now merged into the mainline Exim source.
3837 45. Sometimes the final signoff response after QUIT could fail to get
3838 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3839 < 0 before doing a fflush(). This bug looks as though it goes back to the
3840 introduction of TLS in release 3.20, but "sometimes" must have been rare
3841 because the tests only now provoked it.
3843 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3844 (this can affect the format of dates).
3846 47. exim_tidydb, when checking for the continued existence of a message for
3847 which it has found a message-specific retry record, was not finding
3848 messages that were in split spool directories. Consequently, it was
3849 deleting retry records that should have stayed in existence.
3851 48. Steve fixed some bugs in eximstats.
3853 49. The SPA authentication driver was not abandoning authentication and moving
3854 on to the next authenticator when an expansion was forced to fail,
3855 contradicting the general specification for all authenticators. Instead it
3856 was generating a temporary error. It now behaves as specified.
3858 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3859 (the order specifies the preference for clients). The order is now AES256,
3860 AES128, 3DES, ARCFOUR128.
3862 51. Small patch to Sieve code - explicitly set From: when generating an
3865 52. Exim crashed if a remote delivery caused a very long error message to be
3866 recorded - for instance if somebody sent an entire SpamAssassin report back
3867 as a large number of 550 error lines. This bug was coincidentally fixed by
3868 increasing the size of one of Exim's internal buffers (big_buffer) that
3869 happened as part of the Exiscan merge. However, to be on the safe side, I
3870 have made the code more robust (and fixed the comments that describe what
3873 53. Now that there can be additional text after "Completed" in log lines (if
3874 the queue_time_overall log selector is set), a one-byte patch to exigrep
3875 was needed to allow it to recognize "Completed" as not the last thing in
3878 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3879 patch that reportedly fixes this has been added. I am not expert enough to
3880 create a test for it. This is what the patch creator wrote:
3882 "I found a little strange behaviour of ldap code when working with
3883 Windows 2003 AD Domain, where users was placed in more than one
3884 Organization Units. When I tried to give exim partial DN, the exit code
3885 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
3886 But simultaneously result of request was absolutely normal ldap result,
3887 so I produce this patch..."
3889 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
3890 REFERENCE, so I have modified the code to exclude the patch when that macro
3893 55. Some experimental protocols are using DNS PTR records for new purposes. The
3894 keys for these records are domain names, not reversed IP addresses. The
3895 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3896 leaves it alone. Component reversal etc. now happens only for IP addresses.
3899 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
3901 57. Double the size of the debug message buffer (to 2048) so that more of very
3902 long debug lines gets shown.
3904 58. The exicyclog utility now does better if the number of log files to keep
3905 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
3907 59. Two changes related to the smtp_active_hostname option:
3909 (1) $smtp_active_hostname is now available as a variable.
3910 (2) The default for smtp_banner uses $smtp_active_hostname instead
3911 of $primary_hostname.
3913 60. The host_aton() function is supposed to be passed a string that is known
3914 to be a valid IP address. However, in the case of IPv6 addresses, it was
3915 not checking this. This is a hostage to fortune. Exim now panics and dies
3916 if the condition is not met. A case was found where this could be provoked
3917 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3918 components; fortuitously, this particular loophole had already been fixed
3919 by change 4.50/55 above.
3921 If there are any other similar loopholes, the new check in host_aton()
3922 itself should stop them being exploited. The report I received stated that
3923 data on the command line could provoke the exploit when Exim was running as
3924 exim, but did not say which command line option was involved. All I could
3925 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3926 running as the user.
3929 61. There was a buffer overflow vulnerability in the SPA authentication code
3930 (which came originally from the Samba project). I have added a test to the
3931 spa_base64_to_bits() function which I hope fixes it.
3934 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
3935 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
3937 63. The daemon start-up calls getloadavg() while still root for those OS that
3938 need the first call to be done as root, but it missed one case: when
3939 deliver_queue_load_max is set with deliver_drop_privilege. This is
3940 necessary for the benefit of the queue runner, because there is no re-exec
3941 when deliver_drop_privilege is set.
3943 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
3944 This has been fixed.
3946 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3947 was in use, was not putting the data itself into the right store pool;
3948 consequently, it could be overwritten for a subsequent message in the same
3949 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3952 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
3954 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
3955 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
3956 the function were treating the return as a boolean value, which happened to
3957 work because 0=false and not-0=true, but is not correct code.
3959 68. The host_aton() function was not handling scoped IPv6 addresses (those
3960 with, for example, "%eth0" on the end) correctly.
3962 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
3963 negated items (that is, ~something) in unsigned ints. Some compilers
3964 apparently mutter when there is no cast.
3966 70. If an address verification called from an ACL failed, and did not produce a
3967 user-specific message (i.e. there was only a "system" message), nothing was
3968 put in $acl_verify_message. In this situation, it now puts the system
3971 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3972 session; change 4.31/43 added the unwanted input to the log line - except
3973 that it did not do this in the start of session case. It now does.
3975 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3976 This is wrong and can cause the other end to generate a synchronization
3977 error if it is another Exim or anything else that does the synchronization
3978 check. A QUIT command is no longer sent after a timeout.
3980 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3981 during host lookups.
3983 74. The defer_ok option of callout verification was not working if it was used
3984 when verifying addresses in header lines, that is, for this case:
3986 verify = header_sender/callout=defer_ok
3988 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3989 those file descriptors could be used for SMTP connections. If anything
3990 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3991 could be sent to the SMTP client, causing chaos. The daemon now opens
3992 stdin, stdout, and stderr to /dev/null when it puts itself into the
3995 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3996 log by default. The user can override this with suitable Perl magic.
3998 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3999 the log message when discard triggers, was not working for the DATA ACL or
4000 for the non-SMTP ACL.
4002 78. Error message wording change in sieve.c.
4004 79. If smtp_accept_max_per_host was set, the number of connections could be
4005 restricted to fewer than expected, because the daemon was trying to set up
4006 a new connection before checking whether the processes handling previous
4007 connections had finished. The check for completed processes is now done
4008 earlier. On busy systems, this bug wouldn't be noticed because something
4009 else would have woken the daemon, and it would have reaped the completed
4012 80. If a message was submitted locally by a user whose login name contained one
4013 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4014 It caused a spool format error. I have fixed the spool reading code. A
4015 related problem was that the "from" clause in the Received: line became
4016 illegal because of the space(s). It is now covered by ${quote_local_part.
4018 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4021 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4023 83. Patch from Sieve maintainer to fix unterminated string problem in
4024 "vacation" handling.
4026 84. Some minor changes to the Linux configuration files to help with other
4027 OS variants using glibc.
4029 85. One more patch for Sieve to update vacation handling to latest spec.
4032 ----------------------------------------------------
4033 See the note above about the 4.44 and 4.50 releases.
4034 ----------------------------------------------------
4040 1. Change 4.43/35 introduced a bug that caused file counts to be
4041 incorrectly computed when quota_filecount was set in an appendfile
4044 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4045 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4048 3. Give more explanation in the error message when the command for a transport
4049 filter fails to execute.
4051 4. There are several places where Exim runs a non-Exim command in a
4052 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4053 was being done only for the command run by the queryprogram router. It is
4054 now done for all such subprocesses. The other cases are: ${run, transport
4055 filters, and the commands run by the lmtp and pipe transports.
4057 5. Some older OS have a limit of 256 on the maximum number of file
4058 descriptors. Exim was using setrlimit() to set 1000 as a large value
4059 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4060 systems. I've change it so that if it can't get 1000, it tries for 256.
4062 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4063 was an oversight, and furthermore, ever since the addition of extra
4064 controls (e.g. 4.43/32), the checks on when to allow different forms of
4065 "control" were broken. There should now be diagnostics for all cases when a
4066 control that does not make sense is encountered.
4068 7. $recipients is now available in the predata ACL (oversight).
4070 8. Tidy the search cache before the fork to do a delivery from a message
4071 received from the command line. Otherwise the child will trigger a lookup
4072 failure and thereby defer the delivery if it tries to use (for example) a
4073 cached ldap connection that the parent has called unbind on.
4075 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4076 of $address_data from the recipient verification was clobbered by the
4077 sender verification.
4079 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4080 was its contents. (It was OK if the option was not defined at all.)
4082 11. A "Completed" log line is now written for messages that are removed from
4083 the spool by the -Mrm option.
4085 12. $host_address is now set to the target address during the checking of
4086 ignore_target_hosts.
4088 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4089 being passed; this would have caused $sender_host_name to have been used if
4090 matching the list had actually called for a host name (not very likely,
4091 since this list is usually IP addresses). A host name is now passed as
4094 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4095 code that specifies a non-restarting handler (typically sigaction() in
4096 modern systems) in an attempt to fix a rare and obscure crash bug.
4098 15. Narrowed the window for a race in the daemon that could cause it to ignore
4099 SIGCHLD signals. This is not a major problem, because they are used only to
4100 wake it up if nothing else does.
4102 16. A malformed maildirsize file could cause Exim to calculate negative values
4103 for the mailbox size or file count. Odd effects could occur as a result.
4104 The maildirsize information is now recalculated if the size or filecount
4107 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4108 support for a long time. Removed HAVE_SYS_VFS_H.
4110 18. Updated exipick to current release from John Jetmore.
4112 19. Allow an empty sender to be matched against a lookup in an address list.
4113 Previously the only cases considered were a regular expression, or an
4116 20. Exim went into a mad DNS lookup loop when doing a callout where the
4117 host was specified on the transport, if the DNS lookup yielded more than
4120 21. The RFC2047 encoding function was originally intended for short strings
4121 such as real names; it was not keeping to the 75-character limit for
4122 encoded words that the RFC imposes. It now respects the limit, and
4123 generates multiple encoded words if necessary. To be on the safe side, I
4124 have increased the buffer size for the ${rfc2047: expansion operator from
4127 22. Failure to deliver a bounce message always caused it to be frozen, even if
4128 there was an errors_to setting on the router. The errors_to setting is now
4131 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4132 canonical form (fully expanded) before being placed in
4133 $sender_host_address.
4135 24. Updated eximstats to version 1.33
4137 25. Include certificate and key file names in error message when GnuTLS fails
4138 to set them up, because the GnuTLS error message doesn't include the name
4139 of the failing file when there is a problem reading it.
4141 26. Expand error message when OpenSSL has problems setting up cert/key files.
4144 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4145 (this can affect the format of dates).
4147 28. exim_tidydb, when checking for the continued existence of a message for
4148 which it has found a message-specific retry record, was not finding
4149 messages that were in split spool directories. Consequently, it was
4150 deleting retry records that should have stayed in existence.
4152 29. eximstats updated to version 1.35
4153 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4154 1.35 - bugfix such that pie charts by volume are generated correctly
4156 30. The SPA authentication driver was not abandoning authentication and moving
4157 on to the next authenticator when an expansion was forced to fail,
4158 contradicting the general specification for all authenticators. Instead it
4159 was generating a temporary error. It now behaves as specified.
4161 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4162 (the order specifies the preference for clients). The order is now AES256,
4163 AES128, 3DES, ARCFOUR128.
4165 31. Small patch to Sieve code - explicitly set From: when generating an
4168 32. Exim crashed if a remote delivery caused a very long error message to be
4169 recorded - for instance if somebody sent an entire SpamAssassin report back
4170 as a large number of 550 error lines. This bug was coincidentally fixed by
4171 increasing the size of one of Exim's internal buffers (big_buffer) that
4172 happened as part of the Exiscan merge. However, to be on the safe side, I
4173 have made the code more robust (and fixed the comments that describe what
4176 33. Some experimental protocols are using DNS PTR records for new purposes. The
4177 keys for these records are domain names, not reversed IP addresses. The
4178 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4179 leaves it alone. Component reversal etc. now happens only for IP addresses.
4182 34. The host_aton() function is supposed to be passed a string that is known
4183 to be a valid IP address. However, in the case of IPv6 addresses, it was
4184 not checking this. This is a hostage to fortune. Exim now panics and dies
4185 if the condition is not met. A case was found where this could be provoked
4186 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4187 components; fortuitously, this particular loophole had already been fixed
4188 by change 4.50/55 or 4.44/33 above.
4190 If there are any other similar loopholes, the new check in host_aton()
4191 itself should stop them being exploited. The report I received stated that
4192 data on the command line could provoke the exploit when Exim was running as
4193 exim, but did not say which command line option was involved. All I could
4194 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4195 running as the user.
4198 35. There was a buffer overflow vulnerability in the SPA authentication code
4199 (which came originally from the Samba project). I have added a test to the
4200 spa_base64_to_bits() function which I hope fixes it.
4203 36. The daemon start-up calls getloadavg() while still root for those OS that
4204 need the first call to be done as root, but it missed one case: when
4205 deliver_queue_load_max is set with deliver_drop_privilege. This is
4206 necessary for the benefit of the queue runner, because there is no re-exec
4207 when deliver_drop_privilege is set.
4209 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4210 was in use, was not putting the data itself into the right store pool;
4211 consequently, it could be overwritten for a subsequent message in the same
4212 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4215 38. Sometimes the final signoff response after QUIT could fail to get
4216 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4217 < 0 before doing a fflush(). This bug looks as though it goes back to the
4218 introduction of TLS in release 3.20, but "sometimes" must have been rare
4219 because the tests only now provoked it.
4225 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4226 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4227 Now it returns an integer. A number of calls were still expecting a T/F
4228 return. Fortuitously, in all cases, the tests worked in OK situations,
4229 which is the norm. However, things would have gone wrong on any write
4230 failures on the smtp file descriptor. This function is used when sending
4231 messages over SMTP and also when doing verify callouts.
4233 2. When Exim is called to do synchronous delivery of a locally submitted
4234 message (the -odf or -odi options), it no longer closes stderr before doing
4237 3. Implemented the mua_wrapper option.
4239 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4241 5. Implemented the functions header_remove(), header_testname(),
4242 header_add_at_position(), and receive_remove_recipient(), and exported them
4245 6. If an ACL "warn" statement specified the addition of headers, Exim already
4246 inserted X-ACL-Warn: at the start if there was no header name. However, it
4247 was not making this test for the second and subsequent header lines if
4248 there were newlines in the string. This meant that an invalid header could
4249 be inserted if Exim was badly configured.
4251 7. Allow an ACL "warn" statement to add header lines at the start or after all
4252 the Received: headers, as well as at the end.
4254 8. Added the rcpt_4xx retry error code.
4256 9. Added postmaster_mailfrom=xxx to callout verification option.
4258 10. Added mailfrom=xxxx to the callout verification option, for verify=
4261 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4262 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4264 12. Inserted some casts to stop certain compilers warning when using pointer
4265 differences as field lengths or precisions in printf-type calls (mostly
4266 affecting debugging statements).
4268 13. Added optional readline() support for -be (dynamically loaded).
4270 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4271 same clock tick as a message's arrival, so that its received time was the
4272 same as the "first fail" time on the retry record, and that message
4273 remained on the queue past the ultimate address timeout, every queue runner
4274 would try a delivery (because it was past the ultimate address timeout) but
4275 after another failure, the ultimate address timeout, which should have then
4276 bounced the address, did not kick in. This was a "< instead of <=" error;
4277 in most cases the first failure would have been in the next clock tick
4278 after the received time, and all would be well.
4280 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4281 being recognized when the domain list was tested by the match_domain
4282 condition in an expansion string.
4284 16. Added the ${str2b64: operator.
4286 17. Exim was always calling setrlimit() to set a large limit for the number of
4287 processes, without checking whether the existing limit was already
4288 adequate. (It did check for the limit on file descriptors.) Furthermore,
4289 errors from getrlimit() and setrlimit() were being ignored. Now they are
4290 logged to the main and panic logs, but Exim does carry on, to try to do its
4291 job under whatever limits there are.
4293 18. Imported PCRE 5.0.
4295 19. Trivial typo in log message " temporarily refused connection" (the leading
4298 20. If the log selector return_path_on_delivery was set and an address was
4299 redirected to /dev/null, the delivery process crashed because it assumed
4300 that a return path would always be set for a "successful" delivery. In this
4301 case, the whole delivery is bypassed as an optimization, and therefore no
4304 21. Internal re-arrangement: the function for sending a challenge and reading
4305 a response while authentication was assuming a zero-terminated challenge
4306 string. It's now changed to take a pointer and a length, to allow for
4307 binary data in such strings.
4309 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4311 23. Exim was not respecting finduser_retries when seeking the login of the
4312 uid under which it was called; it was always trying 10 times. (The default
4313 setting of finduser_retries is zero.) Also, it was sleeping after the final
4314 failure, which is pointless.
4316 24. Implemented tls_on_connect_ports.
4318 25. Implemented acl_smtp_predata.
4320 26. If the domain in control=submission is set empty, Exim assumes that the
4321 authenticated id is a complete email address when it generates From: or
4322 Sender: header lines.
4324 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4325 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4326 chown and chgrp in /bin and hostname in /usr/bin.
4328 28. Exim was keeping the "process log" file open after each use, just as it
4329 does for the main log. This opens the possibility of it remaining open for
4330 long periods when the USR1 signal hits a daemon. Occasional processlog
4331 errors were reported, that could have been caused by this. Anyway, it seems
4332 much more sensible not to leave this file open at all, so that is what now
4335 29. The long-running daemon process does not normally write to the log once it
4336 has entered its main loop, and it closes the log before doing so. This is
4337 so that log files can straightforwardly be renamed and moved. However,
4338 there are a couple of unusual error situations where the daemon does write
4339 log entries, and I had neglected to close the log afterwards.
4341 30. The text of an SMTP error response that was received during a remote
4342 delivery was being truncated at 512 bytes. This is too short for some of
4343 the long messages that one sometimes sees. I've increased the limit to
4346 31. It is now possible to make retry rules that apply only when a message has a
4347 specific sender, in particular, an empty sender.
4349 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4350 it possible to be selective about when SMTP synchronization is enforced.
4352 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4354 32. Implemented hosts_connection_nolog.
4356 33. Added an ACL for QUIT.
4358 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4361 35. Added mailbox_size and mailbox_filecount to appendfile.
4363 36. Added control = no_multiline_responses to ACLs.
4365 37. There was a bug in the logic of the code that waits for the clock to tick
4366 in the case where the clock went backwards by a substantial amount such
4367 that the microsecond fraction of "now" was more than the microsecond
4368 fraction of "then" (but the whole seconds number was less).
4370 38. Added support for the libradius Radius client library this is found on
4371 FreeBSD (previously only the radiusclient library was supported).
4377 1. When certain lookups returned multiple values in the form name=value, the
4378 quoting of the values was not always being done properly. Specifically:
4379 (a) If the value started with a double quote, but contained no whitespace,
4381 (b) If the value contained whitespace other than a space character (i.e.
4382 tabs or newlines or carriage returns) it was not quoted.
4383 This fix has been applied to the mysql and pgsql lookups by writing a
4384 separate quoting function and calling it from the lookup code. The fix
4385 should probably also be applied to nisplus, ibase and oracle lookups, but
4386 since I cannot test any of those, I have not disturbed their existing code.
4388 2. A hit in the callout cache for a specific address caused a log line with no
4389 reason for rejecting RCPT. Now it says "Previous (cached) callout
4390 verification failure".
4392 3. There was an off-by-one bug in the queryprogram router. An over-long
4393 return line was truncated at 256 instead of 255 characters, thereby
4394 overflowing its buffer with the terminating zero. As well as fixing this, I
4395 have increased the buffer size to 1024 (and made a note to document this).
4397 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4398 when Exim is waiting for an SMTP response from a remote server, Exim
4399 restarts its select() call on the socket, thereby resetting its timeout.
4400 This is not a problem when such interrupts are rare. Somebody set up a cron
4401 job to run exiwhat every 2 minutes, which is less than the normal select()
4402 timeout (5 or 10 minutes). This meant that the select() timeout never
4403 kicked in because it was always reset. I have fixed this by comparing the
4404 time when an interrupt arrives with the time at the start of the first call
4405 to select(). If more time than the timeout has elapsed, the interrupt is
4406 treated as a timeout.
4408 5. Some internal re-factoring in preparation for the addition of Sieve
4409 extensions (by MH). In particular, the "personal" test is moved to a
4410 separate function, and given an option for scanning Cc: and Bcc: (which is
4411 not set for Exim filters).
4413 6. When Exim created an email address using the login of the caller as the
4414 local part (e.g. when creating a From: or Sender: header line), it was not
4415 quoting the local part when it contained special characters such as @.
4417 7. Installed new OpenBSD configuration files.
4419 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4420 try to make them clearer.
4422 9. Callout options, other than the timeout value, were being ignored when
4423 verifying sender addresses in header lines. For example, when using
4425 verify = header_sender/callout=no_cache
4427 the cache was (incorrectly) being used.
4429 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4430 only the Cygwin environment.
4432 11. When return_path_on_delivery was set as a log selector, if different remote
4433 addresses in the same message used different return paths and parallel
4434 remote delivery occurred, the wrong values would sometimes be logged.
4435 (Whenever a remote delivery process finished, the return path value from
4436 the most recently started remote delivery process was logged.)
4438 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4439 lines when AUTH and/or TLS are in use. This is the "received protocol"
4440 field. Exim used to use "asmtp" for authenticated SMTP, without any
4441 indication (in the protocol name) for TLS use. Now it follows the RFC and
4442 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4443 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4444 names appear in log lines as well as in Received: header lines.
4446 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4447 extensions, and comparison tests, and to fix some bugs.
4449 14. Changes to the "personal" filter test:
4451 (1) The test was buggy in that it was just doing the equivalent of
4452 "contains" tests on header lines. For example, if a user's address was
4453 anne@some.where, the "personal" test would incorrectly be true for
4455 To: susanne@some.where
4457 This test is now done by extracting each address from the header in turn,
4458 and checking the entire address. Other tests that are part of "personal"
4459 are now done using regular expressions (for example, to check local parts
4460 of addresses in From: header lines).
4462 (2) The list of non-personal local parts in From: addresses has been
4463 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4464 taken from the Sieve specification recommendations.
4466 (3) If the message contains any header line starting with "List-" it is
4467 treated as non-personal.
4469 (4) The test for "circular" in the Subject: header line has been removed
4470 because it now seems ill-conceived.
4472 15. Minor typos in src/EDITME comments corrected.
4474 16. Installed latest exipick from John Jetmore.
4476 17. If headers_add on a router specified a text string that was too long for
4477 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4478 of string_sprintf() is now avoided.
4480 18. $message_body_size was not set (it was always zero) when running the DATA
4481 ACL and the local_scan() function.
4483 19. For the "mail" command in an Exim filter, no default was being set for
4484 the once_repeat time, causing a random time value to be used if "once" was
4485 specified. (If the value happened to be <= 0, no repeat happened.) The
4486 default is now 0s, meaning "never repeat". The "vacation" command was OK
4487 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4488 (I found it when inspecting the code).
4490 20. There is now an overall timeout for performing a callout verification. It
4491 defaults to 4 times the callout timeout, which applies to individual SMTP
4492 commands during the callout. The overall timeout applies when there is more
4493 than one host that can be tried. The timeout is checked before trying the
4494 next host. This prevents very long delays if there are a large number of
4495 hosts and all are timing out (e.g. when the network connections are timing
4496 out). The value of the overall timeout can be changed by specifying an
4497 additional sub-option for "callout", called "maxwait". For example:
4499 verify = sender/callout=5s,maxwait=20s
4501 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4502 to the end before writing, but this should make it even safer).
4504 22. Exim was forgetting that it had advertised PIPELINING for the second and
4505 subsequent messages on an SMTP connection. It was also not resetting its
4506 memory on STARTTLS and an internal HELO.
4508 23. When Exim logs an SMTP synchronization error within a session, it now
4509 records whether PIPELINING has been advertised or not.
4511 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4512 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4513 rather than long int.
4515 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4517 26. Added the never_mail option to autoreply.
4523 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4524 crash if the getsockname() call failed; this can happen if a connection is
4525 closed very soon after it is established. The problem was simply in the
4526 order in which certain operations were done, causing Exim to try to write
4527 to the SMTP stream before it had set up the file descriptor. The bug has
4528 been fixed by making things happen in the correct order.
4534 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4535 before the connection was closed, thus losing the rejection response.
4537 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4538 some early Solaris releases, but causes trouble in current releases where
4539 socklen_t is defined.
4541 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4544 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4547 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4548 that releasing the top of it at the end releases what was used for sub-
4549 expansions (unless the block got too big). However, discard this block if
4550 the first thing is a variable or header, so that we can use its block when
4551 it is dynamic (useful for very large $message_headers, for example).
4553 6. Lookups now cache *every* query, not just the most recent. A new, separate
4554 store pool is used for this. It can be recovered when all lookup caches are
4555 flushed. Lookups now release memory at the end of their result strings.
4556 This has involved some general refactoring of the lookup sources.
4558 7. Some code has been added to the store_xxx() functions to reduce the amount
4559 of flapping under certain conditions.
4561 8. log_incoming_interface used to affect only the <= reception log lines. Now
4562 it causes the local interface and port to be added to several more SMTP log
4563 lines, for example "SMTP connection from", and rejection lines.
4565 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4567 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4569 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4570 could be overwritten at the end of the current message (or the start of a
4571 new message if it was set in a HELO ACL). The value is now preserved for
4572 the duration of the SMTP connection.
4574 12. If a transport had a headers_rewrite setting, and a matching header line
4575 contained an unqualified address, that address was qualified, even if it
4576 did not match any rewriting rules. The underlying bug was that the values
4577 of the flags that permit the existence of unqualified sender and recipient
4578 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4579 non-local messages, and by -bnq for local messages) were not being
4580 preserved with the message after it was received.
4582 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4583 "next input=" as part of the text comprising the host identity instead of
4584 the correct text. The code was using the same buffer for two different
4585 strings. However, depending on which order the printing function evaluated
4586 its arguments, the bug did not always show up. Under Linux, for example, my
4587 test suite worked just fine.
4589 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4590 doesn't work with some older versions of Perl. It has been changed to "my",
4591 which in any case is probably the better facility to use.
4593 15. A really picky compiler found some instances of statements for creating
4594 error messages that either had too many or two few arguments for the format
4597 16. The size of the buffer for calls to the DNS resolver has been increased
4598 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4599 for addresses that have a lot of PTR records. This alleviates a problem; it
4600 does not fully solve it.
4602 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4603 buffer now truncates the list and logs the incident, which is the same
4604 action as happens when Exim is looking up a host name and its aliases.
4605 Previously in this situation something unpredictable would happen;
4606 sometimes it was "internal error: store_reset failed".
4608 18. If a server dropped the connection unexpectedly when an Exim client was
4609 using GnuTLS and trying to read a response, the client delivery process
4610 crashed while trying to generate an error log message.
4612 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4613 string, for example:
4615 warn message = H1: something\nH2: something
4617 the text was added as a single header line from Exim's point of view
4618 though it ended up OK in the delivered message. However, searching for the
4619 second and subsequent header lines using $h_h2: did not work. This has been
4620 fixed. Similarly, if a system filter added multiple headers in this way,
4621 the routers could not see them.
4623 20. Expanded the error message when iplsearch is called with an invalid key to
4624 suggest using net-iplsearch in a host list.
4626 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4627 ACLs are no longer actually imposed (and a message to that effect is
4630 22. If a "gecos" field in a passwd entry contained escaped characters, in
4631 particular, if it contained a \" sequence, Exim got it wrong when building
4632 a From: or a Sender: header from that name. A second bug also caused
4633 incorrect handling when an unquoted " was present following a character
4634 that needed quoting.
4636 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4637 was not being matched caselessly.
4639 24. Arranged for all hyphens in the exim.8 source to be escaped with
4642 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4643 properly. Recipient callout cache records were still being keyed to include
4644 the sender, even when use_sender was set false. This led to far more
4645 callouts that were necessary. The sender is no longer included in the key
4646 when use_sender is false.
4648 26. Added "control = submission" modifier to ACLs.
4650 27. Added the ${base62d: operator to decode base 62 numbers.
4652 28. dnsdb lookups can now access SRV records.
4654 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4655 the configuration file.
4657 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4658 (-v) mode. This makes the output for a verbose queue run more intelligible.
4660 31. Added a use_postmaster feature to recipient callouts.
4662 32. Added the $body_zerocount variable, containing the number of binary zero
4663 bytes in the message body.
4665 33. The time of last modification of the "new" subdirectory is now used as the
4666 "mailbox time last read" when there is a quota error for a maildir
4669 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4671 35. Added +ignore_unknown as a special item in host lists.
4673 36. Code for decoding IPv6 addresses in host lists is now included, even if
4674 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4675 address was recognized as an IP address, but was then not correctly decoded
4676 into binary, causing unexpected and incorrect effects when compared with
4683 1. Very minor rewording of debugging text in manualroute to say "list of
4684 hosts" instead of "hostlist".
4686 2. If verify=header_syntax was set, and a header line with an unqualified
4687 address (no domain) and a large number of spaces between the end of the
4688 name and the colon was received, the reception process suffered a buffer
4689 overflow, and (when I tested it) crashed. This was caused by some obsolete
4690 code that should have been removed. The fix is to remove it!
4692 3. When running in the test harness, delay a bit after writing a bounce
4693 message to get a bit more predictability in the log output.
4695 4. Added a call to search_tidyup() just before forking a reception process. In
4696 theory, someone could use a lookup in the expansion of smtp_accept_max_
4697 per_host which, without the tidyup, could leave open a database connection.
4699 5. Added the variables $recipient_data and $sender_data which get set from a
4700 lookup success in an ACL "recipients" or "senders" condition, or a router
4701 "senders" option, similar to $domain_data and $local_part_data.
4703 6. Moved the writing of debug_print from before to after the "senders" test
4706 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4707 problems for message scanning, either using a data ACL, or using
4708 local_scan() because the Received: header was not generated till after they
4709 were called (in order to set the time as the time of reception completion).
4710 I have revised the way this works. The header is now generated after the
4711 body is received, but before the ACL or local_scan() are called. After they
4712 are run, the timestamp in the header is updated.
4718 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4719 before starting a queue runner without re-exec. This happened only when
4720 deliver_drop_privilege was set or when the Exim user was set to root. The
4721 effect of the bug was that timeouts during subsequent deliveries caused
4722 crashes instead of being properly handled. The handler is now left at its
4723 default (and expected) setting.
4725 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4726 message, again when deliver_drop_privilege is set or Exim is run as root.
4727 The bug described in (1) was not present in this case, but the tidying up
4728 of the other signals was missing. I have made the two cases consistent.
4730 3. The ignore_target_hosts setting on a manualroute router was being ignored
4731 for hosts that were looked up using the /MX notation.
4733 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4736 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4737 operated on the sender address. After changing the $sender_address to <>
4738 for the sender address verify, Exim was re-instated it as the original
4739 (before rewriting) address, but remembering that it had rewritten it, so it
4740 wasn't rewriting it again. This bug also had the effect of breaking the
4741 sender address verification caching when the sender address was rewritten.
4743 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4744 This has been changed so that if the ip literal address matches
4745 ignore_target_hosts, the router declines.
4747 7. Added expansion conditions match_domain, match_address, and match_local_
4748 part (NOT match_host).
4750 8. The placeholder for the Received: header didn't have a length field set.
4752 9. Added code to Exim itself and to exim_lock to test for a specific race
4753 condition that could lead to file corruption when using MBX delivery. The
4754 issue is with the lockfile that is created in /tmp. If this file is removed
4755 after a process has opened it but before that process has acquired a lock,
4756 there is the potential for a second process to recreate the file and also
4757 acquire a lock. This could lead to two Exim processes writing to the file
4758 at the same time. The added code performs the same test as UW imapd; it
4759 checks after acquiring the lock that its file descriptor still refers to
4760 the same named file.
4762 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4763 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4766 11. Added the smtp_active_hostname option. If used, this will typically be made
4767 to depend on the incoming interface address. Because $interface_address is
4768 not set up until the daemon has forked a reception process, error responses
4769 that can happen earlier (such as "too many connections") no longer contain
4772 12. If an expansion in a condition on a "warn" statement fails because a lookup
4773 defers, the "warn" statement is abandoned, and the next ACL statement is
4774 processed. Previously this caused the whole ACL to be aborted.
4776 13. Added the iplsearch lookup type.
4778 14. Added ident_timeout as a log selector.
4780 15. Added tls_certificate_verified as a log selector.
4782 16. Added a global option tls_require_ciphers (compare the smtp transport
4783 option of the same name). This controls incoming TLS connections.
4785 17. I finally figured out how to make tls_require_ciphers do a similar thing
4786 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4787 before starting the TLS session.
4789 18. Tabs are now shown as \t in -bP output.
4791 19. If the log selector return_path_on_delivery was set, Exim crashed when
4792 bouncing a message because it had too many Received: header lines.
4794 20. If two routers both had headers_remove settings, and the first one included
4795 a superfluous trailing colon, the final name in the first list and the
4796 first name in the second list were incorrectly joined into one item (with a
4797 colon in the middle).
4803 1. Added -C and -D options to the exinext utility, mainly to make it easier
4804 to include in the automated testing, but these could be helpful when
4805 multiple configurations are in use.
4807 2. The exinext utility was not formatting the output nicely when there was
4808 an alternate port involved in the retry record key, nor when there was a
4809 message id as well (for retries that were specific to a specific message
4810 and a specific host). It was also confused by IPv6 addresses, because of
4811 the additional colons they contain. I have fixed the IPv4 problem, and
4812 patched it up to do a reasonable job for IPv6.
4814 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4815 delivery, the log line now contains "pipelined" if PIPELINING was used.
4817 4. An SMTP transport process used to panic and die if the bind() call to set
4818 an explicit outgoing interface failed. This has been changed; it is now
4819 treated in the same way as a connect() failure.
4821 5. A reference to $sender_host_name in the part of a conditional expansion
4822 that was being skipped was still causing a DNS lookup. This no longer
4825 6. The def: expansion condition was not recognizing references to header lines
4826 that used bh_ and bheader_.
4828 7. Added the _cache feature to named lists.
4830 8. The code for checking quota_filecount in the appendfile transport was
4831 allowing one more file than it should have been.
4833 9. For compatibility with Sendmail, the command line option
4841 and sets the incoming protocol and host name (for trusted callers). The
4842 host name and its colon can be omitted when only the protocol is to be set.
4843 Note the Exim already has two private options, -pd and -ps, that refer to
4844 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4845 "s", but I don't think that's a major issue.
4847 10. A number of refactoring changes to the code, none of which should affect
4850 (a) The number of logging options was getting close to filling up the
4851 32-bit word that was used as a bit map. I have split them into two classes:
4852 those that are passed in the argument to log_write(), and those that are
4853 only ever tested independently outside of that function. These are now in
4854 separate 32-bit words, so there is plenty of room for expansion again.
4855 There is no change in the user interface or the logging behaviour.
4857 (b) When building, for example, log lines, the code previously used a
4858 macro that called string_cat() twice, in order to add two strings. This is
4859 not really sufficiently general. Furthermore, there was one instance where
4860 it was actually wrong because one of the argument was used twice, and in
4861 one call a function was used. (As it happened, calling the function twice
4862 did not affect the overall behaviour.) The macro has been replaced by a
4863 function that can join an arbitrary number of extra strings onto a growing
4866 (c) The code for expansion conditions now uses a table and a binary chop
4867 instead of a serial search (which was left over from when there were very
4868 few conditions). Also, it now recognizes conditions like "pam" even when
4869 the relevant support is not compiled in: a suitably worded error message is
4870 given if an attempt is made to use such a condition.
4872 11. Added ${time_interval:xxxxx}.
4874 12. A bug was causing one of the ddress fields not to be passed back correctly
4875 from remote delivery subprocesses. The field in question was not being
4876 subsequently used, so this caused to problems in practice.
4878 13. Added new log selectors queue_time and deliver_time.
4880 14. Might have fixed a bug in maildirsizefile handling that threw up
4881 "unexpected character" debug warnings, and recalculated the data
4882 unnecessarily. In any case, I expanded the warning message to give more
4885 15. Added the message "Restricted characters in address" to the statements in
4886 the default ACL that block characters like @ and % in local parts.
4888 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
4889 Three changes have been made:
4891 (a) There was a serious bug; a negative response to MAIL caused the whole
4892 recipient domain to be cached as invalid, thereby blocking all messages
4893 to all local parts at the same domain, from all senders. This bug has
4894 been fixed. The domain is no longer cached after a negative response to
4895 MAIL if the sender used is not empty.
4897 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
4900 (c) A new callout option, "use_sender" has been added for people who want
4901 the modified behaviour.
4907 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
4910 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
4911 indeed breaks things for older releases.
4913 3. Added additional logging to the case where there is a problem reading data
4914 from a filter that is running in a subprocess using a pipe, in order to
4915 try to track down a specific problem.
4917 4. Testing facility fudge: when running in the test harness and attempting
4918 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
4919 getting "No route to host". Convert this to a timeout.
4921 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
4924 6. Some OS don't have socklen_t but use size_t instead. This affects the
4925 fifth argument of getsockopt() amongst other things. This is now
4926 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
4927 can be set for individual OS. I have set it for SunOS5, OSF1, and
4928 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
4929 some earlier ones do not.
4931 7. Change 4.30/15 was not doing the test caselessly.
4933 8. The standard form for an IPv6 address literal was being rejected by address
4934 parsing in, for example, MAIL and RCPT commands. An example of this kind of
4935 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4936 this, as well as the form without the "IPv6" on the front (but only when
4937 address literals are enabled, of course).
4939 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
4941 10. Exim crashed if a message with an empty sender address specified by -f
4942 encountered a router with an errors_to setting. This could be provoked only
4943 by a command such as
4947 where an empty string was supplied; "<>" did not hit this bug.
4949 11. Installed PCRE release 4.5.
4951 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
4952 remained set. It is now erased.
4954 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
4955 times from message ids (which are base 36 rather than the normal 62).
4957 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
4958 were being counted as actual protocol errors, and logged if the log
4959 selector +smtp_protocol_error was set. One cannot be perfect in this test,
4960 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
4961 and DATA following a set of rejected RCPTs do not count as protocol errors.
4962 In other words, Exim assumes they were pipelined, though this may not
4963 actually be the case. Of course, in all cases the client gets an
4964 appropriate error code.
4966 15. If a lookup fails in an ACL condition, a message about the failure may
4967 be available; it is used if testing the ACL cannot continue, because most
4968 such messages specify what the cause of the deferral is. However, some
4969 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
4970 that caused an old message to be retained and used if a later statement
4971 caused a defer, replacing the real cause of the deferral.
4973 16. If an IP address had so many PTR records that the DNS lookup buffer
4974 was not large enough to hold them, Exim could crash while trying to process
4975 the truncated data. It now detects and logs this case.
4977 17. Further to 4.21/58, another change has been made: if (and only if) the
4978 first line of a message (the first header line) ends with CRLF, a bare LF
4979 in a subsequent header line has a space inserted after it, so as not to
4980 terminate the header.
4982 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
4983 unnecessarily, used atoi() instead of strtol(), and didn't check the
4984 termination when getting file sizes from file names by regex.
4986 19. Completely re-implemented the support for maildirsize files, in the light
4987 of a number of problems with the previous contributed implementation
4988 (4.30/29). In particular:
4990 . If the quota is zero, the maildirsize file is maintained, but no quota is
4993 . If the maildir directory does not exist, it is created before any attempt
4994 to write a maildirsize file.
4996 . The quota value in the file is just a cache; if the quota is changed in
4997 the transport, the new value overrides.
4999 . A regular expression is available for excluding directories from the
5002 20. The autoreply transport checks the characters in options that define the
5003 message's headers; it allows continued headers, but it was checking with
5004 isspace() after an embedded newline instead of explicitly looking for a
5007 21. If all the "regular" hosts to which an address was routed had passed their
5008 expiry times, and had not reached their retry times, the address was
5009 bounced, even if fallback hosts were defined. Now Exim should go on to try
5012 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5013 equivalent code in the SMTP transport. Some hosts send humungous responses
5014 to HELO/EHLO, more than 1024 it seems.
5016 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5017 gives compiler warnings in some environments. I've now done it "properly",
5020 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5021 (because of problems with the built-in one) was declared to return uschar *
5022 instead of char *, causing compiler failure.
5024 25. Fixed a file descriptor leak when processing alias/forward files.
5026 26. Fixed a minor format string issue in dbfn.c.
5028 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5030 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5031 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5034 29. When a sender address is verified, it is cached, to save repeating the test
5035 when there is more than one recipient in a message. However, when the
5036 verification involves a callout, it is possible for different callout
5037 options to be set for different recipients. It is too complicated to keep
5038 track of this in the cache, so now Exim always runs a verification when a
5039 callout is required, relying on the callout cache for the optimization.
5040 The overhead is duplication of the address routing, but this should not be
5043 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5044 to be verified with callout=postmaster, and the main callout worked but the
5045 postmaster check failed, the verification correctly failed. However, if a
5046 subsequent RCPT command asked for sender verification *without* the
5047 postmaster check, incorrect caching caused this verification also to fail,
5050 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5051 it was not caching the DNS options (qualify_single, search_parents) that
5052 were used when the lookup failed. A subsequent lookup with different
5053 options therefore always gave the same answer, though there were cases
5054 where it should not have. (Example: a "domains = !$mx_any" option on a
5055 dnslookup router: the "domains" option is always processed without any
5056 widening, but the router might have qualify_single set.) Now Exim uses the
5057 cached value only when the same options are set.
5059 32. Added John Jetmore's "exipick" utility to the distribution.
5061 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5062 than a timeout (e.g. a certificate is required, and is not provided), an
5063 Exim server now closes the connection immediately. Previously it waited for
5064 the client to close - but if the client is SSL, it seems that they each
5065 wait for each other, leading to a delay before one of them times out.
5067 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5068 maintained 0.8.x compatibility because I don't think many are using it, and
5069 it is clearly obsolete.
5071 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5074 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5075 client certificate was expired. A simple patch fixes this, though I don't
5076 understand the full logic of why the verify callback is called multiple
5079 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5080 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5081 which causes problems with some clients (such as the Certicom SSL Plus
5082 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5083 disables the coutermeasure allowing Eudora to connect."
5085 38. Exim was not checking that a write() to a log file succeeded. This could
5086 lead to Bad Things if a log got too big, in particular if it hit a file
5087 size limit. Exim now panics and dies if it cannot write to a log file, just
5088 as it does if it cannot open a log file.
5090 39. Modified OS/Makefile-Linux so that it now contains
5092 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5094 The two -D definitions ensure that Exim is compiled with large file
5095 support, which makes it possible to handle log files that are bigger than
5098 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5099 instance) a domain was checked against a named list that involved a lookup,
5100 causing $domain_data to be set, then another domain was checked against the
5101 same list, then the first domain was re-checked, the value of $domain_data
5102 after the final check could be wrong. In particular, if the second check
5103 failed, it could be set empty. This bug probably also applied to
5106 41. The strip_trailing_dot option was not being applied to the address given
5107 with the -f command-line option.
5109 42. The code for reading a message's header from the spool was incrementing
5110 $received_count, but never initializing it. This meant that the value was
5111 incorrect (doubled) while delivering a message in the same process in which
5112 it was received. In the most common configuration of Exim, this never
5113 happens - a fresh exec is done - but it can happen when
5114 deliver_drop_privilege is set.
5116 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5117 it now includes up to 150 characters of the unexpected data in the log
5120 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5121 and building data strings. The size of both of these buffers was 10 000
5122 bytes - far larger than anybody would *ever* want, thought I. Needless to
5123 say, somebody hit the limit. I have increased the maximum line length to
5124 20 000 and the maximum data length of concatenated lines to 100 000. I have
5125 also fixed two bugs, because there was no checking on these buffers. Tsk,
5126 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5127 buffer is too small.
5129 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5130 lsearch lookups. Now it does.
5132 46. When parsing a route_list item in a manualroute router, a fixed-length
5133 buffer was used for the list of hosts. I made this 1024 bytes long,
5134 thinking that nobody would ever have a list of hosts that long. Wrong.
5135 Somebody had a whole pile of complicated expansion conditions, and the
5136 string was silently truncated, leading to an expansion error. It turns out
5137 that it is easier to change to an unlimited length (owing to other changes
5138 that have happened since this code was originally written) than to build
5139 structure for giving a limitation error. The length of the item that
5140 expands into the list of hosts is now unlimited.
5142 47. The lsearch lookup could not handle data where the length of text line was
5143 more than 4095 characters. Such lines were truncated, leading to shortened
5144 data being returned. It should now handle lines of any length.
5146 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5147 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5150 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5151 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5152 no longer happens in comment lines. A list of replacements is now placed
5153 at the head of all of the source files, except those whose only change is
5154 to replace PERL_COMMAND in the very first #! line.
5156 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5157 messages on the queue, with a bottom-up merge sort, using code contributed
5158 by Michael Haardt. This should make operations like -bp somewhat faster on
5159 large queues. It won't affect queue runners, except when queue_run_in_order
5162 51. Installed eximstats 1.31 in the distribution.
5164 52. Added support for SRV lookups to the dnslookup router.
5166 53. If an ACL referred to $message_body or $message_body_end, the value was not
5167 reset for any messages that followed in the same SMTP session.
5169 54. The store-handling optimization for building very long strings was not
5170 differentiating between the different store pools. I don't think this
5171 actually made any difference in practice, but I've tidied it.
5173 55. While running the routers to verify a sender address, $sender_address
5174 was still set to the sender address. This is wrong, because when routing to
5175 send a bounce to the sender, it would be empty. Therefore, I have changed
5176 it so that, while verifying a sender address, $sender_address is set to <>.
5177 (There is no change to what happens when verifying a recipient address.)
5179 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5180 target A or AAAA records (if not already returned) without resetting the
5181 qualify_single or search_parents options of the DNS resolver. These are
5182 inappropriate in this case because the targets of MX and SRV records must
5183 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5184 target that, when qualified, matched something in the local domain. These
5185 two options are now turned off when doing these lookups.
5187 57. It seems that at least some releases of Reiserfs (which does not have the
5188 concept of a fixed number of inodes) returns zero and not -1 for the
5189 number of available inodes. This interacted badly with check_spool_inodes,
5190 which assumed that -1 was the "no such thing" setting. What I have done is
5191 to check that the total number of inodes is greater than zero before doing
5192 the test of how many are available.
5194 58. When a "warn" ACL statement has a log_message modifier, the message is
5195 remembered, and not repeated. This is to avoid a lot of repetition when a
5196 message has many recipients that cause the same warning to be written.
5197 Howewer, Exim was preserving the list of already written lines for an
5198 entire SMTP session, which doesn't seem right. The memory is now reset if a
5199 new message is started.
5201 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5202 debugging output unless log_rewrite was also set.
5204 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5205 of (int)(handle) when we know that handle contains (void *)(-1).
5207 61. The Exim daemon panic-logs an error return when it closes the incoming
5208 connection. However "connection reset by peer" seems to be common, and
5209 isn't really an error worthy of noting specially, so that particular error
5212 62. When Exim is trying to find all the local interfaces, it used to panic and
5213 die if the ioctl to get the interface flags failed. However, it seems that
5214 on at least one OS (Solaris 9) it is possible to have an interface that is
5215 included in the list of interfaces, but for which you get a failure error
5216 for this call. This happens when the interface is not "plumbed" into a
5217 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5218 failure of the "get flags" call assumes that the interface is down.
5220 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5221 makes life easier for people who are doing arithmetic on fields extracted
5222 from dates, where you often get leading zeros that should not be
5223 interpreted as octal.
5225 64. Added qualify_domain to the redirect router, to override the global
5228 65. If a pathologically long header line contained very many addresses (the
5229 report of this problem mentioned 10 000) and each of them was rewritten,
5230 Exim could use up a very large amount of memory. (It kept on making new
5231 copies of the header line as it rewrote, and never released the old ones.)
5232 At the expense of a bit more processing, the header rewriting function has
5233 been changed so that it no longer eats memory in this way.
5235 66. The generation of the Received: header has been moved from the time that a
5236 message starts to be received, to the time that it finishes. The timestamp
5237 in the Received: header should now be very close to that of the <= log
5238 line. There are two side-effects of this change:
5240 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5241 logged header lines no longer include the local Received: line, because
5242 it has not yet been created. The same applies to a copy of the message
5243 that is returned to a non-SMTP sender when a message is rejected.
5245 (b) When a filter file is tested using -bf, no additional Received: header
5246 is added to the test message. After some thought, I decided that this
5249 This change does not affect the value of $received_for. It is still set
5250 after address rewriting, but before local_scan() is called.
5252 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5254 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5255 gave an unhelpful panic error message, and a defer error. I have managed to
5256 change this behaviour so that it now rejects any supplied certificate,
5257 which seems right, as the list of acceptable certificates is empty.
5259 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5260 gave an unhelpful defer error. I have not managed to make this reject any
5261 supplied certificates, but the error message it gives is "no certificate
5262 supplied", which is not helpful.
5264 70. exigrep's output now also includes lines that are not associated with any
5265 message, but which match the given pattern. Implemented by a patch from
5266 Martin Sluka, which also tidied up the Perl a bit.
5268 71. Recipient callout verification, like sender verification, was using <> in
5269 the MAIL FROM command. This isn't really the right thing, since the actual
5270 sender may affect whether the remote host accepts the recipient or not. I
5271 have changed it to use the actual sender in the callout; this means that
5272 the cache record is now keyed on a recipient/sender pair, not just the
5273 recipient address. There doesn't seem to be a real danger of callout loops,
5274 since a callout by the remote host to check the sender would use <>.
5275 [SEE ABOVE: changed after hitting problems.]
5277 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5278 temporary errors. However, in the case of such a code being given after
5279 the end of a data transmission (i.e. after ".") Exim was failing to write
5280 a retry record for the message. (Yes, there was some broken host that was
5281 actually sending 8xx at this point.)
5283 73. An unknown lookup type in a host list could cause Exim to panic-die when
5284 the list was checked. (An example that provoked this was putting <; in the
5285 middle of a list instead of at the start.) If this happened during a DATA
5286 ACL check, a -D file could be left lying around. This kind of configuration
5287 error no longer causes Exim to die; instead it causes a defer errror. The
5288 incident is still logged to the main and panic logs.
5290 74. Buglet left over from Exim 3 conversion. The message "too many messages
5291 in one connection" was written to the rejectlog but not the mainlog, except
5292 when address rewriting (yes!) was being logged.
5294 75. Added write_rejectlog option.
5296 76. When a system filter was run not as root (that is, when system_filter_user
5297 was set), the values of the $n variables were not being returned to the
5298 main process; thus, they were not subsequently available in the $sn
5301 77. Added +return_path_on_delivery log selector.
5303 78. A connection timeout was being treated differently from recipients deferred
5304 when testing hosts_max_try with a message that was older than the host's
5305 retry timeout. (The host should not be counted, thus allowing all hosts to
5306 be tried at least once before bouncing.) This may have been the cause of an
5307 occasionally reported bug whereby a message would remain on the queue
5308 longer than the retry timeout, but would be bounced if a delivery was
5309 forced. I say "may" because I never totally pinned down the problem;
5310 setting up timeout/retry tests is difficult. See also the next item.
5312 79. The ultimate address timeout was not being applied to errors that involved
5313 a combination of host plus message (for example, a timeout on a MAIL
5314 command). When an address resolved to a number of possible hosts, and they
5315 were not all tried for each delivery (e.g. because of hosts_max_try), a
5316 message could remain on the queue longer than the retry timeout.
5318 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5321 81. Fixed an obscure SMTP outgoing bug which required at least the following
5322 conditions: (a) there was another message waiting for the same server;
5323 (b) the server returned 5xx to all RCPT commands in the first message so
5324 that the message was not completed; (c) the server dropped the connection
5325 or gave a negative response to the RSET that Exim sends to abort the
5326 transaction. The observed case was a dropped connection after DATA that had
5327 been sent in pipelining mode. That is, the server had advertised PIPELINING
5328 but was not implementing it correctly. The effect of the bug was incorrect
5329 behaviour, such as trying another host, and this could lead to a crash.
5335 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5336 and daemon.c were passed as pointers to ints; they should have been
5337 pointers to socklen_t variables (which are typically unsigned ints).
5339 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5342 3. Fixed a really odd bug that affected only the testing scheme; patching a
5343 certain fixed string in the binary changed the value of another string that
5344 happened to be identical to the end of the original first string.
5346 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5347 name", it returns that address as the IP address. On some operating
5348 systems (e.g. Solaris), it also passes back the IP address string as the
5349 "host name". However, on others (e.g. Linux), it passes back an empty
5350 string. Exim wasn't checking for this, and was changing the host name to an
5351 empty string, assuming it had been canonicized.
5353 5. Although rare, it is permitted to have more than one PTR record for a given
5354 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5355 all the names associated with an address, because they do in Solaris.
5356 However, it seems that they do not in Linux for data that comes from the
5357 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5358 I found this out when I moved to a new Linux workstation and tried to run
5359 the Exim test suite.
5361 To get round this problem I have changed the code so that it now does its
5362 own call to the DNS to look up PTR records when searching for a host name.
5363 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5364 addresses that are only in /etc/hosts are still found.
5366 This behaviour is, however, controlled by an option called host_lookup_
5367 order, which defaults to "bydns:byaddr". If people want to use the other
5368 order, or indeed, just use one or the other means of lookup, they can
5369 specify it in this variable.
5371 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5372 some operating systems, this comes back from gethostbyaddr() as an empty
5373 string, and this is what Exim used to test for. However, it seems that in
5374 other systems, "." is yielded. Exim now tests for this case too.
5376 7. The values of check_spool_space and check_log_space are now held internally
5377 as a number of kilobytes instead of an absolute number of bytes. If a
5378 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5379 kilobyte. This means that much larger values can be stored.
5381 8. Exim monitor: an attempt to get the action menu when not actually pointing
5382 at a message produces an empty menu entitled "No message selected". This
5383 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5384 no entries in it ("Shell widget menu has zero width and/or height"). So I
5385 have added a single, blank menu entry in this case.
5387 9. Added ${quote_local_part.
5389 10. MIME decoding is now applied to the contents of Subject: header lines when
5392 11. Now that a reference to $sender_host_address automatically causes a reverse
5393 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5394 host lookup before query-style lookups in lists that might use this
5395 variable. This has therefore been abolished, and the "net-" prefix is no
5396 longer necessary for query-style lookups.
5398 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5399 have been a typo for LFLAGS, so it has been changed.
5401 13. The install script calls Exim with "-C /dev/null" in order to find the
5402 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5403 to be output. Howeve, since Exim outputs its version number before the
5404 error, it didn't break the script. It just looked ugly. I fixed this by
5405 always allowing "-C /dev/null" if the caller is root.
5407 14. Ignore overlarge ACL variable number when reading spool file - insurance
5408 against a later release with more variables having written the file.
5410 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5411 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5412 this, as well as the form without the "IPv6" on the front.
5414 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5415 OS/Makefile-Darwin file.
5417 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5418 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5420 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5421 variable. However, some OS use ints for inodes. I've added cast to long int
5422 to get rid of the compiler warning.
5424 19. I had forgotten to lock out "/../" in configuration file names when
5425 ALT_CONFIG_PREFIX was set.
5427 20. Routers used for verification do not need to specify transports. However,
5428 if such a router generated a host list, and callout was configured, Exim
5429 crashed, because it could not find a port number from the (non-existent)
5430 transport. It now assumes port 25 in this circumstance.
5432 21. Added the -t option to exigrep.
5434 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5435 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5436 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5437 them is set, LOOKUP_LSEARCH is forced.
5439 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5441 24. Added sender and host information to the "rejected by local_scan()" log
5442 line; previously there was no indication of these.
5444 25. Added .include_if_exists.
5446 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5447 Linux. It turns out that not all file systems support this. Apparently some
5448 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5449 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5450 not supported on the file descriptor, is now ignored when Exim is trying to
5451 sync a directory. This applies only to Linux.
5453 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5455 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5456 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5459 29. Added support for maildirsize files from supplied patch (modified a bit).
5461 30. The use of :fail: followed by an empty string could lead Exim to respond to
5462 sender verification failures with (e.g.):
5464 550 Verification failed for <xxx>
5465 550 Sender verify failed
5467 where the first response line was missing the '-' that indicates it is not
5468 the final line of the response.
5470 31. The loop for finding the name of the user that called Exim had a hardwired
5471 limit of 10; it now uses the value of finduser_retries, which is used for
5472 all other user lookups.
5474 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5477 33. Exim was neglecting to zero errno before one call of strtol() when
5478 expanding a string and expecting an integer value. On some systems this
5479 resulted in spurious "integer overflow" errors. Also, it was casting the
5480 result into an int without checking.
5482 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5483 did not work. The code looks as if it has *never* worked, though it appears
5484 to have been documented since at least releast 1.62. I have made it work.
5486 35. The "timeout_DNS" error in retry rules, also documented since at least
5487 1.62, also never worked. As it isn't clear exactly what this means, and
5488 clearly it isn't a major issue, I have abolished the feature by treating it
5489 as "timeout", and writing a warning to the main and panic logs.
5491 36. The display of retry rules for -brt wasn't always showing the error code
5494 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5495 timeout_connect_A, timeout_connect_MX.
5497 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5498 to the empty sender.
5500 39. The daemon was not analysing the content of -oX till after it had closed
5501 stderr and disconnected from the controlling terminal. This meant that any
5502 syntax errors were only noted on the panic log, and the return code from
5503 the command was 0. By re-arranging the code a little, I've made the
5504 decoding happen first, so such errors now appear on stderr, and the return
5505 code is 1. However, the actual setting up of the sockets still happens in
5506 the disconnected process, so errors there are still only recorded on the
5509 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5510 connections (as happens on some IP stacks) was logged at start up time as
5511 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5512 it from "IPv6 and IPv4", which means that two separate sockets are being
5515 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5516 says whether AF_INET or AF_INET6 was passed as an argument.
5518 42. Exiwhat output was messed up when time zones were included in log
5521 43. Exiwhat now gives more information about the daemon's listening ports,
5522 and whether -tls-on-connect was used.
5524 44. The "port" option of the smtp transport is now expanded.
5526 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5527 silently ignored. Now an error message is written to the main and panic
5530 46. There's a new ACL modifier called "logwrite" which writes to a log file
5531 as soon as it is encountered.
5533 47. Added $local_user_uid and $local_user_gid at routing time.
5535 48. Exim crashed when trying to verify a sender address that was being
5538 49. Exim was recognizing only a space character after ".include". It now also
5539 recognizes a tab character.
5541 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5542 extracting the relevant information from the specification. The man page no
5543 longer contains scrambled data for the -d option, and I've added a section
5544 at the front about calling Exim under different names.
5546 51. Added "extra_headers" argument to the "mail" command in filter files.
5548 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5551 53. Installed eximstats 1.29.
5553 54. Added transport_filter_timeout as a generic transport option.
5555 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5556 Cc: header lines. This was required by RFC 822, but it not required by RFC
5559 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5560 incoming messages that did not have them. Now it does so only if the
5561 message originates locally, that is, if there is no associated remote host
5562 address. When Resent- header lines are present, this applies to the Resent-
5563 lines rather than the non-Resent- lines.
5565 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5566 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5568 58. Messages for configuration errors now include the name of the main
5569 configuration file - useful now that there may be more than one file in a
5570 list (.included file names were always shown).
5572 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5573 for those rare installations that do not start the daemon as root or run it
5574 setuid root. I've cut out the call to initgroups() if the daemon is not
5577 60. The Exim user and group can now be bound into the binary as text strings
5578 that are looked up at the start of Exim's processing.
5580 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5582 62. Added $mailstore_basename variable.
5584 63. Installed patch to sieve.c from Michael Haardt.
5586 64. When Exim failed to open the panic log after failing to open the main log,
5587 the original message it was trying to log was written to stderr and debug
5588 output, but if they were not available (the usual case in production), it
5589 was lost. Now it is written to syslog before the two lines that record the
5590 failures to open the logs.
5592 65. Users' Exim filters run in subprocesses under the user's uid. It is
5593 possible for a "deliver" command or an alias in a "personal" command to
5594 provoke an address rewrite. If logging of address rewriting is configured,
5595 this fails because the process is not running as root or exim. There may be
5596 a better way of dealing with this, but for the moment (because 4.30 needs
5597 to be released), I have disabled address rewrite logging when running a
5598 filter in a non-root, non-exim process.
5604 1. The buildconfig auxiliary program wasn't quoting the value set for
5605 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5606 not defined. This bug was masked in 4.22 by the effect that was fixed in
5609 2. Some messages that were rejected after a message id was allocated were
5610 shown as "incomplete" by exigrep. It no longer does this for messages that
5611 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5613 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5614 have allow_domain_literals set, the ID did not get logged in the <= line.
5615 Domain literals are now always recognized in Message-ID: header lines.
5617 4. The first argument for a ${extract expansion item is the key name or field
5618 number. Leading and trailing spaces in this item were not being ignored,
5619 causing some misleading effects.
5621 5. When deliver_drop_privilege was set, single queue runner processes started
5622 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5623 same command in the process it spins off) were not dropping privilege.
5625 6. When the daemon running as "exim" started a queue runner, it always
5626 re-executed Exim in the spun-off process. This is a waste of effort when
5627 deliver_drop_privilege is set. The new process now just calls the
5628 queue-runner function directly.
5634 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5637 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5638 ignored. Though the use of -oP was forcing the writing of a pid file, it
5639 was always written to the default place.
5641 3. If the message "no IP address found for host xxxx" is generated during
5642 incoming verification, it is now followed by identification of the incoming
5643 connection (so you can more easily find what provoked it).
5645 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5647 5. Added some features to "harden" Exim a bit more against certain attacks:
5649 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5650 be put in Local/Makefile. This is like the never_users runtime option,
5651 but it cannot be overridden. The default setting is "root".
5653 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5654 prefix string with which any file named in a -C command line option
5657 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5658 is retained for -C and -D only if the caller of Exim is root. Without
5659 it, the exim user may also use -C and -D and retain privilege.
5661 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5662 command line option is disabled.
5664 6. Macro names set by the -D option must start with an upper case letter, just
5665 like macro names defined in the configuration file.
5667 7. Added "dereference=" facility to LDAP.
5669 8. Two instances of the typo "uknown" in the source files are fixed.
5671 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5672 the Configure-Makefile script screwed up while processing it.
5674 10. Incorporated PCRE 4.4.
5676 11. The SMTP synchronization check was not operating right at the start of an
5677 SMTP session. For example, it could not catch a HELO sent before the client
5678 waited for the greeting. There is now a check for outstanding input at the
5679 point when the greeting is written. Because of the duplex, asynchronous
5680 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5681 way, but not yet received, when the check is performed.
5683 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5684 on TCP/IP sockets, because this apparently causes some broken clients to
5687 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5688 unchanged) from the Cygwin maintainer.
5690 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5691 was supported instead of testing for mbx. Effectively a typo.
5693 15. The spa authenticator server code was not checking that the input it
5694 received was valid base64.
5696 16. The debug output line for the "set" modifier in ACLs was not showing the
5697 name of the variable that was being set.
5699 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5701 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5702 lookup on its own. Something else was needed to trigger the lookup. For
5703 example, a match in host_lookup or the need for a host name in a host list.
5704 Now, if $sender_host_name is referenced and the host name has not yet been
5705 looked up, a lookup is performed. If the lookup fails, the variable remains
5706 empty, and $host_lookup_failed is set to "1".
5708 19. Added "eqi" as a case-independent comparison operator.
5710 20. The saslauthd authentication condition could segfault if neither service
5711 nor realm was specified.
5713 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5714 error message that was logged was misleading, and incoming SMTP
5715 connections were dropped. The message is now more accurate, and temporary
5716 errors are given to SMTP connections.
5718 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5719 (except RSET) with a 421 error, until QUIT is received. However, it was
5720 failing to send a response to QUIT.
5722 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5723 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5724 ACL is now run after the helo_try_verify_hosts code.
5726 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5727 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5728 case-independent, so other case variants are also recognized). Apparently
5729 some systems use these upper case variants.
5731 25. If more than two messages were waiting for the same host, and a transport
5732 filter was specified for the transport, Exim sent two messages over the
5733 same TCP/IP connection, and then failed with "socket operation on non-
5734 socket" when it tried to send the third.
5736 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5738 27. The extern definition of crypt16() in expand.c was not being excluded when
5739 the OS had its own crypt16() function.
5741 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5742 as a preferred synonym for return_size_limit, both as an option and as an
5745 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5747 30. Changed the default configuration ACL to relax the local part checking rule
5748 for addresses that are not in any local domains. For these addresses,
5749 slashes and pipe symbols are allowed within local parts, but the sequence
5750 /../ is explicitly forbidden.
5752 31. SPA server authentication was not clearing the challenge buffer before
5755 32. log_message in a "warn" ACL statement was writing to the reject log as
5756 well as to the main log, which contradicts the documentation and doesn't
5757 seem right (because no rejection is happening). So I have stopped it.
5759 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5760 However, I am unable to do any testing of this.
5762 34. Fixed an infelicity in the appendfile transport. When checking directories
5763 for a mailbox, to see if any needed to be created, it was accidentally
5764 using path names with one or more superfluous leading slashes; tracing
5765 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5767 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5768 contents are added to the log line that is written for every discarded
5769 recipient. (Previously a log_message setting was ignored.)
5771 36. The ${quote: operator now quotes the string if it is empty.
5773 37. The install script runs exim in order to find its version number. If for
5774 some reason other than non-existence or emptiness, which it checks, it
5775 could not run './exim', it was installing it with an empty version number,
5776 i.e. as "exim-". This error state is now caught, and the installation is
5779 38. An argument was missing from the function that creates an error message
5780 when Exim fails to connect to the socket for saslauthd authentication.
5781 This could cause Exim to crash, or give a corrupted message.
5783 39. Added isip, isip4, and isip6 to ${if conditions.
5785 40. The ACL variables $acl_xx are now saved with the message, and can be
5786 accessed later in routers, transports, and filters.
5788 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5789 strings in the file are not string-expanded.
5791 42. If a MAIL command specified a SIZE value that was too large to fit into an
5792 int variable, the check against message_size_limit failed. Such values are
5793 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5794 day this will have to be increased, but I don't think I want to be around
5795 when emails are that large.
5802 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5803 iconv() is not standard in FreeBSD.
5805 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5806 IPv6 enabled. The observed symptom was a segmentation fault on return from
5807 the function os_common_find_running_interfaces() in src/os.c.
5809 3. In the check_special_case() function in daemon.c I had used "errno" as an
5810 argument name, which causes warnings on some systems. This was basically a
5811 typo, since it was named "eno" in the comments!
5813 4. The code that waits for the clock to tick (at a resolution of some fraction
5814 of a second) so as to ensure message-id uniqueness was always waiting for
5815 at least one whole tick, when it could have waited for less. [This is
5816 almost certainly not relevant at current processor speeds, where it is
5817 unlikely to ever wait at all. But we try to future-proof.]
5819 5. The function that sleeps for a time interval that includes fractions of a
5820 second contained a race. It did not block SIGALRM between setting the
5821 timer, and suspending (a couple of lines later). If the interval was short
5822 and the sigsuspend() was delayed until after it had expired, the suspension
5823 never ended. On busy systems this could lead to processes getting stuck for
5826 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5827 process, before it forks any delivery processes. The open lookup caching
5828 mechanism meant that the open file or database connection was passed into
5829 the delivery process. The problem was that delivery processes always tidy
5830 up cached lookup data. This could cause a problem for the next delivery
5831 process started by the queue runner, because the external queue runner
5832 process does not know about the closure. So the next delivery process
5833 still has data in the lookup cache. In the case of a file lookup, there was
5834 no problem because closing a file descriptor in a subprocess doesn't affect
5835 the parent. However, if the lookup was caching a connection to a database,
5836 the connection was closed, and the second delivery process was likely to
5837 see errors such as "PGSQL: query failed: server closed the connection
5838 unexpectedly". The problem has been fixed by closing all cached lookups
5839 in a queue runner before running a delivery process.
5841 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5842 seem to have the "const" qualifier which it has on other OS. I've
5845 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5846 *of the same type* (client or server) with the same public name that an
5847 error should be diagnosed.
5849 9. When Exim looked up a host name for an IP address, but failed to find the
5850 original IP address when looking up the host name (a safety check), it
5851 output the message "<ip address> does not match any IP for NULL", which was
5852 confusing, to say the least. The bug was that the host name should have
5853 appeared instead of "NULL".
5855 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5856 user that is built into the binary, and the -C or -D options is used, root
5857 privilege is dropped before the configuration file is read. In addition,
5858 logging is switched to stderr instead of the normal log files. If the
5859 configuration then re-defines the Exim user, the unprivileged environment
5860 is probably not what is expected, so Exim logs a panic warning message (but
5863 However, if deliver_drop_privilege is set, the unprivileged state may well
5864 be exactly what is intended, so the warning has been cut out in that case,
5865 and Exim is allowed to try to write to its normal log files.
5871 1. smtp_return_error_details was not giving details for temporary sender
5872 or receiver verification errors.
5874 2. Diagnose a configuration error if two authenticators have the same public
5877 3. Exim used not to create the message log file for a message until the first
5878 delivery attempt. This could be confusing when incoming messages were held
5879 for policy or load reasons. The message log file is now created at the time
5880 the message is received, and an initial "Received" line is written to it.
5882 4. The automatically generated man page for command line options had a minor
5883 bug that caused no ill effects; however, a more serious problem was that
5884 the procedure for building the man page automatically didn't always
5885 operate. Consequently, release 4.20 contains an out-of-date version. This
5886 shouldn't happen again.
5888 5. When building Exim with embedded Perl support, the script that builds the
5889 Makefile was calling 'perl' to find its compile-time parameters, ignoring
5890 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
5892 6. The freeze_tell option was not being used for messages that were frozen on
5893 arrival, either by an ACL or by local_scan().
5895 7. Added the smtp_incomplete_transaction log selector.
5897 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
5898 was accepting AUTH without a new EHLO.
5900 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
5901 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
5904 10. Logging of TCP/IP connections (when configured) now happens in the main
5905 daemon process instead of the child process, so that the TCP/IP connection
5906 count is more accurate (but it can never be perfect).
5908 11. The use of "drop" in a nested ACL was not being handled correctly in the
5909 outer ACL. Now, if condition failure induced by the nested "drop" causes
5910 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
5911 or "require"), the connection is dropped.
5913 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
5914 that yield "discard" can now be used with an "accept" or a "discard" verb,
5915 but an error is generated for any others (because I can't see a useful way
5916 to define what should happen).
5918 13. When an ACL is read dynamically from a file (or anywhere else), the lines
5919 are now processed in the same way as lines in the Exim configuration file.
5920 In particular, continuation lines are supported.
5922 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
5924 15. Added -ti meaning -t -i.
5926 16. Check for letters, digits, hyphens, and dots in the names of dnslist
5927 domains, and warn by logging if others are found.
5929 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
5930 returned from GIFCONF when Exim is trying to find the list of interfaces on
5931 a host. The code in os.c has been modified to copy each ifreq to an aligned
5932 structure in all cases.
5934 Also, in some cases, the returned ifreq's were being copied to a 'struct
5935 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
5936 means the last couple of bytes of an IPv6 address could be chopped if the
5937 ifreq contained only a normal sockaddr (14 bytes storage).
5939 18. Named domain lists were not supported in the hosts_treat_as_local option.
5940 An entry such as +xxxx was not recognized, and was treated as a literal
5943 19. Ensure that header lines added by a DATA ACL are included in the reject log
5944 if the ACL subsequently rejects the message.
5946 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
5947 MD5 (which is deprecated).
5949 21. When testing a filter file using -bf, Exim was writing a message when it
5950 took the sender from a "From " line in the message, but it was not doing so
5951 when it took $return_path from a Return-Path: header line. It now does.
5953 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
5954 with a valid header line field name (a series of printing characters
5955 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
5957 23. Changed "disc" in the source to "disk" to conform to the documentation and
5958 the book and for uniformity.
5960 24. Ignore Sendmail's -Ooption=value command line item.
5962 25. When execve() failed while trying to run a command in a pipe transport,
5963 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
5964 could be confused with a return value of 69 from the command itself. This
5965 has been changed to 127, the value the shell returns if it is asked to run
5966 a non-existent command. The wording for the related log line suggests a
5967 non-existent command as the problem.
5969 26. If received_header_text expands to an empty string, do not add a Received:
5970 header line to the message. (Well, it adds a token one on the spool, but
5971 marks it "old" so that it doesn't get used or transmitted.)
5973 27. Installed eximstats 1.28 (addition of -nt option).
5975 28. There was no check for failure on the call to getsockname() in the daemon
5976 code. This can fail if there is a shortage of resources on the system, with
5977 ENOMEM, for example. A temporary error is now given on failure.
5979 29. Contrary to the C standard, it seems that in some environments, the
5980 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
5981 program. Exim now does this explicitly; it affects the formatting of
5982 timestamps using strftime().
5984 30. If exiqsumm was given junk data, it threw up some uninitialized variable
5985 complaints. I've now initialized all the variables, to avoid this.
5987 32. Header lines added by a system filter were not being "seen" during
5988 transport-time rewrites.
5990 33. The info_callback() function passed to OpenSSL is set up with type void
5991 (*)(SSL *, int, int), as described somewhere. However, when calling the
5992 function (actually a macro) that sets it up, the type void(*)() is
5993 expected. I've put in a cast to prevent warnings from picky compilers.
5995 34. If a DNS black list lookup found a CNAME record, but there were no A
5996 records associated with the domain it pointed at, Exim crashed.
5998 35. If a DNS black list lookup returned more than one A record, Exim ignored
5999 all but the first. It now scans all returned addresses if a particular IP
6000 value is being sought. In this situation, the contents of the
6001 $dnslist_value variable are a list of all the addresses, separated by a
6004 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6005 to accept a host name and all its aliases if the forward lookup for any of
6006 them yielded the IP address of the incoming connection. Now it accepts only
6007 those names whose forward lookup yields the correct IP address. Any other
6008 names are discarded. This closes a loophole whereby a rogue DNS
6009 administrator could create reverse DNS records to break through a
6010 wildcarded host restriction in an ACL.
6012 37. If a user filter or a system filter that ran in a subprocess used any of
6013 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6014 the wrong values were passed to the pipe command ($thisaddress had the
6015 value of $0, $0 had the value of $1, etc). This bug was introduced by
6016 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6018 38. Improved the line breaking for long SMTP error messages from ACLs.
6019 Previously, if there was no break point between 40 and 75 characters, Exim
6020 left the rest of the message alone. Two changes have been made: (a) I've
6021 reduced the minimum length to 35 characters; (b) if it can't find a break
6022 point between 35 and 75 characters, it looks ahead and uses the first one
6023 that it finds. This may give the occasional overlong line, but at least the
6024 remaining text gets split now.
6026 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6027 file descriptors might be low, and that setting 1000 would always raise it.
6028 It turns out that in some environments, the limit is already over 1000 and
6029 that lowering it causes trouble. So now Exim takes care not to decrease it.
6031 40. When delivering a message, the value of $return_path is set to $sender_
6032 address at the start of routing (routers may change the value). By an
6033 oversight, this default was not being set up when an address was tested by
6034 -bt or -bv, which affected the outcome if any router or filter referred to
6037 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6038 log only when "message" or "log_message" are set. However, if one of the
6039 conditions was an address verification, or a call to a nested ACL, the
6040 messages generated by the underlying test were being passed through. This
6041 no longer happens. The underlying message is available in $acl_verify_
6042 message for both "message" and "log_message" expansions, so it can be
6043 passed through if needed.
6045 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6046 new expansion $bh_ to give the encoded byte string without charset
6047 translation. Translation happens only if iconv() is available; HAVE_ICONV
6048 indicates this at build time. HEADERS_CHARSET gives the charset to
6049 translate to; headers_charset can change it in the configuration, and
6050 "headers charset" can change it in an individual filter file.
6052 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6053 that creates RFC 2047 encoded "words" labels them as that charset instead
6054 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6055 expansion operator; (ii) when Exim creates a From: line for a local
6056 message; (iii) when a header line is rewritten to include a "phrase" part.
6058 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6059 buggy, causing it to skip the first lines of messages whose message ID
6060 ended in 'D'. This would not have bitten before Exim release 4.14, because
6061 message IDs were unlikely to end in 'D' before then. The effect was to have
6062 incorrect size information for certain domains.
6064 45. #include "config.h" was missing at the start of the crypt16.c module. This
6065 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6068 46. If there was a timeout during a "random" callout check, Exim treated it as
6069 a failure of the random address, and carried on sending RSET and the real
6070 address. If the delay was just some slowness somewhere, the response to the
6071 original RCPT would be taken as a response to RSET and so on, causing
6072 mayhem of various kinds.
6074 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6075 when I implemented it. It didn't allow for the fact that some option values
6076 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6077 the right test for positive values.
6079 48. Domain names in DNS records are case-independent. Exim always looks them up
6080 in lower case. Some resolvers return domain names in exactly the case they
6081 appear in the zone file, that is, they may contain uppercase letters. Not
6082 all resolvers do this - some return always lower case. Exim was treating a
6083 change of case by a resolver as a change of domain, similar to a widening
6084 of a domain abbreviation. This triggered its re-routing code and so it was
6085 trying to route what was effectively the same domain again. This normally
6086 caused routing to fail (because the router wouldn't handle the domain
6087 twice). Now Exim checks for this case specially, and just changes the
6088 casing of the domain that it ultimately uses when it transmits the message
6091 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6094 50. If a filter generated a file delivery with a non-absolute name (possible if
6095 no home directory exists for the router), the forbid_file option was not
6098 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6099 the existing equality matching.
6101 52. Exim was using ints instead of ino_t variables in some places where it was
6102 dealing with inode numbers.
6104 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6105 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6106 called TMPDIR, and if it finds it is different, it changes its value.
6108 54. The smtp_printf() function is now made available to local_scan() so
6109 additional output lines can be written before returning. There is also an
6110 smtp_fflush() function to enable the detection of a dropped connection.
6111 The variables smtp_input and smtp_batched_input are exported to
6114 55. Changed the default runtime configuration: the message "Unknown user"
6115 has been removed from the ACL, and instead placed on the localuser router,
6116 using the cannot_route_message feature. This means that any verification
6117 failures that generate their own messages won't get overridden. Similarly,
6118 the "Unrouteable address" message that was in the ACL for unverifiable
6119 relay addresses has also been removed.
6121 56. Added hosts_avoid_esmtp to the smtp transport.
6123 57. The exicyclog script was not checking for the esoteric option
6124 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6125 will work only if exicyclog is run under the appropriate euid.
6127 58. Following a discussion on the list, the rules by which Exim recognises line
6128 endings on incoming messages have been changed. The -dropcr and drop_cr
6129 options are now no-ops, retained only for backwards compatibility. The
6130 following line terminators are recognized: LF CRLF CR. However, special
6131 processing applies to CR:
6133 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6134 nor a local message in the state where . is a terminator.
6136 (ii) If a bare CR is encountered in a header line, an extra space is added
6137 after the line terminator so as not to end the header. The reasoning
6138 behind this is that bare CRs in header lines are most likely either
6139 to be mistakes, or people trying to play silly games.
6141 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6142 was being incorrectly given as 18 bytes larger than it should have been.
6143 This is a VOB (very old bug).
6145 60. This may never have affected anything current, but just in case it has:
6146 When the local host is found other than at the start of a list of hosts,
6147 the local host, those with the same MX, and any that follow, are discarded.
6148 When the list in question was part of a longer list of hosts, the following
6149 hosts (not currently being processed) were also being discarded. This no
6150 longer happens. I'm not sure if this situation could ever has previously
6153 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6156 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6157 header. This is something that is recommended in a new Internet Draft, and
6158 is something that is documented as being done by Sendmail. There are two
6159 possible values. For messages generated by the autoreply transport, Exim
6162 Auto-Submitted: auto-replied
6164 whereas for all other generated messages (e.g. bounces) it adds
6166 Auto-Submitted: auto-generated
6168 63. The "personal" condition in filters now includes a test for the
6169 Auto-Submitted: header. If it contains the string "auto-" the message it
6170 not considered personal.
6172 64. Added rcpt_include_affixes as a generic transport option.
6174 65. Added queue_only_override (default true).
6176 66. Added the syslog_duplication option.
6178 67. If what should have been the first header line of a message consisted of
6179 a space followed by a colon, Exim was mis-interpreting it as a header line.
6180 It isn't of course - it is syntactically invalid and should therefore be
6181 treated as the start of the message body. The misbehaviour could have
6182 caused a number of strange effects, including loss of data in subsequent
6183 header lines, and spool format errors.
6185 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6186 client host had authenticated. This control can now be exercised by an ACL
6187 for more flexibility.
6189 69. By default, callouts do not happen when testing with -bh. There is now a
6190 variant, -bhc, which does actually run the callout code, including
6191 consulting and updating the callout cache.
6193 70. Added support for saslauthd authentication, courtesy of Alexander
6196 71. If statvfs() failed on the spool or log directories while checking their
6197 size for availability, Exim confusingly gave the error "space shortage".
6198 Furthermore, in debugging mode it crashed with a floating point exception.
6199 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6200 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6201 serious problem, Exim now writes to the main and panic logs when this
6202 happens, with details of the failure. It then refuses to accept the
6203 incoming message, giving the message "spool directory problem" or "log
6204 directory problem" with a 421 code for SMTP messages.
6206 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6207 0, 1, and 2 exist, because some OS complain for execs without them (see
6208 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6209 descriptors. However, the code omitted to check that the open succeeded,
6210 causing mysterious errors if for some reason the permissions on /dev/null
6211 got screwed. Now Exim writes a message to the main and panic logs, and
6212 bombs out if it can't open /dev/null.
6214 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6215 interact so that it is all more flexible. It is supposed to remain
6216 backwards compatible. Also added extra_local_interfaces.
6218 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6219 to bomb out with an assertion failure - to the client this appears as a
6220 connection drop. This problem occurs in the part of the code that was taken
6221 from the Samba project. Fortunately, the assertion is in a very simple
6222 function, so I have fixed this by reproducing the function inline in the
6223 one place where it is called, and arranging for authentication to fail
6224 instead of killing the process with assert().
6226 75. The SPA client code was not working when the server requested OEM rather
6227 than Unicode encoding.
6229 76. Added code to make require_files with a specific uid setting more usable in
6230 the case where statting the file as root fails - usually a non-root-mounted
6231 NFS file system. When this happens and the failure is EACCES, Exim now
6232 forks a subprocess and does the per-uid checking as the relevant uid.
6234 77. Added process_log_path.
6236 78. If log_file_path was not explicitly set, a setting of check_log_space or
6237 check_log_inodes was ignored.
6239 79. If a space check for the spool or log partitions fails, the incident is now
6240 logged. Of course, in the latter case the data may get lost...
6242 80. Added the %p formatting code to string_format() so that it can be used to
6243 print addresses in debug_print(). Adjusted all the address printing in the
6244 debugging in store.c to use %p rather than %d.
6246 81. There was a concern that a line of code in smtp_in.c could overflow a
6247 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6248 initially expressed, the concern was not well-founded, because trailing
6249 spaces are removed early. However, if the trailing spaces were followed by
6250 a NULL, they did not get removed, so the overflow was possible. Two fixes
6253 (a) I re-wrote the offending code in a cleaner fashion.
6254 (b) If an incoming SMTP command contains a NULL character, it is rejected
6257 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6258 runs initgroups(), so that if the Exim user is in any additional groups,
6259 they will be used during message reception.
6265 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob