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18 EXPERIMENTAL_CERTNAMES is withdrawn.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).
104 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
106 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
107 when evaluating $sender_host_dnssec.
112 TL/01 When running the test suite, the README says that variables such as
113 no_msglog_check are global and can be placed anywhere in a specific
114 test's script, however it was observed that placement needed to be near
115 the beginning for it to behave that way. Changed the runtest perl
116 script to read through the entire script once to detect and set these
117 variables, reset to the beginning of the script, and then run through
118 the script parsing/test process like normal.
120 TL/02 The BSD's have an arc4random API. One of the functions to induce
121 adding randomness was arc4random_stir(), but it has been removed in
122 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
123 function when detected.
125 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
126 cause callback expansion.
128 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
129 syntax errors in an expansion can be treated as a string instead of
130 logging or causing an error, due to the internal use of bool_lax
131 instead of bool when processing it.
133 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
134 server certificates when making smtp deliveries.
136 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
138 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
140 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
142 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
143 Merged patch from Sebastian Wiedenroth.
145 JH/05 Fix results-pipe from transport process. Several recipients, combined
146 with certificate use, exposed issues where response data items split
147 over buffer boundaries were not parsed properly. This eventually
148 resulted in duplicates being sent. This issue only became common enough
149 to notice due to the introduction of conection certificate information,
150 the item size being so much larger. Found and fixed by Wolfgang Breyha.
152 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
153 size buffer was used, resulting in syntax errors when an expansion
156 JH/07 Add support for directories of certificates when compiled with a GnuTLS
157 version 3.3.6 or later.
159 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
160 is EXPERIMENTAL_EVENT, the main-configuration and transport options
161 both become "event_action", the variables become $event_name, $event_data
162 and $event_defer_errno. There is a new variable $verify_mode, usable in
163 routers, transports and related events. The tls:cert event is now also
164 raised for inbound connections, if the main configuration event_action
167 TL/06 In test suite, disable OCSP for old versions of openssl which contained
168 early OCSP support, but no stapling (appears to be less than 1.0.0).
170 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
171 server certificate names available under the smtp transport option
172 "tls_verify_cert_hostname" now do not permit multi-component wildcard
175 JH/10 Time-related extraction expansions from certificates now use the main
176 option "timezone" setting for output formatting, and are consistent
177 between OpenSSL and GnuTLS compilations. Bug 1541.
179 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
180 encoded parameter in the incoming message. Bug 1558.
182 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
183 include certificate info, eximon was claiming there were spoolfile
186 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
188 JH/14 Log delivery-related information more consistently, using the sequence
189 "H=<name> [<ip>]" wherever possible.
191 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
192 are problematic for Debian distribution, omit them from the release
195 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
197 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
200 JH/17 Fix a null-indirection in certextract expansions when a nondefault
201 output list separator was used.
206 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
207 checkers that were complaining about end of non-void function with no
210 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
211 This was a regression intruduced in 4.83 by another bugfix.
213 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
215 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
216 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
222 TF/01 Correctly close the server side of TLS when forking for delivery.
224 When a message was received over SMTP with TLS, Exim failed to clear up
225 the incoming connection properly after forking off the child process to
226 deliver the message. In some situations the subsequent outgoing
227 delivery connection happened to have the same fd number as the incoming
228 connection previously had. Exim would try to use TLS and fail, logging
229 a "Bad file descriptor" error.
231 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
232 utilities have not been installed.
234 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
235 temporary space as the ACL may create new global variables.
237 TL/01 LDAP support uses per connection or global context settings, depending
238 upon the detected version of the libraries at build time.
240 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
241 to extract and use the src ip:port in logging and expansions as if it
242 were a direct connection from the outside internet. PPv2 support was
243 updated based on HAProxy spec change in May 2014.
245 JH/02 Add ${listextract {number}{list}{success}{fail}}.
247 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
248 Properly escape header and check for NULL return.
250 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
253 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
255 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
256 characters in header names, implemented as a verify condition.
257 Contributed by Michael Fischer v. Mollard.
259 TL/05 Rename SPF condition results err_perm and err_temp to standardized
260 results permerror and temperror. Previous values are deprecated but
261 still accepted. In a future release, err_perm and err_temp will be
262 completely removed, which will be a backward incompatibility if the
263 ACL tests for either of these two old results. Patch contributed by
264 user bes-internal on the mailing list.
266 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
268 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
269 selectors, in both main and reject logs.
271 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
274 JH/07 Add malware type "sock" for talking to simple daemon.
276 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
278 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
279 routers/transports under cutthrough routing.
281 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
282 numbers. Touch up "bool" conditional to keep the same definition.
284 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
286 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
288 JH/12 Expand items in router/transport headers_add or headers_remove lists
289 individually rather than the list as a whole. Bug 1452.
291 Required for reasonable handling of multiple headers_ options when
292 they may be empty; requires that headers_remove items with embedded
293 colons must have them doubled (or the list-separator changed).
295 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
296 view the policy declared in the DMARC record. Currently, $dmarc_status
297 is a combined value of both the record presence and the result of the
300 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
302 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
303 dnslookup router and the smtp transport (applying to the forward
306 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
307 of ldap servers used for a specific lookup. Patch provided by Heiko
310 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
311 New variable $lookup_dnssec_authenticated for observability.
313 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
314 Patch submitted by Lars Timman.
316 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
318 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
319 Requires trusted mode and valid format message id, aborts otherwise.
320 Patch contributed by Heiko Schlichting.
322 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
323 certextract with support for various fields. Bug 1358.
325 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
326 is requested by default, modifiable by smtp transport option
329 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
330 operate on certificate variables to give certificate fingerprints
331 Also new ${sha256:cert_variable}.
333 JH/23 The PRDR feature is moved from being Experimental into the mainline.
335 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
336 Christian Aistleitner.
338 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
340 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
341 file. Patch from Wolfgang Breyha.
343 JH/25 Expand the coverage of the delivery $host and $host_address to
344 client authenticators run in verify callout. Bug 1476.
346 JH/26 Port service names are now accepted for tls_on_connect_ports, to
347 align with daemon_smtp_ports. Bug 72.
349 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
350 support and error reporting did not work properly.
352 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
353 and is readable. Patch from Andrew Colin Kissa.
355 TL/14 Enhance documentation of ${run expansion and how it parses the
356 commandline after expansion, particularly in the case when an
357 unquoted variable expansion results in an empty value.
359 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
361 PP/02 Fix internal collision of T_APL on systems which support RFC3123
362 by renaming away from it. Addresses GH issue 15, reported by
365 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
367 TL/15 SECURITY: prevent double expansion in math comparison functions
368 (can expand unsanitized data). Not remotely exploitable.
375 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
377 PP/02 Make -n do something, by making it not do something.
378 When combined with -bP, the name of an option is not output.
380 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
383 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
384 $sender_host_name and config options to manage this, and basic check
387 PP/05 DSCP support for outbound connections and control modifier for inbound.
389 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
390 (Only plugin which currently uses this is kerberos4, which nobody should
391 be using, but we should make it available and other future plugins might
392 conceivably use it, even though it would break NAT; stuff *should* be
393 using channel bindings instead).
395 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
396 name; added for Sendmail compatibility; requires admin caller.
397 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
398 just ignore it); requires trusted caller.
399 Also parse but ignore: -Ac -Am -X<logfile>
402 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
404 TL/02 Add +smtp_confirmation as a default logging option.
406 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
407 Patch by Magnus Holmgren from 2007-02-20.
409 TL/04 Bugzilla 1281 - Spec typo.
410 Bugzilla 1283 - Spec typo.
411 Bugzilla 1290 - Spec grammar fixes.
413 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
415 TL/06 Add Experimental DMARC support using libopendmarc libraries.
417 TL/07 Fix an out of order global option causing a segfault. Reported to dev
418 mailing list by by Dmitry Isaikin.
420 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
422 JH/02 Support "G" suffix to numbers in ${if comparisons.
424 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
426 NM/01 Bugzilla 1197 - Spec typo
427 Bugzilla 1196 - Spec examples corrections
429 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
431 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
432 gnutls_enable_pkcs11, but renamed to more accurately indicate its
435 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
436 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
438 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
439 "acl {{name}{arg}...}", and optional args on acl condition
442 JH/05 Permit multiple router/transport headers_add/remove lines.
444 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
446 JH/07 Avoid using a waiting database for a single-message-only transport.
447 Performance patch from Paul Fisher. Bugzilla 1262.
449 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
452 JH/09 Add $headers_added variable, with content from use of ACL modifier
453 add_header (but not yet added to the message). Bugzilla 199.
455 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
456 Pulled from Bugzilla 817 by Wolfgang Breyha.
458 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
460 (nb: this is the same fix as in Exim 4.80.1)
462 JH/11 Add A= logging on delivery lines, and a client_set_id option on
465 JH/12 Add optional authenticated_sender logging to A= and a log_selector
468 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
470 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
471 advertise SMTP AUTH mechanism to us, instead of a generic
472 protocol violation error. Also, make Exim more robust to bad
473 data from the Dovecot auth socket.
475 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
477 When a queue runner is handling a message, Exim first routes the
478 recipient addresses, during which it prunes them based on the retry
479 hints database. After that it attempts to deliver the message to
480 any remaining recipients. It then updates the hints database using
483 So if a recipient address works intermittently, it can get repeatedly
484 deferred at routing time. The retry hints record remains fresh so the
485 address never reaches the final cutoff time.
487 This is a fairly common occurrence when a user is bumping up against
488 their storage quota. Exim had some logic in its local delivery code
489 to deal with this. However it did not apply to per-recipient defers
490 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
492 This change adds a proper retry rule check during routing so that the
493 final cutoff time is checked against the message's age. We only do
494 this check if there is an address retry record and there is not a
495 domain retry record; this implies that previous attempts to handle
496 the address had the retry_use_local_parts option turned on. We use
497 this as an approximation for the destination being like a local
498 delivery, as in LMTP.
500 I suspect this new check makes the old local delivery cutoff check
501 redundant, but I have not verified this so I left the code in place.
503 TF/02 Correct gecos expansion when From: is a prefix of the username.
505 Test 0254 submits a message to Exim with the header
509 When I ran the test suite under the user fanf2, Exim expanded
510 the header to contain my full name, whereas it should have added
511 a Resent-Sender: header. It erroneously treats any prefix of the
512 username as equal to the username.
514 This change corrects that bug.
516 GF/01 DCC debug and logging tidyup
517 Error conditions log to paniclog rather than rejectlog.
518 Debug lines prefixed by "DCC: " to remove any ambiguity.
520 TF/03 Avoid unnecessary rebuilds of lookup-related code.
522 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
523 Bug spotted by Jeremy Harris; was flawed since initial commit.
524 Would have resulted in OCSP responses post-SNI triggering an Exim
525 NULL dereference and crash.
527 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
529 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
530 Bug detection, analysis and fix by Samuel Thibault.
531 Bugzilla 1331, Debian bug #698092.
533 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
535 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
536 Server implementation by Todd Lyons, client by JH.
537 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
538 config variable "prdr_enable" controls whether the server
539 advertises the facility. If the client requests PRDR a new
540 acl_data_smtp_prdr ACL is called once for each recipient, after
541 the body content is received and before the acl_smtp_data ACL.
542 The client is controlled by bolth of: a hosts_try_prdr option
543 on the smtp transport, and the server advertisement.
544 Default client logging of deliveries and rejections involving
545 PRDR are flagged with the string "PRDR".
547 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
548 fclose(). Diagnosis by Todd Lyons.
550 PP/17 Update configure.default to handle IPv6 localhost better.
551 Patch by Alain Williams (plus minor tweaks).
554 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
555 This is now consistent with GnuTLS, and is now documented: the
556 previous undocumented portable approach to treating the option as
557 unset was to force an expansion failure. That still works, and
558 an empty string is now equivalent.
560 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
561 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
562 not performing validation itself.
564 PP/20 Added force_command boolean option to pipe transport.
565 Patch from Nick Koston, of cPanel Inc.
567 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
570 TF/04 Added udpsend ACL modifer and hexquote expansion operator
572 PP/21 Fix eximon continuous updating with timestamped log-files.
573 Broken in a format-string cleanup in 4.80, missed when I repaired the
574 other false fix of the same issue.
575 Report and fix from Heiko Schlichting.
578 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
579 Report from Prashanth Katuri.
581 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
582 It's SecureTransport, so affects any MacOS clients which use the
583 system-integrated TLS libraries, including email clients.
585 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
586 using a MIME ACL for non-SMTP local injection.
587 Report and assistance in diagnosis by Warren Baker.
589 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
591 JH/16 Fix comparisons for 64b. Bugzilla 1385.
593 TL/09 Add expansion variable $authenticated_fail_id to keep track of
594 last id that failed so it may be referenced in subsequent ACL's.
596 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
599 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
600 ldap library initialization, allowing self-signed CA's to be
601 used. Also properly sets require_cert option later in code by
602 using NULL (global ldap config) instead of ldap handle (per
603 session). Bug diagnosis and testing by alxgomz.
605 TL/12 Enhanced documentation in the ratelimit.pl script provided in
606 the src/util/ subdirectory.
608 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
609 renamed to Transport Post Delivery Action by Jeremy Harris, as
612 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
613 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
614 redis_servers = needs to be configured which will be used by the redis
615 lookup. Patch from Warren Baker, of The Packet Hub.
617 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
619 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
620 hostname or reverse DNS when processing a host list. Used suggestions
621 from multiple comments on this bug.
623 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
625 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
626 Missed a few lines, added it to make the runtest require no keyboard
629 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
630 contains upper case chars. Make router use caseful_local_part.
632 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
633 support when GnuTLS has been built with p11-kit.
639 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
641 This, or similar/improved, will also be change PP/11 of 4.82.
647 PP/01 Handle short writes when writing local log-files.
648 In practice, only affects FreeBSD (8 onwards).
649 Bugzilla 1053, with thanks to Dmitry Isaikin.
651 NM/01 Bugzilla 949 - Documentation tweak
653 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
656 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
658 PP/02 Implemented gsasl authenticator.
660 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
662 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
663 `pkg-config foo` for cflags/libs.
665 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
666 with rest of GSASL and with heimdal_gssapi.
668 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
669 `pkg-config foo` for cflags/libs for the TLS implementation.
671 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
672 properties get this fed in as external SSF. A number of robustness
673 and debugging improvements to the cyrus_sasl authenticator.
675 PP/08 cyrus_sasl server now expands the server_realm option.
677 PP/09 Bugzilla 1214 - Log authentication information in reject log.
678 Patch by Jeremy Harris.
680 PP/10 Added dbmjz lookup type.
682 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
684 PP/12 MAIL args handles TAB as well as SP, for better interop with
685 non-compliant senders.
686 Analysis and variant patch by Todd Lyons.
688 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
689 Bug report from Lars Müller <lars@samba.org> (via SUSE),
690 Patch from Dirk Mueller <dmueller@suse.com>
692 PP/13 tls_peerdn now print-escaped for spool files.
693 Observed some $tls_peerdn in wild which contained \n, which resulted
694 in spool file corruption.
696 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
697 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
698 or write after TLS renegotiation, which otherwise led to messages
701 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
702 as a tracking header (ie: a signed header comes before the signature).
703 Patch from Wolfgang Breyha.
705 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
706 comma-sep list; embedded commas doubled.
708 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
710 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
712 Report and patch from Dmitry Banschikov.
714 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
715 Removed SSL_clear() after SSL_new() which led to protocol negotiation
716 failures. We appear to now support TLS1.1+ with Exim.
718 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
719 lets Exim select keys and certificates based upon TLS SNI from client.
720 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
721 before an outbound SMTP session. New log_selector, +tls_sni.
723 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
724 NULL dereference. Report and patch from Alun Jones.
726 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
727 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
728 Not seeing resolver debug output on NetBSD, but suspect this is a
729 resolver implementation change.
731 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
732 Left warnings. Added "eximon gdb" invocation mode.
734 PP/21 Defaulting "accept_8bitmime" to true, not false.
736 PP/22 Added -bw for inetd wait mode support.
738 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
739 locate the relevant includes and libraries. Made this the default.
741 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
742 Bugzilla 1246, report and most of solution from Tomasz Kusy.
744 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
745 This may cause build issues on older platforms.
747 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
748 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
749 gnutls_require_mac & gnutls_require_protocols (no longer supported).
750 Added SNI support via GnuTLS too.
751 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
753 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
755 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
756 Applied second patch from Janne, implementing suggestion to default
757 multiple-strings-in-record handling to match SPF spec.
759 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
761 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
762 read-only, out of scope).
763 Patch from Wolfgang Breyha, report from Stuart Northfield.
765 PP/29 Fix three issues highlighted by clang analyser static analysis.
766 Only crash-plausible issue would require the Cambridge-specific
767 iplookup router and a misconfiguration.
768 Report from Marcin Mirosław.
770 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
772 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
773 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
774 As part of this, removing so much warning spew let me fix some minor
775 real issues in debug logging.
777 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
778 assignment on my part. Fixed.
780 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
781 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
782 Janne Snabb (who went above and beyond: thank you).
784 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
785 string otherwise requires a connection and a bunch more work and it's
786 relatively easy to get wrong. Should also expose TLS library linkage
789 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
790 64-bit ${eval} (JH/03).
792 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
793 GNU libc to support some of the 64-bit stuff, should not lead to
794 conflicts. Defined before os.h is pulled in, so if a given platform
795 needs to override this, it can.
797 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
798 protection layer was required, which is not implemented.
799 Bugzilla 1254, patch from Wolfgang Breyha.
801 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
802 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
803 tls_dhparam take prime identifiers. Also unbreak combination of
804 OpenSSL+DH_params+TLSSNI.
806 PP/39 Disable SSLv2 by default in OpenSSL support.
812 PP/01 Solaris build fix for Oracle's LDAP libraries.
813 Bugzilla 1109, patch from Stephen Usher.
815 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
817 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
820 TF/02 Fix a couple more cases where we did not log the error message
821 when unlink() failed. See also change 4.74-TF/03.
823 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
824 lock up or crash if it happened to be inside a call to libc when it
825 got a SIGUSR1 from exiwhat.
827 The SIGUSR1 handler appends the current process status to the process
828 log which is later printed by exiwhat. It used to use the general
829 purpose logging code to do this, but several functions it calls are
830 not safe for signals.
832 The new output code in the SIGUSR1 handler is specific to the process
833 log, and simple enough that it's easy to inspect for signal safety.
834 Removing some special cases also simplifies the general logging code.
835 Removing the spurious timestamps from the process log simplifies
838 TF/04 Improved ratelimit ACL condition.
840 The /noupdate option has been deprecated in favour of /readonly which
841 has clearer semantics. The /leaky, /strict, and /readonly update modes
842 are mutually exclusive. The update mode is no longer included in the
843 database key; it just determines when the database is updated. (This
844 means that when you upgrde Exim will forget old rate measurements.)
846 Exim now checks that the per_* options are used with an update mode that
847 makes sense for the current ACL. For example, when Exim is processing a
848 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
849 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
850 must specify per_mail/readonly. If you omit the update mode it defaults to
851 /leaky where that makes sense (as before) or /readonly where required.
853 The /noupdate option is now undocumented but still supported for
854 backwards compatibility. It is equivalent to /readonly except that in
855 ACLs where /readonly is required you may specify /leaky/noupdate or
856 /strict/noupdate which are treated the same as /readonly.
858 A useful new feature is the /count= option. This is a generalization
859 of the per_byte option, so that you can measure the throughput of other
860 aggregate values. For example, the per_byte option is now equivalent
861 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
863 The per_rcpt option has been generalized using the /count= mechanism
864 (though it's more complicated than the per_byte equivalence). When it is
865 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
866 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
867 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
868 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
869 Note that using per_rcpt with a non-readonly update mode in more than
870 one ACL will cause the recipients to be double-counted. (The per_mail
871 and per_byte options don't have this problem.)
873 The handling of very low rates has changed slightly. If the computed rate
874 is less than the event's count (usually one) then this event is the first
875 after a long gap. In this case the rate is set to the same as this event's
876 count, so that the first message of a spam run is counted properly.
878 The major new feature is a mechanism for counting the rate of unique
879 events. The new per_addr option counts the number of different
880 recipients that someone has sent messages to in the last time period. It
881 behaves like per_rcpt if all the recipient addresses are different, but
882 duplicate recipient addresses do not increase the measured rate. Like
883 the /count= option this is a general mechanism, so the per_addr option
884 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
885 example, measure the rate that a client uses different sender addresses
886 with the options per_mail/unique=$sender_address. There are further
887 details in the main documentation.
889 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
891 TF/06 Removed a few PCRE remnants.
893 TF/07 Automatically extract Exim's version number from tags in the git
894 repository when doing development or release builds.
896 PP/02 Raise smtp_cmd_buffer_size to 16kB.
897 Bugzilla 879. Patch from Paul Fisher.
899 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
900 Heavily based on revision 40f9a89a from Simon Arlott's tree.
903 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
905 PP/05 Variable $av_failed, true if the AV scanner deferred.
906 Bugzilla 1078. Patch from John Horne.
908 PP/06 Stop make process more reliably on build failure.
909 Bugzilla 1087. Patch from Heiko Schlittermann.
911 PP/07 Make maildir_use_size_file an _expandable_ boolean.
912 Bugzilla 1089. Patch from Heiko Schlittermann.
914 PP/08 Handle ${run} returning more data than OS pipe buffer size.
915 Bugzilla 1131. Patch from Holger Weiß.
917 PP/09 Handle IPv6 addresses with SPF.
918 Bugzilla 860. Patch from Wolfgang Breyha.
920 PP/10 GnuTLS: support TLS 1.2 & 1.1.
922 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
925 PP/11 match_* no longer expand right-hand-side by default.
926 New compile-time build option, EXPAND_LISTMATCH_RHS.
927 New expansion conditions, "inlist", "inlisti".
929 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
931 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
933 PP/14 fix log_write() format string regression from TF/03.
934 Bugzilla 1152. Patch from Dmitry Isaikin.
940 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
942 PP/02 Harmonised TLS library version reporting; only show if debugging.
943 Layout now matches that introduced for other libraries in 4.74 PP/03.
945 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
947 PP/04 New "dns_use_edns0" global option.
949 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
952 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
953 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
955 TK/01 Updated PolarSSL code to 0.14.2.
956 Bugzilla 1097. Patch from Andreas Metzler.
958 PP/07 Catch divide-by-zero in ${eval:...}.
961 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
964 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
965 format-string attack -- SECURITY: remote arbitrary code execution.
967 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
968 time unintentionally subject to list matching rules, letting the header
969 cause arbitrary Exim lookups (of items which can occur in lists, *not*
970 arbitrary string expansion). This allowed for information disclosure.
972 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
973 INT_MIN/-1 -- value coerced to INT_MAX.
979 NM/01 Workround for PCRE version dependency in version reporting
982 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
983 This fixes portability to compilers other than gcc, notably
984 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
986 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
987 makefiles for portability to HP-UX and POSIX correctness.
989 PP/01 Permit LOOKUP_foo enabling on the make command-line.
990 Also via indented variable definition in the Makefile.
991 (Debugging by Oliver Heesakkers).
993 PP/02 Restore caching of spamd results with expanded spamd_address.
994 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
996 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
997 Improves build reliability. Fix from: Frank Elsner
999 NM/02 Fix wide character breakage in the rfc2047 coding
1000 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1002 NM/03 Allow underscore in dnslist lookups
1003 Fixes bug 1026. Patch from Graeme Fowler
1005 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1006 Code patches from Adam Ciarcinski of NetBSD.
1008 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1011 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1012 is logged, to avoid truncation. Patch from John Horne.
1014 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1015 Patch from Jakob Hirsch.
1017 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1018 SQL string expansion failure details.
1019 Patch from Andrey Oktyabrski.
1021 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1022 Patch from Simon Arlott.
1024 PP/09 New lookups functionality failed to compile on old gcc which rejects
1025 extern declarations in function scope.
1026 Patch from Oliver Fleischmann
1028 PP/10 Use sig_atomic_t for flags set from signal handlers.
1029 Check getgroups() return and improve debugging.
1030 Fixed developed for diagnosis in bug 927 (which turned out to be
1033 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1034 Patch from Mark Zealey.
1036 PP/12 Bugzilla 1056: Improved spamd server selection.
1037 Patch from Mark Zealey.
1039 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1040 Based on patch from Heiko Schlittermann.
1042 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1043 Patch from Uwe Doering, sign-off by Michael Haardt.
1045 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1046 variable declaration deep within a block. Bug and patch from
1049 PP/15 lookups-Makefile IRIX compatibilty coercion.
1051 PP/16 Make DISABLE_DKIM build knob functional.
1053 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1054 Patch by Simon Arlott
1056 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1057 variable argument macros. Our copy now differs from upstream.
1063 TF/01 Failure to get a lock on a hints database can have serious
1064 consequences so log it to the panic log.
1066 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1067 controlled using the smtp_confirmation log selector.
1069 TF/03 Include the error message when we fail to unlink a spool file.
1071 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1072 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1073 for maintaining out-of-tree patches for some time.
1075 PP/01 Bugzilla 139: Documentation and portability issues.
1076 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1077 Handle per-OS dynamic-module compilation flags.
1079 PP/02 Let /dev/null have normal permissions.
1080 The 4.73 fixes were a little too stringent and complained about the
1081 permissions on /dev/null. Exempt it from some checks.
1082 Reported by Andreas M. Kirchwitz.
1084 PP/03 Report version information for many libraries, including
1085 Exim version information for dynamically loaded libraries. Created
1086 version.h, now support a version extension string for distributors
1087 who patch heavily. Dynamic module ABI change.
1089 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1090 privilege escalation vulnerability whereby the Exim run-time user
1091 can cause root to append content of the attacker's choosing to
1094 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1097 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1098 If dropping privileges for untrusted macros, we disabled normal logging
1099 on the basis that it would fail; for the Exim run-time user, this is not
1100 the case, and it resulted in successful deliveries going unlogged.
1101 Fixed. Reported by Andreas Metzler.
1107 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1108 only prepend for the Resent-* case. Fixes regression introduced in
1109 Exim 4.70 by NM/22 for Bugzilla 607.
1111 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1112 increasing numbers of administrators be bitten by this.
1114 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1116 PP/03 Bugzilla 994: added openssl_options main configuration option.
1118 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1120 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1122 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1124 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1125 without a peer certificate, leading to a segfault because of an
1126 assumption that peers always have certificates. Be a little more
1127 paranoid. Problem reported by Martin Tscholak.
1129 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1130 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1131 NB: ClamAV planning to remove STREAM in "middle of 2010".
1132 CL also introduces -bmalware, various -d+acl logging additions and
1133 more caution in buffer sizes.
1135 PP/09 Implemented reverse_ip expansion operator.
1137 PP/10 Bugzilla 937: provide a "debug" ACL control.
1139 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1141 PP/12 Bugzilla 973: Implement --version.
1143 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1145 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1147 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1149 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1150 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1151 ignore trailing whitespace.
1153 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1155 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1158 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1159 Notification from Dr Andrew Aitchison.
1161 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1162 ExtendedDetectionInfo response format.
1163 Notification from John Horne.
1165 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1168 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1169 XSL and documented dependency on system catalogs, with examples of how
1172 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1175 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1176 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1177 configuration file which is writeable by the Exim user or group.
1179 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1180 of configuration files to cover files specified with the -C option if
1181 they are going to be used with root privileges, not just the default
1184 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1185 option (effectively making it always true).
1187 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1188 files to be used while preserving root privileges.
1190 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1191 that rogue child processes cannot use them.
1193 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1194 run-time user, instead of root.
1196 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1197 Exim run-time user without dropping privileges.
1199 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1200 result string, instead of calling string_vformat() twice with the same
1203 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1204 for other users. Others should always drop root privileges if they use
1205 -C on the command line, even for a whitelisted configure file.
1207 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1209 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1215 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1216 $data_path, and $header_path variables; fixed documentation bugs and
1219 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1220 exipick to access non-standard spools, including the "frozen" queue
1223 NM/01 Bugzilla 965: Support mysql stored procedures.
1224 Patch from Alain Williams
1226 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1228 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1229 Patch from Andreas Metzler
1231 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1232 Patch from Kirill Miazine
1234 NM/05 Bugzilla 671: Added umask to procmail example.
1236 JJ/03 installed exipick 20100323.0, fixing doc bug
1238 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1239 directory. Notification and patch from Dan Rosenberg.
1241 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1243 TK/02 Improve log output when DKIM signing operation fails.
1245 MH/01 Treat the transport option dkim_domain as a colon separated
1246 list, not as a single string, and sign the message with each element,
1247 omitting multiple occurences of the same signer.
1249 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1250 Bugzilla 985, 986. Patch by Simon Arlott
1252 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1253 Patch by Simon Arlott
1255 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1256 MBX locking. Notification from Dan Rosenberg.
1262 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1264 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1266 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1268 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1270 NM/04 Bugzilla 915: Flag broken perl installation during build.
1276 TK/01 Added patch by Johannes Berg that expands the main option
1277 "spamd_address" if it starts with a dollar sign.
1279 TK/02 Write list of recipients to X-Envelope-Sender header when building
1280 the mbox-format spool file for content scanning (suggested by Jakob
1283 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1284 (http://www.dcc-servers.net/) support via dccifd. Activated by
1285 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1287 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1288 by Mark Daniel Reidel <mr@df.eu>.
1290 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1291 When building exim an external PCRE library is now needed -
1292 PCRE is a system library on the majority of modern systems.
1293 See entry on PCRE_LIBS in EDITME file.
1295 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1296 conversation. Added nologin parameter to request.
1297 Patch contributed by Kirill Miazine.
1299 TF/01 Do not log submission mode rewrites if they do not change the address.
1301 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1303 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1304 log files in place. Contributed by Roberto Lima.
1306 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1308 TF/03 Bugzilla 615: When checking the local_parts router precondition
1309 after a local_part_suffix or local_part_prefix option, Exim now
1310 does not use the address's named list lookup cache, since this
1311 contains cached lookups for the whole local part.
1313 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1314 Robert Millan. Documentation is in experimental-spec.txt.
1316 TF/04 Bugzilla 668: Fix parallel build (make -j).
1318 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1320 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1321 Patch provided by Jan Srzednicki.
1323 TF/05 Leading white space used to be stripped from $spam_report which
1324 wrecked the formatting. Now it is preserved.
1326 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1327 that they are available at delivery time.
1329 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1331 TF/08 TLS error reporting now respects the incoming_interface and
1332 incoming_port log selectors.
1334 TF/09 Produce a more useful error message if an SMTP transport's hosts
1335 setting expands to an empty string.
1337 NM/06 Bugzilla 744: EXPN did not work under TLS.
1338 Patch provided by Phil Pennock.
1340 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1341 Patch provided by Richard Godbee.
1343 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1344 acl_smtp_notquit, added index entry.
1346 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1347 Patch provided by Eugene Bujak.
1349 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1350 accept(). Patch provided by Maxim Dounin.
1352 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1353 Patch provided by Phil Pennock.
1355 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1357 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1358 Patch provided by Brad "anomie" Jorsch.
1360 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1361 Patch provided by Dean Brooks.
1363 TK/05 Add native DKIM support (does not depend on external libraries).
1365 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1366 Patch provided by Graeme Fowler.
1368 NM/16 Bugzilla 851: Documentation example syntax fix.
1370 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1372 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1375 NM/19 Bugzilla 745: TLS version reporting.
1376 Patch provided by Phil Pennock.
1378 NM/20 Bugzilla 167: bool: condition support.
1379 Patch provided by Phil Pennock.
1381 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1382 clients. Patch provided by Phil Pennock.
1384 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1385 Patch provided by Brad "anomie" Jorsch.
1387 NM/23 Bugzilla 687: Fix misparses in eximstats.
1388 Patch provided by Heiko Schlittermann.
1390 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1391 Patch provided by Heiko Schlittermann.
1393 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1394 plus update to original patch.
1396 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1398 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1399 Patch provided by David Brownlee.
1401 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1403 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1405 NM/30 Bugzilla 888: TLS documentation bugfixes.
1407 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1409 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1410 Unlike the original bugzilla I have changed all shell scripts in src tree.
1412 NM/33 Bugzilla 898: Transport filter timeout fix.
1413 Patch by Todd Rinaldo.
1415 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1416 Patch by Serge Demonchaux.
1418 NM/35 Bugzilla 39: Base64 decode bug fixes.
1419 Patch by Jakob Hirsch.
1421 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1423 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1425 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1427 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1433 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1434 ALT-N's libdkim that I have put here:
1435 http://duncanthrax.net/exim-experimental/
1437 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1438 were called 'true' and it seems that C99 defines that as a reserved
1439 keyword to be used with 'bool' variable types. That means you could
1440 not include C99-style headers which use bools without triggering
1441 build errors in sieve.c.
1443 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1444 as mailq or other aliases. Changed the --help handling significantly
1445 to do whats expected. exim_usage() emits usage/help information.
1447 SC/01 Added the -bylocaldomain option to eximstats.
1449 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1451 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1453 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1459 PH/01 Another patch from the Sieve maintainer.
1461 PH/02 When an IPv6 address is converted to a string for single-key lookup
1462 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1463 dots are used instead of colons so that keys in lsearch files need not
1464 contain colons. This was done some time before quoting was made available
1465 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1466 (notated using the quote facility) so as to distinguish them from IPv4
1467 keys. This meant that lookups for IP addresses in host lists did not work
1468 for iplsearch lookups.
1470 This has been fixed by arranging for IPv6 addresses to be expressed with
1471 colons if the lookup type is iplsearch. This is not incompatible, because
1472 previously such lookups could never work.
1474 The situation is now rather anomolous, since one *can* have colons in
1475 ordinary lsearch keys. However, making the change in all cases is
1476 incompatible and would probably break a number of configurations.
1478 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1481 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1482 conversion specification without a maximum field width, thereby enabling
1483 a rogue spamd server to cause a buffer overflow. While nobody in their
1484 right mind would setup Exim to query an untrusted spamd server, an
1485 attacker that gains access to a server running spamd could potentially
1486 exploit this vulnerability to run arbitrary code as the Exim user.
1488 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1489 $primary_hostname instead of what libspf2 thinks the hosts name is.
1491 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1492 a directory entry by the name of the lookup key. Previously, if a
1493 symlink pointed to a non-existing file or a file in a directory that
1494 Exim lacked permissions to read, a lookup for a key matching that
1495 symlink would fail. Now it is enough that a matching directory entry
1496 exists, symlink or not. (Bugzilla 503.)
1498 PH/03 The body_linecount and body_zerocount variables are now exported in the
1501 PH/04 Added the $dnslist_matched variable.
1503 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1504 This means they are set thereafter only if the connection becomes
1507 PH/06 Added the client_condition to authenticators so that some can be skipped
1508 by clients under certain conditions.
1510 PH/07 The error message for a badly-placed control=no_multiline_responses left
1511 "_responses" off the end of the name.
1513 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1515 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1516 (without spaces) instead of just copying the configuration text.
1518 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1520 PH/11 Added $max_received_linelength.
1522 PH/12 Added +ignore_defer and +include_defer to host lists.
1524 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1525 way in which PCRE > 7.0 is built.
1527 PH/14 Implemented queue_only_load_latch.
1529 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1530 MAIL command. The effect was to mangle the value on 64-bit systems.
1532 PH/16 Another patch from the Sieve maintainer.
1534 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1536 PH/18 If a system quota error occurred while trying to create the file for
1537 a maildir delivery, the message "Mailbox is full" was not appended to the
1538 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1539 only to a quota excession during the actual writing of the file.
1541 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1542 characters?) which causes problems in log lines. The DN values are now
1543 passed through string_printing() before being added to log lines.
1545 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1546 and InterBase are left for another time.)
1548 PH/21 Added message_body_newlines option.
1550 PH/22 Guard against possible overflow in moan_check_errorcopy().
1552 PH/23 POSIX allows open() to be a macro; guard against that.
1554 PH/24 If the recipient of an error message contained an @ in the local part
1555 (suitably quoted, of course), incorrect values were put in $domain and
1556 $local_part during the evaluation of errors_copy.
1562 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1563 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1566 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1567 issue a MAIL command.
1569 PH/02 In an ACL statement such as
1571 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1573 if a client was not listed at all, or was listed with a value other than
1574 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1575 the condition was not true (as it should be), so access was not denied.
1576 The bug was that the ! inversion was incorrectly passed on to the second
1577 item. This has been fixed.
1579 PH/03 Added additional dnslists conditions == and =& which are different from
1580 = and & when the dns lookup returns more than one IP address.
1582 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1583 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1585 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1586 FSYNC, which compiles an option called disable_fsync that allows for
1587 bypassing fsync(). The documentation is heavily laced with warnings.
1589 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1591 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1592 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1593 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1594 including adding "make clean"; (3) Added -fPIC when compiling the test
1595 dynamically loaded module, to get rid of a warning.
1597 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1598 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1599 = 0s. The bug is otherwise harmless.
1601 PH/07 There was a bug in the dovecot authenticator such that the value of
1602 $auth1 could be overwritten, and so not correctly preserved, after a
1603 successful authentication. This usually meant that the value preserved by
1604 the server_setid option was incorrect.
1606 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1608 PH/09 Installed PCRE release 7.0.
1610 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1611 run for batched SMTP input. It is now run at the start of every message
1612 in the batch. While fixing this I discovered that the process information
1613 (output by running exiwhat) was not always getting set for -bs and -bS
1614 input. This is fixed, and it now also says "batched" for BSMTP.
1616 PH/11 Added control=no_pipelining.
1618 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1619 patch, slightly modified), and move the expansion of helo_data till after
1620 the connection is made in the smtp transport (so it can use these
1623 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1625 PH/14 Added log_selector = +pid.
1627 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1629 PH/16 Add ${if forany and ${if forall.
1631 PH/17 Added dsn_from option to vary the From: line in DSNs.
1633 PH/18 Flush SMTP output before performing a callout, unless control =
1634 no_callout_flush is set.
1636 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1637 was true (the default) a successful delivery failed to delete the retry
1638 item, thus causing premature timeout of the address. The bug is now
1641 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1643 PH/21 Long custom messages for fakedefer and fakereject are now split up
1644 into multiline reponses in the same way that messages for "deny" and
1645 other ACL rejections are.
1647 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1648 with slight modification.
1650 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1651 draft, changing the syntax and factoring some duplicate code".
1653 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1654 for deliveries of the second and subsequent messages over the same SMTP
1657 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1658 ${reduce, with only minor "tidies".
1660 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1662 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1663 expansion side effects.
1665 PH/27 When a message times out after an over-quota error from an Exim-imposed
1666 quota, the bounce message says "mailbox is full". This message was not
1667 being given when it was a system quota that was exceeded. It now should
1670 MH/03 Made $recipients available in local_scan(). local_scan() already has
1671 better access to the recipient list through recipients_list[], but
1672 $recipients can be useful in postmaster-provided expansion strings.
1674 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1675 in the case of a MAIL command with additional options following the
1676 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1677 were accidentally chopped off.
1679 PH/29 SMTP synchronization checks are implemented when a command is read -
1680 there is a check that no more input is waiting when there shouldn't be
1681 any. However, for some commands, a delay in an ACL can mean that it is
1682 some time before the response is written. In this time, more input might
1683 arrive, invalidly. So now there are extra checks after an ACL has run for
1684 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1685 pipelining has not been advertised.
1687 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1689 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1690 returned in the SMTP response when smtp_return_error_details was false.
1691 This has been fixed.
1693 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1694 instead of the C I/O that was originally supplied, because problems were
1695 reported on Solaris.
1697 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1698 Exim which did not show up earlier: it was assuming that a call to
1699 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1700 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1701 was a macro that became an assignment, so it seemed to work. This has
1702 changed to a proper function call with a void return, hence the compile
1703 error. Exim's code has been fixed.
1705 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1708 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1710 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1712 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1713 because it turns out to be too big or there is a timeout) there is no
1714 "Completed" line in the log. When some messages of this type were
1715 selected by exigrep, they were listed as "not completed". Others were
1716 picked up by some special patterns. I have improved the selection
1717 criteria to be more general.
1719 PH/38 The host_find_failed option in the manualroute router can now be set
1720 to "ignore", to completely ignore a host whose IP address cannot be
1721 found. If all hosts are ignored, the behaviour is controlled by the new
1722 host_all_ignored option.
1724 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1725 homing or because of multiple MX records with /mx) generated more than
1726 one IP address, and the following item turned out to be the local host,
1727 all the secondary addresses of the first item were incorrectly removed
1728 from the list, along with the local host and any following hosts (which
1729 is what is supposed to happen).
1731 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1732 whoever called Exim into the -H file. In the case of the daemon it was
1733 behaving confusingly. When first started, it used values for whoever
1734 started the daemon, but after a SIGHUP it used the Exim user (because it
1735 calls itself on a restart). I have changed the code so that it now always
1738 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1739 message are rejected with the same error (e.g. no authentication or bad
1740 sender address), and a DATA command is nevertheless sent (as can happen
1741 with PIPELINING or a stupid MUA), the error message that was given to the
1742 RCPT commands is included in the rejection of the DATA command. This is
1743 intended to be helpful for MUAs that show only the final error to their
1746 PH/42 Another patch from the Sieve maintainer.
1748 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1749 Eximstats - Fixed some broken HTML links and added missing column headers
1751 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1752 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1754 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1756 PH/43 Yet another patch from the Sieve maintainer.
1758 PH/44 I found a way to check for a TCP/IP connection going away before sending
1759 the response to the final '.' that terminates a message, but only in the
1760 case where the client has not sent further data following the '.'
1761 (unfortunately, this is allowed). However, in many cases there won't be
1762 any further data because there won't be any more messages to send. A call
1763 to select() can be used: if it shows that the input is "ready", there is
1764 either input waiting, or the socket has been closed. An attempt to read
1765 the next input character can distinguish the two cases. Previously, Exim
1766 would have sent an OK response which the client would never have see.
1767 This could lead to message repetition. This fix should cure that, at
1768 least in a lot of common cases.
1770 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1771 advertised in response to EHLO.
1777 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1778 fixed by 4.65/MH/01 (is this a record?) are fixed:
1780 (i) An empty string was always treated as zero by the numeric comparison
1781 operators. This behaviour has been restored.
1783 (ii) It is documented that the numeric comparison operators always treat
1784 their arguments as decimal numbers. This was broken in that numbers
1785 starting with 0 were being interpreted as octal.
1787 While fixing these problems I realized that there was another issue that
1788 hadn't been noticed. Values of message_size_limit (both the global option
1789 and the transport option) were treated as octal if they started with 0.
1790 The documentation was vague. These values are now always treated as
1791 decimal, and I will make that clear in the documentation.
1797 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1798 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1801 MH/01 Don't check that the operands of numeric comparison operators are
1802 integers when their expansion is in "skipping" mode (fixes bug
1803 introduced by 4.64-PH/07).
1805 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1806 child addresses, Exim now panics and dies. Previously, because the count
1807 is held in a short int, deliveries were likely to be lost. As such a
1808 large number of recipients for a single message is ridiculous
1809 (performance will be very, very poor), I have chosen to impose a limit
1810 rather than extend the field.
1816 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1817 leftover -K file (the existence of which was triggered by #402).
1818 While we were at it, introduced process PID as part of the -K
1819 filename. This should rule out race conditions when creating
1822 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1823 processing considerably. Previous code took too long for large mails,
1824 triggering a timeout which in turn triggers #401.
1826 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1827 in the DK code in transports.c. sendfile() is not really portable,
1828 hence the _LINUX specificness.
1830 TF/01 In the add_headers option to the mail command in an Exim filter,
1831 there was a bug that Exim would claim a syntax error in any
1832 header after the first one which had an odd number of characters
1835 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1836 callout verification, Exim cached a "reject" for the entire domain. This
1837 is correct for most verifications, but it is not correct for a recipient
1838 verification with use_sender or use_postmaster set, because in that case
1839 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1840 case of MAIL FROM:<> rejection from other early rejections (e.g.
1841 rejection of HELO). When verifying a recipient using a non-null MAIL
1842 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1843 Whatever the result of the callout, the value of the domain cache is
1844 left unchanged (for any other kind of callout, getting as far as trying
1845 RCPT means that the domain itself is ok).
1847 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1850 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1851 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1852 session, the daemon ignores ECONNECT errors and logs others; it now
1853 ignores EPIPE as well.
1855 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1856 (quoted-printable decoding).
1858 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1859 later the small subsequent patch to fix an introduced bug.
1861 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1863 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1865 PH/08 An error is now given if message_size_limit is specified negative.
1867 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1868 to be given (somewhat) arbitrary names.
1870 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1873 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1874 miscellaneous code fixes
1876 PH/10 Added the log_reject_target ACL modifier to specify where to log
1879 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1880 hostname. This is wrong, because it relates to the incoming message (and
1881 probably the interface on which it is arriving) and not to the outgoing
1882 callout (which could be using a different interface). This has been
1883 changed to use the value of the helo_data option from the smtp transport
1884 instead - this is what is used when a message is actually being sent. If
1885 there is no remote transport (possible with a router that sets up host
1886 addresses), $smtp_active_hostname is used.
1888 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1889 tweaks were necessary in order to get it to work (see also 21 below):
1890 (a) The code assumed that strncpy() returns a negative number on buffer
1891 overflow, which isn't the case. Replaced with Exim's string_format()
1893 (b) There were several signed/unsigned issues. I just did the minimum
1894 hacking in of casts. There is scope for a larger refactoring.
1895 (c) The code used strcasecmp() which is not a standard C function.
1896 Replaced with Exim's strcmpic() function.
1897 (d) The code set only $1; it now sets $auth1 as well.
1898 (e) A simple test gave the error "authentication client didn't specify
1899 service in request". It would seem that Dovecot has changed its
1900 interface. Fortunately there's a specification; I followed it and
1901 changed what the client sends and it appears to be working now.
1903 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1906 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1907 address A is aliased to B and C, where B exists and C does not. Without
1908 -v the output is "A verified" because verification stops after a
1909 successful redirection if more than one address is generated. However,
1910 with -v the child addresses are also verified. Exim was outputting "A
1911 failed to verify" and then showing the successful verification for C,
1912 with its parentage. It now outputs "B failed to verify", showing B's
1913 parentage before showing the successful verification of C.
1915 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1916 look up a TXT record in a specific list after matching in a combined
1919 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1920 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1921 they consult the DNS. I had assumed they would set it the way they
1922 wanted; and indeed my experiments on Linux seem to show that in some
1923 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1924 To be on the safe side, however, I have now made the interface to
1925 host_find_byname() similar to host_find_bydns(), with an argument
1926 containing the DNS resolver options. The host_find_byname() function now
1927 sets these options at its start, just as host_find_bydns() does. The smtp
1928 transport options dns_qualify_single and dns_search_parents are passed to
1929 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
1930 of host_find_byname() use the default settings of RES_DEFNAMES
1931 (qualify_single) but not RES_DNSRCH (search_parents).
1933 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
1934 spool_read_header() do less string testing, by means of a preliminary
1935 switch on the second character of optional "-foo" lines. (This is
1936 overdue, caused by the large number of possibilities that now exist.
1937 Originally there were few.) While I was there, I also converted the
1938 str(n)cmp tests so they don't re-test the leading "-" and the first
1939 character, in the hope this might squeeze out yet more improvement.
1941 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
1942 flag allowing group syntax was set by the header_syntax check but not
1943 turned off, possible causing trouble later; (2) The flag was not being
1944 set at all for the header_verify test, causing "group"-style headers to
1945 be rejected. I have now set it in this case, and also caused header_
1946 verify to ignore an empty address taken from a group. While doing this, I
1947 came across some other cases where the code for allowing group syntax
1948 while scanning a header line wasn't quite right (mostly, not resetting
1949 the flag correctly in the right place). These bugs could have caused
1950 trouble for malformed header lines. I hope it is now all correct.
1952 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
1953 with the "reply" argument non-NULL. The code, however (which originally
1954 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
1955 but it didn't always do it. This confused somebody who was copying the
1956 code for some other use. I have removed all the tests.
1958 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
1959 feature that was used to support insecure browsers during the U.S. crypto
1960 embargo. It requires special client support, and Exim is probably the
1961 only MTA that supported it -- and would never use it because real RSA is
1962 always available. This code has been removed, because it had the bad
1963 effect of slowing Exim down by computing (never used) parameters for the
1964 RSA_EXPORT functionality.
1966 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
1967 authenticator to fail if there's a tab character in the incoming data
1968 (there should never be unless someone is messing about, as it's supposed
1969 to be base64-encoded). Also added, on Timo's advice, the "secured" option
1970 if the connection is using TLS or if the remote IP is the same as the
1971 local IP, and the "valid-client-cert option" if a client certificate has
1974 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
1975 authenticators. This can be used for authorization after authentication
1976 succeeds. (In the case of plaintext, it servers for both authentication
1979 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1980 if any retry times were supplied.
1982 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1983 connection, where there is no client IP address to check. In this
1984 situation, the verify now always succeeds.
1986 PH/25 Applied John Jetmore's -Mset patch.
1988 PH/26 Added -bem to be like -Mset, but loading a message from a file.
1990 PH/27 In a string expansion for a processed (not raw) header when multiple
1991 headers of the same name were present, leading whitespace was being
1992 removed from all of them, but trailing whitespace was being removed only
1993 from the last one. Now trailing whitespace is removed from each header
1994 before concatenation. Completely empty headers in a concatenation (as
1995 before) are ignored.
1997 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1998 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2000 PH/29 [Removed. This was a change that I later backed out, and forgot to
2001 correct the ChangeLog entry (that I had efficiently created) before
2002 committing the later change.]
2004 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2005 address errors (4xx response to RCPT) over the same connection as other
2006 messages routed to the same hosts. Such deliveries are always "forced",
2007 so retry times are not inspected. This resulted in far too many retries
2008 for the affected addresses. The effect occurred only when there were more
2009 hosts than the hosts_max_try setting in the smtp transport when it had
2010 the 4xx errors. Those hosts that it had tried were not added to the list
2011 of hosts for which the message was waiting, so if all were tried, there
2012 was no problem. Two fixes have been applied:
2014 (i) If there are any address or message errors in an SMTP delivery, none
2015 of the hosts (tried or untried) are now added to the list of hosts
2016 for which the message is waiting, so the message should not be a
2017 candidate for sending over the same connection that was used for a
2018 successful delivery of some other message. This seems entirely
2019 reasonable: after all the message is NOT "waiting for some host".
2020 This is so "obvious" that I'm not sure why it wasn't done
2021 previously. Hope I haven't missed anything, but it can't do any
2022 harm, as the worst effect is to miss an optimization.
2024 (ii) If, despite (i), such a delivery is accidentally attempted, the
2025 routing retry time is respected, so at least it doesn't keep
2026 hammering the server.
2028 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2029 in ${readsocket because some servers need this prod.
2031 PH/32 Added some extra debug output when updating a wait-xxx database.
2033 PH/33 The hint "could be header name not terminated by colon", which has been
2034 given for certain expansion errors for a long time, was not being given
2035 for the ${if def:h_colon_omitted{... case.
2037 PH/34 The spec says: "With one important exception, whenever a domain list is
2038 being scanned, $domain contains the subject domain." There was at least
2039 one case where this was not true.
2041 PH/35 The error "getsockname() failed: connection reset by peer" was being
2042 written to the panic log as well as the main log, but it isn't really
2043 panic-worthy as it just means the connection died rather early on. I have
2044 removed the panic log writing for the ECONNRESET error when getsockname()
2047 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2048 runs only) independently of the message's sender address. This meant
2049 that, if the 4xx error was in fact related to the sender, a different
2050 message to the same recipient with a different sender could confuse
2051 things. In particualar, this can happen when sending to a greylisting
2052 server, but other circumstances could also provoke similar problems.
2053 I have changed the default so that the retry time for these errors is now
2054 based a combination of the sender and recipient addresses. This change
2055 can be overridden by setting address_retry_include_sender=false in the
2058 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2059 remote server are returned as part of bounce messages. This was not
2060 happening for LMTP over a pipe (the lmtp transport), but now it is the
2061 same for both kinds of LMTP.
2063 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2064 in header lines that were in fact CNAMEs. This is no longer the case.
2066 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2067 and queue runs started by the daemon processed all messages. This has
2068 been fixed so that -R and -S can now usefully be given with -q<time>.
2070 PH/40 Import PCRE release 6.7 (fixes some bugs).
2072 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2074 PH/42 Give an error if -q is specified more than once.
2076 PH/43 Renamed the variables $interface_address and $interface_port as
2077 $received_ip_address and $received_port, to make it clear that these
2078 values apply to message reception, and not to the outgoing interface when
2079 a message is delivered. (The old names remain recognized, of course.)
2081 PH/44 There was no timeout on the connect() call when using a Unix domain
2082 socket in the ${readsocket expansion. There now is.
2084 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2085 be meaningful with "accept".
2087 SC/01 Eximstats V1.43
2088 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2090 SC/02 Eximstats V1.44
2091 Use a glob alias rather than an array ref in the generated
2092 parser. This improves both readability and performance.
2094 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2095 Collect SpamAssassin and rejection statistics.
2096 Don't display local sender or destination tables unless
2097 there is data to show.
2098 Added average volumes into the top table text output.
2100 SC/04 Eximstats V1.46
2101 Collect data on the number of addresses (recipients)
2102 as well as the number of messages.
2104 SC/05 Eximstats V1.47
2105 Added 'Message too big' to the list of mail rejection
2106 reasons (thanks to Marco Gaiarin).
2108 SC/06 Eximstats V1.48
2109 Mainlog lines which have GMT offsets and are too short to
2110 have a flag are now skipped.
2112 SC/07 Eximstats V1.49 (Alain Williams)
2113 Added the -emptyok flag.
2115 SC/08 Eximstats V1.50
2116 Fixes for obtaining the IP address from reject messages.
2118 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2119 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2120 whitesspace changes from 4.64-PH/27
2122 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2125 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2126 are found, allow negative numbers in numeric criteria)
2128 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2130 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2131 to match changes made in 4.64-PH/43
2133 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2135 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2136 "domain" is actually an IP address, and force a failure. This locks out
2137 those revolvers/nameservers that support "A-for-A" lookups, in
2138 contravention of the specifications.
2140 PH/48 When a host name was looked up from an IP address, and the subsequent
2141 forward lookup of the name timed out, the host name was left in
2142 $sender_host_name, contrary to the specification.
2144 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2145 restricted to single-key lookups, Exim was not diagnosing an error if
2146 * or *@ was used with a query-style lookup.
2148 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2150 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2151 long ago, but noone interested enough thought of it. Let's just say that
2152 the "1.1" means that there are some new functions that weren't there at
2153 some point in the past.
2155 PH/51 Error processing for expansion failure of helo_data from an smtp
2156 transport during callout processing was broken.
2158 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2159 tested/used via the -bh/-bhc/-bs options.
2161 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2162 bug, fixed in subsequent PCRE releases).
2164 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2165 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2167 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2173 SC/01 Use a glob alias rather than an array ref in eximstats generated
2174 parser. This improves both readability and performance.
2176 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2177 Don't display local sender or destination tables in eximstats unless
2178 there is data to show.
2179 Added average volumes into the eximstats top table text output.
2181 SC/03 Collect data on the number of addresses (recipients) as well
2182 as the number of messages in eximstats.
2184 TF/01 Correct an error in the documentation for the redirect router. Exim
2185 does (usually) call initgroups() when daemonizing.
2187 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2188 with consistent privilege compared to when running as a daemon.
2190 TF/03 Note in the spec that $authenticated_id is not set for local
2191 submissions from trusted users.
2193 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2194 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2196 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2197 by adding some example configuration directives to the default
2198 configuration file. A little bit of work is required to uncomment the
2199 directives and define how usernames and passwords are checked, but
2200 there is now a framework to start from.
2202 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2203 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2204 without this. I don't know how relevant this is to other LDAP libraries.
2206 PH/02 Add the verb name to the "unknown ACL verb" error.
2208 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2210 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2212 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2213 directory not expanded when it should be if an expanded home directory
2214 was set for the address (which is overridden by the transport).
2216 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2219 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2220 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2221 because it is too late at that time, and has no effect.
2223 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2224 security issue with \' (bugzilla #107). I could not use the
2225 PQescapeStringConn() function, because it needs a PGconn value as one of
2228 PH/08 When testing addresses using -bt, indicate those final addresses that
2229 are duplicates that would not cause an additional delivery. At least one
2230 person was confused, thinking that -bt output corresponded to deliveries.
2231 (Suppressing duplicates isn't a good idea as you lose the information
2232 about possibly different redirections that led to the duplicates.)
2234 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2235 systems where poll() doesn't work, in particular OS X.
2237 PH/10 Added more information to debugging output for retry time not reached.
2239 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2240 operations in malware.c.
2242 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2245 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2246 syslog_duplication set false, log lines that would normally be written
2247 both the the main log and to the reject log were not written to syslog at
2250 PH/14 In the default configuration, change the use of "message" in ACL warn
2251 statements to "add_header".
2253 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2254 not followed by a command (e.g. "seen endif").
2256 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2257 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2260 PH/17 Added extra conditions to the default value of delay_warning_condition
2264 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2265 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2266 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2269 The Auto-Submitted: and various List- headers are standardised, whereas I
2270 don't think Precedence: ever was.
2272 PH/18 Refactored debugging code in route_finduser() to show more information,
2273 in particular, the error code if getpwnam() issues one.
2275 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2276 This is apparently needed in addition to the PH/07 change above to avoid
2277 any possible encoding problems.
2279 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2280 but not after initializing Perl.
2282 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2283 output them only if debugging. By default they are written stderr,
2284 apparently, which is not desirable.
2286 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2289 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2292 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2294 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2295 authenticated or an ident call has been made. Suppress the default
2296 values for $authenticated_id and $authenticated_sender (but permit -oMai
2297 and -oMas) when testing with -bh.
2299 PH/24 Re-jigged the order of the tests in the default configuration so that the
2300 tests for valid domains and recipients precede the DNS black list and CSA
2301 tests, on the grounds that those ones are more expensive.
2303 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2304 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2305 This bug exists in every version of Exim that I still have, right back to
2308 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2309 However, an attempt to turn on case-sensitivity in a regex key by
2310 including (?-i) didn't work because the subject string was already
2311 lowercased, and the effects were non-intuitive. It turns out that a
2312 one-line patch can be used to allow (?-i) to work as expected.
2318 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2319 other effects) broke the use of negated acl sub-conditions.
2321 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2324 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2325 "Deny" causes Exim to reject the incoming connection with a 554 error.
2326 Unfortunately, if there is a major crisis, such as a disk failure,
2327 tcp-wrappers gives "deny", whereas what one would like would be some
2328 kind of temporary error. A kludge has been added to help with this.
2329 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2330 554 error is used if errno is still zero or contains ENOENT (which occurs
2331 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2334 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2336 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2337 errors. Otherwise a message that provokes a temporary error (when other
2338 messages do not) can cause a whole host to time out.
2340 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2341 addresses were routed directly to files or pipes from a redirect router.
2342 File deliveries just didn't batch; pipe deliveries might have suffered
2345 PH/06 A failure to get a lock for a hints database would erroneously always say
2346 "Failed to get write lock", even when it was really a read lock.
2348 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2349 of 0600. This has been changed to use the value of the lockfile_mode
2350 option (which defaults to 0600).
2352 PH/08 Applied small patch from the Sieve maintainer.
2354 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2355 folder from quota calculations, a direct delivery into this folder messed
2356 up the contents of the maildirsize file. This was because the regex was
2357 used only to exclude .Trash (or whatever) when the size of the mailbox
2358 was calculated. There was no check that a delivery was happening into an
2359 excluded directory. This bug has been fixed by ignoring all quota
2360 processing for deliveries into excluded directories.
2362 PH/10 Added the maildirfolder_create_regex option to appendfile.
2368 PH/01 The code for finding all the local interface addresses on a FreeBSD
2369 system running IPv6 was broken. This may well have applied to all BSD
2370 systems, as well as to others that have similar system calls. The broken
2371 code found IPv4 interfaces correctly, but gave incorrect values for the
2372 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2373 that it would not match correctly against @[] and not recognize the IPv6
2376 PH/02 The ipliteral router was not recognizing addresses of the form user@
2377 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2379 PH/03 Added disable_ipv6.
2381 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2382 decoded form, because it is most often used to construct To: headers
2383 lines in autoreplies, and the decoded form may well be syntactically
2384 invalid. However, $reply_address has leading white space removed, and all
2385 newlines turned into spaces so that the autoreply transport does not
2388 PH/05 If group was specified without a user on a router, and no group or user
2389 was specified on a transport, the group from the router was ignored.
2391 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2392 for visible compile-time settings that can be used to change these
2393 numbers, for those that want even more. Backwards compatibility with old
2394 spool files has been maintained. However, going back to a previous Exim
2395 release will lost any variables that are in spool files.
2397 PH/07 Two small changes when running in the test harness: increase delay when
2398 passing a TCP/IP connection to a new process, in case the original
2399 process has to generate a bounce, and remove special handling of
2400 127.0.0.2 (sic), which is no longer necessary.
2402 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2403 be the same on different OS.
2405 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2408 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2409 whether --show-vars was specified or not
2411 JJ/02 exipick: Added support for new ACL variable spool format introduced
2414 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2415 syntactically invalid From: or Reply-to: line, and a filter used this to
2416 generate an autoreply, and therefore failed to obtain an address for the
2417 autoreply, Exim could try to deliver to a non-existent relative file
2418 name, causing unrelated and misleading errors. What now happens is that
2419 it logs this as a hard delivery error, but does not attempt to create a
2422 PH/11 The exinext utility has a -C option for testing purposes, but although
2423 the given file was scanned by exinext itself; it wasn't being passed on
2424 when Exim was called.
2426 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2427 an end-of-file indication when reading a command response.
2429 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2430 compiled. In many other places in Exim, IPv6 addresses are always
2431 recognized, so I have changed this. It also means that IPv4 domain
2432 literals of the form [IPV4:n.n.n.n] are now always recognized.
2434 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2435 used if the router is not running as root, for example, when verifying at
2436 ACL time, or when using -bh. The debugging output from this situation was
2437 non-existent - all you got was a failure to exec. I have made two
2440 (a) Failures to set uid/gid, the current directory, or a process leader
2441 in a subprocess such as that created by queryprogram now generate
2442 suitable debugging ouput when -d is set.
2444 (b) The queryprogram router detects when it is not running as root,
2445 outputs suitable debugging information if -d is set, and then runs
2446 the subprocess without attempting to change uid/gid.
2448 PH/15 Minor change to Makefile for building test_host (undocumented testing
2451 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2452 additional section of a DNS packet that returns MX or SRV records.
2453 Instead, it always explicitly searches for A/AAAA records. This avoids
2454 major problems that occur when a DNS server includes only records of one
2455 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2456 fixed another bug: if SRV records were looked up and the corresponding
2457 address records were *not* found in the additional section, the port
2458 values from the SRV records were lost.
2460 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2461 using the correct key (the original address) when searching the retry
2462 rules in order to find which one to use for generating the retry hint.
2464 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2465 adding the default one. Similarly, if it contains a Reply-To: header, the
2466 errors_reply_to option, if set, is not used.
2468 PH/19 When calculating a retry time, Exim used to measure the "time since
2469 failure" by looking at the "first failed" field in the retry record. Now
2470 it does not use this if it is later than than the arrival time of the
2471 message. Instead it uses the arrival time. This makes for better
2472 behaviour in cases where some deliveries succeed, thus re-setting the
2473 "first failed" field. An example is a quota failure for a huge message
2474 when small messages continue to be delivered. Without this change, the
2475 "time since failure" will always be short, possible causing more frequent
2476 delivery attempts for the huge message than are intended.
2477 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2479 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2480 $1, $2, $3) because the numerical variables can be reset during some
2481 expansion items (e.g. "match"), thereby losing the authentication data.
2483 PH/21 Make -bV show the size of off_t variables so that the test suite can
2484 decide whether to run tests for quotas > 2G.
2486 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2487 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2488 filecount value is greater than 2G or if a quota value is greater than 2G
2489 on a system where the size of off_t is not greater than 4, a panic error
2492 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2493 never match. The debug and -bh output now contains an explicit error
2494 message indicating a malformed IPv4 address or mask.
2496 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2497 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2498 PH/23 above applies.
2500 PH/25 Do not write to syslog when running in the test harness. The only
2501 occasion when this arises is a failure to open the main or panic logs
2502 (for which there is an explicit test).
2504 PH/26 Added the /no_tell option to "control=freeze".
2506 PH/27 If a host name lookup failed very early in a connection, for example, if
2507 the IP address matched host_lookup and the reverse lookup yielded a name
2508 that did not have a forward lookup, an error message of the form "no IP
2509 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2510 could be logged. Now it outputs the IP address instead of "NULL".
2512 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2513 allows the sender and the authenticated sender to be set when
2514 submitting a message from within Exim. Since child_open_exim() is
2515 documented for local_scan(), the new function should be too.
2517 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2518 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2519 results in an empty string is now treated as unset.
2521 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2523 PH/31 Added sender_verify_fail logging option.
2525 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2526 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2527 tidied the source and removed it altogether.
2529 PH/33 When a queue run was abandoned because the load average was too high, a
2530 log line was always written; now it is written only if the queue_run log
2531 selector is set. In addition, the log line for abandonment now contains
2532 information about the queue run such as the pid. This is always present
2533 in "start" and "stop" lines but was omitted from the "abandon" line.
2535 PH/34 Omit spaces between a header name and the colon in the error message that
2536 is given when verify = headers_syntax fails (if there are lots of them,
2537 the message gets confusing).
2539 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2540 names, as there are now some PTR records that contain slashes. This check
2541 is only to protect against broken name servers that fall over on strange
2542 characters, so the fact that it applies to all lookups doesn't matter.
2544 PH/36 Now that the new test suite is complete, we can remove some of the
2545 special code in Exim that was needed for the old test suite. For example,
2546 sorting DNS records because real resolvers return them in an arbitrary
2547 order. The new test suite's fake resolver always returns records in the
2550 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2551 bounces) except when queue_only is set, to avoid logging races between
2552 the different processes.
2554 PH/38 Panic-die if .include specifies a non-absolute path.
2556 PH/39 A tweak to the "H" retry rule from its user.
2558 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2559 a label. They prevented compilation on older perls.
2561 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2562 a warning to be raised on newish perls.
2564 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2565 on queue. Changes to match documented behaviour of showing count of
2566 messages matching specified criteria.
2568 PH/40 Changed the default ident timeout from 30s to 5s.
2570 PH/41 Added support for the use of login_cap features, on those BSD systems
2571 that have them, for controlling the resources used by pipe deliveries.
2573 PH/42 The content-scanning code uses fopen() to create files in which to put
2574 message data. Previously it was not paying any attention to the mode of
2575 the files. Exim runs with umask(0) because the rest of the code creates
2576 files with open(), and sets the required mode explicitly. Thus, these
2577 files were ending up world-writeable. This was not a big issue, because,
2578 being within the spool directory, they were not world-accessible. I have
2579 created a function called modefopen, which takes an additional mode
2580 argument. It sets umask(777), creates the file, chmods it to the required
2581 mode, then resets the umask. All the relevant calls to fopen() in the
2582 content scanning code have been changed to use this function.
2584 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2585 to 24 hours. This avoids potential overflow problems when processing G
2586 and H retry rules. I suspect nobody ever tinkers with this value.
2588 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2590 PH/45 When the plaintext authenticator is running as a client, the server's
2591 challenges are checked to ensure they are valid base64 strings. By
2592 default, the authentication attempt is cancelled if an invalid string is
2593 received. Setting client_ignore_invalid_base64 true ignores these errors.
2594 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2595 they are received. Thus, the responses can be made to depend on the
2596 challenges. If an invalid string is ignored, an empty string is placed in
2599 PH/46 Messages that are created by the autoreply transport now contains a
2600 References: header, in accordance with RFCs 2822 and 3834.
2602 PH/47 Added authenticated_sender_force to the smtp transport.
2604 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2606 PH/49 Installed latest patch from the Sieve maintainer.
2608 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2609 was also set, the appendfile transport was unnecessarily scanning a
2610 directory of message files (e.g. for maildir delivery) to find the count
2611 of files (along with the size), even though it did not need this
2612 information. It now does the scan only if it needs to find either the
2613 size of the count of files.
2615 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2617 PH/52 Two bugs concerned with error handling when the smtp transport is
2620 (i) Exim was not creating retry information for temporary errors given
2621 for individual recipients after the DATA command when the smtp transport
2622 was used in LMTP mode. This meant that they could be retried too
2623 frequently, and not timed out correctly.
2625 (ii) Exim was setting the flag that allows error details to be returned
2626 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2627 recipients that were returned after the DATA command.
2629 PH/53 This is related to PH/52, but is more general: for any failing address,
2630 when detailed error information was permitted to be returned to the
2631 sender, but the error was temporary, then after the final timeout, only
2632 "retry timeout exceeded" was returned. Now it returns the full error as
2633 well as "retry timeout exceeded".
2635 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2636 do this, and (what is worse) MTAs that accept it.
2638 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2639 will now be deprecated.
2641 PH/56 New os.c-cygwin from the Cygwin maintainer.
2643 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2644 formats (previously only available in exim formats via -bpr, -bpru,
2645 and -bpra. Now also available in native and exiqgrep formats)
2647 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2648 with very large, slow to parse queues
2650 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2652 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2654 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2655 responses to authentication challenges, though it was showing the
2656 challenges; (ii) I've removed the CR characters from the debug output for
2659 PH/58 Allow for the insertion of a newline as well as a space when a string
2660 is turned into more than one encoded-word during RFC 2047 encoding. The
2661 Sieve code now uses this.
2663 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2664 data_4xx, lost_connection, tls_required.
2666 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2667 message was being sent as an SMTP response.
2669 PH/61 Add -l and -k options to exicyclog.
2671 PH/62 When verifying, if an address was redirected to one new address, so that
2672 verification continued, and the new address failed or deferred after
2673 having set something in $address_data, the value of $address_data was not
2674 passed back to the ACL. This was different to the case when no
2675 redirection occurred. The value is now passed back in both cases.
2677 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2678 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2679 use login_cap.h, so on its own it isn't the distinguishing feature. The
2680 new name refers directly to the setclassresources() function.
2682 PH/65 Added configuration files for NetBSD3.
2684 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2686 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2687 is preferred over IPv4.
2689 PH/68 The bounce_return_message and bounce_return_body options were not being
2690 honoured for bounces generated during the reception of non-SMTP messages.
2691 In particular, this applied to messages rejected by the ACL. This bug has
2692 been fixed. However, if bounce_return_message is true and bounce_return_
2693 body is false, the headers that are returned for a non-SMTP message
2694 include only those that have been read before the error was detected.
2695 (In the case of an ACL rejection, they have all been read.)
2697 PH/69 The HTML version of the specification is now built in a directory called
2698 spec_html instead of spec.html, because the latter looks like a path with
2699 a MIME-type, and this confuses some software.
2701 PH/70 Catch two compiler warnings in sieve.c.
2703 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2704 function verify_get_ident() calls ip_connect() to connect a socket, but
2705 if the "connect()" function timed out, ip_connect() used to close the
2706 socket. However, verify_get_ident() also closes the socket later, and in
2707 between Exim writes to the log, which may get opened at this point. When
2708 the socket was closed in ip_connect(), the log could get the same file
2709 descriptor number as the socket. This naturally causes chaos. The fix is
2710 not to close the socket in ip_connect(); the socket should be closed by
2711 the function that creates it. There was only one place in the code where
2712 this was missing, in the iplookup router, which I don't think anybody now
2713 uses, but I've fixed it anyway.
2715 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2716 well as to direct DNS lookups. Otherwise the handling of names in host
2717 lists is inconsistent and therefore confusing.
2723 PH/01 Two changes to the default runtime configuration:
2725 (1) Move the checks for relay_from_hosts and authenticated clients from
2726 after to before the (commented out) DNS black list checks.
2728 (2) Add control=submission to the relay_from_hosts and authenticated
2729 clients checks, on the grounds that messages accepted by these
2730 statements are most likely to be submissions.
2732 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2734 (1) Generate an error if the third argument for the ${prvs expansion is
2737 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2740 (3) Reset the variables that are obtained from the first argument of
2741 ${prvscheck and used in the second argument before leaving the code,
2742 because their memory is reclaimed, so using them afterwards may do
2745 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2746 one (it's much easier than Tom thought :-).
2748 (5) Because of (4), we can now allow for the use of $prvscheck_result
2749 inside the third argument.
2751 PH/03 For some reason, the default setting of PATH when running a command from
2752 a pipe transport was just "/usr/bin". I have changed it to
2755 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2756 anything to be listed in the output from -bV.
2758 PH/05 When a filter generated an autoreply, the entire To: header line was
2759 quoted in the delivery log line, like this:
2761 => >A.N.Other <ano@some.domain> <original@ddress> ...
2763 This has been changed so that it extracts the operative address. There
2764 may be more than one such address. If so, they are comma-separated, like
2767 => >ano@some.domain,ona@other.domain <original@ddress> ...
2769 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2770 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2771 not being looked up in the rDNS to get a host name, Exim was showing the
2772 IP address twice in Received: lines, even though the IP addresses were
2773 identical. For example:
2775 Received: from [1.2.3.4] (helo=[1.2.3.4])
2777 However, if the real host name was known, it was omitting the HELO data
2778 if it matched the actual IP address. This has been tidied up so that it
2779 doesn't show the same IP address twice.
2781 PH/07 When both +timestamp and +memory debugging was on, the value given by
2782 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2783 called by the debug printing, thereby overwriting the timestamp buffer.
2784 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2786 PH/08 When the original message was included in an autoreply transport, it
2787 always said "this is a copy of the message, including all the headers",
2788 even if body_only or headers_only was set. It now gives an appropriate
2791 PH/09 Applied a patch from the Sieve maintainer which:
2793 o fixes some comments
2794 o adds the (disabled) notify extension core
2795 o adds some debug output for the result of if/elsif tests
2796 o points to the current vacation draft in the documentation
2797 and documents the missing references header update
2801 o fixes a bug in processing the envelope test (when testing
2802 multiple envelope elements, the last element determinted the
2805 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2806 Electronic Mail") by including:
2808 Auto-submitted: auto-generated
2810 in the messages that it generates (bounce messages and others, such as
2811 warnings). In the case of bounce messages for non-SMTP mesages, there was
2812 also a typo: it was using "Auto_submitted" (underscore instead of
2813 hyphen). Since every message generated by Exim is necessarily in response
2814 to another message, thes have all been changed to:
2816 Auto-Submitted: auto-replied
2818 in accordance with these statements in the RFC:
2820 The auto-replied keyword:
2822 - SHOULD be used on messages sent in direct response to another
2823 message by an automatic process,
2825 - MUST NOT be used on manually-generated messages,
2827 - MAY be used on Delivery Status Notifications (DSNs) and Message
2828 Disposition Notifications (MDNs),
2830 - MUST NOT be used on messages generated by automatic or periodic
2831 processes, except for messages which are automatic responses to
2834 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2835 to the default Received: header definition.
2837 PH/12 Added log selector acl_warn_skipped (default on).
2839 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2840 variables because (a) they are in the wrong storage pool and (b) even if
2841 they were copied, it wouldn't work properly because of the caching.
2843 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2844 checking when decoding. Apparently there are clients that generate
2845 overlong encoded strings. Why am I not surprised?
2847 PH/15 If the first argument of "${if match_address" was not empty, but did not
2848 contain an "@" character, Exim crashed. Now it writes a panic log message
2849 and treats the condition as false.
2851 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2853 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2854 extension "envelope-auth". The code is finished and in agreement with
2855 other implementations, but there is no documentation so far and in fact,
2856 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2857 not changing the active code.
2859 Print executed "if" and "elsif" statements when debugging is used. This
2860 helps a great deal to understand what a filter does.
2862 Document more things not specified clearly in RFC3028. I had all this
2863 sorted out, when out of a sudden new issues came to my mind. Oops."
2865 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2868 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2869 canonical form (as documented). However, after a host name lookup from
2870 the IP address, check_host() was doing a simple string comparison with
2871 addresses acquired from the DNS when checking that the found name did
2872 have the original IP as one of its addresses. Since any found IPv6
2873 addresses are likely to be in abbreviated form, the comparison could
2874 fail. Luckily, there already exists a function for doing the comparison
2875 by converting both addresses to binary, so now that is used instead of
2876 the text comparison.
2878 PH/20 There was another similar case to PH/19, when a complete host name was
2879 given in a host list; looking up its IP address could give an abbreviated
2880 form, whereas the current host's name might or might not be abbreviated.
2881 The same fix has been applied.
2887 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2888 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2891 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2892 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2894 PH/03 Typo: missing ".o" in src/pcre/Makefile.
2896 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2897 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2898 Also, for "Auto-Submitted", treat anything other than "no" as
2899 non-personal, in accordance with RFC 3834. (Previously it treated
2900 anything starting "auto-" as non-personal.)
2902 TF/01 The control=submission/name=... option had a problem with syntax
2903 errors if the name included a slash character. The /name= option
2904 now slurps the rest of the string, so it can include any characters
2905 but it must come last in the list of options (after /sender_retain
2908 PH/05 Some modifications to the interface to the fake nameserver for the new
2916 TK/01 Added the "success_on_redirect" address verification option. See
2917 NewStuff for rationale and an example.
2919 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2921 PH/02 Patch to exigrep to allow it to work on syslog lines.
2923 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2924 fread() to skip over the body file's header line, because in Cygwin the
2925 header line is locked and is inaccessible.
2927 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
2928 co-exist for some time) to make it clear that it is the Exim ID that is
2929 referenced, not the Message-ID: header line.
2931 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
2932 string_format() function, because snprintf() does not exist on all
2935 PH/06 The use of forbid_filter_existstest now also locks out the use of the
2936 ${stat: expansion item.
2938 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
2939 protocol synchronization error", to keep the pedants happy.
2941 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
2942 well as for IRIX systems, when gcc is being used. See the host.c source
2945 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
2947 PH/10 Named domain lists were not working if used in a queue_smtp_domains
2950 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
2951 transport and to the smtp transport in LMTP mode.
2953 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
2955 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
2956 run a filter in a subprocess. This could lead to confusion in subsequent
2957 lookups in the parent process. There should also be a search_tidyup() at
2958 the end of the subprocess.
2960 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
2961 only if the host matched helo_try_verify_hosts, which caused the
2962 verification to occur when the EHLO/HELO command was issued. The ACL just
2963 tested the remembered result. Now, if a previous verification attempt has
2964 not happened, "verify = helo" does it there and then.
2966 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
2968 TK/03 Fix log output including CR from clamd.
2970 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
2971 exist provoked a memory error which could cause a segfault.
2973 PH/15 Installed PCRE 6.2
2975 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2977 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2978 of the problem. Specifically, suggested +O2 rather than +O1 for the
2981 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2983 PH/20 If a delivery was routed to a non-standard port by means of an SRV
2984 record, the port was not correctly logged when the outgoing_port log
2985 selector was set (it logged the transort's default port).
2987 PH/21 Added support for host-specific ports to manualroute, queryprogram,
2988 fallback_hosts, and "hosts" in the smtp transport.
2990 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2991 host errors such as "Connection refused".
2993 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2994 authentication with radiusclient 0.4.9:
2996 - Error returned from rc_read_config was caught wrongly
2997 - Username/password not passed on to radius server due to wrong length.
2999 The presumption is that some radiusclient API changes for 4.51/PH/17
3000 were not taken care of correctly. The code is still untested by me (my
3001 Linux distribution still has 0.3.2 of radiusclient), but it was
3002 contributed by a Radius user.
3004 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3005 expanding the "port" option of the smtp transport.
3007 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3008 while reading a MIME header. Thanks to Tom Hughes for a patch.
3010 PH/24 Include config.h inside local_scan.h so that configuration settings are
3013 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3014 that in an ACL for RCPT (for example), you can examine exactly what was
3017 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3018 commands, but it was not correctly comparing the address with the actual
3019 client host address. Thus, it would show the EHLO address in Received:
3020 header lines when this was not necessary.
3022 PH/27 Added the % operator to ${eval:}.
3024 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3025 it should be ignoring failures (because with -C, for example, it has lost
3026 privilege). It wasn't ignoring creation failures other than "already
3029 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3032 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3033 because an input line was too long, either on its own, or by virtue of
3034 too many continuations, the temporary file was not being removed, and the
3035 return code was incorrect.
3037 PH/31 Missing "BOOL" in function definition in filtertest.c.
3039 PH/32 Applied Sieve patches from the maintainer.
3041 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3043 PH/33 Added "verify = not_blind".
3045 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3046 Local/Makefile (with some defaults set). These are used in built scripts
3047 such as exicyclog, but they have never been used in the exim_install
3048 script (though there are many overriding facilities there). I have
3049 arranged that the exim_install script now takes note of these two
3052 PH/35 Installed configuration files for Dragonfly.
3054 PH/36 When a locally submitted message by a trusted user did not contain a
3055 From: header, and the sender address was obtained from -f or from an SMTP
3056 MAIL command, and the trusted user did not use -F to supply a sender
3057 name, $originator_name was incorrectly used when constructing a From:
3058 header. Furthermore, $originator_name was used for submission mode
3059 messages from external hosts without From: headers in a similar way,
3060 which is clearly wrong.
3062 PH/37 Added control=suppress_local_fixups.
3064 PH/38 When log_selector = +received_sender was set, and the addition of the
3065 sender made the log line's construction buffer exactly full, or one byte
3066 less than full, an overflow happened when the terminating "\n" was
3069 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3070 when the result of a list match is failure because a DNS lookup failed.
3072 PH/40 RM_COMMAND is now used in the building process.
3074 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3075 the "build-* directories that it finds.
3077 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3078 address in a domain literal was a prefix of an interface address.
3080 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3081 when verifying a sender address, unless rewrite_headers is false.
3083 PH/44 Wrote a long comment about why errors_to addresses are verified as
3084 recipients, not senders.
3086 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3087 the ratelimit ACL was added.
3089 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3091 PH/46 Added extra information about PostgreSQL errors to the error string.
3093 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3094 avoiding the need to install special zones in a real server. This is
3095 backwards compatible; if it can't find the fake resolver, it drops back.
3096 Thus, both old and new test suites can be run.
3098 TF/02 Added util/ratelimit.pl
3100 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3101 clock is set back in time.
3103 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3104 Candler <B.Candler@pobox.com>.
3106 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3107 OK for addresses that are the result of redirections.
3109 PH/48 A number of further additions for the benefit of the new test suite,
3110 including a fake gethostbyname() that interfaces to the fake DNS resolver
3113 TF/06 The fix for widen_domains has also been applied to qualify_single and
3114 search_parents which are the other dnslookup options that can cause
3117 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3120 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3122 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3123 service" if it encountered an MX record with an empty target hostname.
3124 The message is now "an MX or SRV record indicated no SMTP service".
3126 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3127 if the DNS of the sending site is misconfigured. This is quite a
3128 common situation. This change restores the behaviour of treating a
3129 helo verification defer as a failure.
3131 PH/51 If self=fail was set on a router, the bounce message did not include the
3132 actual error message.
3138 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3140 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3141 command itself ended in error, the underlying message about the transport
3142 filter timeout was being overwritten with the pipe command error. Now the
3143 underlying error message should be appended to the second error message.
3145 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3147 PH/02 Reduce the amount of output that "make" produces by default. Full output
3148 can still be requested.
3150 PH/03 The warning log line about a condition test deferring for a "warn" verb
3151 was being output only once per connection, rather than after each
3152 occurrence (because it was using the same function as for successful
3153 "warn" verbs). This seems wrong, so I have changed it.
3155 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3156 it should not have, which might have caused a crash in the right
3157 circumstances, but probably never did.
3159 PH/04 Installed a modified version of Tony Finch's patch to make submission
3160 mode fix the return path as well as the Sender: header line, and to
3161 add a /name= option so that you can make the user's friendly name appear
3164 TF/03 Added the control = fakedefer ACL modifier.
3166 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3167 Mark Lowes for thorough testing.
3169 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3171 TK/03 Merged latest SRS patch from Miles Wilton.
3173 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3174 with the definition in sysexits.h (which is #included earlier).
3175 Fortunately, Exim does not actually use EX_OK. The code used to try to
3176 preserve the sysexits.h value, by assumimg that macro definitions were
3177 scanned for macro replacements. I have been disabused of this notion,
3178 so now the code just undefines EX_OK before #including unistd.h.
3180 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3181 in the smtp transport. When a block could not be written in a single
3182 write() function, the timeout was being re-applied to each part-write.
3183 This seems wrong - if the receiver was accepting one byte at a time it
3184 would take for ever. The timeout is now adjusted when this happens. It
3185 doesn't have to be particularly precise.
3187 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3188 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3190 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3191 without a domain if the check to <postmaster@domain> fails.
3193 SC/01 Eximstats: added -xls and the ability to specify output files
3194 (patch written by Frank Heydlauf).
3196 SC/02 Eximstats: use FileHandles for outputing results.
3198 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3200 SC/04 Eximstats: fixed display of large numbers with -nvr option
3202 SC/05 Eximstats: fixed merging of reports with empty tables.
3204 SC/06 Eximstats: added the -include_original_destination flag
3206 SC/07 Eximstats: removed tabs and trailing whitespace.
3208 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3210 TK/06 MBOX spool code: Add real "From " MBOX separator line
3211 so the .eml file is really in mbox format (even though
3212 most programs do not really care). Patch from Alex Miller.
3214 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3215 The latter is generated from $received_to and is only set if the
3216 message has one envelope recipient. SA can use these headers,
3217 obviously out-of-the-box. Patch from Alex Miller.
3219 PH/08 The ${def test on a variable was returning false if the variable's
3220 value was "0", contrary to what the specification has always said!
3221 The result should be true unless the variable is empty.
3223 PH/09 The syntax error of a character other than { following "${if
3224 def:variable_name" (after optional whitespace) was not being diagnosed.
3225 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3226 accidental colon was present, for example, could give incorrect results.
3228 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3229 result is used (not including appendfile, where other changes are about
3232 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3233 This involved changing a lot of size variables from int to off_t. It
3234 should work with maildirs and everything.
3236 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3237 spamd dying while we are connected to it.
3239 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3242 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3243 function for BDB 4.3.
3245 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3247 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3248 the sources for PCRE have been re-arranged and more files are now
3251 PH/15 The code I had for printing potentially long long variables in PH/11
3252 above was not the best (it lost precision). The length of off_t variables
3253 is now inspected at build time, and an appropriate printing format (%ld
3254 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3255 to be "long long int" or "long int". This is needed for the internal
3256 formatting function string_vformat().
3258 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3259 the configuration file to be ":syslog", then the script "guesses" where
3260 the logs files are, rather than using the compiled in default. In our
3261 case the guess is not the same as the compiled default, so the script
3262 suddenly stopped working when I started to use syslog. The patch checks
3263 to see if log_file_path is "". If so, it attempts to read it from exim
3264 with no configuration file to get the compiled in version, before it
3265 falls back to the previous guessing code."
3267 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3268 implementing BATV in an Exim configuration. See NewStuff for the gory
3271 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3272 Makefile that are specific to HP-UX.
3274 PH/18 If the "use_postmaster" option was set for a recipient callout together
3275 with the "random" option, the postmaster address was used as the MAIL
3276 FROM address for the random test, but not for the subsequent recipient
3277 test. It is now used for both.
3279 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3280 patch removes a few documentation additions to RFC 3028, because the
3281 latest draft now contains them. It adds the new en;ascii-case comparator
3282 and a new error check for 8bit text in MIME parts. Comparator and
3283 require names are now matched exactly. I enabled the subaddress
3284 extension, but it is not well tested yet (read: it works for me)."
3286 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3287 rework some of the code of TK/09 above to avoid the hardwired use of
3288 "%lld" and "long long". Replaced the call to snprintf() with a call to
3291 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3292 records point to non-existent hosts", "retry timeout exceeded", and
3293 "retry time not reached for any host after a long failure period".
3295 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3296 experimental DomainKeys support:
3298 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3299 (2) On an error such as an illegally used "control", the wrong name for
3300 the control was given.
3302 These problems did NOT occur unless DomainKeys support was compiled.
3304 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3306 PH/24 Added ${if match_ip condition.
3308 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3309 databases so that it will be absolutely obvious if a crash occurs in the
3310 DB library. This is a regular occurrence (often caused by mis-matched
3313 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3314 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3315 picked up on a user's system that detects such things. There doesn't seem
3316 to be a gcc warning option for this - only an attribute that has to be
3317 put on the function's prototype. It seems that in Fedora Core 4 they have
3318 set this on a number of new functions. No doubt there will be more in due
3321 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3322 specify a transport. However, if an address that was verified by such a
3323 router was the subject of a callout, Exim crashed because it tried to
3324 read the rcpt_include_affixes from the non-existent transport. Now it
3325 just assumes that the setting of that option is false. This bug was
3326 introduced by 4.51/PH/31.
3328 PH/28 Changed -d+all to exclude +memory, because that information is very
3329 rarely of interest, but it makes the output a lot bigger. People tend to
3330 do -d+all out of habit.
3332 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3333 code in os-type was giving problems when libc.so lives in lib64, like on
3336 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3337 aren't the modern standard, and it seems that some systems' include files
3338 don't always have them. Exim was already checking for some of the newer
3339 ones like T_AAAA, and defining it itself. I've added checks for all the
3340 record types that Exim uses.
3342 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3343 not automatically generating a new one, as it is supposed to. This
3344 prevented TLS from working. If the file did exist, but contained invalid
3345 data, a new version was generated, as expected. It was only the case of a
3346 non-existent file that was broken.
3348 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3349 with a change in libdomainkeys > 0.64.
3351 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3352 from DNS. If the selector record carries the flag, it now has
3353 precedence over the domain-wide flag.
3355 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3357 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3358 the use of a port name that isn't defined in /etc/services) occurred, the
3359 message was deferred as in a normal delivery, and thus remained on the
3360 spool, instead of being failed because of the mua_wrapper setting. This
3361 is now fixed, and I tidied up some of the mua_wrapper messages at the
3364 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3365 the messages in a hash of arrays rather than using individual hashes.
3366 This is a bit cleaner and results in dramatic memory savings, albeit
3367 at a slight CPU cost.
3369 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3370 as requested by Marc Sherman.
3372 SC/10 Eximstats: added histograms for user specified patterns as requested
3375 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3377 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3378 fopen() in the content-scanning modules that did not already have it.
3384 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3385 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3387 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3389 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3391 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3392 to test Sieve filters that use "vacation".
3394 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3395 that changes the way the GnuTLS parameters are stored in the cache file.
3396 The new format can be generated externally. For backward compatibility,
3397 if the data in the cache doesn't make sense, Exim assumes it has read an
3398 old-format file, and it generates new data and writes a new file. This
3399 means that you can't go back to an older release without removing the
3402 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3403 work if there are any delivery delays because "one_time" forces the
3404 parent to be marked "delivered", so its unseen clone is never tried
3405 again. For this reason, Exim now forbids the simultaneous setting of
3408 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3409 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3410 entry if you want to know the details. The fix, however, neglected to
3411 consider the case when local delivery batching is involved. The test for
3412 "previously delivered" was not happening when checking to see if an
3413 address could be batched with a previous (undelivered) one; under
3414 certain circumstances this could lead to multiple deliveries to the same
3417 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3418 in its include files, and this causes problems building Exim.
3420 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3421 header_syntax) but Exim was just ignoring anything given after a slash.
3422 In particular, this caused confusion with an attempt to use "verify =
3423 reverse_host_lookup/defer_ok". An error is now given when options are
3424 supplied for verify items that do not have them. (Maybe reverse_host_
3425 lookup should have a defer_ok option, but that's a different point.)
3427 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3428 defined by RFC 821) to 2048, because there were problems with some AUTH
3429 commands, and RFC 1869 says the size should be increased for extended
3430 SMTP commands that take arguments.
3432 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3435 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3436 "unknown" error; now it says that the functionality isn't in the binary.
3438 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3439 an address' error message when a string expansion fails (syntax or
3440 whatever). Otherwise the password may appear in the log. Following change
3441 PH/42 below, there is no longer a chance of it appearing in a bounce
3444 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3446 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3447 of its addresses was ever tried. (Bugzilla bug #2.)
3449 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3450 the result incorrectly in the debug output. (It correctly added a newline
3451 to what was transported.)
3453 TF/01 Added $received_time.
3455 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3456 commented out examples of how to interface to a virus scanner and to
3457 SpamAssassin. Also added commented examples of av_scanner and
3458 spamd_address settings.
3460 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3461 and controls are allowed in which ACLs. There were a couple of minor
3462 errors. Some of the entries in the conditions table (which is a table of
3463 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3464 negation of where the condition IS allowed.
3466 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3468 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3469 header file does not have a version number, so I've had to invent a new
3470 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3471 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3472 radiusclient), but it was contributed by a Radius user.
3474 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3475 files or directories, for OpenSSL.
3477 PH/19 When an Exim process that is running as root has to create an Exim log
3478 file, it does so in a subprocess that runs as exim:exim so as to get the
3479 ownership right at creation (otherwise, other Exim processes might see
3480 the file with the wrong ownership). There was no test for failure of this
3481 fork() call, which would lead to the process getting stuck as it waited
3482 for a non-existent subprocess. Forks do occasionally fail when resources
3483 run out. I reviewed all the other calls to fork(); they all seem to check
3486 PH/20 When checking for unexpected SMTP input at connect time (before writing
3487 the banner), Exim was not dealing correctly with a non-positive return
3488 from the read() function. If the client had disconnected by this time,
3489 the result was a log entry for a synchronization error with an empty
3490 string after "input=" when read() returned zero. If read() returned -1
3491 (an event I could not check), uninitialized data bytes were printed.
3492 There were reports of junk text (parts of files, etc) appearing after
3495 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3497 PH/22 Added support for macro redefinition, and (re)definition in between
3498 driver and ACL definitions.
3500 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3501 forgetting to use the resulting value; it was using the unexpanded value.
3503 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3504 hadn't been configured. The fix is from Juergen Kreileder, who
3505 understands it better than I do:
3507 "Here's what I see happening with three configured cyrus_sasl
3508 authenticators configured (plain, login, cram-md5):
3510 On startup auth_cyrus_sasl_init() gets called for each of these.
3511 This means three calls to sasl_listmech() without a specified mech_list.
3512 => SASL tests which mechs of all available mechs actually work
3513 => three warnings about OTP not working
3514 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3516 With the patch, sasl_listmech() also gets called three times. But now
3517 SASL's mech_list option is set to the server_mech specified in the the
3518 authenticator. Or in other words, the answer from sasl_listmech()
3519 gets limited to just the mech you're testing for (which is different
3521 => the return list contains just 'plain' or 'login', 'cram-md5' or
3522 nothing depending on the value of ob->server_mech.
3524 I've just tested the patch: Authentication still works fine,
3525 unavailable mechs specified in the exim configuration are still
3526 caught, and the auth.log warnings about OTP are gone."
3528 PH/25 When debugging is enabled, the contents of the command line are added
3529 to the debugging output, even when log_selector=+arguments is not
3532 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3533 answer is "GNU", and only if the return is "GNU/something" is the answer
3536 PH/27 $acl_verify_message is now set immediately after the failure of a
3537 verification in an ACL, and so is available in subsequent modifiers. In
3538 particular, the message can be preserved by coding like this:
3540 warn !verify = sender
3541 set acl_m0 = $acl_verify_message
3543 Previously, $acl_verify_message was set only while expanding "message"
3544 and "log_message" when a very denied access.
3546 PH/28 Modified OS/os.c-Linux with
3548 -#ifndef OS_LOAD_AVERAGE
3549 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3551 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3552 nomenclature these days.)
3554 PH/29 Installed patch from the Sieve maintainer that adds the options
3555 sieve_useraddress and sieve_subaddress to the redirect router.
3557 PH/30 In these circumstances:
3558 . Two addresses routed to the same list of hosts;
3559 . First host does not offer TLS;
3560 . First host accepts first address;
3561 . First host gives temporary error to second address;
3562 . Second host offers TLS and a TLS session is established;
3563 . Second host accepts second address.
3564 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3565 and peerdn, if requested) that were in fact used only for the second
3568 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3569 attention to any local part prefix or suffix that was matched by the
3570 router that accepted the address. It now behaves in the same way as it
3571 does for delivery: the affixes are removed from the local part unless
3572 rcpt_include_affixes is set on the transport.
3574 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3575 timeout during the DATA phase of an incoming message.
3577 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3578 applied a patch sanctioned by the Sieve maintainer.
3580 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3581 the uid or gid is negative. A case of a negative gid caused this to be
3582 noticed. The fix allows for either to be negative.
3584 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3585 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3588 PH/36 The change PH/12 above was broken. Fixed it.
3590 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3591 the grounds that routing the same address twice would always produce the
3592 same answer. This might have been true once, but it is certainly no
3593 longer true now. Routing a child address may depend on the previous
3594 routing that produced that child. Some complicated redirection strategies
3595 went wrong when messages had multiple recipients, and made Exim's
3596 behaviour dependent on the order in which the addresses were given.
3598 I have moved the duplicate checking until after the routing is complete.
3599 Exim scans the addresses that are assigned to local and remote
3600 transports, and removes any duplicates. This means that more work will be
3601 done, as duplicates will always all be routed, but duplicates are
3602 presumably rare, so I don't expect this is of any significance.
3604 For deliveries to pipes, files, and autoreplies, the duplicate checking
3605 still happens during the routing process, since they are not going to be
3608 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3609 It corrects a timeout issue with spamd. This is Ian's comment: "The
3610 background is that sometimes spamd either never reads data from a
3611 connection it has accepted, or it never writes response data. The exiscan
3612 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3613 blindly assumes that writes won't block so it may never time out."
3615 PH/39 Allow G after quota size as well as K and M.
3617 PH/40 The value set for $authenticated_id in an authenticator may not contain
3618 binary zeroes or newlines because the value is written to log lines and
3619 to spool files. There was no check on this. Now the value is run through
3620 the string_printing() function so that such characters are converted to
3621 printable escape sequences.
3623 PH/41 $message_linecount is a new variable that contains the total number of
3624 lines in the message. Compare $body_linecount, which is the count for the
3627 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3628 bounce and delay warning messages, except in certain special cases, which
3631 (a) An SMTP error message from a remote host;
3632 (b) A message specified in a :fail: redirection;
3633 (c) A message specified in a "fail" command in a system filter;
3634 (d) A message specified in a FAIL return from the queryprogram router;
3635 (e) A message specified by the cannot_route_message router option.
3637 In these cases only, Exim does include the error details in bounce and
3638 warning messages. There are also a few cases where bland messages such
3639 as "unrouteable address" or "local delivery error" are given.
3641 PH/43 $value is now also set for the "else" part of a ${run expansion.
3643 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3644 being worked on, but at least Exim now implements the latest version to
3647 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3648 process to complete was treated as a delivery failure, a timeout while
3649 writing the message to the pipe was logged, but erroneously treated as a
3650 successful delivery. Such timeouts include transport filter timeouts. For
3651 consistency with the overall process timeout, these timeouts are now
3652 treated as errors, giving rise to delivery failures by default. However,
3653 there is now a new Boolean option for the pipe transport called
3654 timeout_defer, which, if set TRUE, converts the failures into defers for
3655 both kinds of timeout. A transport filter timeout is now identified in
3658 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3659 systems where "make" and "gmake" are different, calling "gmake" at top
3660 level broke things. I've arranged for the value of $(MAKE) to be passed
3661 from the Makefile to this script so that it can call the same version of
3665 A note about Exim versions 4.44 and 4.50
3666 ----------------------------------------
3668 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3669 changes of various kinds. As a consequence, a big documentation update was
3670 needed. This delayed the release for rather longer than seemed good, especially
3671 in the light of a couple of (minor) security issues. Therefore, the changes
3672 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3673 release. So 4.44 and 4.50 are in effect two different branches that both start
3676 I have left the 4.50 change log unchanged; it contains all the changes since
3677 4.43. The change log for 4.44 is below; many of its items are identical to
3678 those for 4.50. This seems to be the most sensible way to preserve the
3679 historical information.
3685 1. Minor wording change to the doc/README.SIEVE file.
3687 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3688 computation of the current number of files was incorrect.
3690 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3691 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3694 4. Give more explanation in the error message when the command for a transport
3695 filter fails to execute.
3697 5. There are several places where Exim runs a non-Exim command in a
3698 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3699 was being done only for the command run by the queryprogram router. It is
3700 now done for all such subprocesses. The other cases are: ${run, transport
3701 filters, and the commands run by the lmtp and pipe transports.
3703 6. Added CONFIGURE_GROUP build-time option.
3705 7. Some older OS have a limit of 256 on the maximum number of file
3706 descriptors. Exim was using setrlimit() to set 1000 as a large value
3707 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3708 systems. I've change it so that if it can't get 1000, it tries for 256.
3710 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3711 was an oversight, and furthermore, ever since the addition of extra
3712 controls (e.g. 4.43/32), the checks on when to allow different forms of
3713 "control" were broken. There should now be diagnostics for all cases when a
3714 control that does not make sense is encountered.
3716 9. Added the /retain_sender option to "control=submission".
3718 10. $recipients is now available in the predata ACL (oversight).
3720 11. Tidy the search cache before the fork to do a delivery from a message
3721 received from the command line. Otherwise the child will trigger a lookup
3722 failure and thereby defer the delivery if it tries to use (for example) a
3723 cached ldap connection that the parent has called unbind on.
3725 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3726 of $address_data from the recipient verification was clobbered by the
3727 sender verification.
3729 13. The value of address_data from a sender verification is now available in
3730 $sender_address_data in subsequent conditions in the ACL statement.
3732 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3734 15. Added a new option "connect=<time>" to callout options, to set a different
3737 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3738 was its contents. (It was OK if the option was not defined at all.)
3740 17. A "Completed" log line is now written for messages that are removed from
3741 the spool by the -Mrm option.
3743 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3744 information about exactly what failed.
3746 19. Added -dd to debug only the daemon process.
3748 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3749 handles timeouts, both on the server side and network timeouts. Renamed the
3750 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3752 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3753 It is now set to "smtps".
3755 22. $host_address is now set to the target address during the checking of
3756 ignore_target_hosts.
3758 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3759 being passed; this would have caused $sender_host_name to have been used if
3760 matching the list had actually called for a host name (not very likely,
3761 since this list is usually IP addresses). A host name is now passed as
3764 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3765 code that specifies a non-restarting handler (typically sigaction() in
3766 modern systems) in an attempt to fix a rare and obscure crash bug.
3768 25. Narrowed the window for a race in the daemon that could cause it to ignore
3769 SIGCHLD signals. This is not a major problem, because they are used only to
3770 wake it up if nothing else does.
3772 26. A malformed maildirsize file could cause Exim to calculate negative values
3773 for the mailbox size or file count. Odd effects could occur as a result.
3774 The maildirsize information is now recalculated if the size or filecount
3777 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3778 support for a long time. Removed HAVE_SYS_VFS_H.
3780 28. Installed the latest version of exipick from John Jetmore.
3782 29. In an address list, if the pattern was not a regular expression, an empty
3783 subject address (from a bounce message) matched only if the pattern was an
3784 empty string. Non-empty patterns were not even tested. This was the wrong
3785 because it is perfectly reasonable to use an empty address as part of a
3786 database query. An empty address is now tested by patterns that are
3787 lookups. However, all the other forms of pattern expect the subject to
3788 contain a local part and a domain, and therefore, for them, an empty
3789 address still always fails if the pattern is not itself empty.
3791 30. Exim went into a mad DNS loop when attempting to do a callout where the
3792 host was specified on an smtp transport, and looking it up yielded more
3793 than one IP address.
3795 31. Re-factored the code for checking spool and log partition space into a
3796 function that finds that data and another that does the check. The former
3797 is then used to implement four new variables: $spool_space, $log_space,
3798 $spool_inodes, and $log_inodes.
3800 32. The RFC2047 encoding function was originally intended for short strings
3801 such as real names; it was not keeping to the 75-character limit for
3802 encoded words that the RFC imposes. It now respects the limit, and
3803 generates multiple encoded words if necessary. To be on the safe side, I
3804 have increased the buffer size for the ${rfc2047: expansion operator from
3807 33. It is now permitted to omit both strings after an "if" condition; if the
3808 condition is true, the result is "true". As before, when the second string
3809 is omitted, a false condition yields an empty string. This makes it less
3810 cumbersome to write custom ACL and router conditions.
3812 34. Failure to deliver a bounce message always caused it to be frozen, even if
3813 there was an errors_to setting on the router. The errors_to setting is now
3816 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3817 canonical form (fully expanded) before being placed in
3818 $sender_host_address.
3820 36. The table in the code that translates DNS record types into text (T_A to
3821 "A" for instance) was missing entries for NS and CNAME. It is just possible
3822 that this could have caused confusion if both these types were looked up
3823 for the same domain, because the text type is used as part of Exim's
3824 per-process caching. But the chance of anyone hitting this buglet seems
3827 37. The dnsdb lookup has been extended in a number of ways.
3829 (1) There is a new type, "zns", which walks up the domain tree until it
3830 finds some nameserver records. It should be used with care.
3832 (2) There is a new type, "mxh", which is like "mx" except that it returns
3833 just the host names, not the priorities.
3835 (3) It is now possible to give a list of domains (or IP addresses) to be
3836 looked up. The behaviour when one of the lookups defers can be
3837 controlled by a keyword.
3839 (4) It is now possible to specify the separator character for use when
3840 multiple records are returned.
3842 38. The dnslists ACL condition has been extended: it is now possible to supply
3843 a list of IP addresses and/or domains to be looked up in a particular DNS
3846 39. Added log_selector=+queue_time_overall.
3848 40. When running the queue in the test harness, wait just a tad after forking a
3849 delivery process, to get repeatability of debugging output.
3851 41. Include certificate and key file names in error message when GnuTLS fails
3852 to set them up, because the GnuTLS error message doesn't include the name
3853 of the failing file when there is a problem reading it.
3855 42. Allow both -bf and -bF in the same test run.
3857 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3859 44. The "Exiscan patch" is now merged into the mainline Exim source.
3861 45. Sometimes the final signoff response after QUIT could fail to get
3862 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3863 < 0 before doing a fflush(). This bug looks as though it goes back to the
3864 introduction of TLS in release 3.20, but "sometimes" must have been rare
3865 because the tests only now provoked it.
3867 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3868 (this can affect the format of dates).
3870 47. exim_tidydb, when checking for the continued existence of a message for
3871 which it has found a message-specific retry record, was not finding
3872 messages that were in split spool directories. Consequently, it was
3873 deleting retry records that should have stayed in existence.
3875 48. Steve fixed some bugs in eximstats.
3877 49. The SPA authentication driver was not abandoning authentication and moving
3878 on to the next authenticator when an expansion was forced to fail,
3879 contradicting the general specification for all authenticators. Instead it
3880 was generating a temporary error. It now behaves as specified.
3882 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3883 (the order specifies the preference for clients). The order is now AES256,
3884 AES128, 3DES, ARCFOUR128.
3886 51. Small patch to Sieve code - explicitly set From: when generating an
3889 52. Exim crashed if a remote delivery caused a very long error message to be
3890 recorded - for instance if somebody sent an entire SpamAssassin report back
3891 as a large number of 550 error lines. This bug was coincidentally fixed by
3892 increasing the size of one of Exim's internal buffers (big_buffer) that
3893 happened as part of the Exiscan merge. However, to be on the safe side, I
3894 have made the code more robust (and fixed the comments that describe what
3897 53. Now that there can be additional text after "Completed" in log lines (if
3898 the queue_time_overall log selector is set), a one-byte patch to exigrep
3899 was needed to allow it to recognize "Completed" as not the last thing in
3902 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3903 patch that reportedly fixes this has been added. I am not expert enough to
3904 create a test for it. This is what the patch creator wrote:
3906 "I found a little strange behaviour of ldap code when working with
3907 Windows 2003 AD Domain, where users was placed in more than one
3908 Organization Units. When I tried to give exim partial DN, the exit code
3909 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
3910 But simultaneously result of request was absolutely normal ldap result,
3911 so I produce this patch..."
3913 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
3914 REFERENCE, so I have modified the code to exclude the patch when that macro
3917 55. Some experimental protocols are using DNS PTR records for new purposes. The
3918 keys for these records are domain names, not reversed IP addresses. The
3919 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3920 leaves it alone. Component reversal etc. now happens only for IP addresses.
3923 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
3925 57. Double the size of the debug message buffer (to 2048) so that more of very
3926 long debug lines gets shown.
3928 58. The exicyclog utility now does better if the number of log files to keep
3929 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
3931 59. Two changes related to the smtp_active_hostname option:
3933 (1) $smtp_active_hostname is now available as a variable.
3934 (2) The default for smtp_banner uses $smtp_active_hostname instead
3935 of $primary_hostname.
3937 60. The host_aton() function is supposed to be passed a string that is known
3938 to be a valid IP address. However, in the case of IPv6 addresses, it was
3939 not checking this. This is a hostage to fortune. Exim now panics and dies
3940 if the condition is not met. A case was found where this could be provoked
3941 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3942 components; fortuitously, this particular loophole had already been fixed
3943 by change 4.50/55 above.
3945 If there are any other similar loopholes, the new check in host_aton()
3946 itself should stop them being exploited. The report I received stated that
3947 data on the command line could provoke the exploit when Exim was running as
3948 exim, but did not say which command line option was involved. All I could
3949 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3950 running as the user.
3953 61. There was a buffer overflow vulnerability in the SPA authentication code
3954 (which came originally from the Samba project). I have added a test to the
3955 spa_base64_to_bits() function which I hope fixes it.
3958 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
3959 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
3961 63. The daemon start-up calls getloadavg() while still root for those OS that
3962 need the first call to be done as root, but it missed one case: when
3963 deliver_queue_load_max is set with deliver_drop_privilege. This is
3964 necessary for the benefit of the queue runner, because there is no re-exec
3965 when deliver_drop_privilege is set.
3967 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
3968 This has been fixed.
3970 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3971 was in use, was not putting the data itself into the right store pool;
3972 consequently, it could be overwritten for a subsequent message in the same
3973 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3976 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
3978 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
3979 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
3980 the function were treating the return as a boolean value, which happened to
3981 work because 0=false and not-0=true, but is not correct code.
3983 68. The host_aton() function was not handling scoped IPv6 addresses (those
3984 with, for example, "%eth0" on the end) correctly.
3986 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
3987 negated items (that is, ~something) in unsigned ints. Some compilers
3988 apparently mutter when there is no cast.
3990 70. If an address verification called from an ACL failed, and did not produce a
3991 user-specific message (i.e. there was only a "system" message), nothing was
3992 put in $acl_verify_message. In this situation, it now puts the system
3995 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3996 session; change 4.31/43 added the unwanted input to the log line - except
3997 that it did not do this in the start of session case. It now does.
3999 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4000 This is wrong and can cause the other end to generate a synchronization
4001 error if it is another Exim or anything else that does the synchronization
4002 check. A QUIT command is no longer sent after a timeout.
4004 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4005 during host lookups.
4007 74. The defer_ok option of callout verification was not working if it was used
4008 when verifying addresses in header lines, that is, for this case:
4010 verify = header_sender/callout=defer_ok
4012 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4013 those file descriptors could be used for SMTP connections. If anything
4014 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4015 could be sent to the SMTP client, causing chaos. The daemon now opens
4016 stdin, stdout, and stderr to /dev/null when it puts itself into the
4019 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4020 log by default. The user can override this with suitable Perl magic.
4022 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4023 the log message when discard triggers, was not working for the DATA ACL or
4024 for the non-SMTP ACL.
4026 78. Error message wording change in sieve.c.
4028 79. If smtp_accept_max_per_host was set, the number of connections could be
4029 restricted to fewer than expected, because the daemon was trying to set up
4030 a new connection before checking whether the processes handling previous
4031 connections had finished. The check for completed processes is now done
4032 earlier. On busy systems, this bug wouldn't be noticed because something
4033 else would have woken the daemon, and it would have reaped the completed
4036 80. If a message was submitted locally by a user whose login name contained one
4037 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4038 It caused a spool format error. I have fixed the spool reading code. A
4039 related problem was that the "from" clause in the Received: line became
4040 illegal because of the space(s). It is now covered by ${quote_local_part.
4042 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4045 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4047 83. Patch from Sieve maintainer to fix unterminated string problem in
4048 "vacation" handling.
4050 84. Some minor changes to the Linux configuration files to help with other
4051 OS variants using glibc.
4053 85. One more patch for Sieve to update vacation handling to latest spec.
4056 ----------------------------------------------------
4057 See the note above about the 4.44 and 4.50 releases.
4058 ----------------------------------------------------
4064 1. Change 4.43/35 introduced a bug that caused file counts to be
4065 incorrectly computed when quota_filecount was set in an appendfile
4068 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4069 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4072 3. Give more explanation in the error message when the command for a transport
4073 filter fails to execute.
4075 4. There are several places where Exim runs a non-Exim command in a
4076 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4077 was being done only for the command run by the queryprogram router. It is
4078 now done for all such subprocesses. The other cases are: ${run, transport
4079 filters, and the commands run by the lmtp and pipe transports.
4081 5. Some older OS have a limit of 256 on the maximum number of file
4082 descriptors. Exim was using setrlimit() to set 1000 as a large value
4083 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4084 systems. I've change it so that if it can't get 1000, it tries for 256.
4086 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4087 was an oversight, and furthermore, ever since the addition of extra
4088 controls (e.g. 4.43/32), the checks on when to allow different forms of
4089 "control" were broken. There should now be diagnostics for all cases when a
4090 control that does not make sense is encountered.
4092 7. $recipients is now available in the predata ACL (oversight).
4094 8. Tidy the search cache before the fork to do a delivery from a message
4095 received from the command line. Otherwise the child will trigger a lookup
4096 failure and thereby defer the delivery if it tries to use (for example) a
4097 cached ldap connection that the parent has called unbind on.
4099 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4100 of $address_data from the recipient verification was clobbered by the
4101 sender verification.
4103 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4104 was its contents. (It was OK if the option was not defined at all.)
4106 11. A "Completed" log line is now written for messages that are removed from
4107 the spool by the -Mrm option.
4109 12. $host_address is now set to the target address during the checking of
4110 ignore_target_hosts.
4112 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4113 being passed; this would have caused $sender_host_name to have been used if
4114 matching the list had actually called for a host name (not very likely,
4115 since this list is usually IP addresses). A host name is now passed as
4118 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4119 code that specifies a non-restarting handler (typically sigaction() in
4120 modern systems) in an attempt to fix a rare and obscure crash bug.
4122 15. Narrowed the window for a race in the daemon that could cause it to ignore
4123 SIGCHLD signals. This is not a major problem, because they are used only to
4124 wake it up if nothing else does.
4126 16. A malformed maildirsize file could cause Exim to calculate negative values
4127 for the mailbox size or file count. Odd effects could occur as a result.
4128 The maildirsize information is now recalculated if the size or filecount
4131 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4132 support for a long time. Removed HAVE_SYS_VFS_H.
4134 18. Updated exipick to current release from John Jetmore.
4136 19. Allow an empty sender to be matched against a lookup in an address list.
4137 Previously the only cases considered were a regular expression, or an
4140 20. Exim went into a mad DNS lookup loop when doing a callout where the
4141 host was specified on the transport, if the DNS lookup yielded more than
4144 21. The RFC2047 encoding function was originally intended for short strings
4145 such as real names; it was not keeping to the 75-character limit for
4146 encoded words that the RFC imposes. It now respects the limit, and
4147 generates multiple encoded words if necessary. To be on the safe side, I
4148 have increased the buffer size for the ${rfc2047: expansion operator from
4151 22. Failure to deliver a bounce message always caused it to be frozen, even if
4152 there was an errors_to setting on the router. The errors_to setting is now
4155 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4156 canonical form (fully expanded) before being placed in
4157 $sender_host_address.
4159 24. Updated eximstats to version 1.33
4161 25. Include certificate and key file names in error message when GnuTLS fails
4162 to set them up, because the GnuTLS error message doesn't include the name
4163 of the failing file when there is a problem reading it.
4165 26. Expand error message when OpenSSL has problems setting up cert/key files.
4168 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4169 (this can affect the format of dates).
4171 28. exim_tidydb, when checking for the continued existence of a message for
4172 which it has found a message-specific retry record, was not finding
4173 messages that were in split spool directories. Consequently, it was
4174 deleting retry records that should have stayed in existence.
4176 29. eximstats updated to version 1.35
4177 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4178 1.35 - bugfix such that pie charts by volume are generated correctly
4180 30. The SPA authentication driver was not abandoning authentication and moving
4181 on to the next authenticator when an expansion was forced to fail,
4182 contradicting the general specification for all authenticators. Instead it
4183 was generating a temporary error. It now behaves as specified.
4185 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4186 (the order specifies the preference for clients). The order is now AES256,
4187 AES128, 3DES, ARCFOUR128.
4189 31. Small patch to Sieve code - explicitly set From: when generating an
4192 32. Exim crashed if a remote delivery caused a very long error message to be
4193 recorded - for instance if somebody sent an entire SpamAssassin report back
4194 as a large number of 550 error lines. This bug was coincidentally fixed by
4195 increasing the size of one of Exim's internal buffers (big_buffer) that
4196 happened as part of the Exiscan merge. However, to be on the safe side, I
4197 have made the code more robust (and fixed the comments that describe what
4200 33. Some experimental protocols are using DNS PTR records for new purposes. The
4201 keys for these records are domain names, not reversed IP addresses. The
4202 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4203 leaves it alone. Component reversal etc. now happens only for IP addresses.
4206 34. The host_aton() function is supposed to be passed a string that is known
4207 to be a valid IP address. However, in the case of IPv6 addresses, it was
4208 not checking this. This is a hostage to fortune. Exim now panics and dies
4209 if the condition is not met. A case was found where this could be provoked
4210 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4211 components; fortuitously, this particular loophole had already been fixed
4212 by change 4.50/55 or 4.44/33 above.
4214 If there are any other similar loopholes, the new check in host_aton()
4215 itself should stop them being exploited. The report I received stated that
4216 data on the command line could provoke the exploit when Exim was running as
4217 exim, but did not say which command line option was involved. All I could
4218 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4219 running as the user.
4222 35. There was a buffer overflow vulnerability in the SPA authentication code
4223 (which came originally from the Samba project). I have added a test to the
4224 spa_base64_to_bits() function which I hope fixes it.
4227 36. The daemon start-up calls getloadavg() while still root for those OS that
4228 need the first call to be done as root, but it missed one case: when
4229 deliver_queue_load_max is set with deliver_drop_privilege. This is
4230 necessary for the benefit of the queue runner, because there is no re-exec
4231 when deliver_drop_privilege is set.
4233 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4234 was in use, was not putting the data itself into the right store pool;
4235 consequently, it could be overwritten for a subsequent message in the same
4236 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4239 38. Sometimes the final signoff response after QUIT could fail to get
4240 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4241 < 0 before doing a fflush(). This bug looks as though it goes back to the
4242 introduction of TLS in release 3.20, but "sometimes" must have been rare
4243 because the tests only now provoked it.
4249 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4250 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4251 Now it returns an integer. A number of calls were still expecting a T/F
4252 return. Fortuitously, in all cases, the tests worked in OK situations,
4253 which is the norm. However, things would have gone wrong on any write
4254 failures on the smtp file descriptor. This function is used when sending
4255 messages over SMTP and also when doing verify callouts.
4257 2. When Exim is called to do synchronous delivery of a locally submitted
4258 message (the -odf or -odi options), it no longer closes stderr before doing
4261 3. Implemented the mua_wrapper option.
4263 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4265 5. Implemented the functions header_remove(), header_testname(),
4266 header_add_at_position(), and receive_remove_recipient(), and exported them
4269 6. If an ACL "warn" statement specified the addition of headers, Exim already
4270 inserted X-ACL-Warn: at the start if there was no header name. However, it
4271 was not making this test for the second and subsequent header lines if
4272 there were newlines in the string. This meant that an invalid header could
4273 be inserted if Exim was badly configured.
4275 7. Allow an ACL "warn" statement to add header lines at the start or after all
4276 the Received: headers, as well as at the end.
4278 8. Added the rcpt_4xx retry error code.
4280 9. Added postmaster_mailfrom=xxx to callout verification option.
4282 10. Added mailfrom=xxxx to the callout verification option, for verify=
4285 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4286 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4288 12. Inserted some casts to stop certain compilers warning when using pointer
4289 differences as field lengths or precisions in printf-type calls (mostly
4290 affecting debugging statements).
4292 13. Added optional readline() support for -be (dynamically loaded).
4294 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4295 same clock tick as a message's arrival, so that its received time was the
4296 same as the "first fail" time on the retry record, and that message
4297 remained on the queue past the ultimate address timeout, every queue runner
4298 would try a delivery (because it was past the ultimate address timeout) but
4299 after another failure, the ultimate address timeout, which should have then
4300 bounced the address, did not kick in. This was a "< instead of <=" error;
4301 in most cases the first failure would have been in the next clock tick
4302 after the received time, and all would be well.
4304 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4305 being recognized when the domain list was tested by the match_domain
4306 condition in an expansion string.
4308 16. Added the ${str2b64: operator.
4310 17. Exim was always calling setrlimit() to set a large limit for the number of
4311 processes, without checking whether the existing limit was already
4312 adequate. (It did check for the limit on file descriptors.) Furthermore,
4313 errors from getrlimit() and setrlimit() were being ignored. Now they are
4314 logged to the main and panic logs, but Exim does carry on, to try to do its
4315 job under whatever limits there are.
4317 18. Imported PCRE 5.0.
4319 19. Trivial typo in log message " temporarily refused connection" (the leading
4322 20. If the log selector return_path_on_delivery was set and an address was
4323 redirected to /dev/null, the delivery process crashed because it assumed
4324 that a return path would always be set for a "successful" delivery. In this
4325 case, the whole delivery is bypassed as an optimization, and therefore no
4328 21. Internal re-arrangement: the function for sending a challenge and reading
4329 a response while authentication was assuming a zero-terminated challenge
4330 string. It's now changed to take a pointer and a length, to allow for
4331 binary data in such strings.
4333 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4335 23. Exim was not respecting finduser_retries when seeking the login of the
4336 uid under which it was called; it was always trying 10 times. (The default
4337 setting of finduser_retries is zero.) Also, it was sleeping after the final
4338 failure, which is pointless.
4340 24. Implemented tls_on_connect_ports.
4342 25. Implemented acl_smtp_predata.
4344 26. If the domain in control=submission is set empty, Exim assumes that the
4345 authenticated id is a complete email address when it generates From: or
4346 Sender: header lines.
4348 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4349 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4350 chown and chgrp in /bin and hostname in /usr/bin.
4352 28. Exim was keeping the "process log" file open after each use, just as it
4353 does for the main log. This opens the possibility of it remaining open for
4354 long periods when the USR1 signal hits a daemon. Occasional processlog
4355 errors were reported, that could have been caused by this. Anyway, it seems
4356 much more sensible not to leave this file open at all, so that is what now
4359 29. The long-running daemon process does not normally write to the log once it
4360 has entered its main loop, and it closes the log before doing so. This is
4361 so that log files can straightforwardly be renamed and moved. However,
4362 there are a couple of unusual error situations where the daemon does write
4363 log entries, and I had neglected to close the log afterwards.
4365 30. The text of an SMTP error response that was received during a remote
4366 delivery was being truncated at 512 bytes. This is too short for some of
4367 the long messages that one sometimes sees. I've increased the limit to
4370 31. It is now possible to make retry rules that apply only when a message has a
4371 specific sender, in particular, an empty sender.
4373 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4374 it possible to be selective about when SMTP synchronization is enforced.
4376 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4378 32. Implemented hosts_connection_nolog.
4380 33. Added an ACL for QUIT.
4382 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4385 35. Added mailbox_size and mailbox_filecount to appendfile.
4387 36. Added control = no_multiline_responses to ACLs.
4389 37. There was a bug in the logic of the code that waits for the clock to tick
4390 in the case where the clock went backwards by a substantial amount such
4391 that the microsecond fraction of "now" was more than the microsecond
4392 fraction of "then" (but the whole seconds number was less).
4394 38. Added support for the libradius Radius client library this is found on
4395 FreeBSD (previously only the radiusclient library was supported).
4401 1. When certain lookups returned multiple values in the form name=value, the
4402 quoting of the values was not always being done properly. Specifically:
4403 (a) If the value started with a double quote, but contained no whitespace,
4405 (b) If the value contained whitespace other than a space character (i.e.
4406 tabs or newlines or carriage returns) it was not quoted.
4407 This fix has been applied to the mysql and pgsql lookups by writing a
4408 separate quoting function and calling it from the lookup code. The fix
4409 should probably also be applied to nisplus, ibase and oracle lookups, but
4410 since I cannot test any of those, I have not disturbed their existing code.
4412 2. A hit in the callout cache for a specific address caused a log line with no
4413 reason for rejecting RCPT. Now it says "Previous (cached) callout
4414 verification failure".
4416 3. There was an off-by-one bug in the queryprogram router. An over-long
4417 return line was truncated at 256 instead of 255 characters, thereby
4418 overflowing its buffer with the terminating zero. As well as fixing this, I
4419 have increased the buffer size to 1024 (and made a note to document this).
4421 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4422 when Exim is waiting for an SMTP response from a remote server, Exim
4423 restarts its select() call on the socket, thereby resetting its timeout.
4424 This is not a problem when such interrupts are rare. Somebody set up a cron
4425 job to run exiwhat every 2 minutes, which is less than the normal select()
4426 timeout (5 or 10 minutes). This meant that the select() timeout never
4427 kicked in because it was always reset. I have fixed this by comparing the
4428 time when an interrupt arrives with the time at the start of the first call
4429 to select(). If more time than the timeout has elapsed, the interrupt is
4430 treated as a timeout.
4432 5. Some internal re-factoring in preparation for the addition of Sieve
4433 extensions (by MH). In particular, the "personal" test is moved to a
4434 separate function, and given an option for scanning Cc: and Bcc: (which is
4435 not set for Exim filters).
4437 6. When Exim created an email address using the login of the caller as the
4438 local part (e.g. when creating a From: or Sender: header line), it was not
4439 quoting the local part when it contained special characters such as @.
4441 7. Installed new OpenBSD configuration files.
4443 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4444 try to make them clearer.
4446 9. Callout options, other than the timeout value, were being ignored when
4447 verifying sender addresses in header lines. For example, when using
4449 verify = header_sender/callout=no_cache
4451 the cache was (incorrectly) being used.
4453 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4454 only the Cygwin environment.
4456 11. When return_path_on_delivery was set as a log selector, if different remote
4457 addresses in the same message used different return paths and parallel
4458 remote delivery occurred, the wrong values would sometimes be logged.
4459 (Whenever a remote delivery process finished, the return path value from
4460 the most recently started remote delivery process was logged.)
4462 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4463 lines when AUTH and/or TLS are in use. This is the "received protocol"
4464 field. Exim used to use "asmtp" for authenticated SMTP, without any
4465 indication (in the protocol name) for TLS use. Now it follows the RFC and
4466 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4467 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4468 names appear in log lines as well as in Received: header lines.
4470 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4471 extensions, and comparison tests, and to fix some bugs.
4473 14. Changes to the "personal" filter test:
4475 (1) The test was buggy in that it was just doing the equivalent of
4476 "contains" tests on header lines. For example, if a user's address was
4477 anne@some.where, the "personal" test would incorrectly be true for
4479 To: susanne@some.where
4481 This test is now done by extracting each address from the header in turn,
4482 and checking the entire address. Other tests that are part of "personal"
4483 are now done using regular expressions (for example, to check local parts
4484 of addresses in From: header lines).
4486 (2) The list of non-personal local parts in From: addresses has been
4487 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4488 taken from the Sieve specification recommendations.
4490 (3) If the message contains any header line starting with "List-" it is
4491 treated as non-personal.
4493 (4) The test for "circular" in the Subject: header line has been removed
4494 because it now seems ill-conceived.
4496 15. Minor typos in src/EDITME comments corrected.
4498 16. Installed latest exipick from John Jetmore.
4500 17. If headers_add on a router specified a text string that was too long for
4501 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4502 of string_sprintf() is now avoided.
4504 18. $message_body_size was not set (it was always zero) when running the DATA
4505 ACL and the local_scan() function.
4507 19. For the "mail" command in an Exim filter, no default was being set for
4508 the once_repeat time, causing a random time value to be used if "once" was
4509 specified. (If the value happened to be <= 0, no repeat happened.) The
4510 default is now 0s, meaning "never repeat". The "vacation" command was OK
4511 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4512 (I found it when inspecting the code).
4514 20. There is now an overall timeout for performing a callout verification. It
4515 defaults to 4 times the callout timeout, which applies to individual SMTP
4516 commands during the callout. The overall timeout applies when there is more
4517 than one host that can be tried. The timeout is checked before trying the
4518 next host. This prevents very long delays if there are a large number of
4519 hosts and all are timing out (e.g. when the network connections are timing
4520 out). The value of the overall timeout can be changed by specifying an
4521 additional sub-option for "callout", called "maxwait". For example:
4523 verify = sender/callout=5s,maxwait=20s
4525 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4526 to the end before writing, but this should make it even safer).
4528 22. Exim was forgetting that it had advertised PIPELINING for the second and
4529 subsequent messages on an SMTP connection. It was also not resetting its
4530 memory on STARTTLS and an internal HELO.
4532 23. When Exim logs an SMTP synchronization error within a session, it now
4533 records whether PIPELINING has been advertised or not.
4535 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4536 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4537 rather than long int.
4539 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4541 26. Added the never_mail option to autoreply.
4547 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4548 crash if the getsockname() call failed; this can happen if a connection is
4549 closed very soon after it is established. The problem was simply in the
4550 order in which certain operations were done, causing Exim to try to write
4551 to the SMTP stream before it had set up the file descriptor. The bug has
4552 been fixed by making things happen in the correct order.
4558 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4559 before the connection was closed, thus losing the rejection response.
4561 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4562 some early Solaris releases, but causes trouble in current releases where
4563 socklen_t is defined.
4565 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4568 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4571 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4572 that releasing the top of it at the end releases what was used for sub-
4573 expansions (unless the block got too big). However, discard this block if
4574 the first thing is a variable or header, so that we can use its block when
4575 it is dynamic (useful for very large $message_headers, for example).
4577 6. Lookups now cache *every* query, not just the most recent. A new, separate
4578 store pool is used for this. It can be recovered when all lookup caches are
4579 flushed. Lookups now release memory at the end of their result strings.
4580 This has involved some general refactoring of the lookup sources.
4582 7. Some code has been added to the store_xxx() functions to reduce the amount
4583 of flapping under certain conditions.
4585 8. log_incoming_interface used to affect only the <= reception log lines. Now
4586 it causes the local interface and port to be added to several more SMTP log
4587 lines, for example "SMTP connection from", and rejection lines.
4589 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4591 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4593 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4594 could be overwritten at the end of the current message (or the start of a
4595 new message if it was set in a HELO ACL). The value is now preserved for
4596 the duration of the SMTP connection.
4598 12. If a transport had a headers_rewrite setting, and a matching header line
4599 contained an unqualified address, that address was qualified, even if it
4600 did not match any rewriting rules. The underlying bug was that the values
4601 of the flags that permit the existence of unqualified sender and recipient
4602 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4603 non-local messages, and by -bnq for local messages) were not being
4604 preserved with the message after it was received.
4606 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4607 "next input=" as part of the text comprising the host identity instead of
4608 the correct text. The code was using the same buffer for two different
4609 strings. However, depending on which order the printing function evaluated
4610 its arguments, the bug did not always show up. Under Linux, for example, my
4611 test suite worked just fine.
4613 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4614 doesn't work with some older versions of Perl. It has been changed to "my",
4615 which in any case is probably the better facility to use.
4617 15. A really picky compiler found some instances of statements for creating
4618 error messages that either had too many or two few arguments for the format
4621 16. The size of the buffer for calls to the DNS resolver has been increased
4622 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4623 for addresses that have a lot of PTR records. This alleviates a problem; it
4624 does not fully solve it.
4626 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4627 buffer now truncates the list and logs the incident, which is the same
4628 action as happens when Exim is looking up a host name and its aliases.
4629 Previously in this situation something unpredictable would happen;
4630 sometimes it was "internal error: store_reset failed".
4632 18. If a server dropped the connection unexpectedly when an Exim client was
4633 using GnuTLS and trying to read a response, the client delivery process
4634 crashed while trying to generate an error log message.
4636 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4637 string, for example:
4639 warn message = H1: something\nH2: something
4641 the text was added as a single header line from Exim's point of view
4642 though it ended up OK in the delivered message. However, searching for the
4643 second and subsequent header lines using $h_h2: did not work. This has been
4644 fixed. Similarly, if a system filter added multiple headers in this way,
4645 the routers could not see them.
4647 20. Expanded the error message when iplsearch is called with an invalid key to
4648 suggest using net-iplsearch in a host list.
4650 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4651 ACLs are no longer actually imposed (and a message to that effect is
4654 22. If a "gecos" field in a passwd entry contained escaped characters, in
4655 particular, if it contained a \" sequence, Exim got it wrong when building
4656 a From: or a Sender: header from that name. A second bug also caused
4657 incorrect handling when an unquoted " was present following a character
4658 that needed quoting.
4660 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4661 was not being matched caselessly.
4663 24. Arranged for all hyphens in the exim.8 source to be escaped with
4666 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4667 properly. Recipient callout cache records were still being keyed to include
4668 the sender, even when use_sender was set false. This led to far more
4669 callouts that were necessary. The sender is no longer included in the key
4670 when use_sender is false.
4672 26. Added "control = submission" modifier to ACLs.
4674 27. Added the ${base62d: operator to decode base 62 numbers.
4676 28. dnsdb lookups can now access SRV records.
4678 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4679 the configuration file.
4681 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4682 (-v) mode. This makes the output for a verbose queue run more intelligible.
4684 31. Added a use_postmaster feature to recipient callouts.
4686 32. Added the $body_zerocount variable, containing the number of binary zero
4687 bytes in the message body.
4689 33. The time of last modification of the "new" subdirectory is now used as the
4690 "mailbox time last read" when there is a quota error for a maildir
4693 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4695 35. Added +ignore_unknown as a special item in host lists.
4697 36. Code for decoding IPv6 addresses in host lists is now included, even if
4698 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4699 address was recognized as an IP address, but was then not correctly decoded
4700 into binary, causing unexpected and incorrect effects when compared with
4707 1. Very minor rewording of debugging text in manualroute to say "list of
4708 hosts" instead of "hostlist".
4710 2. If verify=header_syntax was set, and a header line with an unqualified
4711 address (no domain) and a large number of spaces between the end of the
4712 name and the colon was received, the reception process suffered a buffer
4713 overflow, and (when I tested it) crashed. This was caused by some obsolete
4714 code that should have been removed. The fix is to remove it!
4716 3. When running in the test harness, delay a bit after writing a bounce
4717 message to get a bit more predictability in the log output.
4719 4. Added a call to search_tidyup() just before forking a reception process. In
4720 theory, someone could use a lookup in the expansion of smtp_accept_max_
4721 per_host which, without the tidyup, could leave open a database connection.
4723 5. Added the variables $recipient_data and $sender_data which get set from a
4724 lookup success in an ACL "recipients" or "senders" condition, or a router
4725 "senders" option, similar to $domain_data and $local_part_data.
4727 6. Moved the writing of debug_print from before to after the "senders" test
4730 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4731 problems for message scanning, either using a data ACL, or using
4732 local_scan() because the Received: header was not generated till after they
4733 were called (in order to set the time as the time of reception completion).
4734 I have revised the way this works. The header is now generated after the
4735 body is received, but before the ACL or local_scan() are called. After they
4736 are run, the timestamp in the header is updated.
4742 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4743 before starting a queue runner without re-exec. This happened only when
4744 deliver_drop_privilege was set or when the Exim user was set to root. The
4745 effect of the bug was that timeouts during subsequent deliveries caused
4746 crashes instead of being properly handled. The handler is now left at its
4747 default (and expected) setting.
4749 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4750 message, again when deliver_drop_privilege is set or Exim is run as root.
4751 The bug described in (1) was not present in this case, but the tidying up
4752 of the other signals was missing. I have made the two cases consistent.
4754 3. The ignore_target_hosts setting on a manualroute router was being ignored
4755 for hosts that were looked up using the /MX notation.
4757 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4760 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4761 operated on the sender address. After changing the $sender_address to <>
4762 for the sender address verify, Exim was re-instated it as the original
4763 (before rewriting) address, but remembering that it had rewritten it, so it
4764 wasn't rewriting it again. This bug also had the effect of breaking the
4765 sender address verification caching when the sender address was rewritten.
4767 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4768 This has been changed so that if the ip literal address matches
4769 ignore_target_hosts, the router declines.
4771 7. Added expansion conditions match_domain, match_address, and match_local_
4772 part (NOT match_host).
4774 8. The placeholder for the Received: header didn't have a length field set.
4776 9. Added code to Exim itself and to exim_lock to test for a specific race
4777 condition that could lead to file corruption when using MBX delivery. The
4778 issue is with the lockfile that is created in /tmp. If this file is removed
4779 after a process has opened it but before that process has acquired a lock,
4780 there is the potential for a second process to recreate the file and also
4781 acquire a lock. This could lead to two Exim processes writing to the file
4782 at the same time. The added code performs the same test as UW imapd; it
4783 checks after acquiring the lock that its file descriptor still refers to
4784 the same named file.
4786 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4787 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4790 11. Added the smtp_active_hostname option. If used, this will typically be made
4791 to depend on the incoming interface address. Because $interface_address is
4792 not set up until the daemon has forked a reception process, error responses
4793 that can happen earlier (such as "too many connections") no longer contain
4796 12. If an expansion in a condition on a "warn" statement fails because a lookup
4797 defers, the "warn" statement is abandoned, and the next ACL statement is
4798 processed. Previously this caused the whole ACL to be aborted.
4800 13. Added the iplsearch lookup type.
4802 14. Added ident_timeout as a log selector.
4804 15. Added tls_certificate_verified as a log selector.
4806 16. Added a global option tls_require_ciphers (compare the smtp transport
4807 option of the same name). This controls incoming TLS connections.
4809 17. I finally figured out how to make tls_require_ciphers do a similar thing
4810 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4811 before starting the TLS session.
4813 18. Tabs are now shown as \t in -bP output.
4815 19. If the log selector return_path_on_delivery was set, Exim crashed when
4816 bouncing a message because it had too many Received: header lines.
4818 20. If two routers both had headers_remove settings, and the first one included
4819 a superfluous trailing colon, the final name in the first list and the
4820 first name in the second list were incorrectly joined into one item (with a
4821 colon in the middle).
4827 1. Added -C and -D options to the exinext utility, mainly to make it easier
4828 to include in the automated testing, but these could be helpful when
4829 multiple configurations are in use.
4831 2. The exinext utility was not formatting the output nicely when there was
4832 an alternate port involved in the retry record key, nor when there was a
4833 message id as well (for retries that were specific to a specific message
4834 and a specific host). It was also confused by IPv6 addresses, because of
4835 the additional colons they contain. I have fixed the IPv4 problem, and
4836 patched it up to do a reasonable job for IPv6.
4838 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4839 delivery, the log line now contains "pipelined" if PIPELINING was used.
4841 4. An SMTP transport process used to panic and die if the bind() call to set
4842 an explicit outgoing interface failed. This has been changed; it is now
4843 treated in the same way as a connect() failure.
4845 5. A reference to $sender_host_name in the part of a conditional expansion
4846 that was being skipped was still causing a DNS lookup. This no longer
4849 6. The def: expansion condition was not recognizing references to header lines
4850 that used bh_ and bheader_.
4852 7. Added the _cache feature to named lists.
4854 8. The code for checking quota_filecount in the appendfile transport was
4855 allowing one more file than it should have been.
4857 9. For compatibility with Sendmail, the command line option
4865 and sets the incoming protocol and host name (for trusted callers). The
4866 host name and its colon can be omitted when only the protocol is to be set.
4867 Note the Exim already has two private options, -pd and -ps, that refer to
4868 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4869 "s", but I don't think that's a major issue.
4871 10. A number of refactoring changes to the code, none of which should affect
4874 (a) The number of logging options was getting close to filling up the
4875 32-bit word that was used as a bit map. I have split them into two classes:
4876 those that are passed in the argument to log_write(), and those that are
4877 only ever tested independently outside of that function. These are now in
4878 separate 32-bit words, so there is plenty of room for expansion again.
4879 There is no change in the user interface or the logging behaviour.
4881 (b) When building, for example, log lines, the code previously used a
4882 macro that called string_cat() twice, in order to add two strings. This is
4883 not really sufficiently general. Furthermore, there was one instance where
4884 it was actually wrong because one of the argument was used twice, and in
4885 one call a function was used. (As it happened, calling the function twice
4886 did not affect the overall behaviour.) The macro has been replaced by a
4887 function that can join an arbitrary number of extra strings onto a growing
4890 (c) The code for expansion conditions now uses a table and a binary chop
4891 instead of a serial search (which was left over from when there were very
4892 few conditions). Also, it now recognizes conditions like "pam" even when
4893 the relevant support is not compiled in: a suitably worded error message is
4894 given if an attempt is made to use such a condition.
4896 11. Added ${time_interval:xxxxx}.
4898 12. A bug was causing one of the ddress fields not to be passed back correctly
4899 from remote delivery subprocesses. The field in question was not being
4900 subsequently used, so this caused to problems in practice.
4902 13. Added new log selectors queue_time and deliver_time.
4904 14. Might have fixed a bug in maildirsizefile handling that threw up
4905 "unexpected character" debug warnings, and recalculated the data
4906 unnecessarily. In any case, I expanded the warning message to give more
4909 15. Added the message "Restricted characters in address" to the statements in
4910 the default ACL that block characters like @ and % in local parts.
4912 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
4913 Three changes have been made:
4915 (a) There was a serious bug; a negative response to MAIL caused the whole
4916 recipient domain to be cached as invalid, thereby blocking all messages
4917 to all local parts at the same domain, from all senders. This bug has
4918 been fixed. The domain is no longer cached after a negative response to
4919 MAIL if the sender used is not empty.
4921 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
4924 (c) A new callout option, "use_sender" has been added for people who want
4925 the modified behaviour.
4931 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
4934 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
4935 indeed breaks things for older releases.
4937 3. Added additional logging to the case where there is a problem reading data
4938 from a filter that is running in a subprocess using a pipe, in order to
4939 try to track down a specific problem.
4941 4. Testing facility fudge: when running in the test harness and attempting
4942 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
4943 getting "No route to host". Convert this to a timeout.
4945 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
4948 6. Some OS don't have socklen_t but use size_t instead. This affects the
4949 fifth argument of getsockopt() amongst other things. This is now
4950 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
4951 can be set for individual OS. I have set it for SunOS5, OSF1, and
4952 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
4953 some earlier ones do not.
4955 7. Change 4.30/15 was not doing the test caselessly.
4957 8. The standard form for an IPv6 address literal was being rejected by address
4958 parsing in, for example, MAIL and RCPT commands. An example of this kind of
4959 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4960 this, as well as the form without the "IPv6" on the front (but only when
4961 address literals are enabled, of course).
4963 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
4965 10. Exim crashed if a message with an empty sender address specified by -f
4966 encountered a router with an errors_to setting. This could be provoked only
4967 by a command such as
4971 where an empty string was supplied; "<>" did not hit this bug.
4973 11. Installed PCRE release 4.5.
4975 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
4976 remained set. It is now erased.
4978 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
4979 times from message ids (which are base 36 rather than the normal 62).
4981 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
4982 were being counted as actual protocol errors, and logged if the log
4983 selector +smtp_protocol_error was set. One cannot be perfect in this test,
4984 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
4985 and DATA following a set of rejected RCPTs do not count as protocol errors.
4986 In other words, Exim assumes they were pipelined, though this may not
4987 actually be the case. Of course, in all cases the client gets an
4988 appropriate error code.
4990 15. If a lookup fails in an ACL condition, a message about the failure may
4991 be available; it is used if testing the ACL cannot continue, because most
4992 such messages specify what the cause of the deferral is. However, some
4993 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
4994 that caused an old message to be retained and used if a later statement
4995 caused a defer, replacing the real cause of the deferral.
4997 16. If an IP address had so many PTR records that the DNS lookup buffer
4998 was not large enough to hold them, Exim could crash while trying to process
4999 the truncated data. It now detects and logs this case.
5001 17. Further to 4.21/58, another change has been made: if (and only if) the
5002 first line of a message (the first header line) ends with CRLF, a bare LF
5003 in a subsequent header line has a space inserted after it, so as not to
5004 terminate the header.
5006 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5007 unnecessarily, used atoi() instead of strtol(), and didn't check the
5008 termination when getting file sizes from file names by regex.
5010 19. Completely re-implemented the support for maildirsize files, in the light
5011 of a number of problems with the previous contributed implementation
5012 (4.30/29). In particular:
5014 . If the quota is zero, the maildirsize file is maintained, but no quota is
5017 . If the maildir directory does not exist, it is created before any attempt
5018 to write a maildirsize file.
5020 . The quota value in the file is just a cache; if the quota is changed in
5021 the transport, the new value overrides.
5023 . A regular expression is available for excluding directories from the
5026 20. The autoreply transport checks the characters in options that define the
5027 message's headers; it allows continued headers, but it was checking with
5028 isspace() after an embedded newline instead of explicitly looking for a
5031 21. If all the "regular" hosts to which an address was routed had passed their
5032 expiry times, and had not reached their retry times, the address was
5033 bounced, even if fallback hosts were defined. Now Exim should go on to try
5036 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5037 equivalent code in the SMTP transport. Some hosts send humungous responses
5038 to HELO/EHLO, more than 1024 it seems.
5040 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5041 gives compiler warnings in some environments. I've now done it "properly",
5044 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5045 (because of problems with the built-in one) was declared to return uschar *
5046 instead of char *, causing compiler failure.
5048 25. Fixed a file descriptor leak when processing alias/forward files.
5050 26. Fixed a minor format string issue in dbfn.c.
5052 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5054 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5055 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5058 29. When a sender address is verified, it is cached, to save repeating the test
5059 when there is more than one recipient in a message. However, when the
5060 verification involves a callout, it is possible for different callout
5061 options to be set for different recipients. It is too complicated to keep
5062 track of this in the cache, so now Exim always runs a verification when a
5063 callout is required, relying on the callout cache for the optimization.
5064 The overhead is duplication of the address routing, but this should not be
5067 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5068 to be verified with callout=postmaster, and the main callout worked but the
5069 postmaster check failed, the verification correctly failed. However, if a
5070 subsequent RCPT command asked for sender verification *without* the
5071 postmaster check, incorrect caching caused this verification also to fail,
5074 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5075 it was not caching the DNS options (qualify_single, search_parents) that
5076 were used when the lookup failed. A subsequent lookup with different
5077 options therefore always gave the same answer, though there were cases
5078 where it should not have. (Example: a "domains = !$mx_any" option on a
5079 dnslookup router: the "domains" option is always processed without any
5080 widening, but the router might have qualify_single set.) Now Exim uses the
5081 cached value only when the same options are set.
5083 32. Added John Jetmore's "exipick" utility to the distribution.
5085 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5086 than a timeout (e.g. a certificate is required, and is not provided), an
5087 Exim server now closes the connection immediately. Previously it waited for
5088 the client to close - but if the client is SSL, it seems that they each
5089 wait for each other, leading to a delay before one of them times out.
5091 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5092 maintained 0.8.x compatibility because I don't think many are using it, and
5093 it is clearly obsolete.
5095 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5098 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5099 client certificate was expired. A simple patch fixes this, though I don't
5100 understand the full logic of why the verify callback is called multiple
5103 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5104 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5105 which causes problems with some clients (such as the Certicom SSL Plus
5106 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5107 disables the coutermeasure allowing Eudora to connect."
5109 38. Exim was not checking that a write() to a log file succeeded. This could
5110 lead to Bad Things if a log got too big, in particular if it hit a file
5111 size limit. Exim now panics and dies if it cannot write to a log file, just
5112 as it does if it cannot open a log file.
5114 39. Modified OS/Makefile-Linux so that it now contains
5116 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5118 The two -D definitions ensure that Exim is compiled with large file
5119 support, which makes it possible to handle log files that are bigger than
5122 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5123 instance) a domain was checked against a named list that involved a lookup,
5124 causing $domain_data to be set, then another domain was checked against the
5125 same list, then the first domain was re-checked, the value of $domain_data
5126 after the final check could be wrong. In particular, if the second check
5127 failed, it could be set empty. This bug probably also applied to
5130 41. The strip_trailing_dot option was not being applied to the address given
5131 with the -f command-line option.
5133 42. The code for reading a message's header from the spool was incrementing
5134 $received_count, but never initializing it. This meant that the value was
5135 incorrect (doubled) while delivering a message in the same process in which
5136 it was received. In the most common configuration of Exim, this never
5137 happens - a fresh exec is done - but it can happen when
5138 deliver_drop_privilege is set.
5140 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5141 it now includes up to 150 characters of the unexpected data in the log
5144 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5145 and building data strings. The size of both of these buffers was 10 000
5146 bytes - far larger than anybody would *ever* want, thought I. Needless to
5147 say, somebody hit the limit. I have increased the maximum line length to
5148 20 000 and the maximum data length of concatenated lines to 100 000. I have
5149 also fixed two bugs, because there was no checking on these buffers. Tsk,
5150 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5151 buffer is too small.
5153 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5154 lsearch lookups. Now it does.
5156 46. When parsing a route_list item in a manualroute router, a fixed-length
5157 buffer was used for the list of hosts. I made this 1024 bytes long,
5158 thinking that nobody would ever have a list of hosts that long. Wrong.
5159 Somebody had a whole pile of complicated expansion conditions, and the
5160 string was silently truncated, leading to an expansion error. It turns out
5161 that it is easier to change to an unlimited length (owing to other changes
5162 that have happened since this code was originally written) than to build
5163 structure for giving a limitation error. The length of the item that
5164 expands into the list of hosts is now unlimited.
5166 47. The lsearch lookup could not handle data where the length of text line was
5167 more than 4095 characters. Such lines were truncated, leading to shortened
5168 data being returned. It should now handle lines of any length.
5170 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5171 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5174 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5175 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5176 no longer happens in comment lines. A list of replacements is now placed
5177 at the head of all of the source files, except those whose only change is
5178 to replace PERL_COMMAND in the very first #! line.
5180 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5181 messages on the queue, with a bottom-up merge sort, using code contributed
5182 by Michael Haardt. This should make operations like -bp somewhat faster on
5183 large queues. It won't affect queue runners, except when queue_run_in_order
5186 51. Installed eximstats 1.31 in the distribution.
5188 52. Added support for SRV lookups to the dnslookup router.
5190 53. If an ACL referred to $message_body or $message_body_end, the value was not
5191 reset for any messages that followed in the same SMTP session.
5193 54. The store-handling optimization for building very long strings was not
5194 differentiating between the different store pools. I don't think this
5195 actually made any difference in practice, but I've tidied it.
5197 55. While running the routers to verify a sender address, $sender_address
5198 was still set to the sender address. This is wrong, because when routing to
5199 send a bounce to the sender, it would be empty. Therefore, I have changed
5200 it so that, while verifying a sender address, $sender_address is set to <>.
5201 (There is no change to what happens when verifying a recipient address.)
5203 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5204 target A or AAAA records (if not already returned) without resetting the
5205 qualify_single or search_parents options of the DNS resolver. These are
5206 inappropriate in this case because the targets of MX and SRV records must
5207 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5208 target that, when qualified, matched something in the local domain. These
5209 two options are now turned off when doing these lookups.
5211 57. It seems that at least some releases of Reiserfs (which does not have the
5212 concept of a fixed number of inodes) returns zero and not -1 for the
5213 number of available inodes. This interacted badly with check_spool_inodes,
5214 which assumed that -1 was the "no such thing" setting. What I have done is
5215 to check that the total number of inodes is greater than zero before doing
5216 the test of how many are available.
5218 58. When a "warn" ACL statement has a log_message modifier, the message is
5219 remembered, and not repeated. This is to avoid a lot of repetition when a
5220 message has many recipients that cause the same warning to be written.
5221 Howewer, Exim was preserving the list of already written lines for an
5222 entire SMTP session, which doesn't seem right. The memory is now reset if a
5223 new message is started.
5225 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5226 debugging output unless log_rewrite was also set.
5228 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5229 of (int)(handle) when we know that handle contains (void *)(-1).
5231 61. The Exim daemon panic-logs an error return when it closes the incoming
5232 connection. However "connection reset by peer" seems to be common, and
5233 isn't really an error worthy of noting specially, so that particular error
5236 62. When Exim is trying to find all the local interfaces, it used to panic and
5237 die if the ioctl to get the interface flags failed. However, it seems that
5238 on at least one OS (Solaris 9) it is possible to have an interface that is
5239 included in the list of interfaces, but for which you get a failure error
5240 for this call. This happens when the interface is not "plumbed" into a
5241 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5242 failure of the "get flags" call assumes that the interface is down.
5244 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5245 makes life easier for people who are doing arithmetic on fields extracted
5246 from dates, where you often get leading zeros that should not be
5247 interpreted as octal.
5249 64. Added qualify_domain to the redirect router, to override the global
5252 65. If a pathologically long header line contained very many addresses (the
5253 report of this problem mentioned 10 000) and each of them was rewritten,
5254 Exim could use up a very large amount of memory. (It kept on making new
5255 copies of the header line as it rewrote, and never released the old ones.)
5256 At the expense of a bit more processing, the header rewriting function has
5257 been changed so that it no longer eats memory in this way.
5259 66. The generation of the Received: header has been moved from the time that a
5260 message starts to be received, to the time that it finishes. The timestamp
5261 in the Received: header should now be very close to that of the <= log
5262 line. There are two side-effects of this change:
5264 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5265 logged header lines no longer include the local Received: line, because
5266 it has not yet been created. The same applies to a copy of the message
5267 that is returned to a non-SMTP sender when a message is rejected.
5269 (b) When a filter file is tested using -bf, no additional Received: header
5270 is added to the test message. After some thought, I decided that this
5273 This change does not affect the value of $received_for. It is still set
5274 after address rewriting, but before local_scan() is called.
5276 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5278 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5279 gave an unhelpful panic error message, and a defer error. I have managed to
5280 change this behaviour so that it now rejects any supplied certificate,
5281 which seems right, as the list of acceptable certificates is empty.
5283 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5284 gave an unhelpful defer error. I have not managed to make this reject any
5285 supplied certificates, but the error message it gives is "no certificate
5286 supplied", which is not helpful.
5288 70. exigrep's output now also includes lines that are not associated with any
5289 message, but which match the given pattern. Implemented by a patch from
5290 Martin Sluka, which also tidied up the Perl a bit.
5292 71. Recipient callout verification, like sender verification, was using <> in
5293 the MAIL FROM command. This isn't really the right thing, since the actual
5294 sender may affect whether the remote host accepts the recipient or not. I
5295 have changed it to use the actual sender in the callout; this means that
5296 the cache record is now keyed on a recipient/sender pair, not just the
5297 recipient address. There doesn't seem to be a real danger of callout loops,
5298 since a callout by the remote host to check the sender would use <>.
5299 [SEE ABOVE: changed after hitting problems.]
5301 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5302 temporary errors. However, in the case of such a code being given after
5303 the end of a data transmission (i.e. after ".") Exim was failing to write
5304 a retry record for the message. (Yes, there was some broken host that was
5305 actually sending 8xx at this point.)
5307 73. An unknown lookup type in a host list could cause Exim to panic-die when
5308 the list was checked. (An example that provoked this was putting <; in the
5309 middle of a list instead of at the start.) If this happened during a DATA
5310 ACL check, a -D file could be left lying around. This kind of configuration
5311 error no longer causes Exim to die; instead it causes a defer errror. The
5312 incident is still logged to the main and panic logs.
5314 74. Buglet left over from Exim 3 conversion. The message "too many messages
5315 in one connection" was written to the rejectlog but not the mainlog, except
5316 when address rewriting (yes!) was being logged.
5318 75. Added write_rejectlog option.
5320 76. When a system filter was run not as root (that is, when system_filter_user
5321 was set), the values of the $n variables were not being returned to the
5322 main process; thus, they were not subsequently available in the $sn
5325 77. Added +return_path_on_delivery log selector.
5327 78. A connection timeout was being treated differently from recipients deferred
5328 when testing hosts_max_try with a message that was older than the host's
5329 retry timeout. (The host should not be counted, thus allowing all hosts to
5330 be tried at least once before bouncing.) This may have been the cause of an
5331 occasionally reported bug whereby a message would remain on the queue
5332 longer than the retry timeout, but would be bounced if a delivery was
5333 forced. I say "may" because I never totally pinned down the problem;
5334 setting up timeout/retry tests is difficult. See also the next item.
5336 79. The ultimate address timeout was not being applied to errors that involved
5337 a combination of host plus message (for example, a timeout on a MAIL
5338 command). When an address resolved to a number of possible hosts, and they
5339 were not all tried for each delivery (e.g. because of hosts_max_try), a
5340 message could remain on the queue longer than the retry timeout.
5342 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5345 81. Fixed an obscure SMTP outgoing bug which required at least the following
5346 conditions: (a) there was another message waiting for the same server;
5347 (b) the server returned 5xx to all RCPT commands in the first message so
5348 that the message was not completed; (c) the server dropped the connection
5349 or gave a negative response to the RSET that Exim sends to abort the
5350 transaction. The observed case was a dropped connection after DATA that had
5351 been sent in pipelining mode. That is, the server had advertised PIPELINING
5352 but was not implementing it correctly. The effect of the bug was incorrect
5353 behaviour, such as trying another host, and this could lead to a crash.
5359 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5360 and daemon.c were passed as pointers to ints; they should have been
5361 pointers to socklen_t variables (which are typically unsigned ints).
5363 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5366 3. Fixed a really odd bug that affected only the testing scheme; patching a
5367 certain fixed string in the binary changed the value of another string that
5368 happened to be identical to the end of the original first string.
5370 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5371 name", it returns that address as the IP address. On some operating
5372 systems (e.g. Solaris), it also passes back the IP address string as the
5373 "host name". However, on others (e.g. Linux), it passes back an empty
5374 string. Exim wasn't checking for this, and was changing the host name to an
5375 empty string, assuming it had been canonicized.
5377 5. Although rare, it is permitted to have more than one PTR record for a given
5378 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5379 all the names associated with an address, because they do in Solaris.
5380 However, it seems that they do not in Linux for data that comes from the
5381 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5382 I found this out when I moved to a new Linux workstation and tried to run
5383 the Exim test suite.
5385 To get round this problem I have changed the code so that it now does its
5386 own call to the DNS to look up PTR records when searching for a host name.
5387 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5388 addresses that are only in /etc/hosts are still found.
5390 This behaviour is, however, controlled by an option called host_lookup_
5391 order, which defaults to "bydns:byaddr". If people want to use the other
5392 order, or indeed, just use one or the other means of lookup, they can
5393 specify it in this variable.
5395 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5396 some operating systems, this comes back from gethostbyaddr() as an empty
5397 string, and this is what Exim used to test for. However, it seems that in
5398 other systems, "." is yielded. Exim now tests for this case too.
5400 7. The values of check_spool_space and check_log_space are now held internally
5401 as a number of kilobytes instead of an absolute number of bytes. If a
5402 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5403 kilobyte. This means that much larger values can be stored.
5405 8. Exim monitor: an attempt to get the action menu when not actually pointing
5406 at a message produces an empty menu entitled "No message selected". This
5407 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5408 no entries in it ("Shell widget menu has zero width and/or height"). So I
5409 have added a single, blank menu entry in this case.
5411 9. Added ${quote_local_part.
5413 10. MIME decoding is now applied to the contents of Subject: header lines when
5416 11. Now that a reference to $sender_host_address automatically causes a reverse
5417 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5418 host lookup before query-style lookups in lists that might use this
5419 variable. This has therefore been abolished, and the "net-" prefix is no
5420 longer necessary for query-style lookups.
5422 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5423 have been a typo for LFLAGS, so it has been changed.
5425 13. The install script calls Exim with "-C /dev/null" in order to find the
5426 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5427 to be output. Howeve, since Exim outputs its version number before the
5428 error, it didn't break the script. It just looked ugly. I fixed this by
5429 always allowing "-C /dev/null" if the caller is root.
5431 14. Ignore overlarge ACL variable number when reading spool file - insurance
5432 against a later release with more variables having written the file.
5434 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5435 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5436 this, as well as the form without the "IPv6" on the front.
5438 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5439 OS/Makefile-Darwin file.
5441 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5442 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5444 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5445 variable. However, some OS use ints for inodes. I've added cast to long int
5446 to get rid of the compiler warning.
5448 19. I had forgotten to lock out "/../" in configuration file names when
5449 ALT_CONFIG_PREFIX was set.
5451 20. Routers used for verification do not need to specify transports. However,
5452 if such a router generated a host list, and callout was configured, Exim
5453 crashed, because it could not find a port number from the (non-existent)
5454 transport. It now assumes port 25 in this circumstance.
5456 21. Added the -t option to exigrep.
5458 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5459 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5460 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5461 them is set, LOOKUP_LSEARCH is forced.
5463 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5465 24. Added sender and host information to the "rejected by local_scan()" log
5466 line; previously there was no indication of these.
5468 25. Added .include_if_exists.
5470 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5471 Linux. It turns out that not all file systems support this. Apparently some
5472 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5473 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5474 not supported on the file descriptor, is now ignored when Exim is trying to
5475 sync a directory. This applies only to Linux.
5477 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5479 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5480 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5483 29. Added support for maildirsize files from supplied patch (modified a bit).
5485 30. The use of :fail: followed by an empty string could lead Exim to respond to
5486 sender verification failures with (e.g.):
5488 550 Verification failed for <xxx>
5489 550 Sender verify failed
5491 where the first response line was missing the '-' that indicates it is not
5492 the final line of the response.
5494 31. The loop for finding the name of the user that called Exim had a hardwired
5495 limit of 10; it now uses the value of finduser_retries, which is used for
5496 all other user lookups.
5498 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5501 33. Exim was neglecting to zero errno before one call of strtol() when
5502 expanding a string and expecting an integer value. On some systems this
5503 resulted in spurious "integer overflow" errors. Also, it was casting the
5504 result into an int without checking.
5506 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5507 did not work. The code looks as if it has *never* worked, though it appears
5508 to have been documented since at least releast 1.62. I have made it work.
5510 35. The "timeout_DNS" error in retry rules, also documented since at least
5511 1.62, also never worked. As it isn't clear exactly what this means, and
5512 clearly it isn't a major issue, I have abolished the feature by treating it
5513 as "timeout", and writing a warning to the main and panic logs.
5515 36. The display of retry rules for -brt wasn't always showing the error code
5518 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5519 timeout_connect_A, timeout_connect_MX.
5521 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5522 to the empty sender.
5524 39. The daemon was not analysing the content of -oX till after it had closed
5525 stderr and disconnected from the controlling terminal. This meant that any
5526 syntax errors were only noted on the panic log, and the return code from
5527 the command was 0. By re-arranging the code a little, I've made the
5528 decoding happen first, so such errors now appear on stderr, and the return
5529 code is 1. However, the actual setting up of the sockets still happens in
5530 the disconnected process, so errors there are still only recorded on the
5533 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5534 connections (as happens on some IP stacks) was logged at start up time as
5535 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5536 it from "IPv6 and IPv4", which means that two separate sockets are being
5539 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5540 says whether AF_INET or AF_INET6 was passed as an argument.
5542 42. Exiwhat output was messed up when time zones were included in log
5545 43. Exiwhat now gives more information about the daemon's listening ports,
5546 and whether -tls-on-connect was used.
5548 44. The "port" option of the smtp transport is now expanded.
5550 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5551 silently ignored. Now an error message is written to the main and panic
5554 46. There's a new ACL modifier called "logwrite" which writes to a log file
5555 as soon as it is encountered.
5557 47. Added $local_user_uid and $local_user_gid at routing time.
5559 48. Exim crashed when trying to verify a sender address that was being
5562 49. Exim was recognizing only a space character after ".include". It now also
5563 recognizes a tab character.
5565 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5566 extracting the relevant information from the specification. The man page no
5567 longer contains scrambled data for the -d option, and I've added a section
5568 at the front about calling Exim under different names.
5570 51. Added "extra_headers" argument to the "mail" command in filter files.
5572 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5575 53. Installed eximstats 1.29.
5577 54. Added transport_filter_timeout as a generic transport option.
5579 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5580 Cc: header lines. This was required by RFC 822, but it not required by RFC
5583 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5584 incoming messages that did not have them. Now it does so only if the
5585 message originates locally, that is, if there is no associated remote host
5586 address. When Resent- header lines are present, this applies to the Resent-
5587 lines rather than the non-Resent- lines.
5589 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5590 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5592 58. Messages for configuration errors now include the name of the main
5593 configuration file - useful now that there may be more than one file in a
5594 list (.included file names were always shown).
5596 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5597 for those rare installations that do not start the daemon as root or run it
5598 setuid root. I've cut out the call to initgroups() if the daemon is not
5601 60. The Exim user and group can now be bound into the binary as text strings
5602 that are looked up at the start of Exim's processing.
5604 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5606 62. Added $mailstore_basename variable.
5608 63. Installed patch to sieve.c from Michael Haardt.
5610 64. When Exim failed to open the panic log after failing to open the main log,
5611 the original message it was trying to log was written to stderr and debug
5612 output, but if they were not available (the usual case in production), it
5613 was lost. Now it is written to syslog before the two lines that record the
5614 failures to open the logs.
5616 65. Users' Exim filters run in subprocesses under the user's uid. It is
5617 possible for a "deliver" command or an alias in a "personal" command to
5618 provoke an address rewrite. If logging of address rewriting is configured,
5619 this fails because the process is not running as root or exim. There may be
5620 a better way of dealing with this, but for the moment (because 4.30 needs
5621 to be released), I have disabled address rewrite logging when running a
5622 filter in a non-root, non-exim process.
5628 1. The buildconfig auxiliary program wasn't quoting the value set for
5629 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5630 not defined. This bug was masked in 4.22 by the effect that was fixed in
5633 2. Some messages that were rejected after a message id was allocated were
5634 shown as "incomplete" by exigrep. It no longer does this for messages that
5635 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5637 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5638 have allow_domain_literals set, the ID did not get logged in the <= line.
5639 Domain literals are now always recognized in Message-ID: header lines.
5641 4. The first argument for a ${extract expansion item is the key name or field
5642 number. Leading and trailing spaces in this item were not being ignored,
5643 causing some misleading effects.
5645 5. When deliver_drop_privilege was set, single queue runner processes started
5646 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5647 same command in the process it spins off) were not dropping privilege.
5649 6. When the daemon running as "exim" started a queue runner, it always
5650 re-executed Exim in the spun-off process. This is a waste of effort when
5651 deliver_drop_privilege is set. The new process now just calls the
5652 queue-runner function directly.
5658 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5661 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5662 ignored. Though the use of -oP was forcing the writing of a pid file, it
5663 was always written to the default place.
5665 3. If the message "no IP address found for host xxxx" is generated during
5666 incoming verification, it is now followed by identification of the incoming
5667 connection (so you can more easily find what provoked it).
5669 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5671 5. Added some features to "harden" Exim a bit more against certain attacks:
5673 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5674 be put in Local/Makefile. This is like the never_users runtime option,
5675 but it cannot be overridden. The default setting is "root".
5677 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5678 prefix string with which any file named in a -C command line option
5681 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5682 is retained for -C and -D only if the caller of Exim is root. Without
5683 it, the exim user may also use -C and -D and retain privilege.
5685 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5686 command line option is disabled.
5688 6. Macro names set by the -D option must start with an upper case letter, just
5689 like macro names defined in the configuration file.
5691 7. Added "dereference=" facility to LDAP.
5693 8. Two instances of the typo "uknown" in the source files are fixed.
5695 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5696 the Configure-Makefile script screwed up while processing it.
5698 10. Incorporated PCRE 4.4.
5700 11. The SMTP synchronization check was not operating right at the start of an
5701 SMTP session. For example, it could not catch a HELO sent before the client
5702 waited for the greeting. There is now a check for outstanding input at the
5703 point when the greeting is written. Because of the duplex, asynchronous
5704 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5705 way, but not yet received, when the check is performed.
5707 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5708 on TCP/IP sockets, because this apparently causes some broken clients to
5711 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5712 unchanged) from the Cygwin maintainer.
5714 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5715 was supported instead of testing for mbx. Effectively a typo.
5717 15. The spa authenticator server code was not checking that the input it
5718 received was valid base64.
5720 16. The debug output line for the "set" modifier in ACLs was not showing the
5721 name of the variable that was being set.
5723 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5725 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5726 lookup on its own. Something else was needed to trigger the lookup. For
5727 example, a match in host_lookup or the need for a host name in a host list.
5728 Now, if $sender_host_name is referenced and the host name has not yet been
5729 looked up, a lookup is performed. If the lookup fails, the variable remains
5730 empty, and $host_lookup_failed is set to "1".
5732 19. Added "eqi" as a case-independent comparison operator.
5734 20. The saslauthd authentication condition could segfault if neither service
5735 nor realm was specified.
5737 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5738 error message that was logged was misleading, and incoming SMTP
5739 connections were dropped. The message is now more accurate, and temporary
5740 errors are given to SMTP connections.
5742 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5743 (except RSET) with a 421 error, until QUIT is received. However, it was
5744 failing to send a response to QUIT.
5746 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5747 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5748 ACL is now run after the helo_try_verify_hosts code.
5750 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5751 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5752 case-independent, so other case variants are also recognized). Apparently
5753 some systems use these upper case variants.
5755 25. If more than two messages were waiting for the same host, and a transport
5756 filter was specified for the transport, Exim sent two messages over the
5757 same TCP/IP connection, and then failed with "socket operation on non-
5758 socket" when it tried to send the third.
5760 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5762 27. The extern definition of crypt16() in expand.c was not being excluded when
5763 the OS had its own crypt16() function.
5765 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5766 as a preferred synonym for return_size_limit, both as an option and as an
5769 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5771 30. Changed the default configuration ACL to relax the local part checking rule
5772 for addresses that are not in any local domains. For these addresses,
5773 slashes and pipe symbols are allowed within local parts, but the sequence
5774 /../ is explicitly forbidden.
5776 31. SPA server authentication was not clearing the challenge buffer before
5779 32. log_message in a "warn" ACL statement was writing to the reject log as
5780 well as to the main log, which contradicts the documentation and doesn't
5781 seem right (because no rejection is happening). So I have stopped it.
5783 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5784 However, I am unable to do any testing of this.
5786 34. Fixed an infelicity in the appendfile transport. When checking directories
5787 for a mailbox, to see if any needed to be created, it was accidentally
5788 using path names with one or more superfluous leading slashes; tracing
5789 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5791 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5792 contents are added to the log line that is written for every discarded
5793 recipient. (Previously a log_message setting was ignored.)
5795 36. The ${quote: operator now quotes the string if it is empty.
5797 37. The install script runs exim in order to find its version number. If for
5798 some reason other than non-existence or emptiness, which it checks, it
5799 could not run './exim', it was installing it with an empty version number,
5800 i.e. as "exim-". This error state is now caught, and the installation is
5803 38. An argument was missing from the function that creates an error message
5804 when Exim fails to connect to the socket for saslauthd authentication.
5805 This could cause Exim to crash, or give a corrupted message.
5807 39. Added isip, isip4, and isip6 to ${if conditions.
5809 40. The ACL variables $acl_xx are now saved with the message, and can be
5810 accessed later in routers, transports, and filters.
5812 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5813 strings in the file are not string-expanded.
5815 42. If a MAIL command specified a SIZE value that was too large to fit into an
5816 int variable, the check against message_size_limit failed. Such values are
5817 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5818 day this will have to be increased, but I don't think I want to be around
5819 when emails are that large.
5826 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5827 iconv() is not standard in FreeBSD.
5829 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5830 IPv6 enabled. The observed symptom was a segmentation fault on return from
5831 the function os_common_find_running_interfaces() in src/os.c.
5833 3. In the check_special_case() function in daemon.c I had used "errno" as an
5834 argument name, which causes warnings on some systems. This was basically a
5835 typo, since it was named "eno" in the comments!
5837 4. The code that waits for the clock to tick (at a resolution of some fraction
5838 of a second) so as to ensure message-id uniqueness was always waiting for
5839 at least one whole tick, when it could have waited for less. [This is
5840 almost certainly not relevant at current processor speeds, where it is
5841 unlikely to ever wait at all. But we try to future-proof.]
5843 5. The function that sleeps for a time interval that includes fractions of a
5844 second contained a race. It did not block SIGALRM between setting the
5845 timer, and suspending (a couple of lines later). If the interval was short
5846 and the sigsuspend() was delayed until after it had expired, the suspension
5847 never ended. On busy systems this could lead to processes getting stuck for
5850 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5851 process, before it forks any delivery processes. The open lookup caching
5852 mechanism meant that the open file or database connection was passed into
5853 the delivery process. The problem was that delivery processes always tidy
5854 up cached lookup data. This could cause a problem for the next delivery
5855 process started by the queue runner, because the external queue runner
5856 process does not know about the closure. So the next delivery process
5857 still has data in the lookup cache. In the case of a file lookup, there was
5858 no problem because closing a file descriptor in a subprocess doesn't affect
5859 the parent. However, if the lookup was caching a connection to a database,
5860 the connection was closed, and the second delivery process was likely to
5861 see errors such as "PGSQL: query failed: server closed the connection
5862 unexpectedly". The problem has been fixed by closing all cached lookups
5863 in a queue runner before running a delivery process.
5865 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5866 seem to have the "const" qualifier which it has on other OS. I've
5869 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5870 *of the same type* (client or server) with the same public name that an
5871 error should be diagnosed.
5873 9. When Exim looked up a host name for an IP address, but failed to find the
5874 original IP address when looking up the host name (a safety check), it
5875 output the message "<ip address> does not match any IP for NULL", which was
5876 confusing, to say the least. The bug was that the host name should have
5877 appeared instead of "NULL".
5879 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5880 user that is built into the binary, and the -C or -D options is used, root
5881 privilege is dropped before the configuration file is read. In addition,
5882 logging is switched to stderr instead of the normal log files. If the
5883 configuration then re-defines the Exim user, the unprivileged environment
5884 is probably not what is expected, so Exim logs a panic warning message (but
5887 However, if deliver_drop_privilege is set, the unprivileged state may well
5888 be exactly what is intended, so the warning has been cut out in that case,
5889 and Exim is allowed to try to write to its normal log files.
5895 1. smtp_return_error_details was not giving details for temporary sender
5896 or receiver verification errors.
5898 2. Diagnose a configuration error if two authenticators have the same public
5901 3. Exim used not to create the message log file for a message until the first
5902 delivery attempt. This could be confusing when incoming messages were held
5903 for policy or load reasons. The message log file is now created at the time
5904 the message is received, and an initial "Received" line is written to it.
5906 4. The automatically generated man page for command line options had a minor
5907 bug that caused no ill effects; however, a more serious problem was that
5908 the procedure for building the man page automatically didn't always
5909 operate. Consequently, release 4.20 contains an out-of-date version. This
5910 shouldn't happen again.
5912 5. When building Exim with embedded Perl support, the script that builds the
5913 Makefile was calling 'perl' to find its compile-time parameters, ignoring
5914 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
5916 6. The freeze_tell option was not being used for messages that were frozen on
5917 arrival, either by an ACL or by local_scan().
5919 7. Added the smtp_incomplete_transaction log selector.
5921 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
5922 was accepting AUTH without a new EHLO.
5924 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
5925 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
5928 10. Logging of TCP/IP connections (when configured) now happens in the main
5929 daemon process instead of the child process, so that the TCP/IP connection
5930 count is more accurate (but it can never be perfect).
5932 11. The use of "drop" in a nested ACL was not being handled correctly in the
5933 outer ACL. Now, if condition failure induced by the nested "drop" causes
5934 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
5935 or "require"), the connection is dropped.
5937 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
5938 that yield "discard" can now be used with an "accept" or a "discard" verb,
5939 but an error is generated for any others (because I can't see a useful way
5940 to define what should happen).
5942 13. When an ACL is read dynamically from a file (or anywhere else), the lines
5943 are now processed in the same way as lines in the Exim configuration file.
5944 In particular, continuation lines are supported.
5946 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
5948 15. Added -ti meaning -t -i.
5950 16. Check for letters, digits, hyphens, and dots in the names of dnslist
5951 domains, and warn by logging if others are found.
5953 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
5954 returned from GIFCONF when Exim is trying to find the list of interfaces on
5955 a host. The code in os.c has been modified to copy each ifreq to an aligned
5956 structure in all cases.
5958 Also, in some cases, the returned ifreq's were being copied to a 'struct
5959 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
5960 means the last couple of bytes of an IPv6 address could be chopped if the
5961 ifreq contained only a normal sockaddr (14 bytes storage).
5963 18. Named domain lists were not supported in the hosts_treat_as_local option.
5964 An entry such as +xxxx was not recognized, and was treated as a literal
5967 19. Ensure that header lines added by a DATA ACL are included in the reject log
5968 if the ACL subsequently rejects the message.
5970 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
5971 MD5 (which is deprecated).
5973 21. When testing a filter file using -bf, Exim was writing a message when it
5974 took the sender from a "From " line in the message, but it was not doing so
5975 when it took $return_path from a Return-Path: header line. It now does.
5977 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
5978 with a valid header line field name (a series of printing characters
5979 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
5981 23. Changed "disc" in the source to "disk" to conform to the documentation and
5982 the book and for uniformity.
5984 24. Ignore Sendmail's -Ooption=value command line item.
5986 25. When execve() failed while trying to run a command in a pipe transport,
5987 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
5988 could be confused with a return value of 69 from the command itself. This
5989 has been changed to 127, the value the shell returns if it is asked to run
5990 a non-existent command. The wording for the related log line suggests a
5991 non-existent command as the problem.
5993 26. If received_header_text expands to an empty string, do not add a Received:
5994 header line to the message. (Well, it adds a token one on the spool, but
5995 marks it "old" so that it doesn't get used or transmitted.)
5997 27. Installed eximstats 1.28 (addition of -nt option).
5999 28. There was no check for failure on the call to getsockname() in the daemon
6000 code. This can fail if there is a shortage of resources on the system, with
6001 ENOMEM, for example. A temporary error is now given on failure.
6003 29. Contrary to the C standard, it seems that in some environments, the
6004 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6005 program. Exim now does this explicitly; it affects the formatting of
6006 timestamps using strftime().
6008 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6009 complaints. I've now initialized all the variables, to avoid this.
6011 32. Header lines added by a system filter were not being "seen" during
6012 transport-time rewrites.
6014 33. The info_callback() function passed to OpenSSL is set up with type void
6015 (*)(SSL *, int, int), as described somewhere. However, when calling the
6016 function (actually a macro) that sets it up, the type void(*)() is
6017 expected. I've put in a cast to prevent warnings from picky compilers.
6019 34. If a DNS black list lookup found a CNAME record, but there were no A
6020 records associated with the domain it pointed at, Exim crashed.
6022 35. If a DNS black list lookup returned more than one A record, Exim ignored
6023 all but the first. It now scans all returned addresses if a particular IP
6024 value is being sought. In this situation, the contents of the
6025 $dnslist_value variable are a list of all the addresses, separated by a
6028 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6029 to accept a host name and all its aliases if the forward lookup for any of
6030 them yielded the IP address of the incoming connection. Now it accepts only
6031 those names whose forward lookup yields the correct IP address. Any other
6032 names are discarded. This closes a loophole whereby a rogue DNS
6033 administrator could create reverse DNS records to break through a
6034 wildcarded host restriction in an ACL.
6036 37. If a user filter or a system filter that ran in a subprocess used any of
6037 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6038 the wrong values were passed to the pipe command ($thisaddress had the
6039 value of $0, $0 had the value of $1, etc). This bug was introduced by
6040 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6042 38. Improved the line breaking for long SMTP error messages from ACLs.
6043 Previously, if there was no break point between 40 and 75 characters, Exim
6044 left the rest of the message alone. Two changes have been made: (a) I've
6045 reduced the minimum length to 35 characters; (b) if it can't find a break
6046 point between 35 and 75 characters, it looks ahead and uses the first one
6047 that it finds. This may give the occasional overlong line, but at least the
6048 remaining text gets split now.
6050 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6051 file descriptors might be low, and that setting 1000 would always raise it.
6052 It turns out that in some environments, the limit is already over 1000 and
6053 that lowering it causes trouble. So now Exim takes care not to decrease it.
6055 40. When delivering a message, the value of $return_path is set to $sender_
6056 address at the start of routing (routers may change the value). By an
6057 oversight, this default was not being set up when an address was tested by
6058 -bt or -bv, which affected the outcome if any router or filter referred to
6061 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6062 log only when "message" or "log_message" are set. However, if one of the
6063 conditions was an address verification, or a call to a nested ACL, the
6064 messages generated by the underlying test were being passed through. This
6065 no longer happens. The underlying message is available in $acl_verify_
6066 message for both "message" and "log_message" expansions, so it can be
6067 passed through if needed.
6069 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6070 new expansion $bh_ to give the encoded byte string without charset
6071 translation. Translation happens only if iconv() is available; HAVE_ICONV
6072 indicates this at build time. HEADERS_CHARSET gives the charset to
6073 translate to; headers_charset can change it in the configuration, and
6074 "headers charset" can change it in an individual filter file.
6076 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6077 that creates RFC 2047 encoded "words" labels them as that charset instead
6078 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6079 expansion operator; (ii) when Exim creates a From: line for a local
6080 message; (iii) when a header line is rewritten to include a "phrase" part.
6082 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6083 buggy, causing it to skip the first lines of messages whose message ID
6084 ended in 'D'. This would not have bitten before Exim release 4.14, because
6085 message IDs were unlikely to end in 'D' before then. The effect was to have
6086 incorrect size information for certain domains.
6088 45. #include "config.h" was missing at the start of the crypt16.c module. This
6089 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6092 46. If there was a timeout during a "random" callout check, Exim treated it as
6093 a failure of the random address, and carried on sending RSET and the real
6094 address. If the delay was just some slowness somewhere, the response to the
6095 original RCPT would be taken as a response to RSET and so on, causing
6096 mayhem of various kinds.
6098 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6099 when I implemented it. It didn't allow for the fact that some option values
6100 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6101 the right test for positive values.
6103 48. Domain names in DNS records are case-independent. Exim always looks them up
6104 in lower case. Some resolvers return domain names in exactly the case they
6105 appear in the zone file, that is, they may contain uppercase letters. Not
6106 all resolvers do this - some return always lower case. Exim was treating a
6107 change of case by a resolver as a change of domain, similar to a widening
6108 of a domain abbreviation. This triggered its re-routing code and so it was
6109 trying to route what was effectively the same domain again. This normally
6110 caused routing to fail (because the router wouldn't handle the domain
6111 twice). Now Exim checks for this case specially, and just changes the
6112 casing of the domain that it ultimately uses when it transmits the message
6115 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6118 50. If a filter generated a file delivery with a non-absolute name (possible if
6119 no home directory exists for the router), the forbid_file option was not
6122 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6123 the existing equality matching.
6125 52. Exim was using ints instead of ino_t variables in some places where it was
6126 dealing with inode numbers.
6128 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6129 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6130 called TMPDIR, and if it finds it is different, it changes its value.
6132 54. The smtp_printf() function is now made available to local_scan() so
6133 additional output lines can be written before returning. There is also an
6134 smtp_fflush() function to enable the detection of a dropped connection.
6135 The variables smtp_input and smtp_batched_input are exported to
6138 55. Changed the default runtime configuration: the message "Unknown user"
6139 has been removed from the ACL, and instead placed on the localuser router,
6140 using the cannot_route_message feature. This means that any verification
6141 failures that generate their own messages won't get overridden. Similarly,
6142 the "Unrouteable address" message that was in the ACL for unverifiable
6143 relay addresses has also been removed.
6145 56. Added hosts_avoid_esmtp to the smtp transport.
6147 57. The exicyclog script was not checking for the esoteric option
6148 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6149 will work only if exicyclog is run under the appropriate euid.
6151 58. Following a discussion on the list, the rules by which Exim recognises line
6152 endings on incoming messages have been changed. The -dropcr and drop_cr
6153 options are now no-ops, retained only for backwards compatibility. The
6154 following line terminators are recognized: LF CRLF CR. However, special
6155 processing applies to CR:
6157 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6158 nor a local message in the state where . is a terminator.
6160 (ii) If a bare CR is encountered in a header line, an extra space is added
6161 after the line terminator so as not to end the header. The reasoning
6162 behind this is that bare CRs in header lines are most likely either
6163 to be mistakes, or people trying to play silly games.
6165 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6166 was being incorrectly given as 18 bytes larger than it should have been.
6167 This is a VOB (very old bug).
6169 60. This may never have affected anything current, but just in case it has:
6170 When the local host is found other than at the start of a list of hosts,
6171 the local host, those with the same MX, and any that follow, are discarded.
6172 When the list in question was part of a longer list of hosts, the following
6173 hosts (not currently being processed) were also being discarded. This no
6174 longer happens. I'm not sure if this situation could ever has previously
6177 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6180 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6181 header. This is something that is recommended in a new Internet Draft, and
6182 is something that is documented as being done by Sendmail. There are two
6183 possible values. For messages generated by the autoreply transport, Exim
6186 Auto-Submitted: auto-replied
6188 whereas for all other generated messages (e.g. bounces) it adds
6190 Auto-Submitted: auto-generated
6192 63. The "personal" condition in filters now includes a test for the
6193 Auto-Submitted: header. If it contains the string "auto-" the message it
6194 not considered personal.
6196 64. Added rcpt_include_affixes as a generic transport option.
6198 65. Added queue_only_override (default true).
6200 66. Added the syslog_duplication option.
6202 67. If what should have been the first header line of a message consisted of
6203 a space followed by a colon, Exim was mis-interpreting it as a header line.
6204 It isn't of course - it is syntactically invalid and should therefore be
6205 treated as the start of the message body. The misbehaviour could have
6206 caused a number of strange effects, including loss of data in subsequent
6207 header lines, and spool format errors.
6209 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6210 client host had authenticated. This control can now be exercised by an ACL
6211 for more flexibility.
6213 69. By default, callouts do not happen when testing with -bh. There is now a
6214 variant, -bhc, which does actually run the callout code, including
6215 consulting and updating the callout cache.
6217 70. Added support for saslauthd authentication, courtesy of Alexander
6220 71. If statvfs() failed on the spool or log directories while checking their
6221 size for availability, Exim confusingly gave the error "space shortage".
6222 Furthermore, in debugging mode it crashed with a floating point exception.
6223 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6224 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6225 serious problem, Exim now writes to the main and panic logs when this
6226 happens, with details of the failure. It then refuses to accept the
6227 incoming message, giving the message "spool directory problem" or "log
6228 directory problem" with a 421 code for SMTP messages.
6230 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6231 0, 1, and 2 exist, because some OS complain for execs without them (see
6232 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6233 descriptors. However, the code omitted to check that the open succeeded,
6234 causing mysterious errors if for some reason the permissions on /dev/null
6235 got screwed. Now Exim writes a message to the main and panic logs, and
6236 bombs out if it can't open /dev/null.
6238 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6239 interact so that it is all more flexible. It is supposed to remain
6240 backwards compatible. Also added extra_local_interfaces.
6242 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6243 to bomb out with an assertion failure - to the client this appears as a
6244 connection drop. This problem occurs in the part of the code that was taken
6245 from the Samba project. Fortunately, the assertion is in a very simple
6246 function, so I have fixed this by reproducing the function inline in the
6247 one place where it is called, and arranging for authentication to fail
6248 instead of killing the process with assert().
6250 75. The SPA client code was not working when the server requested OEM rather
6251 than Unicode encoding.
6253 76. Added code to make require_files with a specific uid setting more usable in
6254 the case where statting the file as root fails - usually a non-root-mounted
6255 NFS file system. When this happens and the failure is EACCES, Exim now
6256 forks a subprocess and does the per-uid checking as the relevant uid.
6258 77. Added process_log_path.
6260 78. If log_file_path was not explicitly set, a setting of check_log_space or
6261 check_log_inodes was ignored.
6263 79. If a space check for the spool or log partitions fails, the incident is now
6264 logged. Of course, in the latter case the data may get lost...
6266 80. Added the %p formatting code to string_format() so that it can be used to
6267 print addresses in debug_print(). Adjusted all the address printing in the
6268 debugging in store.c to use %p rather than %d.
6270 81. There was a concern that a line of code in smtp_in.c could overflow a
6271 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6272 initially expressed, the concern was not well-founded, because trailing
6273 spaces are removed early. However, if the trailing spaces were followed by
6274 a NULL, they did not get removed, so the overflow was possible. Two fixes
6277 (a) I re-wrote the offending code in a cleaner fashion.
6278 (b) If an incoming SMTP command contains a NULL character, it is rejected
6281 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6282 runs initgroups(), so that if the Exim user is in any additional groups,
6283 they will be used during message reception.
6289 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob