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18 EXPERIMENTAL_CERTNAMES is withdrawn.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.
84 JH/24 Verification callouts now attempt to use TLS by default.
86 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
87 are generic router options now. The defaults didn't change.
89 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
90 Original patch from Alexander Shikoff, worked over by JH.
92 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
93 files if ZCAT_COMMAND is not executable.
95 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
97 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
99 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
100 Normally benign, it bites when the pair was led to by a CNAME;
101 modern usage is to not canoicalize the domain to a CNAME target
102 (and we were inconsistent anyway for A-only vs AAAA+A).
107 TL/01 When running the test suite, the README says that variables such as
108 no_msglog_check are global and can be placed anywhere in a specific
109 test's script, however it was observed that placement needed to be near
110 the beginning for it to behave that way. Changed the runtest perl
111 script to read through the entire script once to detect and set these
112 variables, reset to the beginning of the script, and then run through
113 the script parsing/test process like normal.
115 TL/02 The BSD's have an arc4random API. One of the functions to induce
116 adding randomness was arc4random_stir(), but it has been removed in
117 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
118 function when detected.
120 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
121 cause callback expansion.
123 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
124 syntax errors in an expansion can be treated as a string instead of
125 logging or causing an error, due to the internal use of bool_lax
126 instead of bool when processing it.
128 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
129 server certificates when making smtp deliveries.
131 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
133 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
135 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
137 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
138 Merged patch from Sebastian Wiedenroth.
140 JH/05 Fix results-pipe from transport process. Several recipients, combined
141 with certificate use, exposed issues where response data items split
142 over buffer boundaries were not parsed properly. This eventually
143 resulted in duplicates being sent. This issue only became common enough
144 to notice due to the introduction of conection certificate information,
145 the item size being so much larger. Found and fixed by Wolfgang Breyha.
147 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
148 size buffer was used, resulting in syntax errors when an expansion
151 JH/07 Add support for directories of certificates when compiled with a GnuTLS
152 version 3.3.6 or later.
154 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
155 is EXPERIMENTAL_EVENT, the main-configuration and transport options
156 both become "event_action", the variables become $event_name, $event_data
157 and $event_defer_errno. There is a new variable $verify_mode, usable in
158 routers, transports and related events. The tls:cert event is now also
159 raised for inbound connections, if the main configuration event_action
162 TL/06 In test suite, disable OCSP for old versions of openssl which contained
163 early OCSP support, but no stapling (appears to be less than 1.0.0).
165 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
166 server certificate names available under the smtp transport option
167 "tls_verify_cert_hostname" now do not permit multi-component wildcard
170 JH/10 Time-related extraction expansions from certificates now use the main
171 option "timezone" setting for output formatting, and are consistent
172 between OpenSSL and GnuTLS compilations. Bug 1541.
174 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
175 encoded parameter in the incoming message. Bug 1558.
177 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
178 include certificate info, eximon was claiming there were spoolfile
181 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
183 JH/14 Log delivery-related information more consistently, using the sequence
184 "H=<name> [<ip>]" wherever possible.
186 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
187 are problematic for Debian distribution, omit them from the release
190 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
192 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
195 JH/17 Fix a null-indirection in certextract expansions when a nondefault
196 output list separator was used.
201 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
202 checkers that were complaining about end of non-void function with no
205 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
206 This was a regression intruduced in 4.83 by another bugfix.
208 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
210 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
211 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
217 TF/01 Correctly close the server side of TLS when forking for delivery.
219 When a message was received over SMTP with TLS, Exim failed to clear up
220 the incoming connection properly after forking off the child process to
221 deliver the message. In some situations the subsequent outgoing
222 delivery connection happened to have the same fd number as the incoming
223 connection previously had. Exim would try to use TLS and fail, logging
224 a "Bad file descriptor" error.
226 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
227 utilities have not been installed.
229 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
230 temporary space as the ACL may create new global variables.
232 TL/01 LDAP support uses per connection or global context settings, depending
233 upon the detected version of the libraries at build time.
235 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
236 to extract and use the src ip:port in logging and expansions as if it
237 were a direct connection from the outside internet. PPv2 support was
238 updated based on HAProxy spec change in May 2014.
240 JH/02 Add ${listextract {number}{list}{success}{fail}}.
242 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
243 Properly escape header and check for NULL return.
245 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
248 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
250 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
251 characters in header names, implemented as a verify condition.
252 Contributed by Michael Fischer v. Mollard.
254 TL/05 Rename SPF condition results err_perm and err_temp to standardized
255 results permerror and temperror. Previous values are deprecated but
256 still accepted. In a future release, err_perm and err_temp will be
257 completely removed, which will be a backward incompatibility if the
258 ACL tests for either of these two old results. Patch contributed by
259 user bes-internal on the mailing list.
261 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
263 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
264 selectors, in both main and reject logs.
266 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
269 JH/07 Add malware type "sock" for talking to simple daemon.
271 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
273 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
274 routers/transports under cutthrough routing.
276 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
277 numbers. Touch up "bool" conditional to keep the same definition.
279 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
281 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
283 JH/12 Expand items in router/transport headers_add or headers_remove lists
284 individually rather than the list as a whole. Bug 1452.
286 Required for reasonable handling of multiple headers_ options when
287 they may be empty; requires that headers_remove items with embedded
288 colons must have them doubled (or the list-separator changed).
290 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
291 view the policy declared in the DMARC record. Currently, $dmarc_status
292 is a combined value of both the record presence and the result of the
295 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
297 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
298 dnslookup router and the smtp transport (applying to the forward
301 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
302 of ldap servers used for a specific lookup. Patch provided by Heiko
305 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
306 New variable $lookup_dnssec_authenticated for observability.
308 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
309 Patch submitted by Lars Timman.
311 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
313 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
314 Requires trusted mode and valid format message id, aborts otherwise.
315 Patch contributed by Heiko Schlichting.
317 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
318 certextract with support for various fields. Bug 1358.
320 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
321 is requested by default, modifiable by smtp transport option
324 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
325 operate on certificate variables to give certificate fingerprints
326 Also new ${sha256:cert_variable}.
328 JH/23 The PRDR feature is moved from being Experimental into the mainline.
330 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
331 Christian Aistleitner.
333 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
335 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
336 file. Patch from Wolfgang Breyha.
338 JH/25 Expand the coverage of the delivery $host and $host_address to
339 client authenticators run in verify callout. Bug 1476.
341 JH/26 Port service names are now accepted for tls_on_connect_ports, to
342 align with daemon_smtp_ports. Bug 72.
344 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
345 support and error reporting did not work properly.
347 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
348 and is readable. Patch from Andrew Colin Kissa.
350 TL/14 Enhance documentation of ${run expansion and how it parses the
351 commandline after expansion, particularly in the case when an
352 unquoted variable expansion results in an empty value.
354 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
356 PP/02 Fix internal collision of T_APL on systems which support RFC3123
357 by renaming away from it. Addresses GH issue 15, reported by
360 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
362 TL/15 SECURITY: prevent double expansion in math comparison functions
363 (can expand unsanitized data). Not remotely exploitable.
370 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
372 PP/02 Make -n do something, by making it not do something.
373 When combined with -bP, the name of an option is not output.
375 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
378 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
379 $sender_host_name and config options to manage this, and basic check
382 PP/05 DSCP support for outbound connections and control modifier for inbound.
384 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
385 (Only plugin which currently uses this is kerberos4, which nobody should
386 be using, but we should make it available and other future plugins might
387 conceivably use it, even though it would break NAT; stuff *should* be
388 using channel bindings instead).
390 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
391 name; added for Sendmail compatibility; requires admin caller.
392 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
393 just ignore it); requires trusted caller.
394 Also parse but ignore: -Ac -Am -X<logfile>
397 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
399 TL/02 Add +smtp_confirmation as a default logging option.
401 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
402 Patch by Magnus Holmgren from 2007-02-20.
404 TL/04 Bugzilla 1281 - Spec typo.
405 Bugzilla 1283 - Spec typo.
406 Bugzilla 1290 - Spec grammar fixes.
408 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
410 TL/06 Add Experimental DMARC support using libopendmarc libraries.
412 TL/07 Fix an out of order global option causing a segfault. Reported to dev
413 mailing list by by Dmitry Isaikin.
415 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
417 JH/02 Support "G" suffix to numbers in ${if comparisons.
419 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
421 NM/01 Bugzilla 1197 - Spec typo
422 Bugzilla 1196 - Spec examples corrections
424 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
426 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
427 gnutls_enable_pkcs11, but renamed to more accurately indicate its
430 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
431 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
433 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
434 "acl {{name}{arg}...}", and optional args on acl condition
437 JH/05 Permit multiple router/transport headers_add/remove lines.
439 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
441 JH/07 Avoid using a waiting database for a single-message-only transport.
442 Performance patch from Paul Fisher. Bugzilla 1262.
444 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
447 JH/09 Add $headers_added variable, with content from use of ACL modifier
448 add_header (but not yet added to the message). Bugzilla 199.
450 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
451 Pulled from Bugzilla 817 by Wolfgang Breyha.
453 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
455 (nb: this is the same fix as in Exim 4.80.1)
457 JH/11 Add A= logging on delivery lines, and a client_set_id option on
460 JH/12 Add optional authenticated_sender logging to A= and a log_selector
463 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
465 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
466 advertise SMTP AUTH mechanism to us, instead of a generic
467 protocol violation error. Also, make Exim more robust to bad
468 data from the Dovecot auth socket.
470 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
472 When a queue runner is handling a message, Exim first routes the
473 recipient addresses, during which it prunes them based on the retry
474 hints database. After that it attempts to deliver the message to
475 any remaining recipients. It then updates the hints database using
478 So if a recipient address works intermittently, it can get repeatedly
479 deferred at routing time. The retry hints record remains fresh so the
480 address never reaches the final cutoff time.
482 This is a fairly common occurrence when a user is bumping up against
483 their storage quota. Exim had some logic in its local delivery code
484 to deal with this. However it did not apply to per-recipient defers
485 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
487 This change adds a proper retry rule check during routing so that the
488 final cutoff time is checked against the message's age. We only do
489 this check if there is an address retry record and there is not a
490 domain retry record; this implies that previous attempts to handle
491 the address had the retry_use_local_parts option turned on. We use
492 this as an approximation for the destination being like a local
493 delivery, as in LMTP.
495 I suspect this new check makes the old local delivery cutoff check
496 redundant, but I have not verified this so I left the code in place.
498 TF/02 Correct gecos expansion when From: is a prefix of the username.
500 Test 0254 submits a message to Exim with the header
504 When I ran the test suite under the user fanf2, Exim expanded
505 the header to contain my full name, whereas it should have added
506 a Resent-Sender: header. It erroneously treats any prefix of the
507 username as equal to the username.
509 This change corrects that bug.
511 GF/01 DCC debug and logging tidyup
512 Error conditions log to paniclog rather than rejectlog.
513 Debug lines prefixed by "DCC: " to remove any ambiguity.
515 TF/03 Avoid unnecessary rebuilds of lookup-related code.
517 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
518 Bug spotted by Jeremy Harris; was flawed since initial commit.
519 Would have resulted in OCSP responses post-SNI triggering an Exim
520 NULL dereference and crash.
522 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
524 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
525 Bug detection, analysis and fix by Samuel Thibault.
526 Bugzilla 1331, Debian bug #698092.
528 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
530 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
531 Server implementation by Todd Lyons, client by JH.
532 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
533 config variable "prdr_enable" controls whether the server
534 advertises the facility. If the client requests PRDR a new
535 acl_data_smtp_prdr ACL is called once for each recipient, after
536 the body content is received and before the acl_smtp_data ACL.
537 The client is controlled by bolth of: a hosts_try_prdr option
538 on the smtp transport, and the server advertisement.
539 Default client logging of deliveries and rejections involving
540 PRDR are flagged with the string "PRDR".
542 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
543 fclose(). Diagnosis by Todd Lyons.
545 PP/17 Update configure.default to handle IPv6 localhost better.
546 Patch by Alain Williams (plus minor tweaks).
549 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
550 This is now consistent with GnuTLS, and is now documented: the
551 previous undocumented portable approach to treating the option as
552 unset was to force an expansion failure. That still works, and
553 an empty string is now equivalent.
555 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
556 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
557 not performing validation itself.
559 PP/20 Added force_command boolean option to pipe transport.
560 Patch from Nick Koston, of cPanel Inc.
562 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
565 TF/04 Added udpsend ACL modifer and hexquote expansion operator
567 PP/21 Fix eximon continuous updating with timestamped log-files.
568 Broken in a format-string cleanup in 4.80, missed when I repaired the
569 other false fix of the same issue.
570 Report and fix from Heiko Schlichting.
573 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
574 Report from Prashanth Katuri.
576 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
577 It's SecureTransport, so affects any MacOS clients which use the
578 system-integrated TLS libraries, including email clients.
580 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
581 using a MIME ACL for non-SMTP local injection.
582 Report and assistance in diagnosis by Warren Baker.
584 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
586 JH/16 Fix comparisons for 64b. Bugzilla 1385.
588 TL/09 Add expansion variable $authenticated_fail_id to keep track of
589 last id that failed so it may be referenced in subsequent ACL's.
591 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
594 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
595 ldap library initialization, allowing self-signed CA's to be
596 used. Also properly sets require_cert option later in code by
597 using NULL (global ldap config) instead of ldap handle (per
598 session). Bug diagnosis and testing by alxgomz.
600 TL/12 Enhanced documentation in the ratelimit.pl script provided in
601 the src/util/ subdirectory.
603 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
604 renamed to Transport Post Delivery Action by Jeremy Harris, as
607 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
608 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
609 redis_servers = needs to be configured which will be used by the redis
610 lookup. Patch from Warren Baker, of The Packet Hub.
612 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
614 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
615 hostname or reverse DNS when processing a host list. Used suggestions
616 from multiple comments on this bug.
618 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
620 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
621 Missed a few lines, added it to make the runtest require no keyboard
624 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
625 contains upper case chars. Make router use caseful_local_part.
627 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
628 support when GnuTLS has been built with p11-kit.
634 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
636 This, or similar/improved, will also be change PP/11 of 4.82.
642 PP/01 Handle short writes when writing local log-files.
643 In practice, only affects FreeBSD (8 onwards).
644 Bugzilla 1053, with thanks to Dmitry Isaikin.
646 NM/01 Bugzilla 949 - Documentation tweak
648 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
651 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
653 PP/02 Implemented gsasl authenticator.
655 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
657 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
658 `pkg-config foo` for cflags/libs.
660 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
661 with rest of GSASL and with heimdal_gssapi.
663 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
664 `pkg-config foo` for cflags/libs for the TLS implementation.
666 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
667 properties get this fed in as external SSF. A number of robustness
668 and debugging improvements to the cyrus_sasl authenticator.
670 PP/08 cyrus_sasl server now expands the server_realm option.
672 PP/09 Bugzilla 1214 - Log authentication information in reject log.
673 Patch by Jeremy Harris.
675 PP/10 Added dbmjz lookup type.
677 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
679 PP/12 MAIL args handles TAB as well as SP, for better interop with
680 non-compliant senders.
681 Analysis and variant patch by Todd Lyons.
683 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
684 Bug report from Lars Müller <lars@samba.org> (via SUSE),
685 Patch from Dirk Mueller <dmueller@suse.com>
687 PP/13 tls_peerdn now print-escaped for spool files.
688 Observed some $tls_peerdn in wild which contained \n, which resulted
689 in spool file corruption.
691 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
692 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
693 or write after TLS renegotiation, which otherwise led to messages
696 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
697 as a tracking header (ie: a signed header comes before the signature).
698 Patch from Wolfgang Breyha.
700 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
701 comma-sep list; embedded commas doubled.
703 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
705 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
707 Report and patch from Dmitry Banschikov.
709 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
710 Removed SSL_clear() after SSL_new() which led to protocol negotiation
711 failures. We appear to now support TLS1.1+ with Exim.
713 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
714 lets Exim select keys and certificates based upon TLS SNI from client.
715 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
716 before an outbound SMTP session. New log_selector, +tls_sni.
718 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
719 NULL dereference. Report and patch from Alun Jones.
721 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
722 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
723 Not seeing resolver debug output on NetBSD, but suspect this is a
724 resolver implementation change.
726 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
727 Left warnings. Added "eximon gdb" invocation mode.
729 PP/21 Defaulting "accept_8bitmime" to true, not false.
731 PP/22 Added -bw for inetd wait mode support.
733 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
734 locate the relevant includes and libraries. Made this the default.
736 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
737 Bugzilla 1246, report and most of solution from Tomasz Kusy.
739 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
740 This may cause build issues on older platforms.
742 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
743 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
744 gnutls_require_mac & gnutls_require_protocols (no longer supported).
745 Added SNI support via GnuTLS too.
746 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
748 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
750 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
751 Applied second patch from Janne, implementing suggestion to default
752 multiple-strings-in-record handling to match SPF spec.
754 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
756 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
757 read-only, out of scope).
758 Patch from Wolfgang Breyha, report from Stuart Northfield.
760 PP/29 Fix three issues highlighted by clang analyser static analysis.
761 Only crash-plausible issue would require the Cambridge-specific
762 iplookup router and a misconfiguration.
763 Report from Marcin Mirosław.
765 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
767 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
768 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
769 As part of this, removing so much warning spew let me fix some minor
770 real issues in debug logging.
772 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
773 assignment on my part. Fixed.
775 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
776 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
777 Janne Snabb (who went above and beyond: thank you).
779 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
780 string otherwise requires a connection and a bunch more work and it's
781 relatively easy to get wrong. Should also expose TLS library linkage
784 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
785 64-bit ${eval} (JH/03).
787 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
788 GNU libc to support some of the 64-bit stuff, should not lead to
789 conflicts. Defined before os.h is pulled in, so if a given platform
790 needs to override this, it can.
792 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
793 protection layer was required, which is not implemented.
794 Bugzilla 1254, patch from Wolfgang Breyha.
796 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
797 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
798 tls_dhparam take prime identifiers. Also unbreak combination of
799 OpenSSL+DH_params+TLSSNI.
801 PP/39 Disable SSLv2 by default in OpenSSL support.
807 PP/01 Solaris build fix for Oracle's LDAP libraries.
808 Bugzilla 1109, patch from Stephen Usher.
810 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
812 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
815 TF/02 Fix a couple more cases where we did not log the error message
816 when unlink() failed. See also change 4.74-TF/03.
818 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
819 lock up or crash if it happened to be inside a call to libc when it
820 got a SIGUSR1 from exiwhat.
822 The SIGUSR1 handler appends the current process status to the process
823 log which is later printed by exiwhat. It used to use the general
824 purpose logging code to do this, but several functions it calls are
825 not safe for signals.
827 The new output code in the SIGUSR1 handler is specific to the process
828 log, and simple enough that it's easy to inspect for signal safety.
829 Removing some special cases also simplifies the general logging code.
830 Removing the spurious timestamps from the process log simplifies
833 TF/04 Improved ratelimit ACL condition.
835 The /noupdate option has been deprecated in favour of /readonly which
836 has clearer semantics. The /leaky, /strict, and /readonly update modes
837 are mutually exclusive. The update mode is no longer included in the
838 database key; it just determines when the database is updated. (This
839 means that when you upgrde Exim will forget old rate measurements.)
841 Exim now checks that the per_* options are used with an update mode that
842 makes sense for the current ACL. For example, when Exim is processing a
843 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
844 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
845 must specify per_mail/readonly. If you omit the update mode it defaults to
846 /leaky where that makes sense (as before) or /readonly where required.
848 The /noupdate option is now undocumented but still supported for
849 backwards compatibility. It is equivalent to /readonly except that in
850 ACLs where /readonly is required you may specify /leaky/noupdate or
851 /strict/noupdate which are treated the same as /readonly.
853 A useful new feature is the /count= option. This is a generalization
854 of the per_byte option, so that you can measure the throughput of other
855 aggregate values. For example, the per_byte option is now equivalent
856 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
858 The per_rcpt option has been generalized using the /count= mechanism
859 (though it's more complicated than the per_byte equivalence). When it is
860 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
861 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
862 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
863 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
864 Note that using per_rcpt with a non-readonly update mode in more than
865 one ACL will cause the recipients to be double-counted. (The per_mail
866 and per_byte options don't have this problem.)
868 The handling of very low rates has changed slightly. If the computed rate
869 is less than the event's count (usually one) then this event is the first
870 after a long gap. In this case the rate is set to the same as this event's
871 count, so that the first message of a spam run is counted properly.
873 The major new feature is a mechanism for counting the rate of unique
874 events. The new per_addr option counts the number of different
875 recipients that someone has sent messages to in the last time period. It
876 behaves like per_rcpt if all the recipient addresses are different, but
877 duplicate recipient addresses do not increase the measured rate. Like
878 the /count= option this is a general mechanism, so the per_addr option
879 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
880 example, measure the rate that a client uses different sender addresses
881 with the options per_mail/unique=$sender_address. There are further
882 details in the main documentation.
884 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
886 TF/06 Removed a few PCRE remnants.
888 TF/07 Automatically extract Exim's version number from tags in the git
889 repository when doing development or release builds.
891 PP/02 Raise smtp_cmd_buffer_size to 16kB.
892 Bugzilla 879. Patch from Paul Fisher.
894 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
895 Heavily based on revision 40f9a89a from Simon Arlott's tree.
898 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
900 PP/05 Variable $av_failed, true if the AV scanner deferred.
901 Bugzilla 1078. Patch from John Horne.
903 PP/06 Stop make process more reliably on build failure.
904 Bugzilla 1087. Patch from Heiko Schlittermann.
906 PP/07 Make maildir_use_size_file an _expandable_ boolean.
907 Bugzilla 1089. Patch from Heiko Schlittermann.
909 PP/08 Handle ${run} returning more data than OS pipe buffer size.
910 Bugzilla 1131. Patch from Holger Weiß.
912 PP/09 Handle IPv6 addresses with SPF.
913 Bugzilla 860. Patch from Wolfgang Breyha.
915 PP/10 GnuTLS: support TLS 1.2 & 1.1.
917 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
920 PP/11 match_* no longer expand right-hand-side by default.
921 New compile-time build option, EXPAND_LISTMATCH_RHS.
922 New expansion conditions, "inlist", "inlisti".
924 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
926 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
928 PP/14 fix log_write() format string regression from TF/03.
929 Bugzilla 1152. Patch from Dmitry Isaikin.
935 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
937 PP/02 Harmonised TLS library version reporting; only show if debugging.
938 Layout now matches that introduced for other libraries in 4.74 PP/03.
940 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
942 PP/04 New "dns_use_edns0" global option.
944 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
947 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
948 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
950 TK/01 Updated PolarSSL code to 0.14.2.
951 Bugzilla 1097. Patch from Andreas Metzler.
953 PP/07 Catch divide-by-zero in ${eval:...}.
956 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
959 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
960 format-string attack -- SECURITY: remote arbitrary code execution.
962 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
963 time unintentionally subject to list matching rules, letting the header
964 cause arbitrary Exim lookups (of items which can occur in lists, *not*
965 arbitrary string expansion). This allowed for information disclosure.
967 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
968 INT_MIN/-1 -- value coerced to INT_MAX.
974 NM/01 Workround for PCRE version dependency in version reporting
977 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
978 This fixes portability to compilers other than gcc, notably
979 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
981 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
982 makefiles for portability to HP-UX and POSIX correctness.
984 PP/01 Permit LOOKUP_foo enabling on the make command-line.
985 Also via indented variable definition in the Makefile.
986 (Debugging by Oliver Heesakkers).
988 PP/02 Restore caching of spamd results with expanded spamd_address.
989 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
991 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
992 Improves build reliability. Fix from: Frank Elsner
994 NM/02 Fix wide character breakage in the rfc2047 coding
995 Fixes bug 1064. Patch from Andrey N. Oktyabrski
997 NM/03 Allow underscore in dnslist lookups
998 Fixes bug 1026. Patch from Graeme Fowler
1000 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1001 Code patches from Adam Ciarcinski of NetBSD.
1003 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1006 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1007 is logged, to avoid truncation. Patch from John Horne.
1009 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1010 Patch from Jakob Hirsch.
1012 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1013 SQL string expansion failure details.
1014 Patch from Andrey Oktyabrski.
1016 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1017 Patch from Simon Arlott.
1019 PP/09 New lookups functionality failed to compile on old gcc which rejects
1020 extern declarations in function scope.
1021 Patch from Oliver Fleischmann
1023 PP/10 Use sig_atomic_t for flags set from signal handlers.
1024 Check getgroups() return and improve debugging.
1025 Fixed developed for diagnosis in bug 927 (which turned out to be
1028 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1029 Patch from Mark Zealey.
1031 PP/12 Bugzilla 1056: Improved spamd server selection.
1032 Patch from Mark Zealey.
1034 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1035 Based on patch from Heiko Schlittermann.
1037 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1038 Patch from Uwe Doering, sign-off by Michael Haardt.
1040 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1041 variable declaration deep within a block. Bug and patch from
1044 PP/15 lookups-Makefile IRIX compatibilty coercion.
1046 PP/16 Make DISABLE_DKIM build knob functional.
1048 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1049 Patch by Simon Arlott
1051 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1052 variable argument macros. Our copy now differs from upstream.
1058 TF/01 Failure to get a lock on a hints database can have serious
1059 consequences so log it to the panic log.
1061 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1062 controlled using the smtp_confirmation log selector.
1064 TF/03 Include the error message when we fail to unlink a spool file.
1066 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1067 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1068 for maintaining out-of-tree patches for some time.
1070 PP/01 Bugzilla 139: Documentation and portability issues.
1071 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1072 Handle per-OS dynamic-module compilation flags.
1074 PP/02 Let /dev/null have normal permissions.
1075 The 4.73 fixes were a little too stringent and complained about the
1076 permissions on /dev/null. Exempt it from some checks.
1077 Reported by Andreas M. Kirchwitz.
1079 PP/03 Report version information for many libraries, including
1080 Exim version information for dynamically loaded libraries. Created
1081 version.h, now support a version extension string for distributors
1082 who patch heavily. Dynamic module ABI change.
1084 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1085 privilege escalation vulnerability whereby the Exim run-time user
1086 can cause root to append content of the attacker's choosing to
1089 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1092 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1093 If dropping privileges for untrusted macros, we disabled normal logging
1094 on the basis that it would fail; for the Exim run-time user, this is not
1095 the case, and it resulted in successful deliveries going unlogged.
1096 Fixed. Reported by Andreas Metzler.
1102 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1103 only prepend for the Resent-* case. Fixes regression introduced in
1104 Exim 4.70 by NM/22 for Bugzilla 607.
1106 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1107 increasing numbers of administrators be bitten by this.
1109 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1111 PP/03 Bugzilla 994: added openssl_options main configuration option.
1113 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1115 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1117 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1119 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1120 without a peer certificate, leading to a segfault because of an
1121 assumption that peers always have certificates. Be a little more
1122 paranoid. Problem reported by Martin Tscholak.
1124 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1125 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1126 NB: ClamAV planning to remove STREAM in "middle of 2010".
1127 CL also introduces -bmalware, various -d+acl logging additions and
1128 more caution in buffer sizes.
1130 PP/09 Implemented reverse_ip expansion operator.
1132 PP/10 Bugzilla 937: provide a "debug" ACL control.
1134 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1136 PP/12 Bugzilla 973: Implement --version.
1138 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1140 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1142 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1144 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1145 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1146 ignore trailing whitespace.
1148 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1150 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1153 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1154 Notification from Dr Andrew Aitchison.
1156 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1157 ExtendedDetectionInfo response format.
1158 Notification from John Horne.
1160 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1163 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1164 XSL and documented dependency on system catalogs, with examples of how
1167 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1170 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1171 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1172 configuration file which is writeable by the Exim user or group.
1174 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1175 of configuration files to cover files specified with the -C option if
1176 they are going to be used with root privileges, not just the default
1179 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1180 option (effectively making it always true).
1182 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1183 files to be used while preserving root privileges.
1185 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1186 that rogue child processes cannot use them.
1188 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1189 run-time user, instead of root.
1191 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1192 Exim run-time user without dropping privileges.
1194 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1195 result string, instead of calling string_vformat() twice with the same
1198 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1199 for other users. Others should always drop root privileges if they use
1200 -C on the command line, even for a whitelisted configure file.
1202 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1204 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1210 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1211 $data_path, and $header_path variables; fixed documentation bugs and
1214 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1215 exipick to access non-standard spools, including the "frozen" queue
1218 NM/01 Bugzilla 965: Support mysql stored procedures.
1219 Patch from Alain Williams
1221 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1223 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1224 Patch from Andreas Metzler
1226 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1227 Patch from Kirill Miazine
1229 NM/05 Bugzilla 671: Added umask to procmail example.
1231 JJ/03 installed exipick 20100323.0, fixing doc bug
1233 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1234 directory. Notification and patch from Dan Rosenberg.
1236 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1238 TK/02 Improve log output when DKIM signing operation fails.
1240 MH/01 Treat the transport option dkim_domain as a colon separated
1241 list, not as a single string, and sign the message with each element,
1242 omitting multiple occurences of the same signer.
1244 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1245 Bugzilla 985, 986. Patch by Simon Arlott
1247 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1248 Patch by Simon Arlott
1250 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1251 MBX locking. Notification from Dan Rosenberg.
1257 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1259 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1261 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1263 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1265 NM/04 Bugzilla 915: Flag broken perl installation during build.
1271 TK/01 Added patch by Johannes Berg that expands the main option
1272 "spamd_address" if it starts with a dollar sign.
1274 TK/02 Write list of recipients to X-Envelope-Sender header when building
1275 the mbox-format spool file for content scanning (suggested by Jakob
1278 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1279 (http://www.dcc-servers.net/) support via dccifd. Activated by
1280 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1282 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1283 by Mark Daniel Reidel <mr@df.eu>.
1285 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1286 When building exim an external PCRE library is now needed -
1287 PCRE is a system library on the majority of modern systems.
1288 See entry on PCRE_LIBS in EDITME file.
1290 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1291 conversation. Added nologin parameter to request.
1292 Patch contributed by Kirill Miazine.
1294 TF/01 Do not log submission mode rewrites if they do not change the address.
1296 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1298 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1299 log files in place. Contributed by Roberto Lima.
1301 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1303 TF/03 Bugzilla 615: When checking the local_parts router precondition
1304 after a local_part_suffix or local_part_prefix option, Exim now
1305 does not use the address's named list lookup cache, since this
1306 contains cached lookups for the whole local part.
1308 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1309 Robert Millan. Documentation is in experimental-spec.txt.
1311 TF/04 Bugzilla 668: Fix parallel build (make -j).
1313 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1315 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1316 Patch provided by Jan Srzednicki.
1318 TF/05 Leading white space used to be stripped from $spam_report which
1319 wrecked the formatting. Now it is preserved.
1321 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1322 that they are available at delivery time.
1324 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1326 TF/08 TLS error reporting now respects the incoming_interface and
1327 incoming_port log selectors.
1329 TF/09 Produce a more useful error message if an SMTP transport's hosts
1330 setting expands to an empty string.
1332 NM/06 Bugzilla 744: EXPN did not work under TLS.
1333 Patch provided by Phil Pennock.
1335 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1336 Patch provided by Richard Godbee.
1338 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1339 acl_smtp_notquit, added index entry.
1341 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1342 Patch provided by Eugene Bujak.
1344 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1345 accept(). Patch provided by Maxim Dounin.
1347 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1348 Patch provided by Phil Pennock.
1350 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1352 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1353 Patch provided by Brad "anomie" Jorsch.
1355 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1356 Patch provided by Dean Brooks.
1358 TK/05 Add native DKIM support (does not depend on external libraries).
1360 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1361 Patch provided by Graeme Fowler.
1363 NM/16 Bugzilla 851: Documentation example syntax fix.
1365 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1367 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1370 NM/19 Bugzilla 745: TLS version reporting.
1371 Patch provided by Phil Pennock.
1373 NM/20 Bugzilla 167: bool: condition support.
1374 Patch provided by Phil Pennock.
1376 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1377 clients. Patch provided by Phil Pennock.
1379 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1380 Patch provided by Brad "anomie" Jorsch.
1382 NM/23 Bugzilla 687: Fix misparses in eximstats.
1383 Patch provided by Heiko Schlittermann.
1385 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1386 Patch provided by Heiko Schlittermann.
1388 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1389 plus update to original patch.
1391 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1393 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1394 Patch provided by David Brownlee.
1396 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1398 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1400 NM/30 Bugzilla 888: TLS documentation bugfixes.
1402 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1404 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1405 Unlike the original bugzilla I have changed all shell scripts in src tree.
1407 NM/33 Bugzilla 898: Transport filter timeout fix.
1408 Patch by Todd Rinaldo.
1410 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1411 Patch by Serge Demonchaux.
1413 NM/35 Bugzilla 39: Base64 decode bug fixes.
1414 Patch by Jakob Hirsch.
1416 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1418 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1420 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1422 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1428 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1429 ALT-N's libdkim that I have put here:
1430 http://duncanthrax.net/exim-experimental/
1432 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1433 were called 'true' and it seems that C99 defines that as a reserved
1434 keyword to be used with 'bool' variable types. That means you could
1435 not include C99-style headers which use bools without triggering
1436 build errors in sieve.c.
1438 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1439 as mailq or other aliases. Changed the --help handling significantly
1440 to do whats expected. exim_usage() emits usage/help information.
1442 SC/01 Added the -bylocaldomain option to eximstats.
1444 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1446 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1448 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1454 PH/01 Another patch from the Sieve maintainer.
1456 PH/02 When an IPv6 address is converted to a string for single-key lookup
1457 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1458 dots are used instead of colons so that keys in lsearch files need not
1459 contain colons. This was done some time before quoting was made available
1460 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1461 (notated using the quote facility) so as to distinguish them from IPv4
1462 keys. This meant that lookups for IP addresses in host lists did not work
1463 for iplsearch lookups.
1465 This has been fixed by arranging for IPv6 addresses to be expressed with
1466 colons if the lookup type is iplsearch. This is not incompatible, because
1467 previously such lookups could never work.
1469 The situation is now rather anomolous, since one *can* have colons in
1470 ordinary lsearch keys. However, making the change in all cases is
1471 incompatible and would probably break a number of configurations.
1473 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1476 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1477 conversion specification without a maximum field width, thereby enabling
1478 a rogue spamd server to cause a buffer overflow. While nobody in their
1479 right mind would setup Exim to query an untrusted spamd server, an
1480 attacker that gains access to a server running spamd could potentially
1481 exploit this vulnerability to run arbitrary code as the Exim user.
1483 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1484 $primary_hostname instead of what libspf2 thinks the hosts name is.
1486 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1487 a directory entry by the name of the lookup key. Previously, if a
1488 symlink pointed to a non-existing file or a file in a directory that
1489 Exim lacked permissions to read, a lookup for a key matching that
1490 symlink would fail. Now it is enough that a matching directory entry
1491 exists, symlink or not. (Bugzilla 503.)
1493 PH/03 The body_linecount and body_zerocount variables are now exported in the
1496 PH/04 Added the $dnslist_matched variable.
1498 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1499 This means they are set thereafter only if the connection becomes
1502 PH/06 Added the client_condition to authenticators so that some can be skipped
1503 by clients under certain conditions.
1505 PH/07 The error message for a badly-placed control=no_multiline_responses left
1506 "_responses" off the end of the name.
1508 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1510 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1511 (without spaces) instead of just copying the configuration text.
1513 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1515 PH/11 Added $max_received_linelength.
1517 PH/12 Added +ignore_defer and +include_defer to host lists.
1519 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1520 way in which PCRE > 7.0 is built.
1522 PH/14 Implemented queue_only_load_latch.
1524 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1525 MAIL command. The effect was to mangle the value on 64-bit systems.
1527 PH/16 Another patch from the Sieve maintainer.
1529 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1531 PH/18 If a system quota error occurred while trying to create the file for
1532 a maildir delivery, the message "Mailbox is full" was not appended to the
1533 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1534 only to a quota excession during the actual writing of the file.
1536 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1537 characters?) which causes problems in log lines. The DN values are now
1538 passed through string_printing() before being added to log lines.
1540 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1541 and InterBase are left for another time.)
1543 PH/21 Added message_body_newlines option.
1545 PH/22 Guard against possible overflow in moan_check_errorcopy().
1547 PH/23 POSIX allows open() to be a macro; guard against that.
1549 PH/24 If the recipient of an error message contained an @ in the local part
1550 (suitably quoted, of course), incorrect values were put in $domain and
1551 $local_part during the evaluation of errors_copy.
1557 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1558 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1561 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1562 issue a MAIL command.
1564 PH/02 In an ACL statement such as
1566 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1568 if a client was not listed at all, or was listed with a value other than
1569 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1570 the condition was not true (as it should be), so access was not denied.
1571 The bug was that the ! inversion was incorrectly passed on to the second
1572 item. This has been fixed.
1574 PH/03 Added additional dnslists conditions == and =& which are different from
1575 = and & when the dns lookup returns more than one IP address.
1577 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1578 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1580 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1581 FSYNC, which compiles an option called disable_fsync that allows for
1582 bypassing fsync(). The documentation is heavily laced with warnings.
1584 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1586 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1587 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1588 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1589 including adding "make clean"; (3) Added -fPIC when compiling the test
1590 dynamically loaded module, to get rid of a warning.
1592 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1593 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1594 = 0s. The bug is otherwise harmless.
1596 PH/07 There was a bug in the dovecot authenticator such that the value of
1597 $auth1 could be overwritten, and so not correctly preserved, after a
1598 successful authentication. This usually meant that the value preserved by
1599 the server_setid option was incorrect.
1601 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1603 PH/09 Installed PCRE release 7.0.
1605 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1606 run for batched SMTP input. It is now run at the start of every message
1607 in the batch. While fixing this I discovered that the process information
1608 (output by running exiwhat) was not always getting set for -bs and -bS
1609 input. This is fixed, and it now also says "batched" for BSMTP.
1611 PH/11 Added control=no_pipelining.
1613 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1614 patch, slightly modified), and move the expansion of helo_data till after
1615 the connection is made in the smtp transport (so it can use these
1618 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1620 PH/14 Added log_selector = +pid.
1622 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1624 PH/16 Add ${if forany and ${if forall.
1626 PH/17 Added dsn_from option to vary the From: line in DSNs.
1628 PH/18 Flush SMTP output before performing a callout, unless control =
1629 no_callout_flush is set.
1631 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1632 was true (the default) a successful delivery failed to delete the retry
1633 item, thus causing premature timeout of the address. The bug is now
1636 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1638 PH/21 Long custom messages for fakedefer and fakereject are now split up
1639 into multiline reponses in the same way that messages for "deny" and
1640 other ACL rejections are.
1642 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1643 with slight modification.
1645 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1646 draft, changing the syntax and factoring some duplicate code".
1648 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1649 for deliveries of the second and subsequent messages over the same SMTP
1652 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1653 ${reduce, with only minor "tidies".
1655 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1657 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1658 expansion side effects.
1660 PH/27 When a message times out after an over-quota error from an Exim-imposed
1661 quota, the bounce message says "mailbox is full". This message was not
1662 being given when it was a system quota that was exceeded. It now should
1665 MH/03 Made $recipients available in local_scan(). local_scan() already has
1666 better access to the recipient list through recipients_list[], but
1667 $recipients can be useful in postmaster-provided expansion strings.
1669 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1670 in the case of a MAIL command with additional options following the
1671 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1672 were accidentally chopped off.
1674 PH/29 SMTP synchronization checks are implemented when a command is read -
1675 there is a check that no more input is waiting when there shouldn't be
1676 any. However, for some commands, a delay in an ACL can mean that it is
1677 some time before the response is written. In this time, more input might
1678 arrive, invalidly. So now there are extra checks after an ACL has run for
1679 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1680 pipelining has not been advertised.
1682 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1684 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1685 returned in the SMTP response when smtp_return_error_details was false.
1686 This has been fixed.
1688 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1689 instead of the C I/O that was originally supplied, because problems were
1690 reported on Solaris.
1692 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1693 Exim which did not show up earlier: it was assuming that a call to
1694 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1695 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1696 was a macro that became an assignment, so it seemed to work. This has
1697 changed to a proper function call with a void return, hence the compile
1698 error. Exim's code has been fixed.
1700 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1703 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1705 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1707 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1708 because it turns out to be too big or there is a timeout) there is no
1709 "Completed" line in the log. When some messages of this type were
1710 selected by exigrep, they were listed as "not completed". Others were
1711 picked up by some special patterns. I have improved the selection
1712 criteria to be more general.
1714 PH/38 The host_find_failed option in the manualroute router can now be set
1715 to "ignore", to completely ignore a host whose IP address cannot be
1716 found. If all hosts are ignored, the behaviour is controlled by the new
1717 host_all_ignored option.
1719 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1720 homing or because of multiple MX records with /mx) generated more than
1721 one IP address, and the following item turned out to be the local host,
1722 all the secondary addresses of the first item were incorrectly removed
1723 from the list, along with the local host and any following hosts (which
1724 is what is supposed to happen).
1726 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1727 whoever called Exim into the -H file. In the case of the daemon it was
1728 behaving confusingly. When first started, it used values for whoever
1729 started the daemon, but after a SIGHUP it used the Exim user (because it
1730 calls itself on a restart). I have changed the code so that it now always
1733 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1734 message are rejected with the same error (e.g. no authentication or bad
1735 sender address), and a DATA command is nevertheless sent (as can happen
1736 with PIPELINING or a stupid MUA), the error message that was given to the
1737 RCPT commands is included in the rejection of the DATA command. This is
1738 intended to be helpful for MUAs that show only the final error to their
1741 PH/42 Another patch from the Sieve maintainer.
1743 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1744 Eximstats - Fixed some broken HTML links and added missing column headers
1746 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1747 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1749 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1751 PH/43 Yet another patch from the Sieve maintainer.
1753 PH/44 I found a way to check for a TCP/IP connection going away before sending
1754 the response to the final '.' that terminates a message, but only in the
1755 case where the client has not sent further data following the '.'
1756 (unfortunately, this is allowed). However, in many cases there won't be
1757 any further data because there won't be any more messages to send. A call
1758 to select() can be used: if it shows that the input is "ready", there is
1759 either input waiting, or the socket has been closed. An attempt to read
1760 the next input character can distinguish the two cases. Previously, Exim
1761 would have sent an OK response which the client would never have see.
1762 This could lead to message repetition. This fix should cure that, at
1763 least in a lot of common cases.
1765 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1766 advertised in response to EHLO.
1772 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1773 fixed by 4.65/MH/01 (is this a record?) are fixed:
1775 (i) An empty string was always treated as zero by the numeric comparison
1776 operators. This behaviour has been restored.
1778 (ii) It is documented that the numeric comparison operators always treat
1779 their arguments as decimal numbers. This was broken in that numbers
1780 starting with 0 were being interpreted as octal.
1782 While fixing these problems I realized that there was another issue that
1783 hadn't been noticed. Values of message_size_limit (both the global option
1784 and the transport option) were treated as octal if they started with 0.
1785 The documentation was vague. These values are now always treated as
1786 decimal, and I will make that clear in the documentation.
1792 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1793 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1796 MH/01 Don't check that the operands of numeric comparison operators are
1797 integers when their expansion is in "skipping" mode (fixes bug
1798 introduced by 4.64-PH/07).
1800 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1801 child addresses, Exim now panics and dies. Previously, because the count
1802 is held in a short int, deliveries were likely to be lost. As such a
1803 large number of recipients for a single message is ridiculous
1804 (performance will be very, very poor), I have chosen to impose a limit
1805 rather than extend the field.
1811 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1812 leftover -K file (the existence of which was triggered by #402).
1813 While we were at it, introduced process PID as part of the -K
1814 filename. This should rule out race conditions when creating
1817 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1818 processing considerably. Previous code took too long for large mails,
1819 triggering a timeout which in turn triggers #401.
1821 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1822 in the DK code in transports.c. sendfile() is not really portable,
1823 hence the _LINUX specificness.
1825 TF/01 In the add_headers option to the mail command in an Exim filter,
1826 there was a bug that Exim would claim a syntax error in any
1827 header after the first one which had an odd number of characters
1830 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1831 callout verification, Exim cached a "reject" for the entire domain. This
1832 is correct for most verifications, but it is not correct for a recipient
1833 verification with use_sender or use_postmaster set, because in that case
1834 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1835 case of MAIL FROM:<> rejection from other early rejections (e.g.
1836 rejection of HELO). When verifying a recipient using a non-null MAIL
1837 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1838 Whatever the result of the callout, the value of the domain cache is
1839 left unchanged (for any other kind of callout, getting as far as trying
1840 RCPT means that the domain itself is ok).
1842 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1845 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1846 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1847 session, the daemon ignores ECONNECT errors and logs others; it now
1848 ignores EPIPE as well.
1850 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1851 (quoted-printable decoding).
1853 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1854 later the small subsequent patch to fix an introduced bug.
1856 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1858 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1860 PH/08 An error is now given if message_size_limit is specified negative.
1862 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1863 to be given (somewhat) arbitrary names.
1865 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1868 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1869 miscellaneous code fixes
1871 PH/10 Added the log_reject_target ACL modifier to specify where to log
1874 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1875 hostname. This is wrong, because it relates to the incoming message (and
1876 probably the interface on which it is arriving) and not to the outgoing
1877 callout (which could be using a different interface). This has been
1878 changed to use the value of the helo_data option from the smtp transport
1879 instead - this is what is used when a message is actually being sent. If
1880 there is no remote transport (possible with a router that sets up host
1881 addresses), $smtp_active_hostname is used.
1883 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1884 tweaks were necessary in order to get it to work (see also 21 below):
1885 (a) The code assumed that strncpy() returns a negative number on buffer
1886 overflow, which isn't the case. Replaced with Exim's string_format()
1888 (b) There were several signed/unsigned issues. I just did the minimum
1889 hacking in of casts. There is scope for a larger refactoring.
1890 (c) The code used strcasecmp() which is not a standard C function.
1891 Replaced with Exim's strcmpic() function.
1892 (d) The code set only $1; it now sets $auth1 as well.
1893 (e) A simple test gave the error "authentication client didn't specify
1894 service in request". It would seem that Dovecot has changed its
1895 interface. Fortunately there's a specification; I followed it and
1896 changed what the client sends and it appears to be working now.
1898 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1901 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1902 address A is aliased to B and C, where B exists and C does not. Without
1903 -v the output is "A verified" because verification stops after a
1904 successful redirection if more than one address is generated. However,
1905 with -v the child addresses are also verified. Exim was outputting "A
1906 failed to verify" and then showing the successful verification for C,
1907 with its parentage. It now outputs "B failed to verify", showing B's
1908 parentage before showing the successful verification of C.
1910 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1911 look up a TXT record in a specific list after matching in a combined
1914 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1915 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1916 they consult the DNS. I had assumed they would set it the way they
1917 wanted; and indeed my experiments on Linux seem to show that in some
1918 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1919 To be on the safe side, however, I have now made the interface to
1920 host_find_byname() similar to host_find_bydns(), with an argument
1921 containing the DNS resolver options. The host_find_byname() function now
1922 sets these options at its start, just as host_find_bydns() does. The smtp
1923 transport options dns_qualify_single and dns_search_parents are passed to
1924 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
1925 of host_find_byname() use the default settings of RES_DEFNAMES
1926 (qualify_single) but not RES_DNSRCH (search_parents).
1928 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
1929 spool_read_header() do less string testing, by means of a preliminary
1930 switch on the second character of optional "-foo" lines. (This is
1931 overdue, caused by the large number of possibilities that now exist.
1932 Originally there were few.) While I was there, I also converted the
1933 str(n)cmp tests so they don't re-test the leading "-" and the first
1934 character, in the hope this might squeeze out yet more improvement.
1936 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
1937 flag allowing group syntax was set by the header_syntax check but not
1938 turned off, possible causing trouble later; (2) The flag was not being
1939 set at all for the header_verify test, causing "group"-style headers to
1940 be rejected. I have now set it in this case, and also caused header_
1941 verify to ignore an empty address taken from a group. While doing this, I
1942 came across some other cases where the code for allowing group syntax
1943 while scanning a header line wasn't quite right (mostly, not resetting
1944 the flag correctly in the right place). These bugs could have caused
1945 trouble for malformed header lines. I hope it is now all correct.
1947 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
1948 with the "reply" argument non-NULL. The code, however (which originally
1949 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
1950 but it didn't always do it. This confused somebody who was copying the
1951 code for some other use. I have removed all the tests.
1953 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
1954 feature that was used to support insecure browsers during the U.S. crypto
1955 embargo. It requires special client support, and Exim is probably the
1956 only MTA that supported it -- and would never use it because real RSA is
1957 always available. This code has been removed, because it had the bad
1958 effect of slowing Exim down by computing (never used) parameters for the
1959 RSA_EXPORT functionality.
1961 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
1962 authenticator to fail if there's a tab character in the incoming data
1963 (there should never be unless someone is messing about, as it's supposed
1964 to be base64-encoded). Also added, on Timo's advice, the "secured" option
1965 if the connection is using TLS or if the remote IP is the same as the
1966 local IP, and the "valid-client-cert option" if a client certificate has
1969 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
1970 authenticators. This can be used for authorization after authentication
1971 succeeds. (In the case of plaintext, it servers for both authentication
1974 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1975 if any retry times were supplied.
1977 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1978 connection, where there is no client IP address to check. In this
1979 situation, the verify now always succeeds.
1981 PH/25 Applied John Jetmore's -Mset patch.
1983 PH/26 Added -bem to be like -Mset, but loading a message from a file.
1985 PH/27 In a string expansion for a processed (not raw) header when multiple
1986 headers of the same name were present, leading whitespace was being
1987 removed from all of them, but trailing whitespace was being removed only
1988 from the last one. Now trailing whitespace is removed from each header
1989 before concatenation. Completely empty headers in a concatenation (as
1990 before) are ignored.
1992 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1993 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1995 PH/29 [Removed. This was a change that I later backed out, and forgot to
1996 correct the ChangeLog entry (that I had efficiently created) before
1997 committing the later change.]
1999 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2000 address errors (4xx response to RCPT) over the same connection as other
2001 messages routed to the same hosts. Such deliveries are always "forced",
2002 so retry times are not inspected. This resulted in far too many retries
2003 for the affected addresses. The effect occurred only when there were more
2004 hosts than the hosts_max_try setting in the smtp transport when it had
2005 the 4xx errors. Those hosts that it had tried were not added to the list
2006 of hosts for which the message was waiting, so if all were tried, there
2007 was no problem. Two fixes have been applied:
2009 (i) If there are any address or message errors in an SMTP delivery, none
2010 of the hosts (tried or untried) are now added to the list of hosts
2011 for which the message is waiting, so the message should not be a
2012 candidate for sending over the same connection that was used for a
2013 successful delivery of some other message. This seems entirely
2014 reasonable: after all the message is NOT "waiting for some host".
2015 This is so "obvious" that I'm not sure why it wasn't done
2016 previously. Hope I haven't missed anything, but it can't do any
2017 harm, as the worst effect is to miss an optimization.
2019 (ii) If, despite (i), such a delivery is accidentally attempted, the
2020 routing retry time is respected, so at least it doesn't keep
2021 hammering the server.
2023 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2024 in ${readsocket because some servers need this prod.
2026 PH/32 Added some extra debug output when updating a wait-xxx database.
2028 PH/33 The hint "could be header name not terminated by colon", which has been
2029 given for certain expansion errors for a long time, was not being given
2030 for the ${if def:h_colon_omitted{... case.
2032 PH/34 The spec says: "With one important exception, whenever a domain list is
2033 being scanned, $domain contains the subject domain." There was at least
2034 one case where this was not true.
2036 PH/35 The error "getsockname() failed: connection reset by peer" was being
2037 written to the panic log as well as the main log, but it isn't really
2038 panic-worthy as it just means the connection died rather early on. I have
2039 removed the panic log writing for the ECONNRESET error when getsockname()
2042 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2043 runs only) independently of the message's sender address. This meant
2044 that, if the 4xx error was in fact related to the sender, a different
2045 message to the same recipient with a different sender could confuse
2046 things. In particualar, this can happen when sending to a greylisting
2047 server, but other circumstances could also provoke similar problems.
2048 I have changed the default so that the retry time for these errors is now
2049 based a combination of the sender and recipient addresses. This change
2050 can be overridden by setting address_retry_include_sender=false in the
2053 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2054 remote server are returned as part of bounce messages. This was not
2055 happening for LMTP over a pipe (the lmtp transport), but now it is the
2056 same for both kinds of LMTP.
2058 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2059 in header lines that were in fact CNAMEs. This is no longer the case.
2061 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2062 and queue runs started by the daemon processed all messages. This has
2063 been fixed so that -R and -S can now usefully be given with -q<time>.
2065 PH/40 Import PCRE release 6.7 (fixes some bugs).
2067 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2069 PH/42 Give an error if -q is specified more than once.
2071 PH/43 Renamed the variables $interface_address and $interface_port as
2072 $received_ip_address and $received_port, to make it clear that these
2073 values apply to message reception, and not to the outgoing interface when
2074 a message is delivered. (The old names remain recognized, of course.)
2076 PH/44 There was no timeout on the connect() call when using a Unix domain
2077 socket in the ${readsocket expansion. There now is.
2079 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2080 be meaningful with "accept".
2082 SC/01 Eximstats V1.43
2083 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2085 SC/02 Eximstats V1.44
2086 Use a glob alias rather than an array ref in the generated
2087 parser. This improves both readability and performance.
2089 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2090 Collect SpamAssassin and rejection statistics.
2091 Don't display local sender or destination tables unless
2092 there is data to show.
2093 Added average volumes into the top table text output.
2095 SC/04 Eximstats V1.46
2096 Collect data on the number of addresses (recipients)
2097 as well as the number of messages.
2099 SC/05 Eximstats V1.47
2100 Added 'Message too big' to the list of mail rejection
2101 reasons (thanks to Marco Gaiarin).
2103 SC/06 Eximstats V1.48
2104 Mainlog lines which have GMT offsets and are too short to
2105 have a flag are now skipped.
2107 SC/07 Eximstats V1.49 (Alain Williams)
2108 Added the -emptyok flag.
2110 SC/08 Eximstats V1.50
2111 Fixes for obtaining the IP address from reject messages.
2113 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2114 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2115 whitesspace changes from 4.64-PH/27
2117 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2120 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2121 are found, allow negative numbers in numeric criteria)
2123 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2125 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2126 to match changes made in 4.64-PH/43
2128 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2130 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2131 "domain" is actually an IP address, and force a failure. This locks out
2132 those revolvers/nameservers that support "A-for-A" lookups, in
2133 contravention of the specifications.
2135 PH/48 When a host name was looked up from an IP address, and the subsequent
2136 forward lookup of the name timed out, the host name was left in
2137 $sender_host_name, contrary to the specification.
2139 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2140 restricted to single-key lookups, Exim was not diagnosing an error if
2141 * or *@ was used with a query-style lookup.
2143 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2145 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2146 long ago, but noone interested enough thought of it. Let's just say that
2147 the "1.1" means that there are some new functions that weren't there at
2148 some point in the past.
2150 PH/51 Error processing for expansion failure of helo_data from an smtp
2151 transport during callout processing was broken.
2153 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2154 tested/used via the -bh/-bhc/-bs options.
2156 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2157 bug, fixed in subsequent PCRE releases).
2159 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2160 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2162 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2168 SC/01 Use a glob alias rather than an array ref in eximstats generated
2169 parser. This improves both readability and performance.
2171 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2172 Don't display local sender or destination tables in eximstats unless
2173 there is data to show.
2174 Added average volumes into the eximstats top table text output.
2176 SC/03 Collect data on the number of addresses (recipients) as well
2177 as the number of messages in eximstats.
2179 TF/01 Correct an error in the documentation for the redirect router. Exim
2180 does (usually) call initgroups() when daemonizing.
2182 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2183 with consistent privilege compared to when running as a daemon.
2185 TF/03 Note in the spec that $authenticated_id is not set for local
2186 submissions from trusted users.
2188 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2189 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2191 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2192 by adding some example configuration directives to the default
2193 configuration file. A little bit of work is required to uncomment the
2194 directives and define how usernames and passwords are checked, but
2195 there is now a framework to start from.
2197 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2198 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2199 without this. I don't know how relevant this is to other LDAP libraries.
2201 PH/02 Add the verb name to the "unknown ACL verb" error.
2203 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2205 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2207 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2208 directory not expanded when it should be if an expanded home directory
2209 was set for the address (which is overridden by the transport).
2211 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2214 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2215 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2216 because it is too late at that time, and has no effect.
2218 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2219 security issue with \' (bugzilla #107). I could not use the
2220 PQescapeStringConn() function, because it needs a PGconn value as one of
2223 PH/08 When testing addresses using -bt, indicate those final addresses that
2224 are duplicates that would not cause an additional delivery. At least one
2225 person was confused, thinking that -bt output corresponded to deliveries.
2226 (Suppressing duplicates isn't a good idea as you lose the information
2227 about possibly different redirections that led to the duplicates.)
2229 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2230 systems where poll() doesn't work, in particular OS X.
2232 PH/10 Added more information to debugging output for retry time not reached.
2234 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2235 operations in malware.c.
2237 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2240 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2241 syslog_duplication set false, log lines that would normally be written
2242 both the the main log and to the reject log were not written to syslog at
2245 PH/14 In the default configuration, change the use of "message" in ACL warn
2246 statements to "add_header".
2248 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2249 not followed by a command (e.g. "seen endif").
2251 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2252 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2255 PH/17 Added extra conditions to the default value of delay_warning_condition
2259 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2260 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2261 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2264 The Auto-Submitted: and various List- headers are standardised, whereas I
2265 don't think Precedence: ever was.
2267 PH/18 Refactored debugging code in route_finduser() to show more information,
2268 in particular, the error code if getpwnam() issues one.
2270 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2271 This is apparently needed in addition to the PH/07 change above to avoid
2272 any possible encoding problems.
2274 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2275 but not after initializing Perl.
2277 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2278 output them only if debugging. By default they are written stderr,
2279 apparently, which is not desirable.
2281 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2284 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2287 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2289 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2290 authenticated or an ident call has been made. Suppress the default
2291 values for $authenticated_id and $authenticated_sender (but permit -oMai
2292 and -oMas) when testing with -bh.
2294 PH/24 Re-jigged the order of the tests in the default configuration so that the
2295 tests for valid domains and recipients precede the DNS black list and CSA
2296 tests, on the grounds that those ones are more expensive.
2298 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2299 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2300 This bug exists in every version of Exim that I still have, right back to
2303 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2304 However, an attempt to turn on case-sensitivity in a regex key by
2305 including (?-i) didn't work because the subject string was already
2306 lowercased, and the effects were non-intuitive. It turns out that a
2307 one-line patch can be used to allow (?-i) to work as expected.
2313 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2314 other effects) broke the use of negated acl sub-conditions.
2316 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2319 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2320 "Deny" causes Exim to reject the incoming connection with a 554 error.
2321 Unfortunately, if there is a major crisis, such as a disk failure,
2322 tcp-wrappers gives "deny", whereas what one would like would be some
2323 kind of temporary error. A kludge has been added to help with this.
2324 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2325 554 error is used if errno is still zero or contains ENOENT (which occurs
2326 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2329 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2331 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2332 errors. Otherwise a message that provokes a temporary error (when other
2333 messages do not) can cause a whole host to time out.
2335 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2336 addresses were routed directly to files or pipes from a redirect router.
2337 File deliveries just didn't batch; pipe deliveries might have suffered
2340 PH/06 A failure to get a lock for a hints database would erroneously always say
2341 "Failed to get write lock", even when it was really a read lock.
2343 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2344 of 0600. This has been changed to use the value of the lockfile_mode
2345 option (which defaults to 0600).
2347 PH/08 Applied small patch from the Sieve maintainer.
2349 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2350 folder from quota calculations, a direct delivery into this folder messed
2351 up the contents of the maildirsize file. This was because the regex was
2352 used only to exclude .Trash (or whatever) when the size of the mailbox
2353 was calculated. There was no check that a delivery was happening into an
2354 excluded directory. This bug has been fixed by ignoring all quota
2355 processing for deliveries into excluded directories.
2357 PH/10 Added the maildirfolder_create_regex option to appendfile.
2363 PH/01 The code for finding all the local interface addresses on a FreeBSD
2364 system running IPv6 was broken. This may well have applied to all BSD
2365 systems, as well as to others that have similar system calls. The broken
2366 code found IPv4 interfaces correctly, but gave incorrect values for the
2367 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2368 that it would not match correctly against @[] and not recognize the IPv6
2371 PH/02 The ipliteral router was not recognizing addresses of the form user@
2372 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2374 PH/03 Added disable_ipv6.
2376 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2377 decoded form, because it is most often used to construct To: headers
2378 lines in autoreplies, and the decoded form may well be syntactically
2379 invalid. However, $reply_address has leading white space removed, and all
2380 newlines turned into spaces so that the autoreply transport does not
2383 PH/05 If group was specified without a user on a router, and no group or user
2384 was specified on a transport, the group from the router was ignored.
2386 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2387 for visible compile-time settings that can be used to change these
2388 numbers, for those that want even more. Backwards compatibility with old
2389 spool files has been maintained. However, going back to a previous Exim
2390 release will lost any variables that are in spool files.
2392 PH/07 Two small changes when running in the test harness: increase delay when
2393 passing a TCP/IP connection to a new process, in case the original
2394 process has to generate a bounce, and remove special handling of
2395 127.0.0.2 (sic), which is no longer necessary.
2397 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2398 be the same on different OS.
2400 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2403 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2404 whether --show-vars was specified or not
2406 JJ/02 exipick: Added support for new ACL variable spool format introduced
2409 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2410 syntactically invalid From: or Reply-to: line, and a filter used this to
2411 generate an autoreply, and therefore failed to obtain an address for the
2412 autoreply, Exim could try to deliver to a non-existent relative file
2413 name, causing unrelated and misleading errors. What now happens is that
2414 it logs this as a hard delivery error, but does not attempt to create a
2417 PH/11 The exinext utility has a -C option for testing purposes, but although
2418 the given file was scanned by exinext itself; it wasn't being passed on
2419 when Exim was called.
2421 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2422 an end-of-file indication when reading a command response.
2424 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2425 compiled. In many other places in Exim, IPv6 addresses are always
2426 recognized, so I have changed this. It also means that IPv4 domain
2427 literals of the form [IPV4:n.n.n.n] are now always recognized.
2429 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2430 used if the router is not running as root, for example, when verifying at
2431 ACL time, or when using -bh. The debugging output from this situation was
2432 non-existent - all you got was a failure to exec. I have made two
2435 (a) Failures to set uid/gid, the current directory, or a process leader
2436 in a subprocess such as that created by queryprogram now generate
2437 suitable debugging ouput when -d is set.
2439 (b) The queryprogram router detects when it is not running as root,
2440 outputs suitable debugging information if -d is set, and then runs
2441 the subprocess without attempting to change uid/gid.
2443 PH/15 Minor change to Makefile for building test_host (undocumented testing
2446 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2447 additional section of a DNS packet that returns MX or SRV records.
2448 Instead, it always explicitly searches for A/AAAA records. This avoids
2449 major problems that occur when a DNS server includes only records of one
2450 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2451 fixed another bug: if SRV records were looked up and the corresponding
2452 address records were *not* found in the additional section, the port
2453 values from the SRV records were lost.
2455 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2456 using the correct key (the original address) when searching the retry
2457 rules in order to find which one to use for generating the retry hint.
2459 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2460 adding the default one. Similarly, if it contains a Reply-To: header, the
2461 errors_reply_to option, if set, is not used.
2463 PH/19 When calculating a retry time, Exim used to measure the "time since
2464 failure" by looking at the "first failed" field in the retry record. Now
2465 it does not use this if it is later than than the arrival time of the
2466 message. Instead it uses the arrival time. This makes for better
2467 behaviour in cases where some deliveries succeed, thus re-setting the
2468 "first failed" field. An example is a quota failure for a huge message
2469 when small messages continue to be delivered. Without this change, the
2470 "time since failure" will always be short, possible causing more frequent
2471 delivery attempts for the huge message than are intended.
2472 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2474 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2475 $1, $2, $3) because the numerical variables can be reset during some
2476 expansion items (e.g. "match"), thereby losing the authentication data.
2478 PH/21 Make -bV show the size of off_t variables so that the test suite can
2479 decide whether to run tests for quotas > 2G.
2481 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2482 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2483 filecount value is greater than 2G or if a quota value is greater than 2G
2484 on a system where the size of off_t is not greater than 4, a panic error
2487 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2488 never match. The debug and -bh output now contains an explicit error
2489 message indicating a malformed IPv4 address or mask.
2491 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2492 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2493 PH/23 above applies.
2495 PH/25 Do not write to syslog when running in the test harness. The only
2496 occasion when this arises is a failure to open the main or panic logs
2497 (for which there is an explicit test).
2499 PH/26 Added the /no_tell option to "control=freeze".
2501 PH/27 If a host name lookup failed very early in a connection, for example, if
2502 the IP address matched host_lookup and the reverse lookup yielded a name
2503 that did not have a forward lookup, an error message of the form "no IP
2504 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2505 could be logged. Now it outputs the IP address instead of "NULL".
2507 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2508 allows the sender and the authenticated sender to be set when
2509 submitting a message from within Exim. Since child_open_exim() is
2510 documented for local_scan(), the new function should be too.
2512 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2513 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2514 results in an empty string is now treated as unset.
2516 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2518 PH/31 Added sender_verify_fail logging option.
2520 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2521 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2522 tidied the source and removed it altogether.
2524 PH/33 When a queue run was abandoned because the load average was too high, a
2525 log line was always written; now it is written only if the queue_run log
2526 selector is set. In addition, the log line for abandonment now contains
2527 information about the queue run such as the pid. This is always present
2528 in "start" and "stop" lines but was omitted from the "abandon" line.
2530 PH/34 Omit spaces between a header name and the colon in the error message that
2531 is given when verify = headers_syntax fails (if there are lots of them,
2532 the message gets confusing).
2534 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2535 names, as there are now some PTR records that contain slashes. This check
2536 is only to protect against broken name servers that fall over on strange
2537 characters, so the fact that it applies to all lookups doesn't matter.
2539 PH/36 Now that the new test suite is complete, we can remove some of the
2540 special code in Exim that was needed for the old test suite. For example,
2541 sorting DNS records because real resolvers return them in an arbitrary
2542 order. The new test suite's fake resolver always returns records in the
2545 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2546 bounces) except when queue_only is set, to avoid logging races between
2547 the different processes.
2549 PH/38 Panic-die if .include specifies a non-absolute path.
2551 PH/39 A tweak to the "H" retry rule from its user.
2553 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2554 a label. They prevented compilation on older perls.
2556 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2557 a warning to be raised on newish perls.
2559 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2560 on queue. Changes to match documented behaviour of showing count of
2561 messages matching specified criteria.
2563 PH/40 Changed the default ident timeout from 30s to 5s.
2565 PH/41 Added support for the use of login_cap features, on those BSD systems
2566 that have them, for controlling the resources used by pipe deliveries.
2568 PH/42 The content-scanning code uses fopen() to create files in which to put
2569 message data. Previously it was not paying any attention to the mode of
2570 the files. Exim runs with umask(0) because the rest of the code creates
2571 files with open(), and sets the required mode explicitly. Thus, these
2572 files were ending up world-writeable. This was not a big issue, because,
2573 being within the spool directory, they were not world-accessible. I have
2574 created a function called modefopen, which takes an additional mode
2575 argument. It sets umask(777), creates the file, chmods it to the required
2576 mode, then resets the umask. All the relevant calls to fopen() in the
2577 content scanning code have been changed to use this function.
2579 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2580 to 24 hours. This avoids potential overflow problems when processing G
2581 and H retry rules. I suspect nobody ever tinkers with this value.
2583 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2585 PH/45 When the plaintext authenticator is running as a client, the server's
2586 challenges are checked to ensure they are valid base64 strings. By
2587 default, the authentication attempt is cancelled if an invalid string is
2588 received. Setting client_ignore_invalid_base64 true ignores these errors.
2589 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2590 they are received. Thus, the responses can be made to depend on the
2591 challenges. If an invalid string is ignored, an empty string is placed in
2594 PH/46 Messages that are created by the autoreply transport now contains a
2595 References: header, in accordance with RFCs 2822 and 3834.
2597 PH/47 Added authenticated_sender_force to the smtp transport.
2599 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2601 PH/49 Installed latest patch from the Sieve maintainer.
2603 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2604 was also set, the appendfile transport was unnecessarily scanning a
2605 directory of message files (e.g. for maildir delivery) to find the count
2606 of files (along with the size), even though it did not need this
2607 information. It now does the scan only if it needs to find either the
2608 size of the count of files.
2610 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2612 PH/52 Two bugs concerned with error handling when the smtp transport is
2615 (i) Exim was not creating retry information for temporary errors given
2616 for individual recipients after the DATA command when the smtp transport
2617 was used in LMTP mode. This meant that they could be retried too
2618 frequently, and not timed out correctly.
2620 (ii) Exim was setting the flag that allows error details to be returned
2621 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2622 recipients that were returned after the DATA command.
2624 PH/53 This is related to PH/52, but is more general: for any failing address,
2625 when detailed error information was permitted to be returned to the
2626 sender, but the error was temporary, then after the final timeout, only
2627 "retry timeout exceeded" was returned. Now it returns the full error as
2628 well as "retry timeout exceeded".
2630 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2631 do this, and (what is worse) MTAs that accept it.
2633 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2634 will now be deprecated.
2636 PH/56 New os.c-cygwin from the Cygwin maintainer.
2638 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2639 formats (previously only available in exim formats via -bpr, -bpru,
2640 and -bpra. Now also available in native and exiqgrep formats)
2642 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2643 with very large, slow to parse queues
2645 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2647 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2649 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2650 responses to authentication challenges, though it was showing the
2651 challenges; (ii) I've removed the CR characters from the debug output for
2654 PH/58 Allow for the insertion of a newline as well as a space when a string
2655 is turned into more than one encoded-word during RFC 2047 encoding. The
2656 Sieve code now uses this.
2658 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2659 data_4xx, lost_connection, tls_required.
2661 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2662 message was being sent as an SMTP response.
2664 PH/61 Add -l and -k options to exicyclog.
2666 PH/62 When verifying, if an address was redirected to one new address, so that
2667 verification continued, and the new address failed or deferred after
2668 having set something in $address_data, the value of $address_data was not
2669 passed back to the ACL. This was different to the case when no
2670 redirection occurred. The value is now passed back in both cases.
2672 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2673 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2674 use login_cap.h, so on its own it isn't the distinguishing feature. The
2675 new name refers directly to the setclassresources() function.
2677 PH/65 Added configuration files for NetBSD3.
2679 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2681 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2682 is preferred over IPv4.
2684 PH/68 The bounce_return_message and bounce_return_body options were not being
2685 honoured for bounces generated during the reception of non-SMTP messages.
2686 In particular, this applied to messages rejected by the ACL. This bug has
2687 been fixed. However, if bounce_return_message is true and bounce_return_
2688 body is false, the headers that are returned for a non-SMTP message
2689 include only those that have been read before the error was detected.
2690 (In the case of an ACL rejection, they have all been read.)
2692 PH/69 The HTML version of the specification is now built in a directory called
2693 spec_html instead of spec.html, because the latter looks like a path with
2694 a MIME-type, and this confuses some software.
2696 PH/70 Catch two compiler warnings in sieve.c.
2698 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2699 function verify_get_ident() calls ip_connect() to connect a socket, but
2700 if the "connect()" function timed out, ip_connect() used to close the
2701 socket. However, verify_get_ident() also closes the socket later, and in
2702 between Exim writes to the log, which may get opened at this point. When
2703 the socket was closed in ip_connect(), the log could get the same file
2704 descriptor number as the socket. This naturally causes chaos. The fix is
2705 not to close the socket in ip_connect(); the socket should be closed by
2706 the function that creates it. There was only one place in the code where
2707 this was missing, in the iplookup router, which I don't think anybody now
2708 uses, but I've fixed it anyway.
2710 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2711 well as to direct DNS lookups. Otherwise the handling of names in host
2712 lists is inconsistent and therefore confusing.
2718 PH/01 Two changes to the default runtime configuration:
2720 (1) Move the checks for relay_from_hosts and authenticated clients from
2721 after to before the (commented out) DNS black list checks.
2723 (2) Add control=submission to the relay_from_hosts and authenticated
2724 clients checks, on the grounds that messages accepted by these
2725 statements are most likely to be submissions.
2727 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2729 (1) Generate an error if the third argument for the ${prvs expansion is
2732 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2735 (3) Reset the variables that are obtained from the first argument of
2736 ${prvscheck and used in the second argument before leaving the code,
2737 because their memory is reclaimed, so using them afterwards may do
2740 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2741 one (it's much easier than Tom thought :-).
2743 (5) Because of (4), we can now allow for the use of $prvscheck_result
2744 inside the third argument.
2746 PH/03 For some reason, the default setting of PATH when running a command from
2747 a pipe transport was just "/usr/bin". I have changed it to
2750 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2751 anything to be listed in the output from -bV.
2753 PH/05 When a filter generated an autoreply, the entire To: header line was
2754 quoted in the delivery log line, like this:
2756 => >A.N.Other <ano@some.domain> <original@ddress> ...
2758 This has been changed so that it extracts the operative address. There
2759 may be more than one such address. If so, they are comma-separated, like
2762 => >ano@some.domain,ona@other.domain <original@ddress> ...
2764 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2765 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2766 not being looked up in the rDNS to get a host name, Exim was showing the
2767 IP address twice in Received: lines, even though the IP addresses were
2768 identical. For example:
2770 Received: from [1.2.3.4] (helo=[1.2.3.4])
2772 However, if the real host name was known, it was omitting the HELO data
2773 if it matched the actual IP address. This has been tidied up so that it
2774 doesn't show the same IP address twice.
2776 PH/07 When both +timestamp and +memory debugging was on, the value given by
2777 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2778 called by the debug printing, thereby overwriting the timestamp buffer.
2779 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2781 PH/08 When the original message was included in an autoreply transport, it
2782 always said "this is a copy of the message, including all the headers",
2783 even if body_only or headers_only was set. It now gives an appropriate
2786 PH/09 Applied a patch from the Sieve maintainer which:
2788 o fixes some comments
2789 o adds the (disabled) notify extension core
2790 o adds some debug output for the result of if/elsif tests
2791 o points to the current vacation draft in the documentation
2792 and documents the missing references header update
2796 o fixes a bug in processing the envelope test (when testing
2797 multiple envelope elements, the last element determinted the
2800 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2801 Electronic Mail") by including:
2803 Auto-submitted: auto-generated
2805 in the messages that it generates (bounce messages and others, such as
2806 warnings). In the case of bounce messages for non-SMTP mesages, there was
2807 also a typo: it was using "Auto_submitted" (underscore instead of
2808 hyphen). Since every message generated by Exim is necessarily in response
2809 to another message, thes have all been changed to:
2811 Auto-Submitted: auto-replied
2813 in accordance with these statements in the RFC:
2815 The auto-replied keyword:
2817 - SHOULD be used on messages sent in direct response to another
2818 message by an automatic process,
2820 - MUST NOT be used on manually-generated messages,
2822 - MAY be used on Delivery Status Notifications (DSNs) and Message
2823 Disposition Notifications (MDNs),
2825 - MUST NOT be used on messages generated by automatic or periodic
2826 processes, except for messages which are automatic responses to
2829 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2830 to the default Received: header definition.
2832 PH/12 Added log selector acl_warn_skipped (default on).
2834 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2835 variables because (a) they are in the wrong storage pool and (b) even if
2836 they were copied, it wouldn't work properly because of the caching.
2838 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2839 checking when decoding. Apparently there are clients that generate
2840 overlong encoded strings. Why am I not surprised?
2842 PH/15 If the first argument of "${if match_address" was not empty, but did not
2843 contain an "@" character, Exim crashed. Now it writes a panic log message
2844 and treats the condition as false.
2846 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2848 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2849 extension "envelope-auth". The code is finished and in agreement with
2850 other implementations, but there is no documentation so far and in fact,
2851 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2852 not changing the active code.
2854 Print executed "if" and "elsif" statements when debugging is used. This
2855 helps a great deal to understand what a filter does.
2857 Document more things not specified clearly in RFC3028. I had all this
2858 sorted out, when out of a sudden new issues came to my mind. Oops."
2860 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2863 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2864 canonical form (as documented). However, after a host name lookup from
2865 the IP address, check_host() was doing a simple string comparison with
2866 addresses acquired from the DNS when checking that the found name did
2867 have the original IP as one of its addresses. Since any found IPv6
2868 addresses are likely to be in abbreviated form, the comparison could
2869 fail. Luckily, there already exists a function for doing the comparison
2870 by converting both addresses to binary, so now that is used instead of
2871 the text comparison.
2873 PH/20 There was another similar case to PH/19, when a complete host name was
2874 given in a host list; looking up its IP address could give an abbreviated
2875 form, whereas the current host's name might or might not be abbreviated.
2876 The same fix has been applied.
2882 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2883 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2886 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2887 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2889 PH/03 Typo: missing ".o" in src/pcre/Makefile.
2891 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2892 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2893 Also, for "Auto-Submitted", treat anything other than "no" as
2894 non-personal, in accordance with RFC 3834. (Previously it treated
2895 anything starting "auto-" as non-personal.)
2897 TF/01 The control=submission/name=... option had a problem with syntax
2898 errors if the name included a slash character. The /name= option
2899 now slurps the rest of the string, so it can include any characters
2900 but it must come last in the list of options (after /sender_retain
2903 PH/05 Some modifications to the interface to the fake nameserver for the new
2911 TK/01 Added the "success_on_redirect" address verification option. See
2912 NewStuff for rationale and an example.
2914 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2916 PH/02 Patch to exigrep to allow it to work on syslog lines.
2918 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2919 fread() to skip over the body file's header line, because in Cygwin the
2920 header line is locked and is inaccessible.
2922 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
2923 co-exist for some time) to make it clear that it is the Exim ID that is
2924 referenced, not the Message-ID: header line.
2926 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
2927 string_format() function, because snprintf() does not exist on all
2930 PH/06 The use of forbid_filter_existstest now also locks out the use of the
2931 ${stat: expansion item.
2933 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
2934 protocol synchronization error", to keep the pedants happy.
2936 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
2937 well as for IRIX systems, when gcc is being used. See the host.c source
2940 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
2942 PH/10 Named domain lists were not working if used in a queue_smtp_domains
2945 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
2946 transport and to the smtp transport in LMTP mode.
2948 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
2950 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
2951 run a filter in a subprocess. This could lead to confusion in subsequent
2952 lookups in the parent process. There should also be a search_tidyup() at
2953 the end of the subprocess.
2955 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
2956 only if the host matched helo_try_verify_hosts, which caused the
2957 verification to occur when the EHLO/HELO command was issued. The ACL just
2958 tested the remembered result. Now, if a previous verification attempt has
2959 not happened, "verify = helo" does it there and then.
2961 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
2963 TK/03 Fix log output including CR from clamd.
2965 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
2966 exist provoked a memory error which could cause a segfault.
2968 PH/15 Installed PCRE 6.2
2970 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2972 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2973 of the problem. Specifically, suggested +O2 rather than +O1 for the
2976 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2978 PH/20 If a delivery was routed to a non-standard port by means of an SRV
2979 record, the port was not correctly logged when the outgoing_port log
2980 selector was set (it logged the transort's default port).
2982 PH/21 Added support for host-specific ports to manualroute, queryprogram,
2983 fallback_hosts, and "hosts" in the smtp transport.
2985 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2986 host errors such as "Connection refused".
2988 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2989 authentication with radiusclient 0.4.9:
2991 - Error returned from rc_read_config was caught wrongly
2992 - Username/password not passed on to radius server due to wrong length.
2994 The presumption is that some radiusclient API changes for 4.51/PH/17
2995 were not taken care of correctly. The code is still untested by me (my
2996 Linux distribution still has 0.3.2 of radiusclient), but it was
2997 contributed by a Radius user.
2999 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3000 expanding the "port" option of the smtp transport.
3002 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3003 while reading a MIME header. Thanks to Tom Hughes for a patch.
3005 PH/24 Include config.h inside local_scan.h so that configuration settings are
3008 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3009 that in an ACL for RCPT (for example), you can examine exactly what was
3012 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3013 commands, but it was not correctly comparing the address with the actual
3014 client host address. Thus, it would show the EHLO address in Received:
3015 header lines when this was not necessary.
3017 PH/27 Added the % operator to ${eval:}.
3019 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3020 it should be ignoring failures (because with -C, for example, it has lost
3021 privilege). It wasn't ignoring creation failures other than "already
3024 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3027 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3028 because an input line was too long, either on its own, or by virtue of
3029 too many continuations, the temporary file was not being removed, and the
3030 return code was incorrect.
3032 PH/31 Missing "BOOL" in function definition in filtertest.c.
3034 PH/32 Applied Sieve patches from the maintainer.
3036 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3038 PH/33 Added "verify = not_blind".
3040 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3041 Local/Makefile (with some defaults set). These are used in built scripts
3042 such as exicyclog, but they have never been used in the exim_install
3043 script (though there are many overriding facilities there). I have
3044 arranged that the exim_install script now takes note of these two
3047 PH/35 Installed configuration files for Dragonfly.
3049 PH/36 When a locally submitted message by a trusted user did not contain a
3050 From: header, and the sender address was obtained from -f or from an SMTP
3051 MAIL command, and the trusted user did not use -F to supply a sender
3052 name, $originator_name was incorrectly used when constructing a From:
3053 header. Furthermore, $originator_name was used for submission mode
3054 messages from external hosts without From: headers in a similar way,
3055 which is clearly wrong.
3057 PH/37 Added control=suppress_local_fixups.
3059 PH/38 When log_selector = +received_sender was set, and the addition of the
3060 sender made the log line's construction buffer exactly full, or one byte
3061 less than full, an overflow happened when the terminating "\n" was
3064 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3065 when the result of a list match is failure because a DNS lookup failed.
3067 PH/40 RM_COMMAND is now used in the building process.
3069 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3070 the "build-* directories that it finds.
3072 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3073 address in a domain literal was a prefix of an interface address.
3075 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3076 when verifying a sender address, unless rewrite_headers is false.
3078 PH/44 Wrote a long comment about why errors_to addresses are verified as
3079 recipients, not senders.
3081 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3082 the ratelimit ACL was added.
3084 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3086 PH/46 Added extra information about PostgreSQL errors to the error string.
3088 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3089 avoiding the need to install special zones in a real server. This is
3090 backwards compatible; if it can't find the fake resolver, it drops back.
3091 Thus, both old and new test suites can be run.
3093 TF/02 Added util/ratelimit.pl
3095 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3096 clock is set back in time.
3098 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3099 Candler <B.Candler@pobox.com>.
3101 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3102 OK for addresses that are the result of redirections.
3104 PH/48 A number of further additions for the benefit of the new test suite,
3105 including a fake gethostbyname() that interfaces to the fake DNS resolver
3108 TF/06 The fix for widen_domains has also been applied to qualify_single and
3109 search_parents which are the other dnslookup options that can cause
3112 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3115 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3117 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3118 service" if it encountered an MX record with an empty target hostname.
3119 The message is now "an MX or SRV record indicated no SMTP service".
3121 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3122 if the DNS of the sending site is misconfigured. This is quite a
3123 common situation. This change restores the behaviour of treating a
3124 helo verification defer as a failure.
3126 PH/51 If self=fail was set on a router, the bounce message did not include the
3127 actual error message.
3133 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3135 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3136 command itself ended in error, the underlying message about the transport
3137 filter timeout was being overwritten with the pipe command error. Now the
3138 underlying error message should be appended to the second error message.
3140 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3142 PH/02 Reduce the amount of output that "make" produces by default. Full output
3143 can still be requested.
3145 PH/03 The warning log line about a condition test deferring for a "warn" verb
3146 was being output only once per connection, rather than after each
3147 occurrence (because it was using the same function as for successful
3148 "warn" verbs). This seems wrong, so I have changed it.
3150 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3151 it should not have, which might have caused a crash in the right
3152 circumstances, but probably never did.
3154 PH/04 Installed a modified version of Tony Finch's patch to make submission
3155 mode fix the return path as well as the Sender: header line, and to
3156 add a /name= option so that you can make the user's friendly name appear
3159 TF/03 Added the control = fakedefer ACL modifier.
3161 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3162 Mark Lowes for thorough testing.
3164 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3166 TK/03 Merged latest SRS patch from Miles Wilton.
3168 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3169 with the definition in sysexits.h (which is #included earlier).
3170 Fortunately, Exim does not actually use EX_OK. The code used to try to
3171 preserve the sysexits.h value, by assumimg that macro definitions were
3172 scanned for macro replacements. I have been disabused of this notion,
3173 so now the code just undefines EX_OK before #including unistd.h.
3175 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3176 in the smtp transport. When a block could not be written in a single
3177 write() function, the timeout was being re-applied to each part-write.
3178 This seems wrong - if the receiver was accepting one byte at a time it
3179 would take for ever. The timeout is now adjusted when this happens. It
3180 doesn't have to be particularly precise.
3182 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3183 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3185 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3186 without a domain if the check to <postmaster@domain> fails.
3188 SC/01 Eximstats: added -xls and the ability to specify output files
3189 (patch written by Frank Heydlauf).
3191 SC/02 Eximstats: use FileHandles for outputing results.
3193 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3195 SC/04 Eximstats: fixed display of large numbers with -nvr option
3197 SC/05 Eximstats: fixed merging of reports with empty tables.
3199 SC/06 Eximstats: added the -include_original_destination flag
3201 SC/07 Eximstats: removed tabs and trailing whitespace.
3203 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3205 TK/06 MBOX spool code: Add real "From " MBOX separator line
3206 so the .eml file is really in mbox format (even though
3207 most programs do not really care). Patch from Alex Miller.
3209 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3210 The latter is generated from $received_to and is only set if the
3211 message has one envelope recipient. SA can use these headers,
3212 obviously out-of-the-box. Patch from Alex Miller.
3214 PH/08 The ${def test on a variable was returning false if the variable's
3215 value was "0", contrary to what the specification has always said!
3216 The result should be true unless the variable is empty.
3218 PH/09 The syntax error of a character other than { following "${if
3219 def:variable_name" (after optional whitespace) was not being diagnosed.
3220 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3221 accidental colon was present, for example, could give incorrect results.
3223 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3224 result is used (not including appendfile, where other changes are about
3227 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3228 This involved changing a lot of size variables from int to off_t. It
3229 should work with maildirs and everything.
3231 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3232 spamd dying while we are connected to it.
3234 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3237 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3238 function for BDB 4.3.
3240 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3242 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3243 the sources for PCRE have been re-arranged and more files are now
3246 PH/15 The code I had for printing potentially long long variables in PH/11
3247 above was not the best (it lost precision). The length of off_t variables
3248 is now inspected at build time, and an appropriate printing format (%ld
3249 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3250 to be "long long int" or "long int". This is needed for the internal
3251 formatting function string_vformat().
3253 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3254 the configuration file to be ":syslog", then the script "guesses" where
3255 the logs files are, rather than using the compiled in default. In our
3256 case the guess is not the same as the compiled default, so the script
3257 suddenly stopped working when I started to use syslog. The patch checks
3258 to see if log_file_path is "". If so, it attempts to read it from exim
3259 with no configuration file to get the compiled in version, before it
3260 falls back to the previous guessing code."
3262 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3263 implementing BATV in an Exim configuration. See NewStuff for the gory
3266 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3267 Makefile that are specific to HP-UX.
3269 PH/18 If the "use_postmaster" option was set for a recipient callout together
3270 with the "random" option, the postmaster address was used as the MAIL
3271 FROM address for the random test, but not for the subsequent recipient
3272 test. It is now used for both.
3274 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3275 patch removes a few documentation additions to RFC 3028, because the
3276 latest draft now contains them. It adds the new en;ascii-case comparator
3277 and a new error check for 8bit text in MIME parts. Comparator and
3278 require names are now matched exactly. I enabled the subaddress
3279 extension, but it is not well tested yet (read: it works for me)."
3281 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3282 rework some of the code of TK/09 above to avoid the hardwired use of
3283 "%lld" and "long long". Replaced the call to snprintf() with a call to
3286 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3287 records point to non-existent hosts", "retry timeout exceeded", and
3288 "retry time not reached for any host after a long failure period".
3290 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3291 experimental DomainKeys support:
3293 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3294 (2) On an error such as an illegally used "control", the wrong name for
3295 the control was given.
3297 These problems did NOT occur unless DomainKeys support was compiled.
3299 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3301 PH/24 Added ${if match_ip condition.
3303 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3304 databases so that it will be absolutely obvious if a crash occurs in the
3305 DB library. This is a regular occurrence (often caused by mis-matched
3308 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3309 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3310 picked up on a user's system that detects such things. There doesn't seem
3311 to be a gcc warning option for this - only an attribute that has to be
3312 put on the function's prototype. It seems that in Fedora Core 4 they have
3313 set this on a number of new functions. No doubt there will be more in due
3316 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3317 specify a transport. However, if an address that was verified by such a
3318 router was the subject of a callout, Exim crashed because it tried to
3319 read the rcpt_include_affixes from the non-existent transport. Now it
3320 just assumes that the setting of that option is false. This bug was
3321 introduced by 4.51/PH/31.
3323 PH/28 Changed -d+all to exclude +memory, because that information is very
3324 rarely of interest, but it makes the output a lot bigger. People tend to
3325 do -d+all out of habit.
3327 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3328 code in os-type was giving problems when libc.so lives in lib64, like on
3331 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3332 aren't the modern standard, and it seems that some systems' include files
3333 don't always have them. Exim was already checking for some of the newer
3334 ones like T_AAAA, and defining it itself. I've added checks for all the
3335 record types that Exim uses.
3337 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3338 not automatically generating a new one, as it is supposed to. This
3339 prevented TLS from working. If the file did exist, but contained invalid
3340 data, a new version was generated, as expected. It was only the case of a
3341 non-existent file that was broken.
3343 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3344 with a change in libdomainkeys > 0.64.
3346 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3347 from DNS. If the selector record carries the flag, it now has
3348 precedence over the domain-wide flag.
3350 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3352 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3353 the use of a port name that isn't defined in /etc/services) occurred, the
3354 message was deferred as in a normal delivery, and thus remained on the
3355 spool, instead of being failed because of the mua_wrapper setting. This
3356 is now fixed, and I tidied up some of the mua_wrapper messages at the
3359 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3360 the messages in a hash of arrays rather than using individual hashes.
3361 This is a bit cleaner and results in dramatic memory savings, albeit
3362 at a slight CPU cost.
3364 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3365 as requested by Marc Sherman.
3367 SC/10 Eximstats: added histograms for user specified patterns as requested
3370 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3372 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3373 fopen() in the content-scanning modules that did not already have it.
3379 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3380 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3382 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3384 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3386 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3387 to test Sieve filters that use "vacation".
3389 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3390 that changes the way the GnuTLS parameters are stored in the cache file.
3391 The new format can be generated externally. For backward compatibility,
3392 if the data in the cache doesn't make sense, Exim assumes it has read an
3393 old-format file, and it generates new data and writes a new file. This
3394 means that you can't go back to an older release without removing the
3397 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3398 work if there are any delivery delays because "one_time" forces the
3399 parent to be marked "delivered", so its unseen clone is never tried
3400 again. For this reason, Exim now forbids the simultaneous setting of
3403 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3404 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3405 entry if you want to know the details. The fix, however, neglected to
3406 consider the case when local delivery batching is involved. The test for
3407 "previously delivered" was not happening when checking to see if an
3408 address could be batched with a previous (undelivered) one; under
3409 certain circumstances this could lead to multiple deliveries to the same
3412 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3413 in its include files, and this causes problems building Exim.
3415 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3416 header_syntax) but Exim was just ignoring anything given after a slash.
3417 In particular, this caused confusion with an attempt to use "verify =
3418 reverse_host_lookup/defer_ok". An error is now given when options are
3419 supplied for verify items that do not have them. (Maybe reverse_host_
3420 lookup should have a defer_ok option, but that's a different point.)
3422 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3423 defined by RFC 821) to 2048, because there were problems with some AUTH
3424 commands, and RFC 1869 says the size should be increased for extended
3425 SMTP commands that take arguments.
3427 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3430 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3431 "unknown" error; now it says that the functionality isn't in the binary.
3433 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3434 an address' error message when a string expansion fails (syntax or
3435 whatever). Otherwise the password may appear in the log. Following change
3436 PH/42 below, there is no longer a chance of it appearing in a bounce
3439 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3441 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3442 of its addresses was ever tried. (Bugzilla bug #2.)
3444 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3445 the result incorrectly in the debug output. (It correctly added a newline
3446 to what was transported.)
3448 TF/01 Added $received_time.
3450 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3451 commented out examples of how to interface to a virus scanner and to
3452 SpamAssassin. Also added commented examples of av_scanner and
3453 spamd_address settings.
3455 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3456 and controls are allowed in which ACLs. There were a couple of minor
3457 errors. Some of the entries in the conditions table (which is a table of
3458 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3459 negation of where the condition IS allowed.
3461 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3463 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3464 header file does not have a version number, so I've had to invent a new
3465 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3466 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3467 radiusclient), but it was contributed by a Radius user.
3469 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3470 files or directories, for OpenSSL.
3472 PH/19 When an Exim process that is running as root has to create an Exim log
3473 file, it does so in a subprocess that runs as exim:exim so as to get the
3474 ownership right at creation (otherwise, other Exim processes might see
3475 the file with the wrong ownership). There was no test for failure of this
3476 fork() call, which would lead to the process getting stuck as it waited
3477 for a non-existent subprocess. Forks do occasionally fail when resources
3478 run out. I reviewed all the other calls to fork(); they all seem to check
3481 PH/20 When checking for unexpected SMTP input at connect time (before writing
3482 the banner), Exim was not dealing correctly with a non-positive return
3483 from the read() function. If the client had disconnected by this time,
3484 the result was a log entry for a synchronization error with an empty
3485 string after "input=" when read() returned zero. If read() returned -1
3486 (an event I could not check), uninitialized data bytes were printed.
3487 There were reports of junk text (parts of files, etc) appearing after
3490 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3492 PH/22 Added support for macro redefinition, and (re)definition in between
3493 driver and ACL definitions.
3495 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3496 forgetting to use the resulting value; it was using the unexpanded value.
3498 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3499 hadn't been configured. The fix is from Juergen Kreileder, who
3500 understands it better than I do:
3502 "Here's what I see happening with three configured cyrus_sasl
3503 authenticators configured (plain, login, cram-md5):
3505 On startup auth_cyrus_sasl_init() gets called for each of these.
3506 This means three calls to sasl_listmech() without a specified mech_list.
3507 => SASL tests which mechs of all available mechs actually work
3508 => three warnings about OTP not working
3509 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3511 With the patch, sasl_listmech() also gets called three times. But now
3512 SASL's mech_list option is set to the server_mech specified in the the
3513 authenticator. Or in other words, the answer from sasl_listmech()
3514 gets limited to just the mech you're testing for (which is different
3516 => the return list contains just 'plain' or 'login', 'cram-md5' or
3517 nothing depending on the value of ob->server_mech.
3519 I've just tested the patch: Authentication still works fine,
3520 unavailable mechs specified in the exim configuration are still
3521 caught, and the auth.log warnings about OTP are gone."
3523 PH/25 When debugging is enabled, the contents of the command line are added
3524 to the debugging output, even when log_selector=+arguments is not
3527 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3528 answer is "GNU", and only if the return is "GNU/something" is the answer
3531 PH/27 $acl_verify_message is now set immediately after the failure of a
3532 verification in an ACL, and so is available in subsequent modifiers. In
3533 particular, the message can be preserved by coding like this:
3535 warn !verify = sender
3536 set acl_m0 = $acl_verify_message
3538 Previously, $acl_verify_message was set only while expanding "message"
3539 and "log_message" when a very denied access.
3541 PH/28 Modified OS/os.c-Linux with
3543 -#ifndef OS_LOAD_AVERAGE
3544 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3546 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3547 nomenclature these days.)
3549 PH/29 Installed patch from the Sieve maintainer that adds the options
3550 sieve_useraddress and sieve_subaddress to the redirect router.
3552 PH/30 In these circumstances:
3553 . Two addresses routed to the same list of hosts;
3554 . First host does not offer TLS;
3555 . First host accepts first address;
3556 . First host gives temporary error to second address;
3557 . Second host offers TLS and a TLS session is established;
3558 . Second host accepts second address.
3559 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3560 and peerdn, if requested) that were in fact used only for the second
3563 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3564 attention to any local part prefix or suffix that was matched by the
3565 router that accepted the address. It now behaves in the same way as it
3566 does for delivery: the affixes are removed from the local part unless
3567 rcpt_include_affixes is set on the transport.
3569 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3570 timeout during the DATA phase of an incoming message.
3572 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3573 applied a patch sanctioned by the Sieve maintainer.
3575 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3576 the uid or gid is negative. A case of a negative gid caused this to be
3577 noticed. The fix allows for either to be negative.
3579 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3580 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3583 PH/36 The change PH/12 above was broken. Fixed it.
3585 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3586 the grounds that routing the same address twice would always produce the
3587 same answer. This might have been true once, but it is certainly no
3588 longer true now. Routing a child address may depend on the previous
3589 routing that produced that child. Some complicated redirection strategies
3590 went wrong when messages had multiple recipients, and made Exim's
3591 behaviour dependent on the order in which the addresses were given.
3593 I have moved the duplicate checking until after the routing is complete.
3594 Exim scans the addresses that are assigned to local and remote
3595 transports, and removes any duplicates. This means that more work will be
3596 done, as duplicates will always all be routed, but duplicates are
3597 presumably rare, so I don't expect this is of any significance.
3599 For deliveries to pipes, files, and autoreplies, the duplicate checking
3600 still happens during the routing process, since they are not going to be
3603 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3604 It corrects a timeout issue with spamd. This is Ian's comment: "The
3605 background is that sometimes spamd either never reads data from a
3606 connection it has accepted, or it never writes response data. The exiscan
3607 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3608 blindly assumes that writes won't block so it may never time out."
3610 PH/39 Allow G after quota size as well as K and M.
3612 PH/40 The value set for $authenticated_id in an authenticator may not contain
3613 binary zeroes or newlines because the value is written to log lines and
3614 to spool files. There was no check on this. Now the value is run through
3615 the string_printing() function so that such characters are converted to
3616 printable escape sequences.
3618 PH/41 $message_linecount is a new variable that contains the total number of
3619 lines in the message. Compare $body_linecount, which is the count for the
3622 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3623 bounce and delay warning messages, except in certain special cases, which
3626 (a) An SMTP error message from a remote host;
3627 (b) A message specified in a :fail: redirection;
3628 (c) A message specified in a "fail" command in a system filter;
3629 (d) A message specified in a FAIL return from the queryprogram router;
3630 (e) A message specified by the cannot_route_message router option.
3632 In these cases only, Exim does include the error details in bounce and
3633 warning messages. There are also a few cases where bland messages such
3634 as "unrouteable address" or "local delivery error" are given.
3636 PH/43 $value is now also set for the "else" part of a ${run expansion.
3638 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3639 being worked on, but at least Exim now implements the latest version to
3642 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3643 process to complete was treated as a delivery failure, a timeout while
3644 writing the message to the pipe was logged, but erroneously treated as a
3645 successful delivery. Such timeouts include transport filter timeouts. For
3646 consistency with the overall process timeout, these timeouts are now
3647 treated as errors, giving rise to delivery failures by default. However,
3648 there is now a new Boolean option for the pipe transport called
3649 timeout_defer, which, if set TRUE, converts the failures into defers for
3650 both kinds of timeout. A transport filter timeout is now identified in
3653 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3654 systems where "make" and "gmake" are different, calling "gmake" at top
3655 level broke things. I've arranged for the value of $(MAKE) to be passed
3656 from the Makefile to this script so that it can call the same version of
3660 A note about Exim versions 4.44 and 4.50
3661 ----------------------------------------
3663 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3664 changes of various kinds. As a consequence, a big documentation update was
3665 needed. This delayed the release for rather longer than seemed good, especially
3666 in the light of a couple of (minor) security issues. Therefore, the changes
3667 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3668 release. So 4.44 and 4.50 are in effect two different branches that both start
3671 I have left the 4.50 change log unchanged; it contains all the changes since
3672 4.43. The change log for 4.44 is below; many of its items are identical to
3673 those for 4.50. This seems to be the most sensible way to preserve the
3674 historical information.
3680 1. Minor wording change to the doc/README.SIEVE file.
3682 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3683 computation of the current number of files was incorrect.
3685 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3686 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3689 4. Give more explanation in the error message when the command for a transport
3690 filter fails to execute.
3692 5. There are several places where Exim runs a non-Exim command in a
3693 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3694 was being done only for the command run by the queryprogram router. It is
3695 now done for all such subprocesses. The other cases are: ${run, transport
3696 filters, and the commands run by the lmtp and pipe transports.
3698 6. Added CONFIGURE_GROUP build-time option.
3700 7. Some older OS have a limit of 256 on the maximum number of file
3701 descriptors. Exim was using setrlimit() to set 1000 as a large value
3702 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3703 systems. I've change it so that if it can't get 1000, it tries for 256.
3705 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3706 was an oversight, and furthermore, ever since the addition of extra
3707 controls (e.g. 4.43/32), the checks on when to allow different forms of
3708 "control" were broken. There should now be diagnostics for all cases when a
3709 control that does not make sense is encountered.
3711 9. Added the /retain_sender option to "control=submission".
3713 10. $recipients is now available in the predata ACL (oversight).
3715 11. Tidy the search cache before the fork to do a delivery from a message
3716 received from the command line. Otherwise the child will trigger a lookup
3717 failure and thereby defer the delivery if it tries to use (for example) a
3718 cached ldap connection that the parent has called unbind on.
3720 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3721 of $address_data from the recipient verification was clobbered by the
3722 sender verification.
3724 13. The value of address_data from a sender verification is now available in
3725 $sender_address_data in subsequent conditions in the ACL statement.
3727 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3729 15. Added a new option "connect=<time>" to callout options, to set a different
3732 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3733 was its contents. (It was OK if the option was not defined at all.)
3735 17. A "Completed" log line is now written for messages that are removed from
3736 the spool by the -Mrm option.
3738 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3739 information about exactly what failed.
3741 19. Added -dd to debug only the daemon process.
3743 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3744 handles timeouts, both on the server side and network timeouts. Renamed the
3745 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3747 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3748 It is now set to "smtps".
3750 22. $host_address is now set to the target address during the checking of
3751 ignore_target_hosts.
3753 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3754 being passed; this would have caused $sender_host_name to have been used if
3755 matching the list had actually called for a host name (not very likely,
3756 since this list is usually IP addresses). A host name is now passed as
3759 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3760 code that specifies a non-restarting handler (typically sigaction() in
3761 modern systems) in an attempt to fix a rare and obscure crash bug.
3763 25. Narrowed the window for a race in the daemon that could cause it to ignore
3764 SIGCHLD signals. This is not a major problem, because they are used only to
3765 wake it up if nothing else does.
3767 26. A malformed maildirsize file could cause Exim to calculate negative values
3768 for the mailbox size or file count. Odd effects could occur as a result.
3769 The maildirsize information is now recalculated if the size or filecount
3772 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3773 support for a long time. Removed HAVE_SYS_VFS_H.
3775 28. Installed the latest version of exipick from John Jetmore.
3777 29. In an address list, if the pattern was not a regular expression, an empty
3778 subject address (from a bounce message) matched only if the pattern was an
3779 empty string. Non-empty patterns were not even tested. This was the wrong
3780 because it is perfectly reasonable to use an empty address as part of a
3781 database query. An empty address is now tested by patterns that are
3782 lookups. However, all the other forms of pattern expect the subject to
3783 contain a local part and a domain, and therefore, for them, an empty
3784 address still always fails if the pattern is not itself empty.
3786 30. Exim went into a mad DNS loop when attempting to do a callout where the
3787 host was specified on an smtp transport, and looking it up yielded more
3788 than one IP address.
3790 31. Re-factored the code for checking spool and log partition space into a
3791 function that finds that data and another that does the check. The former
3792 is then used to implement four new variables: $spool_space, $log_space,
3793 $spool_inodes, and $log_inodes.
3795 32. The RFC2047 encoding function was originally intended for short strings
3796 such as real names; it was not keeping to the 75-character limit for
3797 encoded words that the RFC imposes. It now respects the limit, and
3798 generates multiple encoded words if necessary. To be on the safe side, I
3799 have increased the buffer size for the ${rfc2047: expansion operator from
3802 33. It is now permitted to omit both strings after an "if" condition; if the
3803 condition is true, the result is "true". As before, when the second string
3804 is omitted, a false condition yields an empty string. This makes it less
3805 cumbersome to write custom ACL and router conditions.
3807 34. Failure to deliver a bounce message always caused it to be frozen, even if
3808 there was an errors_to setting on the router. The errors_to setting is now
3811 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3812 canonical form (fully expanded) before being placed in
3813 $sender_host_address.
3815 36. The table in the code that translates DNS record types into text (T_A to
3816 "A" for instance) was missing entries for NS and CNAME. It is just possible
3817 that this could have caused confusion if both these types were looked up
3818 for the same domain, because the text type is used as part of Exim's
3819 per-process caching. But the chance of anyone hitting this buglet seems
3822 37. The dnsdb lookup has been extended in a number of ways.
3824 (1) There is a new type, "zns", which walks up the domain tree until it
3825 finds some nameserver records. It should be used with care.
3827 (2) There is a new type, "mxh", which is like "mx" except that it returns
3828 just the host names, not the priorities.
3830 (3) It is now possible to give a list of domains (or IP addresses) to be
3831 looked up. The behaviour when one of the lookups defers can be
3832 controlled by a keyword.
3834 (4) It is now possible to specify the separator character for use when
3835 multiple records are returned.
3837 38. The dnslists ACL condition has been extended: it is now possible to supply
3838 a list of IP addresses and/or domains to be looked up in a particular DNS
3841 39. Added log_selector=+queue_time_overall.
3843 40. When running the queue in the test harness, wait just a tad after forking a
3844 delivery process, to get repeatability of debugging output.
3846 41. Include certificate and key file names in error message when GnuTLS fails
3847 to set them up, because the GnuTLS error message doesn't include the name
3848 of the failing file when there is a problem reading it.
3850 42. Allow both -bf and -bF in the same test run.
3852 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3854 44. The "Exiscan patch" is now merged into the mainline Exim source.
3856 45. Sometimes the final signoff response after QUIT could fail to get
3857 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3858 < 0 before doing a fflush(). This bug looks as though it goes back to the
3859 introduction of TLS in release 3.20, but "sometimes" must have been rare
3860 because the tests only now provoked it.
3862 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3863 (this can affect the format of dates).
3865 47. exim_tidydb, when checking for the continued existence of a message for
3866 which it has found a message-specific retry record, was not finding
3867 messages that were in split spool directories. Consequently, it was
3868 deleting retry records that should have stayed in existence.
3870 48. Steve fixed some bugs in eximstats.
3872 49. The SPA authentication driver was not abandoning authentication and moving
3873 on to the next authenticator when an expansion was forced to fail,
3874 contradicting the general specification for all authenticators. Instead it
3875 was generating a temporary error. It now behaves as specified.
3877 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3878 (the order specifies the preference for clients). The order is now AES256,
3879 AES128, 3DES, ARCFOUR128.
3881 51. Small patch to Sieve code - explicitly set From: when generating an
3884 52. Exim crashed if a remote delivery caused a very long error message to be
3885 recorded - for instance if somebody sent an entire SpamAssassin report back
3886 as a large number of 550 error lines. This bug was coincidentally fixed by
3887 increasing the size of one of Exim's internal buffers (big_buffer) that
3888 happened as part of the Exiscan merge. However, to be on the safe side, I
3889 have made the code more robust (and fixed the comments that describe what
3892 53. Now that there can be additional text after "Completed" in log lines (if
3893 the queue_time_overall log selector is set), a one-byte patch to exigrep
3894 was needed to allow it to recognize "Completed" as not the last thing in
3897 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3898 patch that reportedly fixes this has been added. I am not expert enough to
3899 create a test for it. This is what the patch creator wrote:
3901 "I found a little strange behaviour of ldap code when working with
3902 Windows 2003 AD Domain, where users was placed in more than one
3903 Organization Units. When I tried to give exim partial DN, the exit code
3904 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
3905 But simultaneously result of request was absolutely normal ldap result,
3906 so I produce this patch..."
3908 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
3909 REFERENCE, so I have modified the code to exclude the patch when that macro
3912 55. Some experimental protocols are using DNS PTR records for new purposes. The
3913 keys for these records are domain names, not reversed IP addresses. The
3914 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3915 leaves it alone. Component reversal etc. now happens only for IP addresses.
3918 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
3920 57. Double the size of the debug message buffer (to 2048) so that more of very
3921 long debug lines gets shown.
3923 58. The exicyclog utility now does better if the number of log files to keep
3924 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
3926 59. Two changes related to the smtp_active_hostname option:
3928 (1) $smtp_active_hostname is now available as a variable.
3929 (2) The default for smtp_banner uses $smtp_active_hostname instead
3930 of $primary_hostname.
3932 60. The host_aton() function is supposed to be passed a string that is known
3933 to be a valid IP address. However, in the case of IPv6 addresses, it was
3934 not checking this. This is a hostage to fortune. Exim now panics and dies
3935 if the condition is not met. A case was found where this could be provoked
3936 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3937 components; fortuitously, this particular loophole had already been fixed
3938 by change 4.50/55 above.
3940 If there are any other similar loopholes, the new check in host_aton()
3941 itself should stop them being exploited. The report I received stated that
3942 data on the command line could provoke the exploit when Exim was running as
3943 exim, but did not say which command line option was involved. All I could
3944 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3945 running as the user.
3948 61. There was a buffer overflow vulnerability in the SPA authentication code
3949 (which came originally from the Samba project). I have added a test to the
3950 spa_base64_to_bits() function which I hope fixes it.
3953 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
3954 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
3956 63. The daemon start-up calls getloadavg() while still root for those OS that
3957 need the first call to be done as root, but it missed one case: when
3958 deliver_queue_load_max is set with deliver_drop_privilege. This is
3959 necessary for the benefit of the queue runner, because there is no re-exec
3960 when deliver_drop_privilege is set.
3962 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
3963 This has been fixed.
3965 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3966 was in use, was not putting the data itself into the right store pool;
3967 consequently, it could be overwritten for a subsequent message in the same
3968 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3971 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
3973 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
3974 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
3975 the function were treating the return as a boolean value, which happened to
3976 work because 0=false and not-0=true, but is not correct code.
3978 68. The host_aton() function was not handling scoped IPv6 addresses (those
3979 with, for example, "%eth0" on the end) correctly.
3981 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
3982 negated items (that is, ~something) in unsigned ints. Some compilers
3983 apparently mutter when there is no cast.
3985 70. If an address verification called from an ACL failed, and did not produce a
3986 user-specific message (i.e. there was only a "system" message), nothing was
3987 put in $acl_verify_message. In this situation, it now puts the system
3990 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3991 session; change 4.31/43 added the unwanted input to the log line - except
3992 that it did not do this in the start of session case. It now does.
3994 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3995 This is wrong and can cause the other end to generate a synchronization
3996 error if it is another Exim or anything else that does the synchronization
3997 check. A QUIT command is no longer sent after a timeout.
3999 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4000 during host lookups.
4002 74. The defer_ok option of callout verification was not working if it was used
4003 when verifying addresses in header lines, that is, for this case:
4005 verify = header_sender/callout=defer_ok
4007 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4008 those file descriptors could be used for SMTP connections. If anything
4009 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4010 could be sent to the SMTP client, causing chaos. The daemon now opens
4011 stdin, stdout, and stderr to /dev/null when it puts itself into the
4014 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4015 log by default. The user can override this with suitable Perl magic.
4017 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4018 the log message when discard triggers, was not working for the DATA ACL or
4019 for the non-SMTP ACL.
4021 78. Error message wording change in sieve.c.
4023 79. If smtp_accept_max_per_host was set, the number of connections could be
4024 restricted to fewer than expected, because the daemon was trying to set up
4025 a new connection before checking whether the processes handling previous
4026 connections had finished. The check for completed processes is now done
4027 earlier. On busy systems, this bug wouldn't be noticed because something
4028 else would have woken the daemon, and it would have reaped the completed
4031 80. If a message was submitted locally by a user whose login name contained one
4032 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4033 It caused a spool format error. I have fixed the spool reading code. A
4034 related problem was that the "from" clause in the Received: line became
4035 illegal because of the space(s). It is now covered by ${quote_local_part.
4037 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4040 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4042 83. Patch from Sieve maintainer to fix unterminated string problem in
4043 "vacation" handling.
4045 84. Some minor changes to the Linux configuration files to help with other
4046 OS variants using glibc.
4048 85. One more patch for Sieve to update vacation handling to latest spec.
4051 ----------------------------------------------------
4052 See the note above about the 4.44 and 4.50 releases.
4053 ----------------------------------------------------
4059 1. Change 4.43/35 introduced a bug that caused file counts to be
4060 incorrectly computed when quota_filecount was set in an appendfile
4063 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4064 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4067 3. Give more explanation in the error message when the command for a transport
4068 filter fails to execute.
4070 4. There are several places where Exim runs a non-Exim command in a
4071 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4072 was being done only for the command run by the queryprogram router. It is
4073 now done for all such subprocesses. The other cases are: ${run, transport
4074 filters, and the commands run by the lmtp and pipe transports.
4076 5. Some older OS have a limit of 256 on the maximum number of file
4077 descriptors. Exim was using setrlimit() to set 1000 as a large value
4078 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4079 systems. I've change it so that if it can't get 1000, it tries for 256.
4081 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4082 was an oversight, and furthermore, ever since the addition of extra
4083 controls (e.g. 4.43/32), the checks on when to allow different forms of
4084 "control" were broken. There should now be diagnostics for all cases when a
4085 control that does not make sense is encountered.
4087 7. $recipients is now available in the predata ACL (oversight).
4089 8. Tidy the search cache before the fork to do a delivery from a message
4090 received from the command line. Otherwise the child will trigger a lookup
4091 failure and thereby defer the delivery if it tries to use (for example) a
4092 cached ldap connection that the parent has called unbind on.
4094 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4095 of $address_data from the recipient verification was clobbered by the
4096 sender verification.
4098 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4099 was its contents. (It was OK if the option was not defined at all.)
4101 11. A "Completed" log line is now written for messages that are removed from
4102 the spool by the -Mrm option.
4104 12. $host_address is now set to the target address during the checking of
4105 ignore_target_hosts.
4107 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4108 being passed; this would have caused $sender_host_name to have been used if
4109 matching the list had actually called for a host name (not very likely,
4110 since this list is usually IP addresses). A host name is now passed as
4113 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4114 code that specifies a non-restarting handler (typically sigaction() in
4115 modern systems) in an attempt to fix a rare and obscure crash bug.
4117 15. Narrowed the window for a race in the daemon that could cause it to ignore
4118 SIGCHLD signals. This is not a major problem, because they are used only to
4119 wake it up if nothing else does.
4121 16. A malformed maildirsize file could cause Exim to calculate negative values
4122 for the mailbox size or file count. Odd effects could occur as a result.
4123 The maildirsize information is now recalculated if the size or filecount
4126 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4127 support for a long time. Removed HAVE_SYS_VFS_H.
4129 18. Updated exipick to current release from John Jetmore.
4131 19. Allow an empty sender to be matched against a lookup in an address list.
4132 Previously the only cases considered were a regular expression, or an
4135 20. Exim went into a mad DNS lookup loop when doing a callout where the
4136 host was specified on the transport, if the DNS lookup yielded more than
4139 21. The RFC2047 encoding function was originally intended for short strings
4140 such as real names; it was not keeping to the 75-character limit for
4141 encoded words that the RFC imposes. It now respects the limit, and
4142 generates multiple encoded words if necessary. To be on the safe side, I
4143 have increased the buffer size for the ${rfc2047: expansion operator from
4146 22. Failure to deliver a bounce message always caused it to be frozen, even if
4147 there was an errors_to setting on the router. The errors_to setting is now
4150 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4151 canonical form (fully expanded) before being placed in
4152 $sender_host_address.
4154 24. Updated eximstats to version 1.33
4156 25. Include certificate and key file names in error message when GnuTLS fails
4157 to set them up, because the GnuTLS error message doesn't include the name
4158 of the failing file when there is a problem reading it.
4160 26. Expand error message when OpenSSL has problems setting up cert/key files.
4163 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4164 (this can affect the format of dates).
4166 28. exim_tidydb, when checking for the continued existence of a message for
4167 which it has found a message-specific retry record, was not finding
4168 messages that were in split spool directories. Consequently, it was
4169 deleting retry records that should have stayed in existence.
4171 29. eximstats updated to version 1.35
4172 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4173 1.35 - bugfix such that pie charts by volume are generated correctly
4175 30. The SPA authentication driver was not abandoning authentication and moving
4176 on to the next authenticator when an expansion was forced to fail,
4177 contradicting the general specification for all authenticators. Instead it
4178 was generating a temporary error. It now behaves as specified.
4180 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4181 (the order specifies the preference for clients). The order is now AES256,
4182 AES128, 3DES, ARCFOUR128.
4184 31. Small patch to Sieve code - explicitly set From: when generating an
4187 32. Exim crashed if a remote delivery caused a very long error message to be
4188 recorded - for instance if somebody sent an entire SpamAssassin report back
4189 as a large number of 550 error lines. This bug was coincidentally fixed by
4190 increasing the size of one of Exim's internal buffers (big_buffer) that
4191 happened as part of the Exiscan merge. However, to be on the safe side, I
4192 have made the code more robust (and fixed the comments that describe what
4195 33. Some experimental protocols are using DNS PTR records for new purposes. The
4196 keys for these records are domain names, not reversed IP addresses. The
4197 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4198 leaves it alone. Component reversal etc. now happens only for IP addresses.
4201 34. The host_aton() function is supposed to be passed a string that is known
4202 to be a valid IP address. However, in the case of IPv6 addresses, it was
4203 not checking this. This is a hostage to fortune. Exim now panics and dies
4204 if the condition is not met. A case was found where this could be provoked
4205 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4206 components; fortuitously, this particular loophole had already been fixed
4207 by change 4.50/55 or 4.44/33 above.
4209 If there are any other similar loopholes, the new check in host_aton()
4210 itself should stop them being exploited. The report I received stated that
4211 data on the command line could provoke the exploit when Exim was running as
4212 exim, but did not say which command line option was involved. All I could
4213 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4214 running as the user.
4217 35. There was a buffer overflow vulnerability in the SPA authentication code
4218 (which came originally from the Samba project). I have added a test to the
4219 spa_base64_to_bits() function which I hope fixes it.
4222 36. The daemon start-up calls getloadavg() while still root for those OS that
4223 need the first call to be done as root, but it missed one case: when
4224 deliver_queue_load_max is set with deliver_drop_privilege. This is
4225 necessary for the benefit of the queue runner, because there is no re-exec
4226 when deliver_drop_privilege is set.
4228 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4229 was in use, was not putting the data itself into the right store pool;
4230 consequently, it could be overwritten for a subsequent message in the same
4231 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4234 38. Sometimes the final signoff response after QUIT could fail to get
4235 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4236 < 0 before doing a fflush(). This bug looks as though it goes back to the
4237 introduction of TLS in release 3.20, but "sometimes" must have been rare
4238 because the tests only now provoked it.
4244 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4245 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4246 Now it returns an integer. A number of calls were still expecting a T/F
4247 return. Fortuitously, in all cases, the tests worked in OK situations,
4248 which is the norm. However, things would have gone wrong on any write
4249 failures on the smtp file descriptor. This function is used when sending
4250 messages over SMTP and also when doing verify callouts.
4252 2. When Exim is called to do synchronous delivery of a locally submitted
4253 message (the -odf or -odi options), it no longer closes stderr before doing
4256 3. Implemented the mua_wrapper option.
4258 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4260 5. Implemented the functions header_remove(), header_testname(),
4261 header_add_at_position(), and receive_remove_recipient(), and exported them
4264 6. If an ACL "warn" statement specified the addition of headers, Exim already
4265 inserted X-ACL-Warn: at the start if there was no header name. However, it
4266 was not making this test for the second and subsequent header lines if
4267 there were newlines in the string. This meant that an invalid header could
4268 be inserted if Exim was badly configured.
4270 7. Allow an ACL "warn" statement to add header lines at the start or after all
4271 the Received: headers, as well as at the end.
4273 8. Added the rcpt_4xx retry error code.
4275 9. Added postmaster_mailfrom=xxx to callout verification option.
4277 10. Added mailfrom=xxxx to the callout verification option, for verify=
4280 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4281 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4283 12. Inserted some casts to stop certain compilers warning when using pointer
4284 differences as field lengths or precisions in printf-type calls (mostly
4285 affecting debugging statements).
4287 13. Added optional readline() support for -be (dynamically loaded).
4289 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4290 same clock tick as a message's arrival, so that its received time was the
4291 same as the "first fail" time on the retry record, and that message
4292 remained on the queue past the ultimate address timeout, every queue runner
4293 would try a delivery (because it was past the ultimate address timeout) but
4294 after another failure, the ultimate address timeout, which should have then
4295 bounced the address, did not kick in. This was a "< instead of <=" error;
4296 in most cases the first failure would have been in the next clock tick
4297 after the received time, and all would be well.
4299 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4300 being recognized when the domain list was tested by the match_domain
4301 condition in an expansion string.
4303 16. Added the ${str2b64: operator.
4305 17. Exim was always calling setrlimit() to set a large limit for the number of
4306 processes, without checking whether the existing limit was already
4307 adequate. (It did check for the limit on file descriptors.) Furthermore,
4308 errors from getrlimit() and setrlimit() were being ignored. Now they are
4309 logged to the main and panic logs, but Exim does carry on, to try to do its
4310 job under whatever limits there are.
4312 18. Imported PCRE 5.0.
4314 19. Trivial typo in log message " temporarily refused connection" (the leading
4317 20. If the log selector return_path_on_delivery was set and an address was
4318 redirected to /dev/null, the delivery process crashed because it assumed
4319 that a return path would always be set for a "successful" delivery. In this
4320 case, the whole delivery is bypassed as an optimization, and therefore no
4323 21. Internal re-arrangement: the function for sending a challenge and reading
4324 a response while authentication was assuming a zero-terminated challenge
4325 string. It's now changed to take a pointer and a length, to allow for
4326 binary data in such strings.
4328 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4330 23. Exim was not respecting finduser_retries when seeking the login of the
4331 uid under which it was called; it was always trying 10 times. (The default
4332 setting of finduser_retries is zero.) Also, it was sleeping after the final
4333 failure, which is pointless.
4335 24. Implemented tls_on_connect_ports.
4337 25. Implemented acl_smtp_predata.
4339 26. If the domain in control=submission is set empty, Exim assumes that the
4340 authenticated id is a complete email address when it generates From: or
4341 Sender: header lines.
4343 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4344 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4345 chown and chgrp in /bin and hostname in /usr/bin.
4347 28. Exim was keeping the "process log" file open after each use, just as it
4348 does for the main log. This opens the possibility of it remaining open for
4349 long periods when the USR1 signal hits a daemon. Occasional processlog
4350 errors were reported, that could have been caused by this. Anyway, it seems
4351 much more sensible not to leave this file open at all, so that is what now
4354 29. The long-running daemon process does not normally write to the log once it
4355 has entered its main loop, and it closes the log before doing so. This is
4356 so that log files can straightforwardly be renamed and moved. However,
4357 there are a couple of unusual error situations where the daemon does write
4358 log entries, and I had neglected to close the log afterwards.
4360 30. The text of an SMTP error response that was received during a remote
4361 delivery was being truncated at 512 bytes. This is too short for some of
4362 the long messages that one sometimes sees. I've increased the limit to
4365 31. It is now possible to make retry rules that apply only when a message has a
4366 specific sender, in particular, an empty sender.
4368 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4369 it possible to be selective about when SMTP synchronization is enforced.
4371 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4373 32. Implemented hosts_connection_nolog.
4375 33. Added an ACL for QUIT.
4377 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4380 35. Added mailbox_size and mailbox_filecount to appendfile.
4382 36. Added control = no_multiline_responses to ACLs.
4384 37. There was a bug in the logic of the code that waits for the clock to tick
4385 in the case where the clock went backwards by a substantial amount such
4386 that the microsecond fraction of "now" was more than the microsecond
4387 fraction of "then" (but the whole seconds number was less).
4389 38. Added support for the libradius Radius client library this is found on
4390 FreeBSD (previously only the radiusclient library was supported).
4396 1. When certain lookups returned multiple values in the form name=value, the
4397 quoting of the values was not always being done properly. Specifically:
4398 (a) If the value started with a double quote, but contained no whitespace,
4400 (b) If the value contained whitespace other than a space character (i.e.
4401 tabs or newlines or carriage returns) it was not quoted.
4402 This fix has been applied to the mysql and pgsql lookups by writing a
4403 separate quoting function and calling it from the lookup code. The fix
4404 should probably also be applied to nisplus, ibase and oracle lookups, but
4405 since I cannot test any of those, I have not disturbed their existing code.
4407 2. A hit in the callout cache for a specific address caused a log line with no
4408 reason for rejecting RCPT. Now it says "Previous (cached) callout
4409 verification failure".
4411 3. There was an off-by-one bug in the queryprogram router. An over-long
4412 return line was truncated at 256 instead of 255 characters, thereby
4413 overflowing its buffer with the terminating zero. As well as fixing this, I
4414 have increased the buffer size to 1024 (and made a note to document this).
4416 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4417 when Exim is waiting for an SMTP response from a remote server, Exim
4418 restarts its select() call on the socket, thereby resetting its timeout.
4419 This is not a problem when such interrupts are rare. Somebody set up a cron
4420 job to run exiwhat every 2 minutes, which is less than the normal select()
4421 timeout (5 or 10 minutes). This meant that the select() timeout never
4422 kicked in because it was always reset. I have fixed this by comparing the
4423 time when an interrupt arrives with the time at the start of the first call
4424 to select(). If more time than the timeout has elapsed, the interrupt is
4425 treated as a timeout.
4427 5. Some internal re-factoring in preparation for the addition of Sieve
4428 extensions (by MH). In particular, the "personal" test is moved to a
4429 separate function, and given an option for scanning Cc: and Bcc: (which is
4430 not set for Exim filters).
4432 6. When Exim created an email address using the login of the caller as the
4433 local part (e.g. when creating a From: or Sender: header line), it was not
4434 quoting the local part when it contained special characters such as @.
4436 7. Installed new OpenBSD configuration files.
4438 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4439 try to make them clearer.
4441 9. Callout options, other than the timeout value, were being ignored when
4442 verifying sender addresses in header lines. For example, when using
4444 verify = header_sender/callout=no_cache
4446 the cache was (incorrectly) being used.
4448 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4449 only the Cygwin environment.
4451 11. When return_path_on_delivery was set as a log selector, if different remote
4452 addresses in the same message used different return paths and parallel
4453 remote delivery occurred, the wrong values would sometimes be logged.
4454 (Whenever a remote delivery process finished, the return path value from
4455 the most recently started remote delivery process was logged.)
4457 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4458 lines when AUTH and/or TLS are in use. This is the "received protocol"
4459 field. Exim used to use "asmtp" for authenticated SMTP, without any
4460 indication (in the protocol name) for TLS use. Now it follows the RFC and
4461 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4462 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4463 names appear in log lines as well as in Received: header lines.
4465 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4466 extensions, and comparison tests, and to fix some bugs.
4468 14. Changes to the "personal" filter test:
4470 (1) The test was buggy in that it was just doing the equivalent of
4471 "contains" tests on header lines. For example, if a user's address was
4472 anne@some.where, the "personal" test would incorrectly be true for
4474 To: susanne@some.where
4476 This test is now done by extracting each address from the header in turn,
4477 and checking the entire address. Other tests that are part of "personal"
4478 are now done using regular expressions (for example, to check local parts
4479 of addresses in From: header lines).
4481 (2) The list of non-personal local parts in From: addresses has been
4482 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4483 taken from the Sieve specification recommendations.
4485 (3) If the message contains any header line starting with "List-" it is
4486 treated as non-personal.
4488 (4) The test for "circular" in the Subject: header line has been removed
4489 because it now seems ill-conceived.
4491 15. Minor typos in src/EDITME comments corrected.
4493 16. Installed latest exipick from John Jetmore.
4495 17. If headers_add on a router specified a text string that was too long for
4496 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4497 of string_sprintf() is now avoided.
4499 18. $message_body_size was not set (it was always zero) when running the DATA
4500 ACL and the local_scan() function.
4502 19. For the "mail" command in an Exim filter, no default was being set for
4503 the once_repeat time, causing a random time value to be used if "once" was
4504 specified. (If the value happened to be <= 0, no repeat happened.) The
4505 default is now 0s, meaning "never repeat". The "vacation" command was OK
4506 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4507 (I found it when inspecting the code).
4509 20. There is now an overall timeout for performing a callout verification. It
4510 defaults to 4 times the callout timeout, which applies to individual SMTP
4511 commands during the callout. The overall timeout applies when there is more
4512 than one host that can be tried. The timeout is checked before trying the
4513 next host. This prevents very long delays if there are a large number of
4514 hosts and all are timing out (e.g. when the network connections are timing
4515 out). The value of the overall timeout can be changed by specifying an
4516 additional sub-option for "callout", called "maxwait". For example:
4518 verify = sender/callout=5s,maxwait=20s
4520 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4521 to the end before writing, but this should make it even safer).
4523 22. Exim was forgetting that it had advertised PIPELINING for the second and
4524 subsequent messages on an SMTP connection. It was also not resetting its
4525 memory on STARTTLS and an internal HELO.
4527 23. When Exim logs an SMTP synchronization error within a session, it now
4528 records whether PIPELINING has been advertised or not.
4530 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4531 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4532 rather than long int.
4534 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4536 26. Added the never_mail option to autoreply.
4542 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4543 crash if the getsockname() call failed; this can happen if a connection is
4544 closed very soon after it is established. The problem was simply in the
4545 order in which certain operations were done, causing Exim to try to write
4546 to the SMTP stream before it had set up the file descriptor. The bug has
4547 been fixed by making things happen in the correct order.
4553 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4554 before the connection was closed, thus losing the rejection response.
4556 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4557 some early Solaris releases, but causes trouble in current releases where
4558 socklen_t is defined.
4560 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4563 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4566 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4567 that releasing the top of it at the end releases what was used for sub-
4568 expansions (unless the block got too big). However, discard this block if
4569 the first thing is a variable or header, so that we can use its block when
4570 it is dynamic (useful for very large $message_headers, for example).
4572 6. Lookups now cache *every* query, not just the most recent. A new, separate
4573 store pool is used for this. It can be recovered when all lookup caches are
4574 flushed. Lookups now release memory at the end of their result strings.
4575 This has involved some general refactoring of the lookup sources.
4577 7. Some code has been added to the store_xxx() functions to reduce the amount
4578 of flapping under certain conditions.
4580 8. log_incoming_interface used to affect only the <= reception log lines. Now
4581 it causes the local interface and port to be added to several more SMTP log
4582 lines, for example "SMTP connection from", and rejection lines.
4584 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4586 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4588 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4589 could be overwritten at the end of the current message (or the start of a
4590 new message if it was set in a HELO ACL). The value is now preserved for
4591 the duration of the SMTP connection.
4593 12. If a transport had a headers_rewrite setting, and a matching header line
4594 contained an unqualified address, that address was qualified, even if it
4595 did not match any rewriting rules. The underlying bug was that the values
4596 of the flags that permit the existence of unqualified sender and recipient
4597 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4598 non-local messages, and by -bnq for local messages) were not being
4599 preserved with the message after it was received.
4601 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4602 "next input=" as part of the text comprising the host identity instead of
4603 the correct text. The code was using the same buffer for two different
4604 strings. However, depending on which order the printing function evaluated
4605 its arguments, the bug did not always show up. Under Linux, for example, my
4606 test suite worked just fine.
4608 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4609 doesn't work with some older versions of Perl. It has been changed to "my",
4610 which in any case is probably the better facility to use.
4612 15. A really picky compiler found some instances of statements for creating
4613 error messages that either had too many or two few arguments for the format
4616 16. The size of the buffer for calls to the DNS resolver has been increased
4617 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4618 for addresses that have a lot of PTR records. This alleviates a problem; it
4619 does not fully solve it.
4621 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4622 buffer now truncates the list and logs the incident, which is the same
4623 action as happens when Exim is looking up a host name and its aliases.
4624 Previously in this situation something unpredictable would happen;
4625 sometimes it was "internal error: store_reset failed".
4627 18. If a server dropped the connection unexpectedly when an Exim client was
4628 using GnuTLS and trying to read a response, the client delivery process
4629 crashed while trying to generate an error log message.
4631 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4632 string, for example:
4634 warn message = H1: something\nH2: something
4636 the text was added as a single header line from Exim's point of view
4637 though it ended up OK in the delivered message. However, searching for the
4638 second and subsequent header lines using $h_h2: did not work. This has been
4639 fixed. Similarly, if a system filter added multiple headers in this way,
4640 the routers could not see them.
4642 20. Expanded the error message when iplsearch is called with an invalid key to
4643 suggest using net-iplsearch in a host list.
4645 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4646 ACLs are no longer actually imposed (and a message to that effect is
4649 22. If a "gecos" field in a passwd entry contained escaped characters, in
4650 particular, if it contained a \" sequence, Exim got it wrong when building
4651 a From: or a Sender: header from that name. A second bug also caused
4652 incorrect handling when an unquoted " was present following a character
4653 that needed quoting.
4655 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4656 was not being matched caselessly.
4658 24. Arranged for all hyphens in the exim.8 source to be escaped with
4661 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4662 properly. Recipient callout cache records were still being keyed to include
4663 the sender, even when use_sender was set false. This led to far more
4664 callouts that were necessary. The sender is no longer included in the key
4665 when use_sender is false.
4667 26. Added "control = submission" modifier to ACLs.
4669 27. Added the ${base62d: operator to decode base 62 numbers.
4671 28. dnsdb lookups can now access SRV records.
4673 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4674 the configuration file.
4676 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4677 (-v) mode. This makes the output for a verbose queue run more intelligible.
4679 31. Added a use_postmaster feature to recipient callouts.
4681 32. Added the $body_zerocount variable, containing the number of binary zero
4682 bytes in the message body.
4684 33. The time of last modification of the "new" subdirectory is now used as the
4685 "mailbox time last read" when there is a quota error for a maildir
4688 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4690 35. Added +ignore_unknown as a special item in host lists.
4692 36. Code for decoding IPv6 addresses in host lists is now included, even if
4693 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4694 address was recognized as an IP address, but was then not correctly decoded
4695 into binary, causing unexpected and incorrect effects when compared with
4702 1. Very minor rewording of debugging text in manualroute to say "list of
4703 hosts" instead of "hostlist".
4705 2. If verify=header_syntax was set, and a header line with an unqualified
4706 address (no domain) and a large number of spaces between the end of the
4707 name and the colon was received, the reception process suffered a buffer
4708 overflow, and (when I tested it) crashed. This was caused by some obsolete
4709 code that should have been removed. The fix is to remove it!
4711 3. When running in the test harness, delay a bit after writing a bounce
4712 message to get a bit more predictability in the log output.
4714 4. Added a call to search_tidyup() just before forking a reception process. In
4715 theory, someone could use a lookup in the expansion of smtp_accept_max_
4716 per_host which, without the tidyup, could leave open a database connection.
4718 5. Added the variables $recipient_data and $sender_data which get set from a
4719 lookup success in an ACL "recipients" or "senders" condition, or a router
4720 "senders" option, similar to $domain_data and $local_part_data.
4722 6. Moved the writing of debug_print from before to after the "senders" test
4725 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4726 problems for message scanning, either using a data ACL, or using
4727 local_scan() because the Received: header was not generated till after they
4728 were called (in order to set the time as the time of reception completion).
4729 I have revised the way this works. The header is now generated after the
4730 body is received, but before the ACL or local_scan() are called. After they
4731 are run, the timestamp in the header is updated.
4737 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4738 before starting a queue runner without re-exec. This happened only when
4739 deliver_drop_privilege was set or when the Exim user was set to root. The
4740 effect of the bug was that timeouts during subsequent deliveries caused
4741 crashes instead of being properly handled. The handler is now left at its
4742 default (and expected) setting.
4744 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4745 message, again when deliver_drop_privilege is set or Exim is run as root.
4746 The bug described in (1) was not present in this case, but the tidying up
4747 of the other signals was missing. I have made the two cases consistent.
4749 3. The ignore_target_hosts setting on a manualroute router was being ignored
4750 for hosts that were looked up using the /MX notation.
4752 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4755 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4756 operated on the sender address. After changing the $sender_address to <>
4757 for the sender address verify, Exim was re-instated it as the original
4758 (before rewriting) address, but remembering that it had rewritten it, so it
4759 wasn't rewriting it again. This bug also had the effect of breaking the
4760 sender address verification caching when the sender address was rewritten.
4762 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4763 This has been changed so that if the ip literal address matches
4764 ignore_target_hosts, the router declines.
4766 7. Added expansion conditions match_domain, match_address, and match_local_
4767 part (NOT match_host).
4769 8. The placeholder for the Received: header didn't have a length field set.
4771 9. Added code to Exim itself and to exim_lock to test for a specific race
4772 condition that could lead to file corruption when using MBX delivery. The
4773 issue is with the lockfile that is created in /tmp. If this file is removed
4774 after a process has opened it but before that process has acquired a lock,
4775 there is the potential for a second process to recreate the file and also
4776 acquire a lock. This could lead to two Exim processes writing to the file
4777 at the same time. The added code performs the same test as UW imapd; it
4778 checks after acquiring the lock that its file descriptor still refers to
4779 the same named file.
4781 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4782 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4785 11. Added the smtp_active_hostname option. If used, this will typically be made
4786 to depend on the incoming interface address. Because $interface_address is
4787 not set up until the daemon has forked a reception process, error responses
4788 that can happen earlier (such as "too many connections") no longer contain
4791 12. If an expansion in a condition on a "warn" statement fails because a lookup
4792 defers, the "warn" statement is abandoned, and the next ACL statement is
4793 processed. Previously this caused the whole ACL to be aborted.
4795 13. Added the iplsearch lookup type.
4797 14. Added ident_timeout as a log selector.
4799 15. Added tls_certificate_verified as a log selector.
4801 16. Added a global option tls_require_ciphers (compare the smtp transport
4802 option of the same name). This controls incoming TLS connections.
4804 17. I finally figured out how to make tls_require_ciphers do a similar thing
4805 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4806 before starting the TLS session.
4808 18. Tabs are now shown as \t in -bP output.
4810 19. If the log selector return_path_on_delivery was set, Exim crashed when
4811 bouncing a message because it had too many Received: header lines.
4813 20. If two routers both had headers_remove settings, and the first one included
4814 a superfluous trailing colon, the final name in the first list and the
4815 first name in the second list were incorrectly joined into one item (with a
4816 colon in the middle).
4822 1. Added -C and -D options to the exinext utility, mainly to make it easier
4823 to include in the automated testing, but these could be helpful when
4824 multiple configurations are in use.
4826 2. The exinext utility was not formatting the output nicely when there was
4827 an alternate port involved in the retry record key, nor when there was a
4828 message id as well (for retries that were specific to a specific message
4829 and a specific host). It was also confused by IPv6 addresses, because of
4830 the additional colons they contain. I have fixed the IPv4 problem, and
4831 patched it up to do a reasonable job for IPv6.
4833 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4834 delivery, the log line now contains "pipelined" if PIPELINING was used.
4836 4. An SMTP transport process used to panic and die if the bind() call to set
4837 an explicit outgoing interface failed. This has been changed; it is now
4838 treated in the same way as a connect() failure.
4840 5. A reference to $sender_host_name in the part of a conditional expansion
4841 that was being skipped was still causing a DNS lookup. This no longer
4844 6. The def: expansion condition was not recognizing references to header lines
4845 that used bh_ and bheader_.
4847 7. Added the _cache feature to named lists.
4849 8. The code for checking quota_filecount in the appendfile transport was
4850 allowing one more file than it should have been.
4852 9. For compatibility with Sendmail, the command line option
4860 and sets the incoming protocol and host name (for trusted callers). The
4861 host name and its colon can be omitted when only the protocol is to be set.
4862 Note the Exim already has two private options, -pd and -ps, that refer to
4863 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4864 "s", but I don't think that's a major issue.
4866 10. A number of refactoring changes to the code, none of which should affect
4869 (a) The number of logging options was getting close to filling up the
4870 32-bit word that was used as a bit map. I have split them into two classes:
4871 those that are passed in the argument to log_write(), and those that are
4872 only ever tested independently outside of that function. These are now in
4873 separate 32-bit words, so there is plenty of room for expansion again.
4874 There is no change in the user interface or the logging behaviour.
4876 (b) When building, for example, log lines, the code previously used a
4877 macro that called string_cat() twice, in order to add two strings. This is
4878 not really sufficiently general. Furthermore, there was one instance where
4879 it was actually wrong because one of the argument was used twice, and in
4880 one call a function was used. (As it happened, calling the function twice
4881 did not affect the overall behaviour.) The macro has been replaced by a
4882 function that can join an arbitrary number of extra strings onto a growing
4885 (c) The code for expansion conditions now uses a table and a binary chop
4886 instead of a serial search (which was left over from when there were very
4887 few conditions). Also, it now recognizes conditions like "pam" even when
4888 the relevant support is not compiled in: a suitably worded error message is
4889 given if an attempt is made to use such a condition.
4891 11. Added ${time_interval:xxxxx}.
4893 12. A bug was causing one of the ddress fields not to be passed back correctly
4894 from remote delivery subprocesses. The field in question was not being
4895 subsequently used, so this caused to problems in practice.
4897 13. Added new log selectors queue_time and deliver_time.
4899 14. Might have fixed a bug in maildirsizefile handling that threw up
4900 "unexpected character" debug warnings, and recalculated the data
4901 unnecessarily. In any case, I expanded the warning message to give more
4904 15. Added the message "Restricted characters in address" to the statements in
4905 the default ACL that block characters like @ and % in local parts.
4907 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
4908 Three changes have been made:
4910 (a) There was a serious bug; a negative response to MAIL caused the whole
4911 recipient domain to be cached as invalid, thereby blocking all messages
4912 to all local parts at the same domain, from all senders. This bug has
4913 been fixed. The domain is no longer cached after a negative response to
4914 MAIL if the sender used is not empty.
4916 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
4919 (c) A new callout option, "use_sender" has been added for people who want
4920 the modified behaviour.
4926 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
4929 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
4930 indeed breaks things for older releases.
4932 3. Added additional logging to the case where there is a problem reading data
4933 from a filter that is running in a subprocess using a pipe, in order to
4934 try to track down a specific problem.
4936 4. Testing facility fudge: when running in the test harness and attempting
4937 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
4938 getting "No route to host". Convert this to a timeout.
4940 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
4943 6. Some OS don't have socklen_t but use size_t instead. This affects the
4944 fifth argument of getsockopt() amongst other things. This is now
4945 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
4946 can be set for individual OS. I have set it for SunOS5, OSF1, and
4947 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
4948 some earlier ones do not.
4950 7. Change 4.30/15 was not doing the test caselessly.
4952 8. The standard form for an IPv6 address literal was being rejected by address
4953 parsing in, for example, MAIL and RCPT commands. An example of this kind of
4954 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4955 this, as well as the form without the "IPv6" on the front (but only when
4956 address literals are enabled, of course).
4958 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
4960 10. Exim crashed if a message with an empty sender address specified by -f
4961 encountered a router with an errors_to setting. This could be provoked only
4962 by a command such as
4966 where an empty string was supplied; "<>" did not hit this bug.
4968 11. Installed PCRE release 4.5.
4970 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
4971 remained set. It is now erased.
4973 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
4974 times from message ids (which are base 36 rather than the normal 62).
4976 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
4977 were being counted as actual protocol errors, and logged if the log
4978 selector +smtp_protocol_error was set. One cannot be perfect in this test,
4979 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
4980 and DATA following a set of rejected RCPTs do not count as protocol errors.
4981 In other words, Exim assumes they were pipelined, though this may not
4982 actually be the case. Of course, in all cases the client gets an
4983 appropriate error code.
4985 15. If a lookup fails in an ACL condition, a message about the failure may
4986 be available; it is used if testing the ACL cannot continue, because most
4987 such messages specify what the cause of the deferral is. However, some
4988 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
4989 that caused an old message to be retained and used if a later statement
4990 caused a defer, replacing the real cause of the deferral.
4992 16. If an IP address had so many PTR records that the DNS lookup buffer
4993 was not large enough to hold them, Exim could crash while trying to process
4994 the truncated data. It now detects and logs this case.
4996 17. Further to 4.21/58, another change has been made: if (and only if) the
4997 first line of a message (the first header line) ends with CRLF, a bare LF
4998 in a subsequent header line has a space inserted after it, so as not to
4999 terminate the header.
5001 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5002 unnecessarily, used atoi() instead of strtol(), and didn't check the
5003 termination when getting file sizes from file names by regex.
5005 19. Completely re-implemented the support for maildirsize files, in the light
5006 of a number of problems with the previous contributed implementation
5007 (4.30/29). In particular:
5009 . If the quota is zero, the maildirsize file is maintained, but no quota is
5012 . If the maildir directory does not exist, it is created before any attempt
5013 to write a maildirsize file.
5015 . The quota value in the file is just a cache; if the quota is changed in
5016 the transport, the new value overrides.
5018 . A regular expression is available for excluding directories from the
5021 20. The autoreply transport checks the characters in options that define the
5022 message's headers; it allows continued headers, but it was checking with
5023 isspace() after an embedded newline instead of explicitly looking for a
5026 21. If all the "regular" hosts to which an address was routed had passed their
5027 expiry times, and had not reached their retry times, the address was
5028 bounced, even if fallback hosts were defined. Now Exim should go on to try
5031 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5032 equivalent code in the SMTP transport. Some hosts send humungous responses
5033 to HELO/EHLO, more than 1024 it seems.
5035 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5036 gives compiler warnings in some environments. I've now done it "properly",
5039 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5040 (because of problems with the built-in one) was declared to return uschar *
5041 instead of char *, causing compiler failure.
5043 25. Fixed a file descriptor leak when processing alias/forward files.
5045 26. Fixed a minor format string issue in dbfn.c.
5047 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5049 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5050 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5053 29. When a sender address is verified, it is cached, to save repeating the test
5054 when there is more than one recipient in a message. However, when the
5055 verification involves a callout, it is possible for different callout
5056 options to be set for different recipients. It is too complicated to keep
5057 track of this in the cache, so now Exim always runs a verification when a
5058 callout is required, relying on the callout cache for the optimization.
5059 The overhead is duplication of the address routing, but this should not be
5062 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5063 to be verified with callout=postmaster, and the main callout worked but the
5064 postmaster check failed, the verification correctly failed. However, if a
5065 subsequent RCPT command asked for sender verification *without* the
5066 postmaster check, incorrect caching caused this verification also to fail,
5069 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5070 it was not caching the DNS options (qualify_single, search_parents) that
5071 were used when the lookup failed. A subsequent lookup with different
5072 options therefore always gave the same answer, though there were cases
5073 where it should not have. (Example: a "domains = !$mx_any" option on a
5074 dnslookup router: the "domains" option is always processed without any
5075 widening, but the router might have qualify_single set.) Now Exim uses the
5076 cached value only when the same options are set.
5078 32. Added John Jetmore's "exipick" utility to the distribution.
5080 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5081 than a timeout (e.g. a certificate is required, and is not provided), an
5082 Exim server now closes the connection immediately. Previously it waited for
5083 the client to close - but if the client is SSL, it seems that they each
5084 wait for each other, leading to a delay before one of them times out.
5086 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5087 maintained 0.8.x compatibility because I don't think many are using it, and
5088 it is clearly obsolete.
5090 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5093 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5094 client certificate was expired. A simple patch fixes this, though I don't
5095 understand the full logic of why the verify callback is called multiple
5098 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5099 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5100 which causes problems with some clients (such as the Certicom SSL Plus
5101 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5102 disables the coutermeasure allowing Eudora to connect."
5104 38. Exim was not checking that a write() to a log file succeeded. This could
5105 lead to Bad Things if a log got too big, in particular if it hit a file
5106 size limit. Exim now panics and dies if it cannot write to a log file, just
5107 as it does if it cannot open a log file.
5109 39. Modified OS/Makefile-Linux so that it now contains
5111 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5113 The two -D definitions ensure that Exim is compiled with large file
5114 support, which makes it possible to handle log files that are bigger than
5117 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5118 instance) a domain was checked against a named list that involved a lookup,
5119 causing $domain_data to be set, then another domain was checked against the
5120 same list, then the first domain was re-checked, the value of $domain_data
5121 after the final check could be wrong. In particular, if the second check
5122 failed, it could be set empty. This bug probably also applied to
5125 41. The strip_trailing_dot option was not being applied to the address given
5126 with the -f command-line option.
5128 42. The code for reading a message's header from the spool was incrementing
5129 $received_count, but never initializing it. This meant that the value was
5130 incorrect (doubled) while delivering a message in the same process in which
5131 it was received. In the most common configuration of Exim, this never
5132 happens - a fresh exec is done - but it can happen when
5133 deliver_drop_privilege is set.
5135 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5136 it now includes up to 150 characters of the unexpected data in the log
5139 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5140 and building data strings. The size of both of these buffers was 10 000
5141 bytes - far larger than anybody would *ever* want, thought I. Needless to
5142 say, somebody hit the limit. I have increased the maximum line length to
5143 20 000 and the maximum data length of concatenated lines to 100 000. I have
5144 also fixed two bugs, because there was no checking on these buffers. Tsk,
5145 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5146 buffer is too small.
5148 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5149 lsearch lookups. Now it does.
5151 46. When parsing a route_list item in a manualroute router, a fixed-length
5152 buffer was used for the list of hosts. I made this 1024 bytes long,
5153 thinking that nobody would ever have a list of hosts that long. Wrong.
5154 Somebody had a whole pile of complicated expansion conditions, and the
5155 string was silently truncated, leading to an expansion error. It turns out
5156 that it is easier to change to an unlimited length (owing to other changes
5157 that have happened since this code was originally written) than to build
5158 structure for giving a limitation error. The length of the item that
5159 expands into the list of hosts is now unlimited.
5161 47. The lsearch lookup could not handle data where the length of text line was
5162 more than 4095 characters. Such lines were truncated, leading to shortened
5163 data being returned. It should now handle lines of any length.
5165 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5166 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5169 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5170 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5171 no longer happens in comment lines. A list of replacements is now placed
5172 at the head of all of the source files, except those whose only change is
5173 to replace PERL_COMMAND in the very first #! line.
5175 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5176 messages on the queue, with a bottom-up merge sort, using code contributed
5177 by Michael Haardt. This should make operations like -bp somewhat faster on
5178 large queues. It won't affect queue runners, except when queue_run_in_order
5181 51. Installed eximstats 1.31 in the distribution.
5183 52. Added support for SRV lookups to the dnslookup router.
5185 53. If an ACL referred to $message_body or $message_body_end, the value was not
5186 reset for any messages that followed in the same SMTP session.
5188 54. The store-handling optimization for building very long strings was not
5189 differentiating between the different store pools. I don't think this
5190 actually made any difference in practice, but I've tidied it.
5192 55. While running the routers to verify a sender address, $sender_address
5193 was still set to the sender address. This is wrong, because when routing to
5194 send a bounce to the sender, it would be empty. Therefore, I have changed
5195 it so that, while verifying a sender address, $sender_address is set to <>.
5196 (There is no change to what happens when verifying a recipient address.)
5198 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5199 target A or AAAA records (if not already returned) without resetting the
5200 qualify_single or search_parents options of the DNS resolver. These are
5201 inappropriate in this case because the targets of MX and SRV records must
5202 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5203 target that, when qualified, matched something in the local domain. These
5204 two options are now turned off when doing these lookups.
5206 57. It seems that at least some releases of Reiserfs (which does not have the
5207 concept of a fixed number of inodes) returns zero and not -1 for the
5208 number of available inodes. This interacted badly with check_spool_inodes,
5209 which assumed that -1 was the "no such thing" setting. What I have done is
5210 to check that the total number of inodes is greater than zero before doing
5211 the test of how many are available.
5213 58. When a "warn" ACL statement has a log_message modifier, the message is
5214 remembered, and not repeated. This is to avoid a lot of repetition when a
5215 message has many recipients that cause the same warning to be written.
5216 Howewer, Exim was preserving the list of already written lines for an
5217 entire SMTP session, which doesn't seem right. The memory is now reset if a
5218 new message is started.
5220 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5221 debugging output unless log_rewrite was also set.
5223 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5224 of (int)(handle) when we know that handle contains (void *)(-1).
5226 61. The Exim daemon panic-logs an error return when it closes the incoming
5227 connection. However "connection reset by peer" seems to be common, and
5228 isn't really an error worthy of noting specially, so that particular error
5231 62. When Exim is trying to find all the local interfaces, it used to panic and
5232 die if the ioctl to get the interface flags failed. However, it seems that
5233 on at least one OS (Solaris 9) it is possible to have an interface that is
5234 included in the list of interfaces, but for which you get a failure error
5235 for this call. This happens when the interface is not "plumbed" into a
5236 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5237 failure of the "get flags" call assumes that the interface is down.
5239 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5240 makes life easier for people who are doing arithmetic on fields extracted
5241 from dates, where you often get leading zeros that should not be
5242 interpreted as octal.
5244 64. Added qualify_domain to the redirect router, to override the global
5247 65. If a pathologically long header line contained very many addresses (the
5248 report of this problem mentioned 10 000) and each of them was rewritten,
5249 Exim could use up a very large amount of memory. (It kept on making new
5250 copies of the header line as it rewrote, and never released the old ones.)
5251 At the expense of a bit more processing, the header rewriting function has
5252 been changed so that it no longer eats memory in this way.
5254 66. The generation of the Received: header has been moved from the time that a
5255 message starts to be received, to the time that it finishes. The timestamp
5256 in the Received: header should now be very close to that of the <= log
5257 line. There are two side-effects of this change:
5259 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5260 logged header lines no longer include the local Received: line, because
5261 it has not yet been created. The same applies to a copy of the message
5262 that is returned to a non-SMTP sender when a message is rejected.
5264 (b) When a filter file is tested using -bf, no additional Received: header
5265 is added to the test message. After some thought, I decided that this
5268 This change does not affect the value of $received_for. It is still set
5269 after address rewriting, but before local_scan() is called.
5271 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5273 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5274 gave an unhelpful panic error message, and a defer error. I have managed to
5275 change this behaviour so that it now rejects any supplied certificate,
5276 which seems right, as the list of acceptable certificates is empty.
5278 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5279 gave an unhelpful defer error. I have not managed to make this reject any
5280 supplied certificates, but the error message it gives is "no certificate
5281 supplied", which is not helpful.
5283 70. exigrep's output now also includes lines that are not associated with any
5284 message, but which match the given pattern. Implemented by a patch from
5285 Martin Sluka, which also tidied up the Perl a bit.
5287 71. Recipient callout verification, like sender verification, was using <> in
5288 the MAIL FROM command. This isn't really the right thing, since the actual
5289 sender may affect whether the remote host accepts the recipient or not. I
5290 have changed it to use the actual sender in the callout; this means that
5291 the cache record is now keyed on a recipient/sender pair, not just the
5292 recipient address. There doesn't seem to be a real danger of callout loops,
5293 since a callout by the remote host to check the sender would use <>.
5294 [SEE ABOVE: changed after hitting problems.]
5296 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5297 temporary errors. However, in the case of such a code being given after
5298 the end of a data transmission (i.e. after ".") Exim was failing to write
5299 a retry record for the message. (Yes, there was some broken host that was
5300 actually sending 8xx at this point.)
5302 73. An unknown lookup type in a host list could cause Exim to panic-die when
5303 the list was checked. (An example that provoked this was putting <; in the
5304 middle of a list instead of at the start.) If this happened during a DATA
5305 ACL check, a -D file could be left lying around. This kind of configuration
5306 error no longer causes Exim to die; instead it causes a defer errror. The
5307 incident is still logged to the main and panic logs.
5309 74. Buglet left over from Exim 3 conversion. The message "too many messages
5310 in one connection" was written to the rejectlog but not the mainlog, except
5311 when address rewriting (yes!) was being logged.
5313 75. Added write_rejectlog option.
5315 76. When a system filter was run not as root (that is, when system_filter_user
5316 was set), the values of the $n variables were not being returned to the
5317 main process; thus, they were not subsequently available in the $sn
5320 77. Added +return_path_on_delivery log selector.
5322 78. A connection timeout was being treated differently from recipients deferred
5323 when testing hosts_max_try with a message that was older than the host's
5324 retry timeout. (The host should not be counted, thus allowing all hosts to
5325 be tried at least once before bouncing.) This may have been the cause of an
5326 occasionally reported bug whereby a message would remain on the queue
5327 longer than the retry timeout, but would be bounced if a delivery was
5328 forced. I say "may" because I never totally pinned down the problem;
5329 setting up timeout/retry tests is difficult. See also the next item.
5331 79. The ultimate address timeout was not being applied to errors that involved
5332 a combination of host plus message (for example, a timeout on a MAIL
5333 command). When an address resolved to a number of possible hosts, and they
5334 were not all tried for each delivery (e.g. because of hosts_max_try), a
5335 message could remain on the queue longer than the retry timeout.
5337 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5340 81. Fixed an obscure SMTP outgoing bug which required at least the following
5341 conditions: (a) there was another message waiting for the same server;
5342 (b) the server returned 5xx to all RCPT commands in the first message so
5343 that the message was not completed; (c) the server dropped the connection
5344 or gave a negative response to the RSET that Exim sends to abort the
5345 transaction. The observed case was a dropped connection after DATA that had
5346 been sent in pipelining mode. That is, the server had advertised PIPELINING
5347 but was not implementing it correctly. The effect of the bug was incorrect
5348 behaviour, such as trying another host, and this could lead to a crash.
5354 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5355 and daemon.c were passed as pointers to ints; they should have been
5356 pointers to socklen_t variables (which are typically unsigned ints).
5358 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5361 3. Fixed a really odd bug that affected only the testing scheme; patching a
5362 certain fixed string in the binary changed the value of another string that
5363 happened to be identical to the end of the original first string.
5365 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5366 name", it returns that address as the IP address. On some operating
5367 systems (e.g. Solaris), it also passes back the IP address string as the
5368 "host name". However, on others (e.g. Linux), it passes back an empty
5369 string. Exim wasn't checking for this, and was changing the host name to an
5370 empty string, assuming it had been canonicized.
5372 5. Although rare, it is permitted to have more than one PTR record for a given
5373 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5374 all the names associated with an address, because they do in Solaris.
5375 However, it seems that they do not in Linux for data that comes from the
5376 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5377 I found this out when I moved to a new Linux workstation and tried to run
5378 the Exim test suite.
5380 To get round this problem I have changed the code so that it now does its
5381 own call to the DNS to look up PTR records when searching for a host name.
5382 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5383 addresses that are only in /etc/hosts are still found.
5385 This behaviour is, however, controlled by an option called host_lookup_
5386 order, which defaults to "bydns:byaddr". If people want to use the other
5387 order, or indeed, just use one or the other means of lookup, they can
5388 specify it in this variable.
5390 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5391 some operating systems, this comes back from gethostbyaddr() as an empty
5392 string, and this is what Exim used to test for. However, it seems that in
5393 other systems, "." is yielded. Exim now tests for this case too.
5395 7. The values of check_spool_space and check_log_space are now held internally
5396 as a number of kilobytes instead of an absolute number of bytes. If a
5397 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5398 kilobyte. This means that much larger values can be stored.
5400 8. Exim monitor: an attempt to get the action menu when not actually pointing
5401 at a message produces an empty menu entitled "No message selected". This
5402 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5403 no entries in it ("Shell widget menu has zero width and/or height"). So I
5404 have added a single, blank menu entry in this case.
5406 9. Added ${quote_local_part.
5408 10. MIME decoding is now applied to the contents of Subject: header lines when
5411 11. Now that a reference to $sender_host_address automatically causes a reverse
5412 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5413 host lookup before query-style lookups in lists that might use this
5414 variable. This has therefore been abolished, and the "net-" prefix is no
5415 longer necessary for query-style lookups.
5417 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5418 have been a typo for LFLAGS, so it has been changed.
5420 13. The install script calls Exim with "-C /dev/null" in order to find the
5421 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5422 to be output. Howeve, since Exim outputs its version number before the
5423 error, it didn't break the script. It just looked ugly. I fixed this by
5424 always allowing "-C /dev/null" if the caller is root.
5426 14. Ignore overlarge ACL variable number when reading spool file - insurance
5427 against a later release with more variables having written the file.
5429 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5430 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5431 this, as well as the form without the "IPv6" on the front.
5433 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5434 OS/Makefile-Darwin file.
5436 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5437 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5439 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5440 variable. However, some OS use ints for inodes. I've added cast to long int
5441 to get rid of the compiler warning.
5443 19. I had forgotten to lock out "/../" in configuration file names when
5444 ALT_CONFIG_PREFIX was set.
5446 20. Routers used for verification do not need to specify transports. However,
5447 if such a router generated a host list, and callout was configured, Exim
5448 crashed, because it could not find a port number from the (non-existent)
5449 transport. It now assumes port 25 in this circumstance.
5451 21. Added the -t option to exigrep.
5453 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5454 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5455 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5456 them is set, LOOKUP_LSEARCH is forced.
5458 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5460 24. Added sender and host information to the "rejected by local_scan()" log
5461 line; previously there was no indication of these.
5463 25. Added .include_if_exists.
5465 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5466 Linux. It turns out that not all file systems support this. Apparently some
5467 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5468 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5469 not supported on the file descriptor, is now ignored when Exim is trying to
5470 sync a directory. This applies only to Linux.
5472 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5474 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5475 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5478 29. Added support for maildirsize files from supplied patch (modified a bit).
5480 30. The use of :fail: followed by an empty string could lead Exim to respond to
5481 sender verification failures with (e.g.):
5483 550 Verification failed for <xxx>
5484 550 Sender verify failed
5486 where the first response line was missing the '-' that indicates it is not
5487 the final line of the response.
5489 31. The loop for finding the name of the user that called Exim had a hardwired
5490 limit of 10; it now uses the value of finduser_retries, which is used for
5491 all other user lookups.
5493 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5496 33. Exim was neglecting to zero errno before one call of strtol() when
5497 expanding a string and expecting an integer value. On some systems this
5498 resulted in spurious "integer overflow" errors. Also, it was casting the
5499 result into an int without checking.
5501 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5502 did not work. The code looks as if it has *never* worked, though it appears
5503 to have been documented since at least releast 1.62. I have made it work.
5505 35. The "timeout_DNS" error in retry rules, also documented since at least
5506 1.62, also never worked. As it isn't clear exactly what this means, and
5507 clearly it isn't a major issue, I have abolished the feature by treating it
5508 as "timeout", and writing a warning to the main and panic logs.
5510 36. The display of retry rules for -brt wasn't always showing the error code
5513 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5514 timeout_connect_A, timeout_connect_MX.
5516 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5517 to the empty sender.
5519 39. The daemon was not analysing the content of -oX till after it had closed
5520 stderr and disconnected from the controlling terminal. This meant that any
5521 syntax errors were only noted on the panic log, and the return code from
5522 the command was 0. By re-arranging the code a little, I've made the
5523 decoding happen first, so such errors now appear on stderr, and the return
5524 code is 1. However, the actual setting up of the sockets still happens in
5525 the disconnected process, so errors there are still only recorded on the
5528 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5529 connections (as happens on some IP stacks) was logged at start up time as
5530 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5531 it from "IPv6 and IPv4", which means that two separate sockets are being
5534 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5535 says whether AF_INET or AF_INET6 was passed as an argument.
5537 42. Exiwhat output was messed up when time zones were included in log
5540 43. Exiwhat now gives more information about the daemon's listening ports,
5541 and whether -tls-on-connect was used.
5543 44. The "port" option of the smtp transport is now expanded.
5545 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5546 silently ignored. Now an error message is written to the main and panic
5549 46. There's a new ACL modifier called "logwrite" which writes to a log file
5550 as soon as it is encountered.
5552 47. Added $local_user_uid and $local_user_gid at routing time.
5554 48. Exim crashed when trying to verify a sender address that was being
5557 49. Exim was recognizing only a space character after ".include". It now also
5558 recognizes a tab character.
5560 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5561 extracting the relevant information from the specification. The man page no
5562 longer contains scrambled data for the -d option, and I've added a section
5563 at the front about calling Exim under different names.
5565 51. Added "extra_headers" argument to the "mail" command in filter files.
5567 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5570 53. Installed eximstats 1.29.
5572 54. Added transport_filter_timeout as a generic transport option.
5574 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5575 Cc: header lines. This was required by RFC 822, but it not required by RFC
5578 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5579 incoming messages that did not have them. Now it does so only if the
5580 message originates locally, that is, if there is no associated remote host
5581 address. When Resent- header lines are present, this applies to the Resent-
5582 lines rather than the non-Resent- lines.
5584 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5585 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5587 58. Messages for configuration errors now include the name of the main
5588 configuration file - useful now that there may be more than one file in a
5589 list (.included file names were always shown).
5591 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5592 for those rare installations that do not start the daemon as root or run it
5593 setuid root. I've cut out the call to initgroups() if the daemon is not
5596 60. The Exim user and group can now be bound into the binary as text strings
5597 that are looked up at the start of Exim's processing.
5599 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5601 62. Added $mailstore_basename variable.
5603 63. Installed patch to sieve.c from Michael Haardt.
5605 64. When Exim failed to open the panic log after failing to open the main log,
5606 the original message it was trying to log was written to stderr and debug
5607 output, but if they were not available (the usual case in production), it
5608 was lost. Now it is written to syslog before the two lines that record the
5609 failures to open the logs.
5611 65. Users' Exim filters run in subprocesses under the user's uid. It is
5612 possible for a "deliver" command or an alias in a "personal" command to
5613 provoke an address rewrite. If logging of address rewriting is configured,
5614 this fails because the process is not running as root or exim. There may be
5615 a better way of dealing with this, but for the moment (because 4.30 needs
5616 to be released), I have disabled address rewrite logging when running a
5617 filter in a non-root, non-exim process.
5623 1. The buildconfig auxiliary program wasn't quoting the value set for
5624 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5625 not defined. This bug was masked in 4.22 by the effect that was fixed in
5628 2. Some messages that were rejected after a message id was allocated were
5629 shown as "incomplete" by exigrep. It no longer does this for messages that
5630 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5632 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5633 have allow_domain_literals set, the ID did not get logged in the <= line.
5634 Domain literals are now always recognized in Message-ID: header lines.
5636 4. The first argument for a ${extract expansion item is the key name or field
5637 number. Leading and trailing spaces in this item were not being ignored,
5638 causing some misleading effects.
5640 5. When deliver_drop_privilege was set, single queue runner processes started
5641 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5642 same command in the process it spins off) were not dropping privilege.
5644 6. When the daemon running as "exim" started a queue runner, it always
5645 re-executed Exim in the spun-off process. This is a waste of effort when
5646 deliver_drop_privilege is set. The new process now just calls the
5647 queue-runner function directly.
5653 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5656 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5657 ignored. Though the use of -oP was forcing the writing of a pid file, it
5658 was always written to the default place.
5660 3. If the message "no IP address found for host xxxx" is generated during
5661 incoming verification, it is now followed by identification of the incoming
5662 connection (so you can more easily find what provoked it).
5664 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5666 5. Added some features to "harden" Exim a bit more against certain attacks:
5668 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5669 be put in Local/Makefile. This is like the never_users runtime option,
5670 but it cannot be overridden. The default setting is "root".
5672 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5673 prefix string with which any file named in a -C command line option
5676 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5677 is retained for -C and -D only if the caller of Exim is root. Without
5678 it, the exim user may also use -C and -D and retain privilege.
5680 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5681 command line option is disabled.
5683 6. Macro names set by the -D option must start with an upper case letter, just
5684 like macro names defined in the configuration file.
5686 7. Added "dereference=" facility to LDAP.
5688 8. Two instances of the typo "uknown" in the source files are fixed.
5690 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5691 the Configure-Makefile script screwed up while processing it.
5693 10. Incorporated PCRE 4.4.
5695 11. The SMTP synchronization check was not operating right at the start of an
5696 SMTP session. For example, it could not catch a HELO sent before the client
5697 waited for the greeting. There is now a check for outstanding input at the
5698 point when the greeting is written. Because of the duplex, asynchronous
5699 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5700 way, but not yet received, when the check is performed.
5702 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5703 on TCP/IP sockets, because this apparently causes some broken clients to
5706 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5707 unchanged) from the Cygwin maintainer.
5709 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5710 was supported instead of testing for mbx. Effectively a typo.
5712 15. The spa authenticator server code was not checking that the input it
5713 received was valid base64.
5715 16. The debug output line for the "set" modifier in ACLs was not showing the
5716 name of the variable that was being set.
5718 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5720 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5721 lookup on its own. Something else was needed to trigger the lookup. For
5722 example, a match in host_lookup or the need for a host name in a host list.
5723 Now, if $sender_host_name is referenced and the host name has not yet been
5724 looked up, a lookup is performed. If the lookup fails, the variable remains
5725 empty, and $host_lookup_failed is set to "1".
5727 19. Added "eqi" as a case-independent comparison operator.
5729 20. The saslauthd authentication condition could segfault if neither service
5730 nor realm was specified.
5732 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5733 error message that was logged was misleading, and incoming SMTP
5734 connections were dropped. The message is now more accurate, and temporary
5735 errors are given to SMTP connections.
5737 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5738 (except RSET) with a 421 error, until QUIT is received. However, it was
5739 failing to send a response to QUIT.
5741 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5742 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5743 ACL is now run after the helo_try_verify_hosts code.
5745 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5746 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5747 case-independent, so other case variants are also recognized). Apparently
5748 some systems use these upper case variants.
5750 25. If more than two messages were waiting for the same host, and a transport
5751 filter was specified for the transport, Exim sent two messages over the
5752 same TCP/IP connection, and then failed with "socket operation on non-
5753 socket" when it tried to send the third.
5755 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5757 27. The extern definition of crypt16() in expand.c was not being excluded when
5758 the OS had its own crypt16() function.
5760 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5761 as a preferred synonym for return_size_limit, both as an option and as an
5764 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5766 30. Changed the default configuration ACL to relax the local part checking rule
5767 for addresses that are not in any local domains. For these addresses,
5768 slashes and pipe symbols are allowed within local parts, but the sequence
5769 /../ is explicitly forbidden.
5771 31. SPA server authentication was not clearing the challenge buffer before
5774 32. log_message in a "warn" ACL statement was writing to the reject log as
5775 well as to the main log, which contradicts the documentation and doesn't
5776 seem right (because no rejection is happening). So I have stopped it.
5778 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5779 However, I am unable to do any testing of this.
5781 34. Fixed an infelicity in the appendfile transport. When checking directories
5782 for a mailbox, to see if any needed to be created, it was accidentally
5783 using path names with one or more superfluous leading slashes; tracing
5784 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5786 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5787 contents are added to the log line that is written for every discarded
5788 recipient. (Previously a log_message setting was ignored.)
5790 36. The ${quote: operator now quotes the string if it is empty.
5792 37. The install script runs exim in order to find its version number. If for
5793 some reason other than non-existence or emptiness, which it checks, it
5794 could not run './exim', it was installing it with an empty version number,
5795 i.e. as "exim-". This error state is now caught, and the installation is
5798 38. An argument was missing from the function that creates an error message
5799 when Exim fails to connect to the socket for saslauthd authentication.
5800 This could cause Exim to crash, or give a corrupted message.
5802 39. Added isip, isip4, and isip6 to ${if conditions.
5804 40. The ACL variables $acl_xx are now saved with the message, and can be
5805 accessed later in routers, transports, and filters.
5807 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5808 strings in the file are not string-expanded.
5810 42. If a MAIL command specified a SIZE value that was too large to fit into an
5811 int variable, the check against message_size_limit failed. Such values are
5812 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5813 day this will have to be increased, but I don't think I want to be around
5814 when emails are that large.
5821 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5822 iconv() is not standard in FreeBSD.
5824 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5825 IPv6 enabled. The observed symptom was a segmentation fault on return from
5826 the function os_common_find_running_interfaces() in src/os.c.
5828 3. In the check_special_case() function in daemon.c I had used "errno" as an
5829 argument name, which causes warnings on some systems. This was basically a
5830 typo, since it was named "eno" in the comments!
5832 4. The code that waits for the clock to tick (at a resolution of some fraction
5833 of a second) so as to ensure message-id uniqueness was always waiting for
5834 at least one whole tick, when it could have waited for less. [This is
5835 almost certainly not relevant at current processor speeds, where it is
5836 unlikely to ever wait at all. But we try to future-proof.]
5838 5. The function that sleeps for a time interval that includes fractions of a
5839 second contained a race. It did not block SIGALRM between setting the
5840 timer, and suspending (a couple of lines later). If the interval was short
5841 and the sigsuspend() was delayed until after it had expired, the suspension
5842 never ended. On busy systems this could lead to processes getting stuck for
5845 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5846 process, before it forks any delivery processes. The open lookup caching
5847 mechanism meant that the open file or database connection was passed into
5848 the delivery process. The problem was that delivery processes always tidy
5849 up cached lookup data. This could cause a problem for the next delivery
5850 process started by the queue runner, because the external queue runner
5851 process does not know about the closure. So the next delivery process
5852 still has data in the lookup cache. In the case of a file lookup, there was
5853 no problem because closing a file descriptor in a subprocess doesn't affect
5854 the parent. However, if the lookup was caching a connection to a database,
5855 the connection was closed, and the second delivery process was likely to
5856 see errors such as "PGSQL: query failed: server closed the connection
5857 unexpectedly". The problem has been fixed by closing all cached lookups
5858 in a queue runner before running a delivery process.
5860 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5861 seem to have the "const" qualifier which it has on other OS. I've
5864 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5865 *of the same type* (client or server) with the same public name that an
5866 error should be diagnosed.
5868 9. When Exim looked up a host name for an IP address, but failed to find the
5869 original IP address when looking up the host name (a safety check), it
5870 output the message "<ip address> does not match any IP for NULL", which was
5871 confusing, to say the least. The bug was that the host name should have
5872 appeared instead of "NULL".
5874 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5875 user that is built into the binary, and the -C or -D options is used, root
5876 privilege is dropped before the configuration file is read. In addition,
5877 logging is switched to stderr instead of the normal log files. If the
5878 configuration then re-defines the Exim user, the unprivileged environment
5879 is probably not what is expected, so Exim logs a panic warning message (but
5882 However, if deliver_drop_privilege is set, the unprivileged state may well
5883 be exactly what is intended, so the warning has been cut out in that case,
5884 and Exim is allowed to try to write to its normal log files.
5890 1. smtp_return_error_details was not giving details for temporary sender
5891 or receiver verification errors.
5893 2. Diagnose a configuration error if two authenticators have the same public
5896 3. Exim used not to create the message log file for a message until the first
5897 delivery attempt. This could be confusing when incoming messages were held
5898 for policy or load reasons. The message log file is now created at the time
5899 the message is received, and an initial "Received" line is written to it.
5901 4. The automatically generated man page for command line options had a minor
5902 bug that caused no ill effects; however, a more serious problem was that
5903 the procedure for building the man page automatically didn't always
5904 operate. Consequently, release 4.20 contains an out-of-date version. This
5905 shouldn't happen again.
5907 5. When building Exim with embedded Perl support, the script that builds the
5908 Makefile was calling 'perl' to find its compile-time parameters, ignoring
5909 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
5911 6. The freeze_tell option was not being used for messages that were frozen on
5912 arrival, either by an ACL or by local_scan().
5914 7. Added the smtp_incomplete_transaction log selector.
5916 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
5917 was accepting AUTH without a new EHLO.
5919 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
5920 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
5923 10. Logging of TCP/IP connections (when configured) now happens in the main
5924 daemon process instead of the child process, so that the TCP/IP connection
5925 count is more accurate (but it can never be perfect).
5927 11. The use of "drop" in a nested ACL was not being handled correctly in the
5928 outer ACL. Now, if condition failure induced by the nested "drop" causes
5929 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
5930 or "require"), the connection is dropped.
5932 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
5933 that yield "discard" can now be used with an "accept" or a "discard" verb,
5934 but an error is generated for any others (because I can't see a useful way
5935 to define what should happen).
5937 13. When an ACL is read dynamically from a file (or anywhere else), the lines
5938 are now processed in the same way as lines in the Exim configuration file.
5939 In particular, continuation lines are supported.
5941 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
5943 15. Added -ti meaning -t -i.
5945 16. Check for letters, digits, hyphens, and dots in the names of dnslist
5946 domains, and warn by logging if others are found.
5948 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
5949 returned from GIFCONF when Exim is trying to find the list of interfaces on
5950 a host. The code in os.c has been modified to copy each ifreq to an aligned
5951 structure in all cases.
5953 Also, in some cases, the returned ifreq's were being copied to a 'struct
5954 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
5955 means the last couple of bytes of an IPv6 address could be chopped if the
5956 ifreq contained only a normal sockaddr (14 bytes storage).
5958 18. Named domain lists were not supported in the hosts_treat_as_local option.
5959 An entry such as +xxxx was not recognized, and was treated as a literal
5962 19. Ensure that header lines added by a DATA ACL are included in the reject log
5963 if the ACL subsequently rejects the message.
5965 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
5966 MD5 (which is deprecated).
5968 21. When testing a filter file using -bf, Exim was writing a message when it
5969 took the sender from a "From " line in the message, but it was not doing so
5970 when it took $return_path from a Return-Path: header line. It now does.
5972 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
5973 with a valid header line field name (a series of printing characters
5974 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
5976 23. Changed "disc" in the source to "disk" to conform to the documentation and
5977 the book and for uniformity.
5979 24. Ignore Sendmail's -Ooption=value command line item.
5981 25. When execve() failed while trying to run a command in a pipe transport,
5982 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
5983 could be confused with a return value of 69 from the command itself. This
5984 has been changed to 127, the value the shell returns if it is asked to run
5985 a non-existent command. The wording for the related log line suggests a
5986 non-existent command as the problem.
5988 26. If received_header_text expands to an empty string, do not add a Received:
5989 header line to the message. (Well, it adds a token one on the spool, but
5990 marks it "old" so that it doesn't get used or transmitted.)
5992 27. Installed eximstats 1.28 (addition of -nt option).
5994 28. There was no check for failure on the call to getsockname() in the daemon
5995 code. This can fail if there is a shortage of resources on the system, with
5996 ENOMEM, for example. A temporary error is now given on failure.
5998 29. Contrary to the C standard, it seems that in some environments, the
5999 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6000 program. Exim now does this explicitly; it affects the formatting of
6001 timestamps using strftime().
6003 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6004 complaints. I've now initialized all the variables, to avoid this.
6006 32. Header lines added by a system filter were not being "seen" during
6007 transport-time rewrites.
6009 33. The info_callback() function passed to OpenSSL is set up with type void
6010 (*)(SSL *, int, int), as described somewhere. However, when calling the
6011 function (actually a macro) that sets it up, the type void(*)() is
6012 expected. I've put in a cast to prevent warnings from picky compilers.
6014 34. If a DNS black list lookup found a CNAME record, but there were no A
6015 records associated with the domain it pointed at, Exim crashed.
6017 35. If a DNS black list lookup returned more than one A record, Exim ignored
6018 all but the first. It now scans all returned addresses if a particular IP
6019 value is being sought. In this situation, the contents of the
6020 $dnslist_value variable are a list of all the addresses, separated by a
6023 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6024 to accept a host name and all its aliases if the forward lookup for any of
6025 them yielded the IP address of the incoming connection. Now it accepts only
6026 those names whose forward lookup yields the correct IP address. Any other
6027 names are discarded. This closes a loophole whereby a rogue DNS
6028 administrator could create reverse DNS records to break through a
6029 wildcarded host restriction in an ACL.
6031 37. If a user filter or a system filter that ran in a subprocess used any of
6032 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6033 the wrong values were passed to the pipe command ($thisaddress had the
6034 value of $0, $0 had the value of $1, etc). This bug was introduced by
6035 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6037 38. Improved the line breaking for long SMTP error messages from ACLs.
6038 Previously, if there was no break point between 40 and 75 characters, Exim
6039 left the rest of the message alone. Two changes have been made: (a) I've
6040 reduced the minimum length to 35 characters; (b) if it can't find a break
6041 point between 35 and 75 characters, it looks ahead and uses the first one
6042 that it finds. This may give the occasional overlong line, but at least the
6043 remaining text gets split now.
6045 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6046 file descriptors might be low, and that setting 1000 would always raise it.
6047 It turns out that in some environments, the limit is already over 1000 and
6048 that lowering it causes trouble. So now Exim takes care not to decrease it.
6050 40. When delivering a message, the value of $return_path is set to $sender_
6051 address at the start of routing (routers may change the value). By an
6052 oversight, this default was not being set up when an address was tested by
6053 -bt or -bv, which affected the outcome if any router or filter referred to
6056 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6057 log only when "message" or "log_message" are set. However, if one of the
6058 conditions was an address verification, or a call to a nested ACL, the
6059 messages generated by the underlying test were being passed through. This
6060 no longer happens. The underlying message is available in $acl_verify_
6061 message for both "message" and "log_message" expansions, so it can be
6062 passed through if needed.
6064 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6065 new expansion $bh_ to give the encoded byte string without charset
6066 translation. Translation happens only if iconv() is available; HAVE_ICONV
6067 indicates this at build time. HEADERS_CHARSET gives the charset to
6068 translate to; headers_charset can change it in the configuration, and
6069 "headers charset" can change it in an individual filter file.
6071 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6072 that creates RFC 2047 encoded "words" labels them as that charset instead
6073 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6074 expansion operator; (ii) when Exim creates a From: line for a local
6075 message; (iii) when a header line is rewritten to include a "phrase" part.
6077 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6078 buggy, causing it to skip the first lines of messages whose message ID
6079 ended in 'D'. This would not have bitten before Exim release 4.14, because
6080 message IDs were unlikely to end in 'D' before then. The effect was to have
6081 incorrect size information for certain domains.
6083 45. #include "config.h" was missing at the start of the crypt16.c module. This
6084 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6087 46. If there was a timeout during a "random" callout check, Exim treated it as
6088 a failure of the random address, and carried on sending RSET and the real
6089 address. If the delay was just some slowness somewhere, the response to the
6090 original RCPT would be taken as a response to RSET and so on, causing
6091 mayhem of various kinds.
6093 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6094 when I implemented it. It didn't allow for the fact that some option values
6095 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6096 the right test for positive values.
6098 48. Domain names in DNS records are case-independent. Exim always looks them up
6099 in lower case. Some resolvers return domain names in exactly the case they
6100 appear in the zone file, that is, they may contain uppercase letters. Not
6101 all resolvers do this - some return always lower case. Exim was treating a
6102 change of case by a resolver as a change of domain, similar to a widening
6103 of a domain abbreviation. This triggered its re-routing code and so it was
6104 trying to route what was effectively the same domain again. This normally
6105 caused routing to fail (because the router wouldn't handle the domain
6106 twice). Now Exim checks for this case specially, and just changes the
6107 casing of the domain that it ultimately uses when it transmits the message
6110 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6113 50. If a filter generated a file delivery with a non-absolute name (possible if
6114 no home directory exists for the router), the forbid_file option was not
6117 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6118 the existing equality matching.
6120 52. Exim was using ints instead of ino_t variables in some places where it was
6121 dealing with inode numbers.
6123 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6124 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6125 called TMPDIR, and if it finds it is different, it changes its value.
6127 54. The smtp_printf() function is now made available to local_scan() so
6128 additional output lines can be written before returning. There is also an
6129 smtp_fflush() function to enable the detection of a dropped connection.
6130 The variables smtp_input and smtp_batched_input are exported to
6133 55. Changed the default runtime configuration: the message "Unknown user"
6134 has been removed from the ACL, and instead placed on the localuser router,
6135 using the cannot_route_message feature. This means that any verification
6136 failures that generate their own messages won't get overridden. Similarly,
6137 the "Unrouteable address" message that was in the ACL for unverifiable
6138 relay addresses has also been removed.
6140 56. Added hosts_avoid_esmtp to the smtp transport.
6142 57. The exicyclog script was not checking for the esoteric option
6143 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6144 will work only if exicyclog is run under the appropriate euid.
6146 58. Following a discussion on the list, the rules by which Exim recognises line
6147 endings on incoming messages have been changed. The -dropcr and drop_cr
6148 options are now no-ops, retained only for backwards compatibility. The
6149 following line terminators are recognized: LF CRLF CR. However, special
6150 processing applies to CR:
6152 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6153 nor a local message in the state where . is a terminator.
6155 (ii) If a bare CR is encountered in a header line, an extra space is added
6156 after the line terminator so as not to end the header. The reasoning
6157 behind this is that bare CRs in header lines are most likely either
6158 to be mistakes, or people trying to play silly games.
6160 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6161 was being incorrectly given as 18 bytes larger than it should have been.
6162 This is a VOB (very old bug).
6164 60. This may never have affected anything current, but just in case it has:
6165 When the local host is found other than at the start of a list of hosts,
6166 the local host, those with the same MX, and any that follow, are discarded.
6167 When the list in question was part of a longer list of hosts, the following
6168 hosts (not currently being processed) were also being discarded. This no
6169 longer happens. I'm not sure if this situation could ever has previously
6172 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6175 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6176 header. This is something that is recommended in a new Internet Draft, and
6177 is something that is documented as being done by Sendmail. There are two
6178 possible values. For messages generated by the autoreply transport, Exim
6181 Auto-Submitted: auto-replied
6183 whereas for all other generated messages (e.g. bounces) it adds
6185 Auto-Submitted: auto-generated
6187 63. The "personal" condition in filters now includes a test for the
6188 Auto-Submitted: header. If it contains the string "auto-" the message it
6189 not considered personal.
6191 64. Added rcpt_include_affixes as a generic transport option.
6193 65. Added queue_only_override (default true).
6195 66. Added the syslog_duplication option.
6197 67. If what should have been the first header line of a message consisted of
6198 a space followed by a colon, Exim was mis-interpreting it as a header line.
6199 It isn't of course - it is syntactically invalid and should therefore be
6200 treated as the start of the message body. The misbehaviour could have
6201 caused a number of strange effects, including loss of data in subsequent
6202 header lines, and spool format errors.
6204 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6205 client host had authenticated. This control can now be exercised by an ACL
6206 for more flexibility.
6208 69. By default, callouts do not happen when testing with -bh. There is now a
6209 variant, -bhc, which does actually run the callout code, including
6210 consulting and updating the callout cache.
6212 70. Added support for saslauthd authentication, courtesy of Alexander
6215 71. If statvfs() failed on the spool or log directories while checking their
6216 size for availability, Exim confusingly gave the error "space shortage".
6217 Furthermore, in debugging mode it crashed with a floating point exception.
6218 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6219 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6220 serious problem, Exim now writes to the main and panic logs when this
6221 happens, with details of the failure. It then refuses to accept the
6222 incoming message, giving the message "spool directory problem" or "log
6223 directory problem" with a 421 code for SMTP messages.
6225 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6226 0, 1, and 2 exist, because some OS complain for execs without them (see
6227 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6228 descriptors. However, the code omitted to check that the open succeeded,
6229 causing mysterious errors if for some reason the permissions on /dev/null
6230 got screwed. Now Exim writes a message to the main and panic logs, and
6231 bombs out if it can't open /dev/null.
6233 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6234 interact so that it is all more flexible. It is supposed to remain
6235 backwards compatible. Also added extra_local_interfaces.
6237 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6238 to bomb out with an assertion failure - to the client this appears as a
6239 connection drop. This problem occurs in the part of the code that was taken
6240 from the Samba project. Fortunately, the assertion is in a very simple
6241 function, so I have fixed this by reproducing the function inline in the
6242 one place where it is called, and arranging for authentication to fail
6243 instead of killing the process with assert().
6245 75. The SPA client code was not working when the server requested OEM rather
6246 than Unicode encoding.
6248 76. Added code to make require_files with a specific uid setting more usable in
6249 the case where statting the file as root fails - usually a non-root-mounted
6250 NFS file system. When this happens and the failure is EACCES, Exim now
6251 forks a subprocess and does the per-uid checking as the relevant uid.
6253 77. Added process_log_path.
6255 78. If log_file_path was not explicitly set, a setting of check_log_space or
6256 check_log_inodes was ignored.
6258 79. If a space check for the spool or log partitions fails, the incident is now
6259 logged. Of course, in the latter case the data may get lost...
6261 80. Added the %p formatting code to string_format() so that it can be used to
6262 print addresses in debug_print(). Adjusted all the address printing in the
6263 debugging in store.c to use %p rather than %d.
6265 81. There was a concern that a line of code in smtp_in.c could overflow a
6266 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6267 initially expressed, the concern was not well-founded, because trailing
6268 spaces are removed early. However, if the trailing spaces were followed by
6269 a NULL, they did not get removed, so the overflow was possible. Two fixes
6272 (a) I re-wrote the offending code in a cleaner fashion.
6273 (b) If an incoming SMTP command contains a NULL character, it is rejected
6276 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6277 runs initgroups(), so that if the Exim user is in any additional groups,
6278 they will be used during message reception.
6284 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / jgh / exim.git / blob