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_hostnames
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18 EXPERIMENTAL_CERTNAMES is withdrawn.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.
109 JH/31 Check the HELO verification lookup for DNSSEC, adding new
110 $sender_helo_dnssec variable.
112 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
114 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
116 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
118 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
119 documented as working, but never had. Support all but $spam_report.
121 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
122 added for tls authenticator.
127 TL/01 When running the test suite, the README says that variables such as
128 no_msglog_check are global and can be placed anywhere in a specific
129 test's script, however it was observed that placement needed to be near
130 the beginning for it to behave that way. Changed the runtest perl
131 script to read through the entire script once to detect and set these
132 variables, reset to the beginning of the script, and then run through
133 the script parsing/test process like normal.
135 TL/02 The BSD's have an arc4random API. One of the functions to induce
136 adding randomness was arc4random_stir(), but it has been removed in
137 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
138 function when detected.
140 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
141 cause callback expansion.
143 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
144 syntax errors in an expansion can be treated as a string instead of
145 logging or causing an error, due to the internal use of bool_lax
146 instead of bool when processing it.
148 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
149 server certificates when making smtp deliveries.
151 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
153 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
155 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
157 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
158 Merged patch from Sebastian Wiedenroth.
160 JH/05 Fix results-pipe from transport process. Several recipients, combined
161 with certificate use, exposed issues where response data items split
162 over buffer boundaries were not parsed properly. This eventually
163 resulted in duplicates being sent. This issue only became common enough
164 to notice due to the introduction of conection certificate information,
165 the item size being so much larger. Found and fixed by Wolfgang Breyha.
167 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
168 size buffer was used, resulting in syntax errors when an expansion
171 JH/07 Add support for directories of certificates when compiled with a GnuTLS
172 version 3.3.6 or later.
174 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
175 is EXPERIMENTAL_EVENT, the main-configuration and transport options
176 both become "event_action", the variables become $event_name, $event_data
177 and $event_defer_errno. There is a new variable $verify_mode, usable in
178 routers, transports and related events. The tls:cert event is now also
179 raised for inbound connections, if the main configuration event_action
182 TL/06 In test suite, disable OCSP for old versions of openssl which contained
183 early OCSP support, but no stapling (appears to be less than 1.0.0).
185 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
186 server certificate names available under the smtp transport option
187 "tls_verify_cert_hostname" now do not permit multi-component wildcard
190 JH/10 Time-related extraction expansions from certificates now use the main
191 option "timezone" setting for output formatting, and are consistent
192 between OpenSSL and GnuTLS compilations. Bug 1541.
194 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
195 encoded parameter in the incoming message. Bug 1558.
197 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
198 include certificate info, eximon was claiming there were spoolfile
201 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
203 JH/14 Log delivery-related information more consistently, using the sequence
204 "H=<name> [<ip>]" wherever possible.
206 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
207 are problematic for Debian distribution, omit them from the release
210 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
212 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
215 JH/17 Fix a null-indirection in certextract expansions when a nondefault
216 output list separator was used.
221 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
222 checkers that were complaining about end of non-void function with no
225 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
226 This was a regression intruduced in 4.83 by another bugfix.
228 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
230 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
231 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
237 TF/01 Correctly close the server side of TLS when forking for delivery.
239 When a message was received over SMTP with TLS, Exim failed to clear up
240 the incoming connection properly after forking off the child process to
241 deliver the message. In some situations the subsequent outgoing
242 delivery connection happened to have the same fd number as the incoming
243 connection previously had. Exim would try to use TLS and fail, logging
244 a "Bad file descriptor" error.
246 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
247 utilities have not been installed.
249 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
250 temporary space as the ACL may create new global variables.
252 TL/01 LDAP support uses per connection or global context settings, depending
253 upon the detected version of the libraries at build time.
255 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
256 to extract and use the src ip:port in logging and expansions as if it
257 were a direct connection from the outside internet. PPv2 support was
258 updated based on HAProxy spec change in May 2014.
260 JH/02 Add ${listextract {number}{list}{success}{fail}}.
262 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
263 Properly escape header and check for NULL return.
265 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
268 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
270 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
271 characters in header names, implemented as a verify condition.
272 Contributed by Michael Fischer v. Mollard.
274 TL/05 Rename SPF condition results err_perm and err_temp to standardized
275 results permerror and temperror. Previous values are deprecated but
276 still accepted. In a future release, err_perm and err_temp will be
277 completely removed, which will be a backward incompatibility if the
278 ACL tests for either of these two old results. Patch contributed by
279 user bes-internal on the mailing list.
281 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
283 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
284 selectors, in both main and reject logs.
286 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
289 JH/07 Add malware type "sock" for talking to simple daemon.
291 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
293 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
294 routers/transports under cutthrough routing.
296 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
297 numbers. Touch up "bool" conditional to keep the same definition.
299 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
301 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
303 JH/12 Expand items in router/transport headers_add or headers_remove lists
304 individually rather than the list as a whole. Bug 1452.
306 Required for reasonable handling of multiple headers_ options when
307 they may be empty; requires that headers_remove items with embedded
308 colons must have them doubled (or the list-separator changed).
310 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
311 view the policy declared in the DMARC record. Currently, $dmarc_status
312 is a combined value of both the record presence and the result of the
315 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
317 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
318 dnslookup router and the smtp transport (applying to the forward
321 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
322 of ldap servers used for a specific lookup. Patch provided by Heiko
325 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
326 New variable $lookup_dnssec_authenticated for observability.
328 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
329 Patch submitted by Lars Timman.
331 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
333 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
334 Requires trusted mode and valid format message id, aborts otherwise.
335 Patch contributed by Heiko Schlichting.
337 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
338 certextract with support for various fields. Bug 1358.
340 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
341 is requested by default, modifiable by smtp transport option
344 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
345 operate on certificate variables to give certificate fingerprints
346 Also new ${sha256:cert_variable}.
348 JH/23 The PRDR feature is moved from being Experimental into the mainline.
350 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
351 Christian Aistleitner.
353 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
355 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
356 file. Patch from Wolfgang Breyha.
358 JH/25 Expand the coverage of the delivery $host and $host_address to
359 client authenticators run in verify callout. Bug 1476.
361 JH/26 Port service names are now accepted for tls_on_connect_ports, to
362 align with daemon_smtp_ports. Bug 72.
364 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
365 support and error reporting did not work properly.
367 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
368 and is readable. Patch from Andrew Colin Kissa.
370 TL/14 Enhance documentation of ${run expansion and how it parses the
371 commandline after expansion, particularly in the case when an
372 unquoted variable expansion results in an empty value.
374 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
376 PP/02 Fix internal collision of T_APL on systems which support RFC3123
377 by renaming away from it. Addresses GH issue 15, reported by
380 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
382 TL/15 SECURITY: prevent double expansion in math comparison functions
383 (can expand unsanitized data). Not remotely exploitable.
390 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
392 PP/02 Make -n do something, by making it not do something.
393 When combined with -bP, the name of an option is not output.
395 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
398 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
399 $sender_host_name and config options to manage this, and basic check
402 PP/05 DSCP support for outbound connections and control modifier for inbound.
404 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
405 (Only plugin which currently uses this is kerberos4, which nobody should
406 be using, but we should make it available and other future plugins might
407 conceivably use it, even though it would break NAT; stuff *should* be
408 using channel bindings instead).
410 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
411 name; added for Sendmail compatibility; requires admin caller.
412 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
413 just ignore it); requires trusted caller.
414 Also parse but ignore: -Ac -Am -X<logfile>
417 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
419 TL/02 Add +smtp_confirmation as a default logging option.
421 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
422 Patch by Magnus Holmgren from 2007-02-20.
424 TL/04 Bugzilla 1281 - Spec typo.
425 Bugzilla 1283 - Spec typo.
426 Bugzilla 1290 - Spec grammar fixes.
428 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
430 TL/06 Add Experimental DMARC support using libopendmarc libraries.
432 TL/07 Fix an out of order global option causing a segfault. Reported to dev
433 mailing list by by Dmitry Isaikin.
435 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
437 JH/02 Support "G" suffix to numbers in ${if comparisons.
439 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
441 NM/01 Bugzilla 1197 - Spec typo
442 Bugzilla 1196 - Spec examples corrections
444 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
446 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
447 gnutls_enable_pkcs11, but renamed to more accurately indicate its
450 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
451 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
453 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
454 "acl {{name}{arg}...}", and optional args on acl condition
457 JH/05 Permit multiple router/transport headers_add/remove lines.
459 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
461 JH/07 Avoid using a waiting database for a single-message-only transport.
462 Performance patch from Paul Fisher. Bugzilla 1262.
464 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
467 JH/09 Add $headers_added variable, with content from use of ACL modifier
468 add_header (but not yet added to the message). Bugzilla 199.
470 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
471 Pulled from Bugzilla 817 by Wolfgang Breyha.
473 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
475 (nb: this is the same fix as in Exim 4.80.1)
477 JH/11 Add A= logging on delivery lines, and a client_set_id option on
480 JH/12 Add optional authenticated_sender logging to A= and a log_selector
483 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
485 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
486 advertise SMTP AUTH mechanism to us, instead of a generic
487 protocol violation error. Also, make Exim more robust to bad
488 data from the Dovecot auth socket.
490 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
492 When a queue runner is handling a message, Exim first routes the
493 recipient addresses, during which it prunes them based on the retry
494 hints database. After that it attempts to deliver the message to
495 any remaining recipients. It then updates the hints database using
498 So if a recipient address works intermittently, it can get repeatedly
499 deferred at routing time. The retry hints record remains fresh so the
500 address never reaches the final cutoff time.
502 This is a fairly common occurrence when a user is bumping up against
503 their storage quota. Exim had some logic in its local delivery code
504 to deal with this. However it did not apply to per-recipient defers
505 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
507 This change adds a proper retry rule check during routing so that the
508 final cutoff time is checked against the message's age. We only do
509 this check if there is an address retry record and there is not a
510 domain retry record; this implies that previous attempts to handle
511 the address had the retry_use_local_parts option turned on. We use
512 this as an approximation for the destination being like a local
513 delivery, as in LMTP.
515 I suspect this new check makes the old local delivery cutoff check
516 redundant, but I have not verified this so I left the code in place.
518 TF/02 Correct gecos expansion when From: is a prefix of the username.
520 Test 0254 submits a message to Exim with the header
524 When I ran the test suite under the user fanf2, Exim expanded
525 the header to contain my full name, whereas it should have added
526 a Resent-Sender: header. It erroneously treats any prefix of the
527 username as equal to the username.
529 This change corrects that bug.
531 GF/01 DCC debug and logging tidyup
532 Error conditions log to paniclog rather than rejectlog.
533 Debug lines prefixed by "DCC: " to remove any ambiguity.
535 TF/03 Avoid unnecessary rebuilds of lookup-related code.
537 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
538 Bug spotted by Jeremy Harris; was flawed since initial commit.
539 Would have resulted in OCSP responses post-SNI triggering an Exim
540 NULL dereference and crash.
542 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
544 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
545 Bug detection, analysis and fix by Samuel Thibault.
546 Bugzilla 1331, Debian bug #698092.
548 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
550 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
551 Server implementation by Todd Lyons, client by JH.
552 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
553 config variable "prdr_enable" controls whether the server
554 advertises the facility. If the client requests PRDR a new
555 acl_data_smtp_prdr ACL is called once for each recipient, after
556 the body content is received and before the acl_smtp_data ACL.
557 The client is controlled by bolth of: a hosts_try_prdr option
558 on the smtp transport, and the server advertisement.
559 Default client logging of deliveries and rejections involving
560 PRDR are flagged with the string "PRDR".
562 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
563 fclose(). Diagnosis by Todd Lyons.
565 PP/17 Update configure.default to handle IPv6 localhost better.
566 Patch by Alain Williams (plus minor tweaks).
569 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
570 This is now consistent with GnuTLS, and is now documented: the
571 previous undocumented portable approach to treating the option as
572 unset was to force an expansion failure. That still works, and
573 an empty string is now equivalent.
575 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
576 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
577 not performing validation itself.
579 PP/20 Added force_command boolean option to pipe transport.
580 Patch from Nick Koston, of cPanel Inc.
582 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
585 TF/04 Added udpsend ACL modifer and hexquote expansion operator
587 PP/21 Fix eximon continuous updating with timestamped log-files.
588 Broken in a format-string cleanup in 4.80, missed when I repaired the
589 other false fix of the same issue.
590 Report and fix from Heiko Schlichting.
593 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
594 Report from Prashanth Katuri.
596 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
597 It's SecureTransport, so affects any MacOS clients which use the
598 system-integrated TLS libraries, including email clients.
600 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
601 using a MIME ACL for non-SMTP local injection.
602 Report and assistance in diagnosis by Warren Baker.
604 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
606 JH/16 Fix comparisons for 64b. Bugzilla 1385.
608 TL/09 Add expansion variable $authenticated_fail_id to keep track of
609 last id that failed so it may be referenced in subsequent ACL's.
611 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
614 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
615 ldap library initialization, allowing self-signed CA's to be
616 used. Also properly sets require_cert option later in code by
617 using NULL (global ldap config) instead of ldap handle (per
618 session). Bug diagnosis and testing by alxgomz.
620 TL/12 Enhanced documentation in the ratelimit.pl script provided in
621 the src/util/ subdirectory.
623 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
624 renamed to Transport Post Delivery Action by Jeremy Harris, as
627 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
628 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
629 redis_servers = needs to be configured which will be used by the redis
630 lookup. Patch from Warren Baker, of The Packet Hub.
632 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
634 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
635 hostname or reverse DNS when processing a host list. Used suggestions
636 from multiple comments on this bug.
638 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
640 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
641 Missed a few lines, added it to make the runtest require no keyboard
644 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
645 contains upper case chars. Make router use caseful_local_part.
647 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
648 support when GnuTLS has been built with p11-kit.
654 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
656 This, or similar/improved, will also be change PP/11 of 4.82.
662 PP/01 Handle short writes when writing local log-files.
663 In practice, only affects FreeBSD (8 onwards).
664 Bugzilla 1053, with thanks to Dmitry Isaikin.
666 NM/01 Bugzilla 949 - Documentation tweak
668 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
671 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
673 PP/02 Implemented gsasl authenticator.
675 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
677 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
678 `pkg-config foo` for cflags/libs.
680 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
681 with rest of GSASL and with heimdal_gssapi.
683 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
684 `pkg-config foo` for cflags/libs for the TLS implementation.
686 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
687 properties get this fed in as external SSF. A number of robustness
688 and debugging improvements to the cyrus_sasl authenticator.
690 PP/08 cyrus_sasl server now expands the server_realm option.
692 PP/09 Bugzilla 1214 - Log authentication information in reject log.
693 Patch by Jeremy Harris.
695 PP/10 Added dbmjz lookup type.
697 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
699 PP/12 MAIL args handles TAB as well as SP, for better interop with
700 non-compliant senders.
701 Analysis and variant patch by Todd Lyons.
703 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
704 Bug report from Lars Müller <lars@samba.org> (via SUSE),
705 Patch from Dirk Mueller <dmueller@suse.com>
707 PP/13 tls_peerdn now print-escaped for spool files.
708 Observed some $tls_peerdn in wild which contained \n, which resulted
709 in spool file corruption.
711 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
712 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
713 or write after TLS renegotiation, which otherwise led to messages
716 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
717 as a tracking header (ie: a signed header comes before the signature).
718 Patch from Wolfgang Breyha.
720 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
721 comma-sep list; embedded commas doubled.
723 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
725 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
727 Report and patch from Dmitry Banschikov.
729 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
730 Removed SSL_clear() after SSL_new() which led to protocol negotiation
731 failures. We appear to now support TLS1.1+ with Exim.
733 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
734 lets Exim select keys and certificates based upon TLS SNI from client.
735 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
736 before an outbound SMTP session. New log_selector, +tls_sni.
738 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
739 NULL dereference. Report and patch from Alun Jones.
741 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
742 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
743 Not seeing resolver debug output on NetBSD, but suspect this is a
744 resolver implementation change.
746 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
747 Left warnings. Added "eximon gdb" invocation mode.
749 PP/21 Defaulting "accept_8bitmime" to true, not false.
751 PP/22 Added -bw for inetd wait mode support.
753 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
754 locate the relevant includes and libraries. Made this the default.
756 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
757 Bugzilla 1246, report and most of solution from Tomasz Kusy.
759 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
760 This may cause build issues on older platforms.
762 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
763 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
764 gnutls_require_mac & gnutls_require_protocols (no longer supported).
765 Added SNI support via GnuTLS too.
766 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
768 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
770 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
771 Applied second patch from Janne, implementing suggestion to default
772 multiple-strings-in-record handling to match SPF spec.
774 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
776 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
777 read-only, out of scope).
778 Patch from Wolfgang Breyha, report from Stuart Northfield.
780 PP/29 Fix three issues highlighted by clang analyser static analysis.
781 Only crash-plausible issue would require the Cambridge-specific
782 iplookup router and a misconfiguration.
783 Report from Marcin Mirosław.
785 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
787 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
788 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
789 As part of this, removing so much warning spew let me fix some minor
790 real issues in debug logging.
792 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
793 assignment on my part. Fixed.
795 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
796 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
797 Janne Snabb (who went above and beyond: thank you).
799 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
800 string otherwise requires a connection and a bunch more work and it's
801 relatively easy to get wrong. Should also expose TLS library linkage
804 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
805 64-bit ${eval} (JH/03).
807 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
808 GNU libc to support some of the 64-bit stuff, should not lead to
809 conflicts. Defined before os.h is pulled in, so if a given platform
810 needs to override this, it can.
812 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
813 protection layer was required, which is not implemented.
814 Bugzilla 1254, patch from Wolfgang Breyha.
816 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
817 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
818 tls_dhparam take prime identifiers. Also unbreak combination of
819 OpenSSL+DH_params+TLSSNI.
821 PP/39 Disable SSLv2 by default in OpenSSL support.
827 PP/01 Solaris build fix for Oracle's LDAP libraries.
828 Bugzilla 1109, patch from Stephen Usher.
830 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
832 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
835 TF/02 Fix a couple more cases where we did not log the error message
836 when unlink() failed. See also change 4.74-TF/03.
838 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
839 lock up or crash if it happened to be inside a call to libc when it
840 got a SIGUSR1 from exiwhat.
842 The SIGUSR1 handler appends the current process status to the process
843 log which is later printed by exiwhat. It used to use the general
844 purpose logging code to do this, but several functions it calls are
845 not safe for signals.
847 The new output code in the SIGUSR1 handler is specific to the process
848 log, and simple enough that it's easy to inspect for signal safety.
849 Removing some special cases also simplifies the general logging code.
850 Removing the spurious timestamps from the process log simplifies
853 TF/04 Improved ratelimit ACL condition.
855 The /noupdate option has been deprecated in favour of /readonly which
856 has clearer semantics. The /leaky, /strict, and /readonly update modes
857 are mutually exclusive. The update mode is no longer included in the
858 database key; it just determines when the database is updated. (This
859 means that when you upgrde Exim will forget old rate measurements.)
861 Exim now checks that the per_* options are used with an update mode that
862 makes sense for the current ACL. For example, when Exim is processing a
863 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
864 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
865 must specify per_mail/readonly. If you omit the update mode it defaults to
866 /leaky where that makes sense (as before) or /readonly where required.
868 The /noupdate option is now undocumented but still supported for
869 backwards compatibility. It is equivalent to /readonly except that in
870 ACLs where /readonly is required you may specify /leaky/noupdate or
871 /strict/noupdate which are treated the same as /readonly.
873 A useful new feature is the /count= option. This is a generalization
874 of the per_byte option, so that you can measure the throughput of other
875 aggregate values. For example, the per_byte option is now equivalent
876 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
878 The per_rcpt option has been generalized using the /count= mechanism
879 (though it's more complicated than the per_byte equivalence). When it is
880 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
881 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
882 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
883 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
884 Note that using per_rcpt with a non-readonly update mode in more than
885 one ACL will cause the recipients to be double-counted. (The per_mail
886 and per_byte options don't have this problem.)
888 The handling of very low rates has changed slightly. If the computed rate
889 is less than the event's count (usually one) then this event is the first
890 after a long gap. In this case the rate is set to the same as this event's
891 count, so that the first message of a spam run is counted properly.
893 The major new feature is a mechanism for counting the rate of unique
894 events. The new per_addr option counts the number of different
895 recipients that someone has sent messages to in the last time period. It
896 behaves like per_rcpt if all the recipient addresses are different, but
897 duplicate recipient addresses do not increase the measured rate. Like
898 the /count= option this is a general mechanism, so the per_addr option
899 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
900 example, measure the rate that a client uses different sender addresses
901 with the options per_mail/unique=$sender_address. There are further
902 details in the main documentation.
904 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
906 TF/06 Removed a few PCRE remnants.
908 TF/07 Automatically extract Exim's version number from tags in the git
909 repository when doing development or release builds.
911 PP/02 Raise smtp_cmd_buffer_size to 16kB.
912 Bugzilla 879. Patch from Paul Fisher.
914 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
915 Heavily based on revision 40f9a89a from Simon Arlott's tree.
918 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
920 PP/05 Variable $av_failed, true if the AV scanner deferred.
921 Bugzilla 1078. Patch from John Horne.
923 PP/06 Stop make process more reliably on build failure.
924 Bugzilla 1087. Patch from Heiko Schlittermann.
926 PP/07 Make maildir_use_size_file an _expandable_ boolean.
927 Bugzilla 1089. Patch from Heiko Schlittermann.
929 PP/08 Handle ${run} returning more data than OS pipe buffer size.
930 Bugzilla 1131. Patch from Holger Weiß.
932 PP/09 Handle IPv6 addresses with SPF.
933 Bugzilla 860. Patch from Wolfgang Breyha.
935 PP/10 GnuTLS: support TLS 1.2 & 1.1.
937 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
940 PP/11 match_* no longer expand right-hand-side by default.
941 New compile-time build option, EXPAND_LISTMATCH_RHS.
942 New expansion conditions, "inlist", "inlisti".
944 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
946 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
948 PP/14 fix log_write() format string regression from TF/03.
949 Bugzilla 1152. Patch from Dmitry Isaikin.
955 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
957 PP/02 Harmonised TLS library version reporting; only show if debugging.
958 Layout now matches that introduced for other libraries in 4.74 PP/03.
960 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
962 PP/04 New "dns_use_edns0" global option.
964 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
967 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
968 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
970 TK/01 Updated PolarSSL code to 0.14.2.
971 Bugzilla 1097. Patch from Andreas Metzler.
973 PP/07 Catch divide-by-zero in ${eval:...}.
976 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
979 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
980 format-string attack -- SECURITY: remote arbitrary code execution.
982 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
983 time unintentionally subject to list matching rules, letting the header
984 cause arbitrary Exim lookups (of items which can occur in lists, *not*
985 arbitrary string expansion). This allowed for information disclosure.
987 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
988 INT_MIN/-1 -- value coerced to INT_MAX.
994 NM/01 Workround for PCRE version dependency in version reporting
997 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
998 This fixes portability to compilers other than gcc, notably
999 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1001 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1002 makefiles for portability to HP-UX and POSIX correctness.
1004 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1005 Also via indented variable definition in the Makefile.
1006 (Debugging by Oliver Heesakkers).
1008 PP/02 Restore caching of spamd results with expanded spamd_address.
1009 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1011 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1012 Improves build reliability. Fix from: Frank Elsner
1014 NM/02 Fix wide character breakage in the rfc2047 coding
1015 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1017 NM/03 Allow underscore in dnslist lookups
1018 Fixes bug 1026. Patch from Graeme Fowler
1020 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1021 Code patches from Adam Ciarcinski of NetBSD.
1023 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1026 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1027 is logged, to avoid truncation. Patch from John Horne.
1029 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1030 Patch from Jakob Hirsch.
1032 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1033 SQL string expansion failure details.
1034 Patch from Andrey Oktyabrski.
1036 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1037 Patch from Simon Arlott.
1039 PP/09 New lookups functionality failed to compile on old gcc which rejects
1040 extern declarations in function scope.
1041 Patch from Oliver Fleischmann
1043 PP/10 Use sig_atomic_t for flags set from signal handlers.
1044 Check getgroups() return and improve debugging.
1045 Fixed developed for diagnosis in bug 927 (which turned out to be
1048 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1049 Patch from Mark Zealey.
1051 PP/12 Bugzilla 1056: Improved spamd server selection.
1052 Patch from Mark Zealey.
1054 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1055 Based on patch from Heiko Schlittermann.
1057 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1058 Patch from Uwe Doering, sign-off by Michael Haardt.
1060 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1061 variable declaration deep within a block. Bug and patch from
1064 PP/15 lookups-Makefile IRIX compatibilty coercion.
1066 PP/16 Make DISABLE_DKIM build knob functional.
1068 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1069 Patch by Simon Arlott
1071 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1072 variable argument macros. Our copy now differs from upstream.
1078 TF/01 Failure to get a lock on a hints database can have serious
1079 consequences so log it to the panic log.
1081 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1082 controlled using the smtp_confirmation log selector.
1084 TF/03 Include the error message when we fail to unlink a spool file.
1086 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1087 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1088 for maintaining out-of-tree patches for some time.
1090 PP/01 Bugzilla 139: Documentation and portability issues.
1091 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1092 Handle per-OS dynamic-module compilation flags.
1094 PP/02 Let /dev/null have normal permissions.
1095 The 4.73 fixes were a little too stringent and complained about the
1096 permissions on /dev/null. Exempt it from some checks.
1097 Reported by Andreas M. Kirchwitz.
1099 PP/03 Report version information for many libraries, including
1100 Exim version information for dynamically loaded libraries. Created
1101 version.h, now support a version extension string for distributors
1102 who patch heavily. Dynamic module ABI change.
1104 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1105 privilege escalation vulnerability whereby the Exim run-time user
1106 can cause root to append content of the attacker's choosing to
1109 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1112 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1113 If dropping privileges for untrusted macros, we disabled normal logging
1114 on the basis that it would fail; for the Exim run-time user, this is not
1115 the case, and it resulted in successful deliveries going unlogged.
1116 Fixed. Reported by Andreas Metzler.
1122 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1123 only prepend for the Resent-* case. Fixes regression introduced in
1124 Exim 4.70 by NM/22 for Bugzilla 607.
1126 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1127 increasing numbers of administrators be bitten by this.
1129 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1131 PP/03 Bugzilla 994: added openssl_options main configuration option.
1133 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1135 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1137 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1139 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1140 without a peer certificate, leading to a segfault because of an
1141 assumption that peers always have certificates. Be a little more
1142 paranoid. Problem reported by Martin Tscholak.
1144 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1145 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1146 NB: ClamAV planning to remove STREAM in "middle of 2010".
1147 CL also introduces -bmalware, various -d+acl logging additions and
1148 more caution in buffer sizes.
1150 PP/09 Implemented reverse_ip expansion operator.
1152 PP/10 Bugzilla 937: provide a "debug" ACL control.
1154 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1156 PP/12 Bugzilla 973: Implement --version.
1158 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1160 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1162 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1164 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1165 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1166 ignore trailing whitespace.
1168 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1170 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1173 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1174 Notification from Dr Andrew Aitchison.
1176 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1177 ExtendedDetectionInfo response format.
1178 Notification from John Horne.
1180 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1183 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1184 XSL and documented dependency on system catalogs, with examples of how
1187 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1190 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1191 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1192 configuration file which is writeable by the Exim user or group.
1194 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1195 of configuration files to cover files specified with the -C option if
1196 they are going to be used with root privileges, not just the default
1199 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1200 option (effectively making it always true).
1202 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1203 files to be used while preserving root privileges.
1205 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1206 that rogue child processes cannot use them.
1208 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1209 run-time user, instead of root.
1211 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1212 Exim run-time user without dropping privileges.
1214 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1215 result string, instead of calling string_vformat() twice with the same
1218 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1219 for other users. Others should always drop root privileges if they use
1220 -C on the command line, even for a whitelisted configure file.
1222 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1224 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1230 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1231 $data_path, and $header_path variables; fixed documentation bugs and
1234 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1235 exipick to access non-standard spools, including the "frozen" queue
1238 NM/01 Bugzilla 965: Support mysql stored procedures.
1239 Patch from Alain Williams
1241 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1243 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1244 Patch from Andreas Metzler
1246 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1247 Patch from Kirill Miazine
1249 NM/05 Bugzilla 671: Added umask to procmail example.
1251 JJ/03 installed exipick 20100323.0, fixing doc bug
1253 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1254 directory. Notification and patch from Dan Rosenberg.
1256 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1258 TK/02 Improve log output when DKIM signing operation fails.
1260 MH/01 Treat the transport option dkim_domain as a colon separated
1261 list, not as a single string, and sign the message with each element,
1262 omitting multiple occurences of the same signer.
1264 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1265 Bugzilla 985, 986. Patch by Simon Arlott
1267 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1268 Patch by Simon Arlott
1270 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1271 MBX locking. Notification from Dan Rosenberg.
1277 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1279 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1281 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1283 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1285 NM/04 Bugzilla 915: Flag broken perl installation during build.
1291 TK/01 Added patch by Johannes Berg that expands the main option
1292 "spamd_address" if it starts with a dollar sign.
1294 TK/02 Write list of recipients to X-Envelope-Sender header when building
1295 the mbox-format spool file for content scanning (suggested by Jakob
1298 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1299 (http://www.dcc-servers.net/) support via dccifd. Activated by
1300 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1302 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1303 by Mark Daniel Reidel <mr@df.eu>.
1305 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1306 When building exim an external PCRE library is now needed -
1307 PCRE is a system library on the majority of modern systems.
1308 See entry on PCRE_LIBS in EDITME file.
1310 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1311 conversation. Added nologin parameter to request.
1312 Patch contributed by Kirill Miazine.
1314 TF/01 Do not log submission mode rewrites if they do not change the address.
1316 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1318 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1319 log files in place. Contributed by Roberto Lima.
1321 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1323 TF/03 Bugzilla 615: When checking the local_parts router precondition
1324 after a local_part_suffix or local_part_prefix option, Exim now
1325 does not use the address's named list lookup cache, since this
1326 contains cached lookups for the whole local part.
1328 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1329 Robert Millan. Documentation is in experimental-spec.txt.
1331 TF/04 Bugzilla 668: Fix parallel build (make -j).
1333 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1335 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1336 Patch provided by Jan Srzednicki.
1338 TF/05 Leading white space used to be stripped from $spam_report which
1339 wrecked the formatting. Now it is preserved.
1341 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1342 that they are available at delivery time.
1344 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1346 TF/08 TLS error reporting now respects the incoming_interface and
1347 incoming_port log selectors.
1349 TF/09 Produce a more useful error message if an SMTP transport's hosts
1350 setting expands to an empty string.
1352 NM/06 Bugzilla 744: EXPN did not work under TLS.
1353 Patch provided by Phil Pennock.
1355 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1356 Patch provided by Richard Godbee.
1358 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1359 acl_smtp_notquit, added index entry.
1361 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1362 Patch provided by Eugene Bujak.
1364 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1365 accept(). Patch provided by Maxim Dounin.
1367 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1368 Patch provided by Phil Pennock.
1370 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1372 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1373 Patch provided by Brad "anomie" Jorsch.
1375 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1376 Patch provided by Dean Brooks.
1378 TK/05 Add native DKIM support (does not depend on external libraries).
1380 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1381 Patch provided by Graeme Fowler.
1383 NM/16 Bugzilla 851: Documentation example syntax fix.
1385 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1387 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1390 NM/19 Bugzilla 745: TLS version reporting.
1391 Patch provided by Phil Pennock.
1393 NM/20 Bugzilla 167: bool: condition support.
1394 Patch provided by Phil Pennock.
1396 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1397 clients. Patch provided by Phil Pennock.
1399 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1400 Patch provided by Brad "anomie" Jorsch.
1402 NM/23 Bugzilla 687: Fix misparses in eximstats.
1403 Patch provided by Heiko Schlittermann.
1405 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1406 Patch provided by Heiko Schlittermann.
1408 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1409 plus update to original patch.
1411 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1413 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1414 Patch provided by David Brownlee.
1416 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1418 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1420 NM/30 Bugzilla 888: TLS documentation bugfixes.
1422 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1424 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1425 Unlike the original bugzilla I have changed all shell scripts in src tree.
1427 NM/33 Bugzilla 898: Transport filter timeout fix.
1428 Patch by Todd Rinaldo.
1430 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1431 Patch by Serge Demonchaux.
1433 NM/35 Bugzilla 39: Base64 decode bug fixes.
1434 Patch by Jakob Hirsch.
1436 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1438 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1440 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1442 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1448 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1449 ALT-N's libdkim that I have put here:
1450 http://duncanthrax.net/exim-experimental/
1452 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1453 were called 'true' and it seems that C99 defines that as a reserved
1454 keyword to be used with 'bool' variable types. That means you could
1455 not include C99-style headers which use bools without triggering
1456 build errors in sieve.c.
1458 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1459 as mailq or other aliases. Changed the --help handling significantly
1460 to do whats expected. exim_usage() emits usage/help information.
1462 SC/01 Added the -bylocaldomain option to eximstats.
1464 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1466 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1468 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1474 PH/01 Another patch from the Sieve maintainer.
1476 PH/02 When an IPv6 address is converted to a string for single-key lookup
1477 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1478 dots are used instead of colons so that keys in lsearch files need not
1479 contain colons. This was done some time before quoting was made available
1480 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1481 (notated using the quote facility) so as to distinguish them from IPv4
1482 keys. This meant that lookups for IP addresses in host lists did not work
1483 for iplsearch lookups.
1485 This has been fixed by arranging for IPv6 addresses to be expressed with
1486 colons if the lookup type is iplsearch. This is not incompatible, because
1487 previously such lookups could never work.
1489 The situation is now rather anomolous, since one *can* have colons in
1490 ordinary lsearch keys. However, making the change in all cases is
1491 incompatible and would probably break a number of configurations.
1493 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1496 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1497 conversion specification without a maximum field width, thereby enabling
1498 a rogue spamd server to cause a buffer overflow. While nobody in their
1499 right mind would setup Exim to query an untrusted spamd server, an
1500 attacker that gains access to a server running spamd could potentially
1501 exploit this vulnerability to run arbitrary code as the Exim user.
1503 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1504 $primary_hostname instead of what libspf2 thinks the hosts name is.
1506 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1507 a directory entry by the name of the lookup key. Previously, if a
1508 symlink pointed to a non-existing file or a file in a directory that
1509 Exim lacked permissions to read, a lookup for a key matching that
1510 symlink would fail. Now it is enough that a matching directory entry
1511 exists, symlink or not. (Bugzilla 503.)
1513 PH/03 The body_linecount and body_zerocount variables are now exported in the
1516 PH/04 Added the $dnslist_matched variable.
1518 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1519 This means they are set thereafter only if the connection becomes
1522 PH/06 Added the client_condition to authenticators so that some can be skipped
1523 by clients under certain conditions.
1525 PH/07 The error message for a badly-placed control=no_multiline_responses left
1526 "_responses" off the end of the name.
1528 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1530 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1531 (without spaces) instead of just copying the configuration text.
1533 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1535 PH/11 Added $max_received_linelength.
1537 PH/12 Added +ignore_defer and +include_defer to host lists.
1539 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1540 way in which PCRE > 7.0 is built.
1542 PH/14 Implemented queue_only_load_latch.
1544 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1545 MAIL command. The effect was to mangle the value on 64-bit systems.
1547 PH/16 Another patch from the Sieve maintainer.
1549 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1551 PH/18 If a system quota error occurred while trying to create the file for
1552 a maildir delivery, the message "Mailbox is full" was not appended to the
1553 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1554 only to a quota excession during the actual writing of the file.
1556 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1557 characters?) which causes problems in log lines. The DN values are now
1558 passed through string_printing() before being added to log lines.
1560 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1561 and InterBase are left for another time.)
1563 PH/21 Added message_body_newlines option.
1565 PH/22 Guard against possible overflow in moan_check_errorcopy().
1567 PH/23 POSIX allows open() to be a macro; guard against that.
1569 PH/24 If the recipient of an error message contained an @ in the local part
1570 (suitably quoted, of course), incorrect values were put in $domain and
1571 $local_part during the evaluation of errors_copy.
1577 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1578 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1581 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1582 issue a MAIL command.
1584 PH/02 In an ACL statement such as
1586 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1588 if a client was not listed at all, or was listed with a value other than
1589 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1590 the condition was not true (as it should be), so access was not denied.
1591 The bug was that the ! inversion was incorrectly passed on to the second
1592 item. This has been fixed.
1594 PH/03 Added additional dnslists conditions == and =& which are different from
1595 = and & when the dns lookup returns more than one IP address.
1597 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1598 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1600 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1601 FSYNC, which compiles an option called disable_fsync that allows for
1602 bypassing fsync(). The documentation is heavily laced with warnings.
1604 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1606 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1607 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1608 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1609 including adding "make clean"; (3) Added -fPIC when compiling the test
1610 dynamically loaded module, to get rid of a warning.
1612 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1613 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1614 = 0s. The bug is otherwise harmless.
1616 PH/07 There was a bug in the dovecot authenticator such that the value of
1617 $auth1 could be overwritten, and so not correctly preserved, after a
1618 successful authentication. This usually meant that the value preserved by
1619 the server_setid option was incorrect.
1621 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1623 PH/09 Installed PCRE release 7.0.
1625 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1626 run for batched SMTP input. It is now run at the start of every message
1627 in the batch. While fixing this I discovered that the process information
1628 (output by running exiwhat) was not always getting set for -bs and -bS
1629 input. This is fixed, and it now also says "batched" for BSMTP.
1631 PH/11 Added control=no_pipelining.
1633 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1634 patch, slightly modified), and move the expansion of helo_data till after
1635 the connection is made in the smtp transport (so it can use these
1638 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1640 PH/14 Added log_selector = +pid.
1642 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1644 PH/16 Add ${if forany and ${if forall.
1646 PH/17 Added dsn_from option to vary the From: line in DSNs.
1648 PH/18 Flush SMTP output before performing a callout, unless control =
1649 no_callout_flush is set.
1651 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1652 was true (the default) a successful delivery failed to delete the retry
1653 item, thus causing premature timeout of the address. The bug is now
1656 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1658 PH/21 Long custom messages for fakedefer and fakereject are now split up
1659 into multiline reponses in the same way that messages for "deny" and
1660 other ACL rejections are.
1662 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1663 with slight modification.
1665 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1666 draft, changing the syntax and factoring some duplicate code".
1668 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1669 for deliveries of the second and subsequent messages over the same SMTP
1672 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1673 ${reduce, with only minor "tidies".
1675 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1677 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1678 expansion side effects.
1680 PH/27 When a message times out after an over-quota error from an Exim-imposed
1681 quota, the bounce message says "mailbox is full". This message was not
1682 being given when it was a system quota that was exceeded. It now should
1685 MH/03 Made $recipients available in local_scan(). local_scan() already has
1686 better access to the recipient list through recipients_list[], but
1687 $recipients can be useful in postmaster-provided expansion strings.
1689 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1690 in the case of a MAIL command with additional options following the
1691 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1692 were accidentally chopped off.
1694 PH/29 SMTP synchronization checks are implemented when a command is read -
1695 there is a check that no more input is waiting when there shouldn't be
1696 any. However, for some commands, a delay in an ACL can mean that it is
1697 some time before the response is written. In this time, more input might
1698 arrive, invalidly. So now there are extra checks after an ACL has run for
1699 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1700 pipelining has not been advertised.
1702 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1704 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1705 returned in the SMTP response when smtp_return_error_details was false.
1706 This has been fixed.
1708 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1709 instead of the C I/O that was originally supplied, because problems were
1710 reported on Solaris.
1712 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1713 Exim which did not show up earlier: it was assuming that a call to
1714 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1715 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1716 was a macro that became an assignment, so it seemed to work. This has
1717 changed to a proper function call with a void return, hence the compile
1718 error. Exim's code has been fixed.
1720 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1723 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1725 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1727 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1728 because it turns out to be too big or there is a timeout) there is no
1729 "Completed" line in the log. When some messages of this type were
1730 selected by exigrep, they were listed as "not completed". Others were
1731 picked up by some special patterns. I have improved the selection
1732 criteria to be more general.
1734 PH/38 The host_find_failed option in the manualroute router can now be set
1735 to "ignore", to completely ignore a host whose IP address cannot be
1736 found. If all hosts are ignored, the behaviour is controlled by the new
1737 host_all_ignored option.
1739 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1740 homing or because of multiple MX records with /mx) generated more than
1741 one IP address, and the following item turned out to be the local host,
1742 all the secondary addresses of the first item were incorrectly removed
1743 from the list, along with the local host and any following hosts (which
1744 is what is supposed to happen).
1746 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1747 whoever called Exim into the -H file. In the case of the daemon it was
1748 behaving confusingly. When first started, it used values for whoever
1749 started the daemon, but after a SIGHUP it used the Exim user (because it
1750 calls itself on a restart). I have changed the code so that it now always
1753 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1754 message are rejected with the same error (e.g. no authentication or bad
1755 sender address), and a DATA command is nevertheless sent (as can happen
1756 with PIPELINING or a stupid MUA), the error message that was given to the
1757 RCPT commands is included in the rejection of the DATA command. This is
1758 intended to be helpful for MUAs that show only the final error to their
1761 PH/42 Another patch from the Sieve maintainer.
1763 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1764 Eximstats - Fixed some broken HTML links and added missing column headers
1766 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1767 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1769 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1771 PH/43 Yet another patch from the Sieve maintainer.
1773 PH/44 I found a way to check for a TCP/IP connection going away before sending
1774 the response to the final '.' that terminates a message, but only in the
1775 case where the client has not sent further data following the '.'
1776 (unfortunately, this is allowed). However, in many cases there won't be
1777 any further data because there won't be any more messages to send. A call
1778 to select() can be used: if it shows that the input is "ready", there is
1779 either input waiting, or the socket has been closed. An attempt to read
1780 the next input character can distinguish the two cases. Previously, Exim
1781 would have sent an OK response which the client would never have see.
1782 This could lead to message repetition. This fix should cure that, at
1783 least in a lot of common cases.
1785 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1786 advertised in response to EHLO.
1792 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1793 fixed by 4.65/MH/01 (is this a record?) are fixed:
1795 (i) An empty string was always treated as zero by the numeric comparison
1796 operators. This behaviour has been restored.
1798 (ii) It is documented that the numeric comparison operators always treat
1799 their arguments as decimal numbers. This was broken in that numbers
1800 starting with 0 were being interpreted as octal.
1802 While fixing these problems I realized that there was another issue that
1803 hadn't been noticed. Values of message_size_limit (both the global option
1804 and the transport option) were treated as octal if they started with 0.
1805 The documentation was vague. These values are now always treated as
1806 decimal, and I will make that clear in the documentation.
1812 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1813 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1816 MH/01 Don't check that the operands of numeric comparison operators are
1817 integers when their expansion is in "skipping" mode (fixes bug
1818 introduced by 4.64-PH/07).
1820 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1821 child addresses, Exim now panics and dies. Previously, because the count
1822 is held in a short int, deliveries were likely to be lost. As such a
1823 large number of recipients for a single message is ridiculous
1824 (performance will be very, very poor), I have chosen to impose a limit
1825 rather than extend the field.
1831 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1832 leftover -K file (the existence of which was triggered by #402).
1833 While we were at it, introduced process PID as part of the -K
1834 filename. This should rule out race conditions when creating
1837 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1838 processing considerably. Previous code took too long for large mails,
1839 triggering a timeout which in turn triggers #401.
1841 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1842 in the DK code in transports.c. sendfile() is not really portable,
1843 hence the _LINUX specificness.
1845 TF/01 In the add_headers option to the mail command in an Exim filter,
1846 there was a bug that Exim would claim a syntax error in any
1847 header after the first one which had an odd number of characters
1850 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1851 callout verification, Exim cached a "reject" for the entire domain. This
1852 is correct for most verifications, but it is not correct for a recipient
1853 verification with use_sender or use_postmaster set, because in that case
1854 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1855 case of MAIL FROM:<> rejection from other early rejections (e.g.
1856 rejection of HELO). When verifying a recipient using a non-null MAIL
1857 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1858 Whatever the result of the callout, the value of the domain cache is
1859 left unchanged (for any other kind of callout, getting as far as trying
1860 RCPT means that the domain itself is ok).
1862 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1865 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1866 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1867 session, the daemon ignores ECONNECT errors and logs others; it now
1868 ignores EPIPE as well.
1870 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1871 (quoted-printable decoding).
1873 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1874 later the small subsequent patch to fix an introduced bug.
1876 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1878 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1880 PH/08 An error is now given if message_size_limit is specified negative.
1882 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1883 to be given (somewhat) arbitrary names.
1885 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1888 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1889 miscellaneous code fixes
1891 PH/10 Added the log_reject_target ACL modifier to specify where to log
1894 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1895 hostname. This is wrong, because it relates to the incoming message (and
1896 probably the interface on which it is arriving) and not to the outgoing
1897 callout (which could be using a different interface). This has been
1898 changed to use the value of the helo_data option from the smtp transport
1899 instead - this is what is used when a message is actually being sent. If
1900 there is no remote transport (possible with a router that sets up host
1901 addresses), $smtp_active_hostname is used.
1903 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1904 tweaks were necessary in order to get it to work (see also 21 below):
1905 (a) The code assumed that strncpy() returns a negative number on buffer
1906 overflow, which isn't the case. Replaced with Exim's string_format()
1908 (b) There were several signed/unsigned issues. I just did the minimum
1909 hacking in of casts. There is scope for a larger refactoring.
1910 (c) The code used strcasecmp() which is not a standard C function.
1911 Replaced with Exim's strcmpic() function.
1912 (d) The code set only $1; it now sets $auth1 as well.
1913 (e) A simple test gave the error "authentication client didn't specify
1914 service in request". It would seem that Dovecot has changed its
1915 interface. Fortunately there's a specification; I followed it and
1916 changed what the client sends and it appears to be working now.
1918 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1921 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1922 address A is aliased to B and C, where B exists and C does not. Without
1923 -v the output is "A verified" because verification stops after a
1924 successful redirection if more than one address is generated. However,
1925 with -v the child addresses are also verified. Exim was outputting "A
1926 failed to verify" and then showing the successful verification for C,
1927 with its parentage. It now outputs "B failed to verify", showing B's
1928 parentage before showing the successful verification of C.
1930 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1931 look up a TXT record in a specific list after matching in a combined
1934 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1935 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1936 they consult the DNS. I had assumed they would set it the way they
1937 wanted; and indeed my experiments on Linux seem to show that in some
1938 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1939 To be on the safe side, however, I have now made the interface to
1940 host_find_byname() similar to host_find_bydns(), with an argument
1941 containing the DNS resolver options. The host_find_byname() function now
1942 sets these options at its start, just as host_find_bydns() does. The smtp
1943 transport options dns_qualify_single and dns_search_parents are passed to
1944 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
1945 of host_find_byname() use the default settings of RES_DEFNAMES
1946 (qualify_single) but not RES_DNSRCH (search_parents).
1948 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
1949 spool_read_header() do less string testing, by means of a preliminary
1950 switch on the second character of optional "-foo" lines. (This is
1951 overdue, caused by the large number of possibilities that now exist.
1952 Originally there were few.) While I was there, I also converted the
1953 str(n)cmp tests so they don't re-test the leading "-" and the first
1954 character, in the hope this might squeeze out yet more improvement.
1956 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
1957 flag allowing group syntax was set by the header_syntax check but not
1958 turned off, possible causing trouble later; (2) The flag was not being
1959 set at all for the header_verify test, causing "group"-style headers to
1960 be rejected. I have now set it in this case, and also caused header_
1961 verify to ignore an empty address taken from a group. While doing this, I
1962 came across some other cases where the code for allowing group syntax
1963 while scanning a header line wasn't quite right (mostly, not resetting
1964 the flag correctly in the right place). These bugs could have caused
1965 trouble for malformed header lines. I hope it is now all correct.
1967 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
1968 with the "reply" argument non-NULL. The code, however (which originally
1969 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
1970 but it didn't always do it. This confused somebody who was copying the
1971 code for some other use. I have removed all the tests.
1973 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
1974 feature that was used to support insecure browsers during the U.S. crypto
1975 embargo. It requires special client support, and Exim is probably the
1976 only MTA that supported it -- and would never use it because real RSA is
1977 always available. This code has been removed, because it had the bad
1978 effect of slowing Exim down by computing (never used) parameters for the
1979 RSA_EXPORT functionality.
1981 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
1982 authenticator to fail if there's a tab character in the incoming data
1983 (there should never be unless someone is messing about, as it's supposed
1984 to be base64-encoded). Also added, on Timo's advice, the "secured" option
1985 if the connection is using TLS or if the remote IP is the same as the
1986 local IP, and the "valid-client-cert option" if a client certificate has
1989 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
1990 authenticators. This can be used for authorization after authentication
1991 succeeds. (In the case of plaintext, it servers for both authentication
1994 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1995 if any retry times were supplied.
1997 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1998 connection, where there is no client IP address to check. In this
1999 situation, the verify now always succeeds.
2001 PH/25 Applied John Jetmore's -Mset patch.
2003 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2005 PH/27 In a string expansion for a processed (not raw) header when multiple
2006 headers of the same name were present, leading whitespace was being
2007 removed from all of them, but trailing whitespace was being removed only
2008 from the last one. Now trailing whitespace is removed from each header
2009 before concatenation. Completely empty headers in a concatenation (as
2010 before) are ignored.
2012 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2013 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2015 PH/29 [Removed. This was a change that I later backed out, and forgot to
2016 correct the ChangeLog entry (that I had efficiently created) before
2017 committing the later change.]
2019 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2020 address errors (4xx response to RCPT) over the same connection as other
2021 messages routed to the same hosts. Such deliveries are always "forced",
2022 so retry times are not inspected. This resulted in far too many retries
2023 for the affected addresses. The effect occurred only when there were more
2024 hosts than the hosts_max_try setting in the smtp transport when it had
2025 the 4xx errors. Those hosts that it had tried were not added to the list
2026 of hosts for which the message was waiting, so if all were tried, there
2027 was no problem. Two fixes have been applied:
2029 (i) If there are any address or message errors in an SMTP delivery, none
2030 of the hosts (tried or untried) are now added to the list of hosts
2031 for which the message is waiting, so the message should not be a
2032 candidate for sending over the same connection that was used for a
2033 successful delivery of some other message. This seems entirely
2034 reasonable: after all the message is NOT "waiting for some host".
2035 This is so "obvious" that I'm not sure why it wasn't done
2036 previously. Hope I haven't missed anything, but it can't do any
2037 harm, as the worst effect is to miss an optimization.
2039 (ii) If, despite (i), such a delivery is accidentally attempted, the
2040 routing retry time is respected, so at least it doesn't keep
2041 hammering the server.
2043 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2044 in ${readsocket because some servers need this prod.
2046 PH/32 Added some extra debug output when updating a wait-xxx database.
2048 PH/33 The hint "could be header name not terminated by colon", which has been
2049 given for certain expansion errors for a long time, was not being given
2050 for the ${if def:h_colon_omitted{... case.
2052 PH/34 The spec says: "With one important exception, whenever a domain list is
2053 being scanned, $domain contains the subject domain." There was at least
2054 one case where this was not true.
2056 PH/35 The error "getsockname() failed: connection reset by peer" was being
2057 written to the panic log as well as the main log, but it isn't really
2058 panic-worthy as it just means the connection died rather early on. I have
2059 removed the panic log writing for the ECONNRESET error when getsockname()
2062 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2063 runs only) independently of the message's sender address. This meant
2064 that, if the 4xx error was in fact related to the sender, a different
2065 message to the same recipient with a different sender could confuse
2066 things. In particualar, this can happen when sending to a greylisting
2067 server, but other circumstances could also provoke similar problems.
2068 I have changed the default so that the retry time for these errors is now
2069 based a combination of the sender and recipient addresses. This change
2070 can be overridden by setting address_retry_include_sender=false in the
2073 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2074 remote server are returned as part of bounce messages. This was not
2075 happening for LMTP over a pipe (the lmtp transport), but now it is the
2076 same for both kinds of LMTP.
2078 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2079 in header lines that were in fact CNAMEs. This is no longer the case.
2081 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2082 and queue runs started by the daemon processed all messages. This has
2083 been fixed so that -R and -S can now usefully be given with -q<time>.
2085 PH/40 Import PCRE release 6.7 (fixes some bugs).
2087 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2089 PH/42 Give an error if -q is specified more than once.
2091 PH/43 Renamed the variables $interface_address and $interface_port as
2092 $received_ip_address and $received_port, to make it clear that these
2093 values apply to message reception, and not to the outgoing interface when
2094 a message is delivered. (The old names remain recognized, of course.)
2096 PH/44 There was no timeout on the connect() call when using a Unix domain
2097 socket in the ${readsocket expansion. There now is.
2099 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2100 be meaningful with "accept".
2102 SC/01 Eximstats V1.43
2103 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2105 SC/02 Eximstats V1.44
2106 Use a glob alias rather than an array ref in the generated
2107 parser. This improves both readability and performance.
2109 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2110 Collect SpamAssassin and rejection statistics.
2111 Don't display local sender or destination tables unless
2112 there is data to show.
2113 Added average volumes into the top table text output.
2115 SC/04 Eximstats V1.46
2116 Collect data on the number of addresses (recipients)
2117 as well as the number of messages.
2119 SC/05 Eximstats V1.47
2120 Added 'Message too big' to the list of mail rejection
2121 reasons (thanks to Marco Gaiarin).
2123 SC/06 Eximstats V1.48
2124 Mainlog lines which have GMT offsets and are too short to
2125 have a flag are now skipped.
2127 SC/07 Eximstats V1.49 (Alain Williams)
2128 Added the -emptyok flag.
2130 SC/08 Eximstats V1.50
2131 Fixes for obtaining the IP address from reject messages.
2133 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2134 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2135 whitesspace changes from 4.64-PH/27
2137 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2140 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2141 are found, allow negative numbers in numeric criteria)
2143 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2145 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2146 to match changes made in 4.64-PH/43
2148 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2150 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2151 "domain" is actually an IP address, and force a failure. This locks out
2152 those revolvers/nameservers that support "A-for-A" lookups, in
2153 contravention of the specifications.
2155 PH/48 When a host name was looked up from an IP address, and the subsequent
2156 forward lookup of the name timed out, the host name was left in
2157 $sender_host_name, contrary to the specification.
2159 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2160 restricted to single-key lookups, Exim was not diagnosing an error if
2161 * or *@ was used with a query-style lookup.
2163 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2165 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2166 long ago, but noone interested enough thought of it. Let's just say that
2167 the "1.1" means that there are some new functions that weren't there at
2168 some point in the past.
2170 PH/51 Error processing for expansion failure of helo_data from an smtp
2171 transport during callout processing was broken.
2173 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2174 tested/used via the -bh/-bhc/-bs options.
2176 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2177 bug, fixed in subsequent PCRE releases).
2179 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2180 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2182 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2188 SC/01 Use a glob alias rather than an array ref in eximstats generated
2189 parser. This improves both readability and performance.
2191 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2192 Don't display local sender or destination tables in eximstats unless
2193 there is data to show.
2194 Added average volumes into the eximstats top table text output.
2196 SC/03 Collect data on the number of addresses (recipients) as well
2197 as the number of messages in eximstats.
2199 TF/01 Correct an error in the documentation for the redirect router. Exim
2200 does (usually) call initgroups() when daemonizing.
2202 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2203 with consistent privilege compared to when running as a daemon.
2205 TF/03 Note in the spec that $authenticated_id is not set for local
2206 submissions from trusted users.
2208 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2209 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2211 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2212 by adding some example configuration directives to the default
2213 configuration file. A little bit of work is required to uncomment the
2214 directives and define how usernames and passwords are checked, but
2215 there is now a framework to start from.
2217 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2218 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2219 without this. I don't know how relevant this is to other LDAP libraries.
2221 PH/02 Add the verb name to the "unknown ACL verb" error.
2223 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2225 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2227 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2228 directory not expanded when it should be if an expanded home directory
2229 was set for the address (which is overridden by the transport).
2231 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2234 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2235 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2236 because it is too late at that time, and has no effect.
2238 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2239 security issue with \' (bugzilla #107). I could not use the
2240 PQescapeStringConn() function, because it needs a PGconn value as one of
2243 PH/08 When testing addresses using -bt, indicate those final addresses that
2244 are duplicates that would not cause an additional delivery. At least one
2245 person was confused, thinking that -bt output corresponded to deliveries.
2246 (Suppressing duplicates isn't a good idea as you lose the information
2247 about possibly different redirections that led to the duplicates.)
2249 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2250 systems where poll() doesn't work, in particular OS X.
2252 PH/10 Added more information to debugging output for retry time not reached.
2254 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2255 operations in malware.c.
2257 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2260 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2261 syslog_duplication set false, log lines that would normally be written
2262 both the the main log and to the reject log were not written to syslog at
2265 PH/14 In the default configuration, change the use of "message" in ACL warn
2266 statements to "add_header".
2268 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2269 not followed by a command (e.g. "seen endif").
2271 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2272 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2275 PH/17 Added extra conditions to the default value of delay_warning_condition
2279 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2280 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2281 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2284 The Auto-Submitted: and various List- headers are standardised, whereas I
2285 don't think Precedence: ever was.
2287 PH/18 Refactored debugging code in route_finduser() to show more information,
2288 in particular, the error code if getpwnam() issues one.
2290 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2291 This is apparently needed in addition to the PH/07 change above to avoid
2292 any possible encoding problems.
2294 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2295 but not after initializing Perl.
2297 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2298 output them only if debugging. By default they are written stderr,
2299 apparently, which is not desirable.
2301 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2304 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2307 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2309 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2310 authenticated or an ident call has been made. Suppress the default
2311 values for $authenticated_id and $authenticated_sender (but permit -oMai
2312 and -oMas) when testing with -bh.
2314 PH/24 Re-jigged the order of the tests in the default configuration so that the
2315 tests for valid domains and recipients precede the DNS black list and CSA
2316 tests, on the grounds that those ones are more expensive.
2318 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2319 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2320 This bug exists in every version of Exim that I still have, right back to
2323 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2324 However, an attempt to turn on case-sensitivity in a regex key by
2325 including (?-i) didn't work because the subject string was already
2326 lowercased, and the effects were non-intuitive. It turns out that a
2327 one-line patch can be used to allow (?-i) to work as expected.
2333 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2334 other effects) broke the use of negated acl sub-conditions.
2336 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2339 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2340 "Deny" causes Exim to reject the incoming connection with a 554 error.
2341 Unfortunately, if there is a major crisis, such as a disk failure,
2342 tcp-wrappers gives "deny", whereas what one would like would be some
2343 kind of temporary error. A kludge has been added to help with this.
2344 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2345 554 error is used if errno is still zero or contains ENOENT (which occurs
2346 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2349 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2351 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2352 errors. Otherwise a message that provokes a temporary error (when other
2353 messages do not) can cause a whole host to time out.
2355 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2356 addresses were routed directly to files or pipes from a redirect router.
2357 File deliveries just didn't batch; pipe deliveries might have suffered
2360 PH/06 A failure to get a lock for a hints database would erroneously always say
2361 "Failed to get write lock", even when it was really a read lock.
2363 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2364 of 0600. This has been changed to use the value of the lockfile_mode
2365 option (which defaults to 0600).
2367 PH/08 Applied small patch from the Sieve maintainer.
2369 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2370 folder from quota calculations, a direct delivery into this folder messed
2371 up the contents of the maildirsize file. This was because the regex was
2372 used only to exclude .Trash (or whatever) when the size of the mailbox
2373 was calculated. There was no check that a delivery was happening into an
2374 excluded directory. This bug has been fixed by ignoring all quota
2375 processing for deliveries into excluded directories.
2377 PH/10 Added the maildirfolder_create_regex option to appendfile.
2383 PH/01 The code for finding all the local interface addresses on a FreeBSD
2384 system running IPv6 was broken. This may well have applied to all BSD
2385 systems, as well as to others that have similar system calls. The broken
2386 code found IPv4 interfaces correctly, but gave incorrect values for the
2387 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2388 that it would not match correctly against @[] and not recognize the IPv6
2391 PH/02 The ipliteral router was not recognizing addresses of the form user@
2392 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2394 PH/03 Added disable_ipv6.
2396 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2397 decoded form, because it is most often used to construct To: headers
2398 lines in autoreplies, and the decoded form may well be syntactically
2399 invalid. However, $reply_address has leading white space removed, and all
2400 newlines turned into spaces so that the autoreply transport does not
2403 PH/05 If group was specified without a user on a router, and no group or user
2404 was specified on a transport, the group from the router was ignored.
2406 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2407 for visible compile-time settings that can be used to change these
2408 numbers, for those that want even more. Backwards compatibility with old
2409 spool files has been maintained. However, going back to a previous Exim
2410 release will lost any variables that are in spool files.
2412 PH/07 Two small changes when running in the test harness: increase delay when
2413 passing a TCP/IP connection to a new process, in case the original
2414 process has to generate a bounce, and remove special handling of
2415 127.0.0.2 (sic), which is no longer necessary.
2417 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2418 be the same on different OS.
2420 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2423 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2424 whether --show-vars was specified or not
2426 JJ/02 exipick: Added support for new ACL variable spool format introduced
2429 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2430 syntactically invalid From: or Reply-to: line, and a filter used this to
2431 generate an autoreply, and therefore failed to obtain an address for the
2432 autoreply, Exim could try to deliver to a non-existent relative file
2433 name, causing unrelated and misleading errors. What now happens is that
2434 it logs this as a hard delivery error, but does not attempt to create a
2437 PH/11 The exinext utility has a -C option for testing purposes, but although
2438 the given file was scanned by exinext itself; it wasn't being passed on
2439 when Exim was called.
2441 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2442 an end-of-file indication when reading a command response.
2444 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2445 compiled. In many other places in Exim, IPv6 addresses are always
2446 recognized, so I have changed this. It also means that IPv4 domain
2447 literals of the form [IPV4:n.n.n.n] are now always recognized.
2449 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2450 used if the router is not running as root, for example, when verifying at
2451 ACL time, or when using -bh. The debugging output from this situation was
2452 non-existent - all you got was a failure to exec. I have made two
2455 (a) Failures to set uid/gid, the current directory, or a process leader
2456 in a subprocess such as that created by queryprogram now generate
2457 suitable debugging ouput when -d is set.
2459 (b) The queryprogram router detects when it is not running as root,
2460 outputs suitable debugging information if -d is set, and then runs
2461 the subprocess without attempting to change uid/gid.
2463 PH/15 Minor change to Makefile for building test_host (undocumented testing
2466 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2467 additional section of a DNS packet that returns MX or SRV records.
2468 Instead, it always explicitly searches for A/AAAA records. This avoids
2469 major problems that occur when a DNS server includes only records of one
2470 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2471 fixed another bug: if SRV records were looked up and the corresponding
2472 address records were *not* found in the additional section, the port
2473 values from the SRV records were lost.
2475 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2476 using the correct key (the original address) when searching the retry
2477 rules in order to find which one to use for generating the retry hint.
2479 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2480 adding the default one. Similarly, if it contains a Reply-To: header, the
2481 errors_reply_to option, if set, is not used.
2483 PH/19 When calculating a retry time, Exim used to measure the "time since
2484 failure" by looking at the "first failed" field in the retry record. Now
2485 it does not use this if it is later than than the arrival time of the
2486 message. Instead it uses the arrival time. This makes for better
2487 behaviour in cases where some deliveries succeed, thus re-setting the
2488 "first failed" field. An example is a quota failure for a huge message
2489 when small messages continue to be delivered. Without this change, the
2490 "time since failure" will always be short, possible causing more frequent
2491 delivery attempts for the huge message than are intended.
2492 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2494 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2495 $1, $2, $3) because the numerical variables can be reset during some
2496 expansion items (e.g. "match"), thereby losing the authentication data.
2498 PH/21 Make -bV show the size of off_t variables so that the test suite can
2499 decide whether to run tests for quotas > 2G.
2501 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2502 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2503 filecount value is greater than 2G or if a quota value is greater than 2G
2504 on a system where the size of off_t is not greater than 4, a panic error
2507 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2508 never match. The debug and -bh output now contains an explicit error
2509 message indicating a malformed IPv4 address or mask.
2511 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2512 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2513 PH/23 above applies.
2515 PH/25 Do not write to syslog when running in the test harness. The only
2516 occasion when this arises is a failure to open the main or panic logs
2517 (for which there is an explicit test).
2519 PH/26 Added the /no_tell option to "control=freeze".
2521 PH/27 If a host name lookup failed very early in a connection, for example, if
2522 the IP address matched host_lookup and the reverse lookup yielded a name
2523 that did not have a forward lookup, an error message of the form "no IP
2524 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2525 could be logged. Now it outputs the IP address instead of "NULL".
2527 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2528 allows the sender and the authenticated sender to be set when
2529 submitting a message from within Exim. Since child_open_exim() is
2530 documented for local_scan(), the new function should be too.
2532 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2533 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2534 results in an empty string is now treated as unset.
2536 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2538 PH/31 Added sender_verify_fail logging option.
2540 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2541 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2542 tidied the source and removed it altogether.
2544 PH/33 When a queue run was abandoned because the load average was too high, a
2545 log line was always written; now it is written only if the queue_run log
2546 selector is set. In addition, the log line for abandonment now contains
2547 information about the queue run such as the pid. This is always present
2548 in "start" and "stop" lines but was omitted from the "abandon" line.
2550 PH/34 Omit spaces between a header name and the colon in the error message that
2551 is given when verify = headers_syntax fails (if there are lots of them,
2552 the message gets confusing).
2554 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2555 names, as there are now some PTR records that contain slashes. This check
2556 is only to protect against broken name servers that fall over on strange
2557 characters, so the fact that it applies to all lookups doesn't matter.
2559 PH/36 Now that the new test suite is complete, we can remove some of the
2560 special code in Exim that was needed for the old test suite. For example,
2561 sorting DNS records because real resolvers return them in an arbitrary
2562 order. The new test suite's fake resolver always returns records in the
2565 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2566 bounces) except when queue_only is set, to avoid logging races between
2567 the different processes.
2569 PH/38 Panic-die if .include specifies a non-absolute path.
2571 PH/39 A tweak to the "H" retry rule from its user.
2573 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2574 a label. They prevented compilation on older perls.
2576 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2577 a warning to be raised on newish perls.
2579 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2580 on queue. Changes to match documented behaviour of showing count of
2581 messages matching specified criteria.
2583 PH/40 Changed the default ident timeout from 30s to 5s.
2585 PH/41 Added support for the use of login_cap features, on those BSD systems
2586 that have them, for controlling the resources used by pipe deliveries.
2588 PH/42 The content-scanning code uses fopen() to create files in which to put
2589 message data. Previously it was not paying any attention to the mode of
2590 the files. Exim runs with umask(0) because the rest of the code creates
2591 files with open(), and sets the required mode explicitly. Thus, these
2592 files were ending up world-writeable. This was not a big issue, because,
2593 being within the spool directory, they were not world-accessible. I have
2594 created a function called modefopen, which takes an additional mode
2595 argument. It sets umask(777), creates the file, chmods it to the required
2596 mode, then resets the umask. All the relevant calls to fopen() in the
2597 content scanning code have been changed to use this function.
2599 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2600 to 24 hours. This avoids potential overflow problems when processing G
2601 and H retry rules. I suspect nobody ever tinkers with this value.
2603 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2605 PH/45 When the plaintext authenticator is running as a client, the server's
2606 challenges are checked to ensure they are valid base64 strings. By
2607 default, the authentication attempt is cancelled if an invalid string is
2608 received. Setting client_ignore_invalid_base64 true ignores these errors.
2609 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2610 they are received. Thus, the responses can be made to depend on the
2611 challenges. If an invalid string is ignored, an empty string is placed in
2614 PH/46 Messages that are created by the autoreply transport now contains a
2615 References: header, in accordance with RFCs 2822 and 3834.
2617 PH/47 Added authenticated_sender_force to the smtp transport.
2619 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2621 PH/49 Installed latest patch from the Sieve maintainer.
2623 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2624 was also set, the appendfile transport was unnecessarily scanning a
2625 directory of message files (e.g. for maildir delivery) to find the count
2626 of files (along with the size), even though it did not need this
2627 information. It now does the scan only if it needs to find either the
2628 size of the count of files.
2630 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2632 PH/52 Two bugs concerned with error handling when the smtp transport is
2635 (i) Exim was not creating retry information for temporary errors given
2636 for individual recipients after the DATA command when the smtp transport
2637 was used in LMTP mode. This meant that they could be retried too
2638 frequently, and not timed out correctly.
2640 (ii) Exim was setting the flag that allows error details to be returned
2641 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2642 recipients that were returned after the DATA command.
2644 PH/53 This is related to PH/52, but is more general: for any failing address,
2645 when detailed error information was permitted to be returned to the
2646 sender, but the error was temporary, then after the final timeout, only
2647 "retry timeout exceeded" was returned. Now it returns the full error as
2648 well as "retry timeout exceeded".
2650 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2651 do this, and (what is worse) MTAs that accept it.
2653 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2654 will now be deprecated.
2656 PH/56 New os.c-cygwin from the Cygwin maintainer.
2658 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2659 formats (previously only available in exim formats via -bpr, -bpru,
2660 and -bpra. Now also available in native and exiqgrep formats)
2662 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2663 with very large, slow to parse queues
2665 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2667 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2669 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2670 responses to authentication challenges, though it was showing the
2671 challenges; (ii) I've removed the CR characters from the debug output for
2674 PH/58 Allow for the insertion of a newline as well as a space when a string
2675 is turned into more than one encoded-word during RFC 2047 encoding. The
2676 Sieve code now uses this.
2678 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2679 data_4xx, lost_connection, tls_required.
2681 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2682 message was being sent as an SMTP response.
2684 PH/61 Add -l and -k options to exicyclog.
2686 PH/62 When verifying, if an address was redirected to one new address, so that
2687 verification continued, and the new address failed or deferred after
2688 having set something in $address_data, the value of $address_data was not
2689 passed back to the ACL. This was different to the case when no
2690 redirection occurred. The value is now passed back in both cases.
2692 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2693 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2694 use login_cap.h, so on its own it isn't the distinguishing feature. The
2695 new name refers directly to the setclassresources() function.
2697 PH/65 Added configuration files for NetBSD3.
2699 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2701 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2702 is preferred over IPv4.
2704 PH/68 The bounce_return_message and bounce_return_body options were not being
2705 honoured for bounces generated during the reception of non-SMTP messages.
2706 In particular, this applied to messages rejected by the ACL. This bug has
2707 been fixed. However, if bounce_return_message is true and bounce_return_
2708 body is false, the headers that are returned for a non-SMTP message
2709 include only those that have been read before the error was detected.
2710 (In the case of an ACL rejection, they have all been read.)
2712 PH/69 The HTML version of the specification is now built in a directory called
2713 spec_html instead of spec.html, because the latter looks like a path with
2714 a MIME-type, and this confuses some software.
2716 PH/70 Catch two compiler warnings in sieve.c.
2718 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2719 function verify_get_ident() calls ip_connect() to connect a socket, but
2720 if the "connect()" function timed out, ip_connect() used to close the
2721 socket. However, verify_get_ident() also closes the socket later, and in
2722 between Exim writes to the log, which may get opened at this point. When
2723 the socket was closed in ip_connect(), the log could get the same file
2724 descriptor number as the socket. This naturally causes chaos. The fix is
2725 not to close the socket in ip_connect(); the socket should be closed by
2726 the function that creates it. There was only one place in the code where
2727 this was missing, in the iplookup router, which I don't think anybody now
2728 uses, but I've fixed it anyway.
2730 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2731 well as to direct DNS lookups. Otherwise the handling of names in host
2732 lists is inconsistent and therefore confusing.
2738 PH/01 Two changes to the default runtime configuration:
2740 (1) Move the checks for relay_from_hosts and authenticated clients from
2741 after to before the (commented out) DNS black list checks.
2743 (2) Add control=submission to the relay_from_hosts and authenticated
2744 clients checks, on the grounds that messages accepted by these
2745 statements are most likely to be submissions.
2747 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2749 (1) Generate an error if the third argument for the ${prvs expansion is
2752 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2755 (3) Reset the variables that are obtained from the first argument of
2756 ${prvscheck and used in the second argument before leaving the code,
2757 because their memory is reclaimed, so using them afterwards may do
2760 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2761 one (it's much easier than Tom thought :-).
2763 (5) Because of (4), we can now allow for the use of $prvscheck_result
2764 inside the third argument.
2766 PH/03 For some reason, the default setting of PATH when running a command from
2767 a pipe transport was just "/usr/bin". I have changed it to
2770 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2771 anything to be listed in the output from -bV.
2773 PH/05 When a filter generated an autoreply, the entire To: header line was
2774 quoted in the delivery log line, like this:
2776 => >A.N.Other <ano@some.domain> <original@ddress> ...
2778 This has been changed so that it extracts the operative address. There
2779 may be more than one such address. If so, they are comma-separated, like
2782 => >ano@some.domain,ona@other.domain <original@ddress> ...
2784 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2785 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2786 not being looked up in the rDNS to get a host name, Exim was showing the
2787 IP address twice in Received: lines, even though the IP addresses were
2788 identical. For example:
2790 Received: from [1.2.3.4] (helo=[1.2.3.4])
2792 However, if the real host name was known, it was omitting the HELO data
2793 if it matched the actual IP address. This has been tidied up so that it
2794 doesn't show the same IP address twice.
2796 PH/07 When both +timestamp and +memory debugging was on, the value given by
2797 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2798 called by the debug printing, thereby overwriting the timestamp buffer.
2799 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2801 PH/08 When the original message was included in an autoreply transport, it
2802 always said "this is a copy of the message, including all the headers",
2803 even if body_only or headers_only was set. It now gives an appropriate
2806 PH/09 Applied a patch from the Sieve maintainer which:
2808 o fixes some comments
2809 o adds the (disabled) notify extension core
2810 o adds some debug output for the result of if/elsif tests
2811 o points to the current vacation draft in the documentation
2812 and documents the missing references header update
2816 o fixes a bug in processing the envelope test (when testing
2817 multiple envelope elements, the last element determinted the
2820 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2821 Electronic Mail") by including:
2823 Auto-submitted: auto-generated
2825 in the messages that it generates (bounce messages and others, such as
2826 warnings). In the case of bounce messages for non-SMTP mesages, there was
2827 also a typo: it was using "Auto_submitted" (underscore instead of
2828 hyphen). Since every message generated by Exim is necessarily in response
2829 to another message, thes have all been changed to:
2831 Auto-Submitted: auto-replied
2833 in accordance with these statements in the RFC:
2835 The auto-replied keyword:
2837 - SHOULD be used on messages sent in direct response to another
2838 message by an automatic process,
2840 - MUST NOT be used on manually-generated messages,
2842 - MAY be used on Delivery Status Notifications (DSNs) and Message
2843 Disposition Notifications (MDNs),
2845 - MUST NOT be used on messages generated by automatic or periodic
2846 processes, except for messages which are automatic responses to
2849 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2850 to the default Received: header definition.
2852 PH/12 Added log selector acl_warn_skipped (default on).
2854 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2855 variables because (a) they are in the wrong storage pool and (b) even if
2856 they were copied, it wouldn't work properly because of the caching.
2858 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2859 checking when decoding. Apparently there are clients that generate
2860 overlong encoded strings. Why am I not surprised?
2862 PH/15 If the first argument of "${if match_address" was not empty, but did not
2863 contain an "@" character, Exim crashed. Now it writes a panic log message
2864 and treats the condition as false.
2866 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2868 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2869 extension "envelope-auth". The code is finished and in agreement with
2870 other implementations, but there is no documentation so far and in fact,
2871 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2872 not changing the active code.
2874 Print executed "if" and "elsif" statements when debugging is used. This
2875 helps a great deal to understand what a filter does.
2877 Document more things not specified clearly in RFC3028. I had all this
2878 sorted out, when out of a sudden new issues came to my mind. Oops."
2880 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2883 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2884 canonical form (as documented). However, after a host name lookup from
2885 the IP address, check_host() was doing a simple string comparison with
2886 addresses acquired from the DNS when checking that the found name did
2887 have the original IP as one of its addresses. Since any found IPv6
2888 addresses are likely to be in abbreviated form, the comparison could
2889 fail. Luckily, there already exists a function for doing the comparison
2890 by converting both addresses to binary, so now that is used instead of
2891 the text comparison.
2893 PH/20 There was another similar case to PH/19, when a complete host name was
2894 given in a host list; looking up its IP address could give an abbreviated
2895 form, whereas the current host's name might or might not be abbreviated.
2896 The same fix has been applied.
2902 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2903 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2906 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2907 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2909 PH/03 Typo: missing ".o" in src/pcre/Makefile.
2911 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2912 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2913 Also, for "Auto-Submitted", treat anything other than "no" as
2914 non-personal, in accordance with RFC 3834. (Previously it treated
2915 anything starting "auto-" as non-personal.)
2917 TF/01 The control=submission/name=... option had a problem with syntax
2918 errors if the name included a slash character. The /name= option
2919 now slurps the rest of the string, so it can include any characters
2920 but it must come last in the list of options (after /sender_retain
2923 PH/05 Some modifications to the interface to the fake nameserver for the new
2931 TK/01 Added the "success_on_redirect" address verification option. See
2932 NewStuff for rationale and an example.
2934 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2936 PH/02 Patch to exigrep to allow it to work on syslog lines.
2938 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2939 fread() to skip over the body file's header line, because in Cygwin the
2940 header line is locked and is inaccessible.
2942 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
2943 co-exist for some time) to make it clear that it is the Exim ID that is
2944 referenced, not the Message-ID: header line.
2946 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
2947 string_format() function, because snprintf() does not exist on all
2950 PH/06 The use of forbid_filter_existstest now also locks out the use of the
2951 ${stat: expansion item.
2953 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
2954 protocol synchronization error", to keep the pedants happy.
2956 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
2957 well as for IRIX systems, when gcc is being used. See the host.c source
2960 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
2962 PH/10 Named domain lists were not working if used in a queue_smtp_domains
2965 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
2966 transport and to the smtp transport in LMTP mode.
2968 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
2970 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
2971 run a filter in a subprocess. This could lead to confusion in subsequent
2972 lookups in the parent process. There should also be a search_tidyup() at
2973 the end of the subprocess.
2975 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
2976 only if the host matched helo_try_verify_hosts, which caused the
2977 verification to occur when the EHLO/HELO command was issued. The ACL just
2978 tested the remembered result. Now, if a previous verification attempt has
2979 not happened, "verify = helo" does it there and then.
2981 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
2983 TK/03 Fix log output including CR from clamd.
2985 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
2986 exist provoked a memory error which could cause a segfault.
2988 PH/15 Installed PCRE 6.2
2990 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2992 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2993 of the problem. Specifically, suggested +O2 rather than +O1 for the
2996 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2998 PH/20 If a delivery was routed to a non-standard port by means of an SRV
2999 record, the port was not correctly logged when the outgoing_port log
3000 selector was set (it logged the transort's default port).
3002 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3003 fallback_hosts, and "hosts" in the smtp transport.
3005 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3006 host errors such as "Connection refused".
3008 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3009 authentication with radiusclient 0.4.9:
3011 - Error returned from rc_read_config was caught wrongly
3012 - Username/password not passed on to radius server due to wrong length.
3014 The presumption is that some radiusclient API changes for 4.51/PH/17
3015 were not taken care of correctly. The code is still untested by me (my
3016 Linux distribution still has 0.3.2 of radiusclient), but it was
3017 contributed by a Radius user.
3019 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3020 expanding the "port" option of the smtp transport.
3022 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3023 while reading a MIME header. Thanks to Tom Hughes for a patch.
3025 PH/24 Include config.h inside local_scan.h so that configuration settings are
3028 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3029 that in an ACL for RCPT (for example), you can examine exactly what was
3032 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3033 commands, but it was not correctly comparing the address with the actual
3034 client host address. Thus, it would show the EHLO address in Received:
3035 header lines when this was not necessary.
3037 PH/27 Added the % operator to ${eval:}.
3039 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3040 it should be ignoring failures (because with -C, for example, it has lost
3041 privilege). It wasn't ignoring creation failures other than "already
3044 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3047 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3048 because an input line was too long, either on its own, or by virtue of
3049 too many continuations, the temporary file was not being removed, and the
3050 return code was incorrect.
3052 PH/31 Missing "BOOL" in function definition in filtertest.c.
3054 PH/32 Applied Sieve patches from the maintainer.
3056 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3058 PH/33 Added "verify = not_blind".
3060 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3061 Local/Makefile (with some defaults set). These are used in built scripts
3062 such as exicyclog, but they have never been used in the exim_install
3063 script (though there are many overriding facilities there). I have
3064 arranged that the exim_install script now takes note of these two
3067 PH/35 Installed configuration files for Dragonfly.
3069 PH/36 When a locally submitted message by a trusted user did not contain a
3070 From: header, and the sender address was obtained from -f or from an SMTP
3071 MAIL command, and the trusted user did not use -F to supply a sender
3072 name, $originator_name was incorrectly used when constructing a From:
3073 header. Furthermore, $originator_name was used for submission mode
3074 messages from external hosts without From: headers in a similar way,
3075 which is clearly wrong.
3077 PH/37 Added control=suppress_local_fixups.
3079 PH/38 When log_selector = +received_sender was set, and the addition of the
3080 sender made the log line's construction buffer exactly full, or one byte
3081 less than full, an overflow happened when the terminating "\n" was
3084 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3085 when the result of a list match is failure because a DNS lookup failed.
3087 PH/40 RM_COMMAND is now used in the building process.
3089 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3090 the "build-* directories that it finds.
3092 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3093 address in a domain literal was a prefix of an interface address.
3095 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3096 when verifying a sender address, unless rewrite_headers is false.
3098 PH/44 Wrote a long comment about why errors_to addresses are verified as
3099 recipients, not senders.
3101 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3102 the ratelimit ACL was added.
3104 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3106 PH/46 Added extra information about PostgreSQL errors to the error string.
3108 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3109 avoiding the need to install special zones in a real server. This is
3110 backwards compatible; if it can't find the fake resolver, it drops back.
3111 Thus, both old and new test suites can be run.
3113 TF/02 Added util/ratelimit.pl
3115 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3116 clock is set back in time.
3118 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3119 Candler <B.Candler@pobox.com>.
3121 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3122 OK for addresses that are the result of redirections.
3124 PH/48 A number of further additions for the benefit of the new test suite,
3125 including a fake gethostbyname() that interfaces to the fake DNS resolver
3128 TF/06 The fix for widen_domains has also been applied to qualify_single and
3129 search_parents which are the other dnslookup options that can cause
3132 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3135 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3137 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3138 service" if it encountered an MX record with an empty target hostname.
3139 The message is now "an MX or SRV record indicated no SMTP service".
3141 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3142 if the DNS of the sending site is misconfigured. This is quite a
3143 common situation. This change restores the behaviour of treating a
3144 helo verification defer as a failure.
3146 PH/51 If self=fail was set on a router, the bounce message did not include the
3147 actual error message.
3153 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3155 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3156 command itself ended in error, the underlying message about the transport
3157 filter timeout was being overwritten with the pipe command error. Now the
3158 underlying error message should be appended to the second error message.
3160 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3162 PH/02 Reduce the amount of output that "make" produces by default. Full output
3163 can still be requested.
3165 PH/03 The warning log line about a condition test deferring for a "warn" verb
3166 was being output only once per connection, rather than after each
3167 occurrence (because it was using the same function as for successful
3168 "warn" verbs). This seems wrong, so I have changed it.
3170 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3171 it should not have, which might have caused a crash in the right
3172 circumstances, but probably never did.
3174 PH/04 Installed a modified version of Tony Finch's patch to make submission
3175 mode fix the return path as well as the Sender: header line, and to
3176 add a /name= option so that you can make the user's friendly name appear
3179 TF/03 Added the control = fakedefer ACL modifier.
3181 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3182 Mark Lowes for thorough testing.
3184 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3186 TK/03 Merged latest SRS patch from Miles Wilton.
3188 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3189 with the definition in sysexits.h (which is #included earlier).
3190 Fortunately, Exim does not actually use EX_OK. The code used to try to
3191 preserve the sysexits.h value, by assumimg that macro definitions were
3192 scanned for macro replacements. I have been disabused of this notion,
3193 so now the code just undefines EX_OK before #including unistd.h.
3195 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3196 in the smtp transport. When a block could not be written in a single
3197 write() function, the timeout was being re-applied to each part-write.
3198 This seems wrong - if the receiver was accepting one byte at a time it
3199 would take for ever. The timeout is now adjusted when this happens. It
3200 doesn't have to be particularly precise.
3202 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3203 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3205 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3206 without a domain if the check to <postmaster@domain> fails.
3208 SC/01 Eximstats: added -xls and the ability to specify output files
3209 (patch written by Frank Heydlauf).
3211 SC/02 Eximstats: use FileHandles for outputing results.
3213 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3215 SC/04 Eximstats: fixed display of large numbers with -nvr option
3217 SC/05 Eximstats: fixed merging of reports with empty tables.
3219 SC/06 Eximstats: added the -include_original_destination flag
3221 SC/07 Eximstats: removed tabs and trailing whitespace.
3223 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3225 TK/06 MBOX spool code: Add real "From " MBOX separator line
3226 so the .eml file is really in mbox format (even though
3227 most programs do not really care). Patch from Alex Miller.
3229 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3230 The latter is generated from $received_to and is only set if the
3231 message has one envelope recipient. SA can use these headers,
3232 obviously out-of-the-box. Patch from Alex Miller.
3234 PH/08 The ${def test on a variable was returning false if the variable's
3235 value was "0", contrary to what the specification has always said!
3236 The result should be true unless the variable is empty.
3238 PH/09 The syntax error of a character other than { following "${if
3239 def:variable_name" (after optional whitespace) was not being diagnosed.
3240 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3241 accidental colon was present, for example, could give incorrect results.
3243 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3244 result is used (not including appendfile, where other changes are about
3247 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3248 This involved changing a lot of size variables from int to off_t. It
3249 should work with maildirs and everything.
3251 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3252 spamd dying while we are connected to it.
3254 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3257 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3258 function for BDB 4.3.
3260 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3262 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3263 the sources for PCRE have been re-arranged and more files are now
3266 PH/15 The code I had for printing potentially long long variables in PH/11
3267 above was not the best (it lost precision). The length of off_t variables
3268 is now inspected at build time, and an appropriate printing format (%ld
3269 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3270 to be "long long int" or "long int". This is needed for the internal
3271 formatting function string_vformat().
3273 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3274 the configuration file to be ":syslog", then the script "guesses" where
3275 the logs files are, rather than using the compiled in default. In our
3276 case the guess is not the same as the compiled default, so the script
3277 suddenly stopped working when I started to use syslog. The patch checks
3278 to see if log_file_path is "". If so, it attempts to read it from exim
3279 with no configuration file to get the compiled in version, before it
3280 falls back to the previous guessing code."
3282 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3283 implementing BATV in an Exim configuration. See NewStuff for the gory
3286 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3287 Makefile that are specific to HP-UX.
3289 PH/18 If the "use_postmaster" option was set for a recipient callout together
3290 with the "random" option, the postmaster address was used as the MAIL
3291 FROM address for the random test, but not for the subsequent recipient
3292 test. It is now used for both.
3294 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3295 patch removes a few documentation additions to RFC 3028, because the
3296 latest draft now contains them. It adds the new en;ascii-case comparator
3297 and a new error check for 8bit text in MIME parts. Comparator and
3298 require names are now matched exactly. I enabled the subaddress
3299 extension, but it is not well tested yet (read: it works for me)."
3301 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3302 rework some of the code of TK/09 above to avoid the hardwired use of
3303 "%lld" and "long long". Replaced the call to snprintf() with a call to
3306 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3307 records point to non-existent hosts", "retry timeout exceeded", and
3308 "retry time not reached for any host after a long failure period".
3310 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3311 experimental DomainKeys support:
3313 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3314 (2) On an error such as an illegally used "control", the wrong name for
3315 the control was given.
3317 These problems did NOT occur unless DomainKeys support was compiled.
3319 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3321 PH/24 Added ${if match_ip condition.
3323 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3324 databases so that it will be absolutely obvious if a crash occurs in the
3325 DB library. This is a regular occurrence (often caused by mis-matched
3328 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3329 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3330 picked up on a user's system that detects such things. There doesn't seem
3331 to be a gcc warning option for this - only an attribute that has to be
3332 put on the function's prototype. It seems that in Fedora Core 4 they have
3333 set this on a number of new functions. No doubt there will be more in due
3336 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3337 specify a transport. However, if an address that was verified by such a
3338 router was the subject of a callout, Exim crashed because it tried to
3339 read the rcpt_include_affixes from the non-existent transport. Now it
3340 just assumes that the setting of that option is false. This bug was
3341 introduced by 4.51/PH/31.
3343 PH/28 Changed -d+all to exclude +memory, because that information is very
3344 rarely of interest, but it makes the output a lot bigger. People tend to
3345 do -d+all out of habit.
3347 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3348 code in os-type was giving problems when libc.so lives in lib64, like on
3351 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3352 aren't the modern standard, and it seems that some systems' include files
3353 don't always have them. Exim was already checking for some of the newer
3354 ones like T_AAAA, and defining it itself. I've added checks for all the
3355 record types that Exim uses.
3357 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3358 not automatically generating a new one, as it is supposed to. This
3359 prevented TLS from working. If the file did exist, but contained invalid
3360 data, a new version was generated, as expected. It was only the case of a
3361 non-existent file that was broken.
3363 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3364 with a change in libdomainkeys > 0.64.
3366 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3367 from DNS. If the selector record carries the flag, it now has
3368 precedence over the domain-wide flag.
3370 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3372 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3373 the use of a port name that isn't defined in /etc/services) occurred, the
3374 message was deferred as in a normal delivery, and thus remained on the
3375 spool, instead of being failed because of the mua_wrapper setting. This
3376 is now fixed, and I tidied up some of the mua_wrapper messages at the
3379 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3380 the messages in a hash of arrays rather than using individual hashes.
3381 This is a bit cleaner and results in dramatic memory savings, albeit
3382 at a slight CPU cost.
3384 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3385 as requested by Marc Sherman.
3387 SC/10 Eximstats: added histograms for user specified patterns as requested
3390 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3392 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3393 fopen() in the content-scanning modules that did not already have it.
3399 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3400 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3402 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3404 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3406 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3407 to test Sieve filters that use "vacation".
3409 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3410 that changes the way the GnuTLS parameters are stored in the cache file.
3411 The new format can be generated externally. For backward compatibility,
3412 if the data in the cache doesn't make sense, Exim assumes it has read an
3413 old-format file, and it generates new data and writes a new file. This
3414 means that you can't go back to an older release without removing the
3417 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3418 work if there are any delivery delays because "one_time" forces the
3419 parent to be marked "delivered", so its unseen clone is never tried
3420 again. For this reason, Exim now forbids the simultaneous setting of
3423 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3424 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3425 entry if you want to know the details. The fix, however, neglected to
3426 consider the case when local delivery batching is involved. The test for
3427 "previously delivered" was not happening when checking to see if an
3428 address could be batched with a previous (undelivered) one; under
3429 certain circumstances this could lead to multiple deliveries to the same
3432 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3433 in its include files, and this causes problems building Exim.
3435 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3436 header_syntax) but Exim was just ignoring anything given after a slash.
3437 In particular, this caused confusion with an attempt to use "verify =
3438 reverse_host_lookup/defer_ok". An error is now given when options are
3439 supplied for verify items that do not have them. (Maybe reverse_host_
3440 lookup should have a defer_ok option, but that's a different point.)
3442 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3443 defined by RFC 821) to 2048, because there were problems with some AUTH
3444 commands, and RFC 1869 says the size should be increased for extended
3445 SMTP commands that take arguments.
3447 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3450 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3451 "unknown" error; now it says that the functionality isn't in the binary.
3453 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3454 an address' error message when a string expansion fails (syntax or
3455 whatever). Otherwise the password may appear in the log. Following change
3456 PH/42 below, there is no longer a chance of it appearing in a bounce
3459 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3461 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3462 of its addresses was ever tried. (Bugzilla bug #2.)
3464 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3465 the result incorrectly in the debug output. (It correctly added a newline
3466 to what was transported.)
3468 TF/01 Added $received_time.
3470 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3471 commented out examples of how to interface to a virus scanner and to
3472 SpamAssassin. Also added commented examples of av_scanner and
3473 spamd_address settings.
3475 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3476 and controls are allowed in which ACLs. There were a couple of minor
3477 errors. Some of the entries in the conditions table (which is a table of
3478 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3479 negation of where the condition IS allowed.
3481 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3483 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3484 header file does not have a version number, so I've had to invent a new
3485 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3486 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3487 radiusclient), but it was contributed by a Radius user.
3489 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3490 files or directories, for OpenSSL.
3492 PH/19 When an Exim process that is running as root has to create an Exim log
3493 file, it does so in a subprocess that runs as exim:exim so as to get the
3494 ownership right at creation (otherwise, other Exim processes might see
3495 the file with the wrong ownership). There was no test for failure of this
3496 fork() call, which would lead to the process getting stuck as it waited
3497 for a non-existent subprocess. Forks do occasionally fail when resources
3498 run out. I reviewed all the other calls to fork(); they all seem to check
3501 PH/20 When checking for unexpected SMTP input at connect time (before writing
3502 the banner), Exim was not dealing correctly with a non-positive return
3503 from the read() function. If the client had disconnected by this time,
3504 the result was a log entry for a synchronization error with an empty
3505 string after "input=" when read() returned zero. If read() returned -1
3506 (an event I could not check), uninitialized data bytes were printed.
3507 There were reports of junk text (parts of files, etc) appearing after
3510 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3512 PH/22 Added support for macro redefinition, and (re)definition in between
3513 driver and ACL definitions.
3515 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3516 forgetting to use the resulting value; it was using the unexpanded value.
3518 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3519 hadn't been configured. The fix is from Juergen Kreileder, who
3520 understands it better than I do:
3522 "Here's what I see happening with three configured cyrus_sasl
3523 authenticators configured (plain, login, cram-md5):
3525 On startup auth_cyrus_sasl_init() gets called for each of these.
3526 This means three calls to sasl_listmech() without a specified mech_list.
3527 => SASL tests which mechs of all available mechs actually work
3528 => three warnings about OTP not working
3529 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3531 With the patch, sasl_listmech() also gets called three times. But now
3532 SASL's mech_list option is set to the server_mech specified in the the
3533 authenticator. Or in other words, the answer from sasl_listmech()
3534 gets limited to just the mech you're testing for (which is different
3536 => the return list contains just 'plain' or 'login', 'cram-md5' or
3537 nothing depending on the value of ob->server_mech.
3539 I've just tested the patch: Authentication still works fine,
3540 unavailable mechs specified in the exim configuration are still
3541 caught, and the auth.log warnings about OTP are gone."
3543 PH/25 When debugging is enabled, the contents of the command line are added
3544 to the debugging output, even when log_selector=+arguments is not
3547 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3548 answer is "GNU", and only if the return is "GNU/something" is the answer
3551 PH/27 $acl_verify_message is now set immediately after the failure of a
3552 verification in an ACL, and so is available in subsequent modifiers. In
3553 particular, the message can be preserved by coding like this:
3555 warn !verify = sender
3556 set acl_m0 = $acl_verify_message
3558 Previously, $acl_verify_message was set only while expanding "message"
3559 and "log_message" when a very denied access.
3561 PH/28 Modified OS/os.c-Linux with
3563 -#ifndef OS_LOAD_AVERAGE
3564 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3566 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3567 nomenclature these days.)
3569 PH/29 Installed patch from the Sieve maintainer that adds the options
3570 sieve_useraddress and sieve_subaddress to the redirect router.
3572 PH/30 In these circumstances:
3573 . Two addresses routed to the same list of hosts;
3574 . First host does not offer TLS;
3575 . First host accepts first address;
3576 . First host gives temporary error to second address;
3577 . Second host offers TLS and a TLS session is established;
3578 . Second host accepts second address.
3579 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3580 and peerdn, if requested) that were in fact used only for the second
3583 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3584 attention to any local part prefix or suffix that was matched by the
3585 router that accepted the address. It now behaves in the same way as it
3586 does for delivery: the affixes are removed from the local part unless
3587 rcpt_include_affixes is set on the transport.
3589 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3590 timeout during the DATA phase of an incoming message.
3592 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3593 applied a patch sanctioned by the Sieve maintainer.
3595 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3596 the uid or gid is negative. A case of a negative gid caused this to be
3597 noticed. The fix allows for either to be negative.
3599 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3600 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3603 PH/36 The change PH/12 above was broken. Fixed it.
3605 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3606 the grounds that routing the same address twice would always produce the
3607 same answer. This might have been true once, but it is certainly no
3608 longer true now. Routing a child address may depend on the previous
3609 routing that produced that child. Some complicated redirection strategies
3610 went wrong when messages had multiple recipients, and made Exim's
3611 behaviour dependent on the order in which the addresses were given.
3613 I have moved the duplicate checking until after the routing is complete.
3614 Exim scans the addresses that are assigned to local and remote
3615 transports, and removes any duplicates. This means that more work will be
3616 done, as duplicates will always all be routed, but duplicates are
3617 presumably rare, so I don't expect this is of any significance.
3619 For deliveries to pipes, files, and autoreplies, the duplicate checking
3620 still happens during the routing process, since they are not going to be
3623 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3624 It corrects a timeout issue with spamd. This is Ian's comment: "The
3625 background is that sometimes spamd either never reads data from a
3626 connection it has accepted, or it never writes response data. The exiscan
3627 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3628 blindly assumes that writes won't block so it may never time out."
3630 PH/39 Allow G after quota size as well as K and M.
3632 PH/40 The value set for $authenticated_id in an authenticator may not contain
3633 binary zeroes or newlines because the value is written to log lines and
3634 to spool files. There was no check on this. Now the value is run through
3635 the string_printing() function so that such characters are converted to
3636 printable escape sequences.
3638 PH/41 $message_linecount is a new variable that contains the total number of
3639 lines in the message. Compare $body_linecount, which is the count for the
3642 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3643 bounce and delay warning messages, except in certain special cases, which
3646 (a) An SMTP error message from a remote host;
3647 (b) A message specified in a :fail: redirection;
3648 (c) A message specified in a "fail" command in a system filter;
3649 (d) A message specified in a FAIL return from the queryprogram router;
3650 (e) A message specified by the cannot_route_message router option.
3652 In these cases only, Exim does include the error details in bounce and
3653 warning messages. There are also a few cases where bland messages such
3654 as "unrouteable address" or "local delivery error" are given.
3656 PH/43 $value is now also set for the "else" part of a ${run expansion.
3658 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3659 being worked on, but at least Exim now implements the latest version to
3662 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3663 process to complete was treated as a delivery failure, a timeout while
3664 writing the message to the pipe was logged, but erroneously treated as a
3665 successful delivery. Such timeouts include transport filter timeouts. For
3666 consistency with the overall process timeout, these timeouts are now
3667 treated as errors, giving rise to delivery failures by default. However,
3668 there is now a new Boolean option for the pipe transport called
3669 timeout_defer, which, if set TRUE, converts the failures into defers for
3670 both kinds of timeout. A transport filter timeout is now identified in
3673 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3674 systems where "make" and "gmake" are different, calling "gmake" at top
3675 level broke things. I've arranged for the value of $(MAKE) to be passed
3676 from the Makefile to this script so that it can call the same version of
3680 A note about Exim versions 4.44 and 4.50
3681 ----------------------------------------
3683 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3684 changes of various kinds. As a consequence, a big documentation update was
3685 needed. This delayed the release for rather longer than seemed good, especially
3686 in the light of a couple of (minor) security issues. Therefore, the changes
3687 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3688 release. So 4.44 and 4.50 are in effect two different branches that both start
3691 I have left the 4.50 change log unchanged; it contains all the changes since
3692 4.43. The change log for 4.44 is below; many of its items are identical to
3693 those for 4.50. This seems to be the most sensible way to preserve the
3694 historical information.
3700 1. Minor wording change to the doc/README.SIEVE file.
3702 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3703 computation of the current number of files was incorrect.
3705 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3706 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3709 4. Give more explanation in the error message when the command for a transport
3710 filter fails to execute.
3712 5. There are several places where Exim runs a non-Exim command in a
3713 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3714 was being done only for the command run by the queryprogram router. It is
3715 now done for all such subprocesses. The other cases are: ${run, transport
3716 filters, and the commands run by the lmtp and pipe transports.
3718 6. Added CONFIGURE_GROUP build-time option.
3720 7. Some older OS have a limit of 256 on the maximum number of file
3721 descriptors. Exim was using setrlimit() to set 1000 as a large value
3722 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3723 systems. I've change it so that if it can't get 1000, it tries for 256.
3725 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3726 was an oversight, and furthermore, ever since the addition of extra
3727 controls (e.g. 4.43/32), the checks on when to allow different forms of
3728 "control" were broken. There should now be diagnostics for all cases when a
3729 control that does not make sense is encountered.
3731 9. Added the /retain_sender option to "control=submission".
3733 10. $recipients is now available in the predata ACL (oversight).
3735 11. Tidy the search cache before the fork to do a delivery from a message
3736 received from the command line. Otherwise the child will trigger a lookup
3737 failure and thereby defer the delivery if it tries to use (for example) a
3738 cached ldap connection that the parent has called unbind on.
3740 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3741 of $address_data from the recipient verification was clobbered by the
3742 sender verification.
3744 13. The value of address_data from a sender verification is now available in
3745 $sender_address_data in subsequent conditions in the ACL statement.
3747 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3749 15. Added a new option "connect=<time>" to callout options, to set a different
3752 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3753 was its contents. (It was OK if the option was not defined at all.)
3755 17. A "Completed" log line is now written for messages that are removed from
3756 the spool by the -Mrm option.
3758 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3759 information about exactly what failed.
3761 19. Added -dd to debug only the daemon process.
3763 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3764 handles timeouts, both on the server side and network timeouts. Renamed the
3765 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3767 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3768 It is now set to "smtps".
3770 22. $host_address is now set to the target address during the checking of
3771 ignore_target_hosts.
3773 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3774 being passed; this would have caused $sender_host_name to have been used if
3775 matching the list had actually called for a host name (not very likely,
3776 since this list is usually IP addresses). A host name is now passed as
3779 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3780 code that specifies a non-restarting handler (typically sigaction() in
3781 modern systems) in an attempt to fix a rare and obscure crash bug.
3783 25. Narrowed the window for a race in the daemon that could cause it to ignore
3784 SIGCHLD signals. This is not a major problem, because they are used only to
3785 wake it up if nothing else does.
3787 26. A malformed maildirsize file could cause Exim to calculate negative values
3788 for the mailbox size or file count. Odd effects could occur as a result.
3789 The maildirsize information is now recalculated if the size or filecount
3792 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3793 support for a long time. Removed HAVE_SYS_VFS_H.
3795 28. Installed the latest version of exipick from John Jetmore.
3797 29. In an address list, if the pattern was not a regular expression, an empty
3798 subject address (from a bounce message) matched only if the pattern was an
3799 empty string. Non-empty patterns were not even tested. This was the wrong
3800 because it is perfectly reasonable to use an empty address as part of a
3801 database query. An empty address is now tested by patterns that are
3802 lookups. However, all the other forms of pattern expect the subject to
3803 contain a local part and a domain, and therefore, for them, an empty
3804 address still always fails if the pattern is not itself empty.
3806 30. Exim went into a mad DNS loop when attempting to do a callout where the
3807 host was specified on an smtp transport, and looking it up yielded more
3808 than one IP address.
3810 31. Re-factored the code for checking spool and log partition space into a
3811 function that finds that data and another that does the check. The former
3812 is then used to implement four new variables: $spool_space, $log_space,
3813 $spool_inodes, and $log_inodes.
3815 32. The RFC2047 encoding function was originally intended for short strings
3816 such as real names; it was not keeping to the 75-character limit for
3817 encoded words that the RFC imposes. It now respects the limit, and
3818 generates multiple encoded words if necessary. To be on the safe side, I
3819 have increased the buffer size for the ${rfc2047: expansion operator from
3822 33. It is now permitted to omit both strings after an "if" condition; if the
3823 condition is true, the result is "true". As before, when the second string
3824 is omitted, a false condition yields an empty string. This makes it less
3825 cumbersome to write custom ACL and router conditions.
3827 34. Failure to deliver a bounce message always caused it to be frozen, even if
3828 there was an errors_to setting on the router. The errors_to setting is now
3831 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3832 canonical form (fully expanded) before being placed in
3833 $sender_host_address.
3835 36. The table in the code that translates DNS record types into text (T_A to
3836 "A" for instance) was missing entries for NS and CNAME. It is just possible
3837 that this could have caused confusion if both these types were looked up
3838 for the same domain, because the text type is used as part of Exim's
3839 per-process caching. But the chance of anyone hitting this buglet seems
3842 37. The dnsdb lookup has been extended in a number of ways.
3844 (1) There is a new type, "zns", which walks up the domain tree until it
3845 finds some nameserver records. It should be used with care.
3847 (2) There is a new type, "mxh", which is like "mx" except that it returns
3848 just the host names, not the priorities.
3850 (3) It is now possible to give a list of domains (or IP addresses) to be
3851 looked up. The behaviour when one of the lookups defers can be
3852 controlled by a keyword.
3854 (4) It is now possible to specify the separator character for use when
3855 multiple records are returned.
3857 38. The dnslists ACL condition has been extended: it is now possible to supply
3858 a list of IP addresses and/or domains to be looked up in a particular DNS
3861 39. Added log_selector=+queue_time_overall.
3863 40. When running the queue in the test harness, wait just a tad after forking a
3864 delivery process, to get repeatability of debugging output.
3866 41. Include certificate and key file names in error message when GnuTLS fails
3867 to set them up, because the GnuTLS error message doesn't include the name
3868 of the failing file when there is a problem reading it.
3870 42. Allow both -bf and -bF in the same test run.
3872 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3874 44. The "Exiscan patch" is now merged into the mainline Exim source.
3876 45. Sometimes the final signoff response after QUIT could fail to get
3877 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3878 < 0 before doing a fflush(). This bug looks as though it goes back to the
3879 introduction of TLS in release 3.20, but "sometimes" must have been rare
3880 because the tests only now provoked it.
3882 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3883 (this can affect the format of dates).
3885 47. exim_tidydb, when checking for the continued existence of a message for
3886 which it has found a message-specific retry record, was not finding
3887 messages that were in split spool directories. Consequently, it was
3888 deleting retry records that should have stayed in existence.
3890 48. Steve fixed some bugs in eximstats.
3892 49. The SPA authentication driver was not abandoning authentication and moving
3893 on to the next authenticator when an expansion was forced to fail,
3894 contradicting the general specification for all authenticators. Instead it
3895 was generating a temporary error. It now behaves as specified.
3897 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3898 (the order specifies the preference for clients). The order is now AES256,
3899 AES128, 3DES, ARCFOUR128.
3901 51. Small patch to Sieve code - explicitly set From: when generating an
3904 52. Exim crashed if a remote delivery caused a very long error message to be
3905 recorded - for instance if somebody sent an entire SpamAssassin report back
3906 as a large number of 550 error lines. This bug was coincidentally fixed by
3907 increasing the size of one of Exim's internal buffers (big_buffer) that
3908 happened as part of the Exiscan merge. However, to be on the safe side, I
3909 have made the code more robust (and fixed the comments that describe what
3912 53. Now that there can be additional text after "Completed" in log lines (if
3913 the queue_time_overall log selector is set), a one-byte patch to exigrep
3914 was needed to allow it to recognize "Completed" as not the last thing in
3917 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3918 patch that reportedly fixes this has been added. I am not expert enough to
3919 create a test for it. This is what the patch creator wrote:
3921 "I found a little strange behaviour of ldap code when working with
3922 Windows 2003 AD Domain, where users was placed in more than one
3923 Organization Units. When I tried to give exim partial DN, the exit code
3924 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
3925 But simultaneously result of request was absolutely normal ldap result,
3926 so I produce this patch..."
3928 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
3929 REFERENCE, so I have modified the code to exclude the patch when that macro
3932 55. Some experimental protocols are using DNS PTR records for new purposes. The
3933 keys for these records are domain names, not reversed IP addresses. The
3934 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3935 leaves it alone. Component reversal etc. now happens only for IP addresses.
3938 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
3940 57. Double the size of the debug message buffer (to 2048) so that more of very
3941 long debug lines gets shown.
3943 58. The exicyclog utility now does better if the number of log files to keep
3944 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
3946 59. Two changes related to the smtp_active_hostname option:
3948 (1) $smtp_active_hostname is now available as a variable.
3949 (2) The default for smtp_banner uses $smtp_active_hostname instead
3950 of $primary_hostname.
3952 60. The host_aton() function is supposed to be passed a string that is known
3953 to be a valid IP address. However, in the case of IPv6 addresses, it was
3954 not checking this. This is a hostage to fortune. Exim now panics and dies
3955 if the condition is not met. A case was found where this could be provoked
3956 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3957 components; fortuitously, this particular loophole had already been fixed
3958 by change 4.50/55 above.
3960 If there are any other similar loopholes, the new check in host_aton()
3961 itself should stop them being exploited. The report I received stated that
3962 data on the command line could provoke the exploit when Exim was running as
3963 exim, but did not say which command line option was involved. All I could
3964 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3965 running as the user.
3968 61. There was a buffer overflow vulnerability in the SPA authentication code
3969 (which came originally from the Samba project). I have added a test to the
3970 spa_base64_to_bits() function which I hope fixes it.
3973 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
3974 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
3976 63. The daemon start-up calls getloadavg() while still root for those OS that
3977 need the first call to be done as root, but it missed one case: when
3978 deliver_queue_load_max is set with deliver_drop_privilege. This is
3979 necessary for the benefit of the queue runner, because there is no re-exec
3980 when deliver_drop_privilege is set.
3982 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
3983 This has been fixed.
3985 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3986 was in use, was not putting the data itself into the right store pool;
3987 consequently, it could be overwritten for a subsequent message in the same
3988 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3991 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
3993 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
3994 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
3995 the function were treating the return as a boolean value, which happened to
3996 work because 0=false and not-0=true, but is not correct code.
3998 68. The host_aton() function was not handling scoped IPv6 addresses (those
3999 with, for example, "%eth0" on the end) correctly.
4001 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4002 negated items (that is, ~something) in unsigned ints. Some compilers
4003 apparently mutter when there is no cast.
4005 70. If an address verification called from an ACL failed, and did not produce a
4006 user-specific message (i.e. there was only a "system" message), nothing was
4007 put in $acl_verify_message. In this situation, it now puts the system
4010 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4011 session; change 4.31/43 added the unwanted input to the log line - except
4012 that it did not do this in the start of session case. It now does.
4014 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4015 This is wrong and can cause the other end to generate a synchronization
4016 error if it is another Exim or anything else that does the synchronization
4017 check. A QUIT command is no longer sent after a timeout.
4019 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4020 during host lookups.
4022 74. The defer_ok option of callout verification was not working if it was used
4023 when verifying addresses in header lines, that is, for this case:
4025 verify = header_sender/callout=defer_ok
4027 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4028 those file descriptors could be used for SMTP connections. If anything
4029 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4030 could be sent to the SMTP client, causing chaos. The daemon now opens
4031 stdin, stdout, and stderr to /dev/null when it puts itself into the
4034 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4035 log by default. The user can override this with suitable Perl magic.
4037 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4038 the log message when discard triggers, was not working for the DATA ACL or
4039 for the non-SMTP ACL.
4041 78. Error message wording change in sieve.c.
4043 79. If smtp_accept_max_per_host was set, the number of connections could be
4044 restricted to fewer than expected, because the daemon was trying to set up
4045 a new connection before checking whether the processes handling previous
4046 connections had finished. The check for completed processes is now done
4047 earlier. On busy systems, this bug wouldn't be noticed because something
4048 else would have woken the daemon, and it would have reaped the completed
4051 80. If a message was submitted locally by a user whose login name contained one
4052 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4053 It caused a spool format error. I have fixed the spool reading code. A
4054 related problem was that the "from" clause in the Received: line became
4055 illegal because of the space(s). It is now covered by ${quote_local_part.
4057 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4060 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4062 83. Patch from Sieve maintainer to fix unterminated string problem in
4063 "vacation" handling.
4065 84. Some minor changes to the Linux configuration files to help with other
4066 OS variants using glibc.
4068 85. One more patch for Sieve to update vacation handling to latest spec.
4071 ----------------------------------------------------
4072 See the note above about the 4.44 and 4.50 releases.
4073 ----------------------------------------------------
4079 1. Change 4.43/35 introduced a bug that caused file counts to be
4080 incorrectly computed when quota_filecount was set in an appendfile
4083 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4084 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4087 3. Give more explanation in the error message when the command for a transport
4088 filter fails to execute.
4090 4. There are several places where Exim runs a non-Exim command in a
4091 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4092 was being done only for the command run by the queryprogram router. It is
4093 now done for all such subprocesses. The other cases are: ${run, transport
4094 filters, and the commands run by the lmtp and pipe transports.
4096 5. Some older OS have a limit of 256 on the maximum number of file
4097 descriptors. Exim was using setrlimit() to set 1000 as a large value
4098 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4099 systems. I've change it so that if it can't get 1000, it tries for 256.
4101 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4102 was an oversight, and furthermore, ever since the addition of extra
4103 controls (e.g. 4.43/32), the checks on when to allow different forms of
4104 "control" were broken. There should now be diagnostics for all cases when a
4105 control that does not make sense is encountered.
4107 7. $recipients is now available in the predata ACL (oversight).
4109 8. Tidy the search cache before the fork to do a delivery from a message
4110 received from the command line. Otherwise the child will trigger a lookup
4111 failure and thereby defer the delivery if it tries to use (for example) a
4112 cached ldap connection that the parent has called unbind on.
4114 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4115 of $address_data from the recipient verification was clobbered by the
4116 sender verification.
4118 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4119 was its contents. (It was OK if the option was not defined at all.)
4121 11. A "Completed" log line is now written for messages that are removed from
4122 the spool by the -Mrm option.
4124 12. $host_address is now set to the target address during the checking of
4125 ignore_target_hosts.
4127 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4128 being passed; this would have caused $sender_host_name to have been used if
4129 matching the list had actually called for a host name (not very likely,
4130 since this list is usually IP addresses). A host name is now passed as
4133 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4134 code that specifies a non-restarting handler (typically sigaction() in
4135 modern systems) in an attempt to fix a rare and obscure crash bug.
4137 15. Narrowed the window for a race in the daemon that could cause it to ignore
4138 SIGCHLD signals. This is not a major problem, because they are used only to
4139 wake it up if nothing else does.
4141 16. A malformed maildirsize file could cause Exim to calculate negative values
4142 for the mailbox size or file count. Odd effects could occur as a result.
4143 The maildirsize information is now recalculated if the size or filecount
4146 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4147 support for a long time. Removed HAVE_SYS_VFS_H.
4149 18. Updated exipick to current release from John Jetmore.
4151 19. Allow an empty sender to be matched against a lookup in an address list.
4152 Previously the only cases considered were a regular expression, or an
4155 20. Exim went into a mad DNS lookup loop when doing a callout where the
4156 host was specified on the transport, if the DNS lookup yielded more than
4159 21. The RFC2047 encoding function was originally intended for short strings
4160 such as real names; it was not keeping to the 75-character limit for
4161 encoded words that the RFC imposes. It now respects the limit, and
4162 generates multiple encoded words if necessary. To be on the safe side, I
4163 have increased the buffer size for the ${rfc2047: expansion operator from
4166 22. Failure to deliver a bounce message always caused it to be frozen, even if
4167 there was an errors_to setting on the router. The errors_to setting is now
4170 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4171 canonical form (fully expanded) before being placed in
4172 $sender_host_address.
4174 24. Updated eximstats to version 1.33
4176 25. Include certificate and key file names in error message when GnuTLS fails
4177 to set them up, because the GnuTLS error message doesn't include the name
4178 of the failing file when there is a problem reading it.
4180 26. Expand error message when OpenSSL has problems setting up cert/key files.
4183 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4184 (this can affect the format of dates).
4186 28. exim_tidydb, when checking for the continued existence of a message for
4187 which it has found a message-specific retry record, was not finding
4188 messages that were in split spool directories. Consequently, it was
4189 deleting retry records that should have stayed in existence.
4191 29. eximstats updated to version 1.35
4192 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4193 1.35 - bugfix such that pie charts by volume are generated correctly
4195 30. The SPA authentication driver was not abandoning authentication and moving
4196 on to the next authenticator when an expansion was forced to fail,
4197 contradicting the general specification for all authenticators. Instead it
4198 was generating a temporary error. It now behaves as specified.
4200 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4201 (the order specifies the preference for clients). The order is now AES256,
4202 AES128, 3DES, ARCFOUR128.
4204 31. Small patch to Sieve code - explicitly set From: when generating an
4207 32. Exim crashed if a remote delivery caused a very long error message to be
4208 recorded - for instance if somebody sent an entire SpamAssassin report back
4209 as a large number of 550 error lines. This bug was coincidentally fixed by
4210 increasing the size of one of Exim's internal buffers (big_buffer) that
4211 happened as part of the Exiscan merge. However, to be on the safe side, I
4212 have made the code more robust (and fixed the comments that describe what
4215 33. Some experimental protocols are using DNS PTR records for new purposes. The
4216 keys for these records are domain names, not reversed IP addresses. The
4217 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4218 leaves it alone. Component reversal etc. now happens only for IP addresses.
4221 34. The host_aton() function is supposed to be passed a string that is known
4222 to be a valid IP address. However, in the case of IPv6 addresses, it was
4223 not checking this. This is a hostage to fortune. Exim now panics and dies
4224 if the condition is not met. A case was found where this could be provoked
4225 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4226 components; fortuitously, this particular loophole had already been fixed
4227 by change 4.50/55 or 4.44/33 above.
4229 If there are any other similar loopholes, the new check in host_aton()
4230 itself should stop them being exploited. The report I received stated that
4231 data on the command line could provoke the exploit when Exim was running as
4232 exim, but did not say which command line option was involved. All I could
4233 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4234 running as the user.
4237 35. There was a buffer overflow vulnerability in the SPA authentication code
4238 (which came originally from the Samba project). I have added a test to the
4239 spa_base64_to_bits() function which I hope fixes it.
4242 36. The daemon start-up calls getloadavg() while still root for those OS that
4243 need the first call to be done as root, but it missed one case: when
4244 deliver_queue_load_max is set with deliver_drop_privilege. This is
4245 necessary for the benefit of the queue runner, because there is no re-exec
4246 when deliver_drop_privilege is set.
4248 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4249 was in use, was not putting the data itself into the right store pool;
4250 consequently, it could be overwritten for a subsequent message in the same
4251 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4254 38. Sometimes the final signoff response after QUIT could fail to get
4255 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4256 < 0 before doing a fflush(). This bug looks as though it goes back to the
4257 introduction of TLS in release 3.20, but "sometimes" must have been rare
4258 because the tests only now provoked it.
4264 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4265 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4266 Now it returns an integer. A number of calls were still expecting a T/F
4267 return. Fortuitously, in all cases, the tests worked in OK situations,
4268 which is the norm. However, things would have gone wrong on any write
4269 failures on the smtp file descriptor. This function is used when sending
4270 messages over SMTP and also when doing verify callouts.
4272 2. When Exim is called to do synchronous delivery of a locally submitted
4273 message (the -odf or -odi options), it no longer closes stderr before doing
4276 3. Implemented the mua_wrapper option.
4278 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4280 5. Implemented the functions header_remove(), header_testname(),
4281 header_add_at_position(), and receive_remove_recipient(), and exported them
4284 6. If an ACL "warn" statement specified the addition of headers, Exim already
4285 inserted X-ACL-Warn: at the start if there was no header name. However, it
4286 was not making this test for the second and subsequent header lines if
4287 there were newlines in the string. This meant that an invalid header could
4288 be inserted if Exim was badly configured.
4290 7. Allow an ACL "warn" statement to add header lines at the start or after all
4291 the Received: headers, as well as at the end.
4293 8. Added the rcpt_4xx retry error code.
4295 9. Added postmaster_mailfrom=xxx to callout verification option.
4297 10. Added mailfrom=xxxx to the callout verification option, for verify=
4300 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4301 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4303 12. Inserted some casts to stop certain compilers warning when using pointer
4304 differences as field lengths or precisions in printf-type calls (mostly
4305 affecting debugging statements).
4307 13. Added optional readline() support for -be (dynamically loaded).
4309 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4310 same clock tick as a message's arrival, so that its received time was the
4311 same as the "first fail" time on the retry record, and that message
4312 remained on the queue past the ultimate address timeout, every queue runner
4313 would try a delivery (because it was past the ultimate address timeout) but
4314 after another failure, the ultimate address timeout, which should have then
4315 bounced the address, did not kick in. This was a "< instead of <=" error;
4316 in most cases the first failure would have been in the next clock tick
4317 after the received time, and all would be well.
4319 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4320 being recognized when the domain list was tested by the match_domain
4321 condition in an expansion string.
4323 16. Added the ${str2b64: operator.
4325 17. Exim was always calling setrlimit() to set a large limit for the number of
4326 processes, without checking whether the existing limit was already
4327 adequate. (It did check for the limit on file descriptors.) Furthermore,
4328 errors from getrlimit() and setrlimit() were being ignored. Now they are
4329 logged to the main and panic logs, but Exim does carry on, to try to do its
4330 job under whatever limits there are.
4332 18. Imported PCRE 5.0.
4334 19. Trivial typo in log message " temporarily refused connection" (the leading
4337 20. If the log selector return_path_on_delivery was set and an address was
4338 redirected to /dev/null, the delivery process crashed because it assumed
4339 that a return path would always be set for a "successful" delivery. In this
4340 case, the whole delivery is bypassed as an optimization, and therefore no
4343 21. Internal re-arrangement: the function for sending a challenge and reading
4344 a response while authentication was assuming a zero-terminated challenge
4345 string. It's now changed to take a pointer and a length, to allow for
4346 binary data in such strings.
4348 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4350 23. Exim was not respecting finduser_retries when seeking the login of the
4351 uid under which it was called; it was always trying 10 times. (The default
4352 setting of finduser_retries is zero.) Also, it was sleeping after the final
4353 failure, which is pointless.
4355 24. Implemented tls_on_connect_ports.
4357 25. Implemented acl_smtp_predata.
4359 26. If the domain in control=submission is set empty, Exim assumes that the
4360 authenticated id is a complete email address when it generates From: or
4361 Sender: header lines.
4363 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4364 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4365 chown and chgrp in /bin and hostname in /usr/bin.
4367 28. Exim was keeping the "process log" file open after each use, just as it
4368 does for the main log. This opens the possibility of it remaining open for
4369 long periods when the USR1 signal hits a daemon. Occasional processlog
4370 errors were reported, that could have been caused by this. Anyway, it seems
4371 much more sensible not to leave this file open at all, so that is what now
4374 29. The long-running daemon process does not normally write to the log once it
4375 has entered its main loop, and it closes the log before doing so. This is
4376 so that log files can straightforwardly be renamed and moved. However,
4377 there are a couple of unusual error situations where the daemon does write
4378 log entries, and I had neglected to close the log afterwards.
4380 30. The text of an SMTP error response that was received during a remote
4381 delivery was being truncated at 512 bytes. This is too short for some of
4382 the long messages that one sometimes sees. I've increased the limit to
4385 31. It is now possible to make retry rules that apply only when a message has a
4386 specific sender, in particular, an empty sender.
4388 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4389 it possible to be selective about when SMTP synchronization is enforced.
4391 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4393 32. Implemented hosts_connection_nolog.
4395 33. Added an ACL for QUIT.
4397 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4400 35. Added mailbox_size and mailbox_filecount to appendfile.
4402 36. Added control = no_multiline_responses to ACLs.
4404 37. There was a bug in the logic of the code that waits for the clock to tick
4405 in the case where the clock went backwards by a substantial amount such
4406 that the microsecond fraction of "now" was more than the microsecond
4407 fraction of "then" (but the whole seconds number was less).
4409 38. Added support for the libradius Radius client library this is found on
4410 FreeBSD (previously only the radiusclient library was supported).
4416 1. When certain lookups returned multiple values in the form name=value, the
4417 quoting of the values was not always being done properly. Specifically:
4418 (a) If the value started with a double quote, but contained no whitespace,
4420 (b) If the value contained whitespace other than a space character (i.e.
4421 tabs or newlines or carriage returns) it was not quoted.
4422 This fix has been applied to the mysql and pgsql lookups by writing a
4423 separate quoting function and calling it from the lookup code. The fix
4424 should probably also be applied to nisplus, ibase and oracle lookups, but
4425 since I cannot test any of those, I have not disturbed their existing code.
4427 2. A hit in the callout cache for a specific address caused a log line with no
4428 reason for rejecting RCPT. Now it says "Previous (cached) callout
4429 verification failure".
4431 3. There was an off-by-one bug in the queryprogram router. An over-long
4432 return line was truncated at 256 instead of 255 characters, thereby
4433 overflowing its buffer with the terminating zero. As well as fixing this, I
4434 have increased the buffer size to 1024 (and made a note to document this).
4436 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4437 when Exim is waiting for an SMTP response from a remote server, Exim
4438 restarts its select() call on the socket, thereby resetting its timeout.
4439 This is not a problem when such interrupts are rare. Somebody set up a cron
4440 job to run exiwhat every 2 minutes, which is less than the normal select()
4441 timeout (5 or 10 minutes). This meant that the select() timeout never
4442 kicked in because it was always reset. I have fixed this by comparing the
4443 time when an interrupt arrives with the time at the start of the first call
4444 to select(). If more time than the timeout has elapsed, the interrupt is
4445 treated as a timeout.
4447 5. Some internal re-factoring in preparation for the addition of Sieve
4448 extensions (by MH). In particular, the "personal" test is moved to a
4449 separate function, and given an option for scanning Cc: and Bcc: (which is
4450 not set for Exim filters).
4452 6. When Exim created an email address using the login of the caller as the
4453 local part (e.g. when creating a From: or Sender: header line), it was not
4454 quoting the local part when it contained special characters such as @.
4456 7. Installed new OpenBSD configuration files.
4458 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4459 try to make them clearer.
4461 9. Callout options, other than the timeout value, were being ignored when
4462 verifying sender addresses in header lines. For example, when using
4464 verify = header_sender/callout=no_cache
4466 the cache was (incorrectly) being used.
4468 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4469 only the Cygwin environment.
4471 11. When return_path_on_delivery was set as a log selector, if different remote
4472 addresses in the same message used different return paths and parallel
4473 remote delivery occurred, the wrong values would sometimes be logged.
4474 (Whenever a remote delivery process finished, the return path value from
4475 the most recently started remote delivery process was logged.)
4477 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4478 lines when AUTH and/or TLS are in use. This is the "received protocol"
4479 field. Exim used to use "asmtp" for authenticated SMTP, without any
4480 indication (in the protocol name) for TLS use. Now it follows the RFC and
4481 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4482 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4483 names appear in log lines as well as in Received: header lines.
4485 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4486 extensions, and comparison tests, and to fix some bugs.
4488 14. Changes to the "personal" filter test:
4490 (1) The test was buggy in that it was just doing the equivalent of
4491 "contains" tests on header lines. For example, if a user's address was
4492 anne@some.where, the "personal" test would incorrectly be true for
4494 To: susanne@some.where
4496 This test is now done by extracting each address from the header in turn,
4497 and checking the entire address. Other tests that are part of "personal"
4498 are now done using regular expressions (for example, to check local parts
4499 of addresses in From: header lines).
4501 (2) The list of non-personal local parts in From: addresses has been
4502 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4503 taken from the Sieve specification recommendations.
4505 (3) If the message contains any header line starting with "List-" it is
4506 treated as non-personal.
4508 (4) The test for "circular" in the Subject: header line has been removed
4509 because it now seems ill-conceived.
4511 15. Minor typos in src/EDITME comments corrected.
4513 16. Installed latest exipick from John Jetmore.
4515 17. If headers_add on a router specified a text string that was too long for
4516 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4517 of string_sprintf() is now avoided.
4519 18. $message_body_size was not set (it was always zero) when running the DATA
4520 ACL and the local_scan() function.
4522 19. For the "mail" command in an Exim filter, no default was being set for
4523 the once_repeat time, causing a random time value to be used if "once" was
4524 specified. (If the value happened to be <= 0, no repeat happened.) The
4525 default is now 0s, meaning "never repeat". The "vacation" command was OK
4526 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4527 (I found it when inspecting the code).
4529 20. There is now an overall timeout for performing a callout verification. It
4530 defaults to 4 times the callout timeout, which applies to individual SMTP
4531 commands during the callout. The overall timeout applies when there is more
4532 than one host that can be tried. The timeout is checked before trying the
4533 next host. This prevents very long delays if there are a large number of
4534 hosts and all are timing out (e.g. when the network connections are timing
4535 out). The value of the overall timeout can be changed by specifying an
4536 additional sub-option for "callout", called "maxwait". For example:
4538 verify = sender/callout=5s,maxwait=20s
4540 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4541 to the end before writing, but this should make it even safer).
4543 22. Exim was forgetting that it had advertised PIPELINING for the second and
4544 subsequent messages on an SMTP connection. It was also not resetting its
4545 memory on STARTTLS and an internal HELO.
4547 23. When Exim logs an SMTP synchronization error within a session, it now
4548 records whether PIPELINING has been advertised or not.
4550 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4551 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4552 rather than long int.
4554 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4556 26. Added the never_mail option to autoreply.
4562 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4563 crash if the getsockname() call failed; this can happen if a connection is
4564 closed very soon after it is established. The problem was simply in the
4565 order in which certain operations were done, causing Exim to try to write
4566 to the SMTP stream before it had set up the file descriptor. The bug has
4567 been fixed by making things happen in the correct order.
4573 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4574 before the connection was closed, thus losing the rejection response.
4576 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4577 some early Solaris releases, but causes trouble in current releases where
4578 socklen_t is defined.
4580 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4583 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4586 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4587 that releasing the top of it at the end releases what was used for sub-
4588 expansions (unless the block got too big). However, discard this block if
4589 the first thing is a variable or header, so that we can use its block when
4590 it is dynamic (useful for very large $message_headers, for example).
4592 6. Lookups now cache *every* query, not just the most recent. A new, separate
4593 store pool is used for this. It can be recovered when all lookup caches are
4594 flushed. Lookups now release memory at the end of their result strings.
4595 This has involved some general refactoring of the lookup sources.
4597 7. Some code has been added to the store_xxx() functions to reduce the amount
4598 of flapping under certain conditions.
4600 8. log_incoming_interface used to affect only the <= reception log lines. Now
4601 it causes the local interface and port to be added to several more SMTP log
4602 lines, for example "SMTP connection from", and rejection lines.
4604 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4606 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4608 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4609 could be overwritten at the end of the current message (or the start of a
4610 new message if it was set in a HELO ACL). The value is now preserved for
4611 the duration of the SMTP connection.
4613 12. If a transport had a headers_rewrite setting, and a matching header line
4614 contained an unqualified address, that address was qualified, even if it
4615 did not match any rewriting rules. The underlying bug was that the values
4616 of the flags that permit the existence of unqualified sender and recipient
4617 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4618 non-local messages, and by -bnq for local messages) were not being
4619 preserved with the message after it was received.
4621 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4622 "next input=" as part of the text comprising the host identity instead of
4623 the correct text. The code was using the same buffer for two different
4624 strings. However, depending on which order the printing function evaluated
4625 its arguments, the bug did not always show up. Under Linux, for example, my
4626 test suite worked just fine.
4628 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4629 doesn't work with some older versions of Perl. It has been changed to "my",
4630 which in any case is probably the better facility to use.
4632 15. A really picky compiler found some instances of statements for creating
4633 error messages that either had too many or two few arguments for the format
4636 16. The size of the buffer for calls to the DNS resolver has been increased
4637 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4638 for addresses that have a lot of PTR records. This alleviates a problem; it
4639 does not fully solve it.
4641 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4642 buffer now truncates the list and logs the incident, which is the same
4643 action as happens when Exim is looking up a host name and its aliases.
4644 Previously in this situation something unpredictable would happen;
4645 sometimes it was "internal error: store_reset failed".
4647 18. If a server dropped the connection unexpectedly when an Exim client was
4648 using GnuTLS and trying to read a response, the client delivery process
4649 crashed while trying to generate an error log message.
4651 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4652 string, for example:
4654 warn message = H1: something\nH2: something
4656 the text was added as a single header line from Exim's point of view
4657 though it ended up OK in the delivered message. However, searching for the
4658 second and subsequent header lines using $h_h2: did not work. This has been
4659 fixed. Similarly, if a system filter added multiple headers in this way,
4660 the routers could not see them.
4662 20. Expanded the error message when iplsearch is called with an invalid key to
4663 suggest using net-iplsearch in a host list.
4665 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4666 ACLs are no longer actually imposed (and a message to that effect is
4669 22. If a "gecos" field in a passwd entry contained escaped characters, in
4670 particular, if it contained a \" sequence, Exim got it wrong when building
4671 a From: or a Sender: header from that name. A second bug also caused
4672 incorrect handling when an unquoted " was present following a character
4673 that needed quoting.
4675 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4676 was not being matched caselessly.
4678 24. Arranged for all hyphens in the exim.8 source to be escaped with
4681 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4682 properly. Recipient callout cache records were still being keyed to include
4683 the sender, even when use_sender was set false. This led to far more
4684 callouts that were necessary. The sender is no longer included in the key
4685 when use_sender is false.
4687 26. Added "control = submission" modifier to ACLs.
4689 27. Added the ${base62d: operator to decode base 62 numbers.
4691 28. dnsdb lookups can now access SRV records.
4693 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4694 the configuration file.
4696 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4697 (-v) mode. This makes the output for a verbose queue run more intelligible.
4699 31. Added a use_postmaster feature to recipient callouts.
4701 32. Added the $body_zerocount variable, containing the number of binary zero
4702 bytes in the message body.
4704 33. The time of last modification of the "new" subdirectory is now used as the
4705 "mailbox time last read" when there is a quota error for a maildir
4708 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4710 35. Added +ignore_unknown as a special item in host lists.
4712 36. Code for decoding IPv6 addresses in host lists is now included, even if
4713 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4714 address was recognized as an IP address, but was then not correctly decoded
4715 into binary, causing unexpected and incorrect effects when compared with
4722 1. Very minor rewording of debugging text in manualroute to say "list of
4723 hosts" instead of "hostlist".
4725 2. If verify=header_syntax was set, and a header line with an unqualified
4726 address (no domain) and a large number of spaces between the end of the
4727 name and the colon was received, the reception process suffered a buffer
4728 overflow, and (when I tested it) crashed. This was caused by some obsolete
4729 code that should have been removed. The fix is to remove it!
4731 3. When running in the test harness, delay a bit after writing a bounce
4732 message to get a bit more predictability in the log output.
4734 4. Added a call to search_tidyup() just before forking a reception process. In
4735 theory, someone could use a lookup in the expansion of smtp_accept_max_
4736 per_host which, without the tidyup, could leave open a database connection.
4738 5. Added the variables $recipient_data and $sender_data which get set from a
4739 lookup success in an ACL "recipients" or "senders" condition, or a router
4740 "senders" option, similar to $domain_data and $local_part_data.
4742 6. Moved the writing of debug_print from before to after the "senders" test
4745 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4746 problems for message scanning, either using a data ACL, or using
4747 local_scan() because the Received: header was not generated till after they
4748 were called (in order to set the time as the time of reception completion).
4749 I have revised the way this works. The header is now generated after the
4750 body is received, but before the ACL or local_scan() are called. After they
4751 are run, the timestamp in the header is updated.
4757 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4758 before starting a queue runner without re-exec. This happened only when
4759 deliver_drop_privilege was set or when the Exim user was set to root. The
4760 effect of the bug was that timeouts during subsequent deliveries caused
4761 crashes instead of being properly handled. The handler is now left at its
4762 default (and expected) setting.
4764 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4765 message, again when deliver_drop_privilege is set or Exim is run as root.
4766 The bug described in (1) was not present in this case, but the tidying up
4767 of the other signals was missing. I have made the two cases consistent.
4769 3. The ignore_target_hosts setting on a manualroute router was being ignored
4770 for hosts that were looked up using the /MX notation.
4772 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4775 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4776 operated on the sender address. After changing the $sender_address to <>
4777 for the sender address verify, Exim was re-instated it as the original
4778 (before rewriting) address, but remembering that it had rewritten it, so it
4779 wasn't rewriting it again. This bug also had the effect of breaking the
4780 sender address verification caching when the sender address was rewritten.
4782 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4783 This has been changed so that if the ip literal address matches
4784 ignore_target_hosts, the router declines.
4786 7. Added expansion conditions match_domain, match_address, and match_local_
4787 part (NOT match_host).
4789 8. The placeholder for the Received: header didn't have a length field set.
4791 9. Added code to Exim itself and to exim_lock to test for a specific race
4792 condition that could lead to file corruption when using MBX delivery. The
4793 issue is with the lockfile that is created in /tmp. If this file is removed
4794 after a process has opened it but before that process has acquired a lock,
4795 there is the potential for a second process to recreate the file and also
4796 acquire a lock. This could lead to two Exim processes writing to the file
4797 at the same time. The added code performs the same test as UW imapd; it
4798 checks after acquiring the lock that its file descriptor still refers to
4799 the same named file.
4801 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4802 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4805 11. Added the smtp_active_hostname option. If used, this will typically be made
4806 to depend on the incoming interface address. Because $interface_address is
4807 not set up until the daemon has forked a reception process, error responses
4808 that can happen earlier (such as "too many connections") no longer contain
4811 12. If an expansion in a condition on a "warn" statement fails because a lookup
4812 defers, the "warn" statement is abandoned, and the next ACL statement is
4813 processed. Previously this caused the whole ACL to be aborted.
4815 13. Added the iplsearch lookup type.
4817 14. Added ident_timeout as a log selector.
4819 15. Added tls_certificate_verified as a log selector.
4821 16. Added a global option tls_require_ciphers (compare the smtp transport
4822 option of the same name). This controls incoming TLS connections.
4824 17. I finally figured out how to make tls_require_ciphers do a similar thing
4825 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4826 before starting the TLS session.
4828 18. Tabs are now shown as \t in -bP output.
4830 19. If the log selector return_path_on_delivery was set, Exim crashed when
4831 bouncing a message because it had too many Received: header lines.
4833 20. If two routers both had headers_remove settings, and the first one included
4834 a superfluous trailing colon, the final name in the first list and the
4835 first name in the second list were incorrectly joined into one item (with a
4836 colon in the middle).
4842 1. Added -C and -D options to the exinext utility, mainly to make it easier
4843 to include in the automated testing, but these could be helpful when
4844 multiple configurations are in use.
4846 2. The exinext utility was not formatting the output nicely when there was
4847 an alternate port involved in the retry record key, nor when there was a
4848 message id as well (for retries that were specific to a specific message
4849 and a specific host). It was also confused by IPv6 addresses, because of
4850 the additional colons they contain. I have fixed the IPv4 problem, and
4851 patched it up to do a reasonable job for IPv6.
4853 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4854 delivery, the log line now contains "pipelined" if PIPELINING was used.
4856 4. An SMTP transport process used to panic and die if the bind() call to set
4857 an explicit outgoing interface failed. This has been changed; it is now
4858 treated in the same way as a connect() failure.
4860 5. A reference to $sender_host_name in the part of a conditional expansion
4861 that was being skipped was still causing a DNS lookup. This no longer
4864 6. The def: expansion condition was not recognizing references to header lines
4865 that used bh_ and bheader_.
4867 7. Added the _cache feature to named lists.
4869 8. The code for checking quota_filecount in the appendfile transport was
4870 allowing one more file than it should have been.
4872 9. For compatibility with Sendmail, the command line option
4880 and sets the incoming protocol and host name (for trusted callers). The
4881 host name and its colon can be omitted when only the protocol is to be set.
4882 Note the Exim already has two private options, -pd and -ps, that refer to
4883 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4884 "s", but I don't think that's a major issue.
4886 10. A number of refactoring changes to the code, none of which should affect
4889 (a) The number of logging options was getting close to filling up the
4890 32-bit word that was used as a bit map. I have split them into two classes:
4891 those that are passed in the argument to log_write(), and those that are
4892 only ever tested independently outside of that function. These are now in
4893 separate 32-bit words, so there is plenty of room for expansion again.
4894 There is no change in the user interface or the logging behaviour.
4896 (b) When building, for example, log lines, the code previously used a
4897 macro that called string_cat() twice, in order to add two strings. This is
4898 not really sufficiently general. Furthermore, there was one instance where
4899 it was actually wrong because one of the argument was used twice, and in
4900 one call a function was used. (As it happened, calling the function twice
4901 did not affect the overall behaviour.) The macro has been replaced by a
4902 function that can join an arbitrary number of extra strings onto a growing
4905 (c) The code for expansion conditions now uses a table and a binary chop
4906 instead of a serial search (which was left over from when there were very
4907 few conditions). Also, it now recognizes conditions like "pam" even when
4908 the relevant support is not compiled in: a suitably worded error message is
4909 given if an attempt is made to use such a condition.
4911 11. Added ${time_interval:xxxxx}.
4913 12. A bug was causing one of the ddress fields not to be passed back correctly
4914 from remote delivery subprocesses. The field in question was not being
4915 subsequently used, so this caused to problems in practice.
4917 13. Added new log selectors queue_time and deliver_time.
4919 14. Might have fixed a bug in maildirsizefile handling that threw up
4920 "unexpected character" debug warnings, and recalculated the data
4921 unnecessarily. In any case, I expanded the warning message to give more
4924 15. Added the message "Restricted characters in address" to the statements in
4925 the default ACL that block characters like @ and % in local parts.
4927 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
4928 Three changes have been made:
4930 (a) There was a serious bug; a negative response to MAIL caused the whole
4931 recipient domain to be cached as invalid, thereby blocking all messages
4932 to all local parts at the same domain, from all senders. This bug has
4933 been fixed. The domain is no longer cached after a negative response to
4934 MAIL if the sender used is not empty.
4936 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
4939 (c) A new callout option, "use_sender" has been added for people who want
4940 the modified behaviour.
4946 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
4949 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
4950 indeed breaks things for older releases.
4952 3. Added additional logging to the case where there is a problem reading data
4953 from a filter that is running in a subprocess using a pipe, in order to
4954 try to track down a specific problem.
4956 4. Testing facility fudge: when running in the test harness and attempting
4957 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
4958 getting "No route to host". Convert this to a timeout.
4960 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
4963 6. Some OS don't have socklen_t but use size_t instead. This affects the
4964 fifth argument of getsockopt() amongst other things. This is now
4965 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
4966 can be set for individual OS. I have set it for SunOS5, OSF1, and
4967 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
4968 some earlier ones do not.
4970 7. Change 4.30/15 was not doing the test caselessly.
4972 8. The standard form for an IPv6 address literal was being rejected by address
4973 parsing in, for example, MAIL and RCPT commands. An example of this kind of
4974 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4975 this, as well as the form without the "IPv6" on the front (but only when
4976 address literals are enabled, of course).
4978 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
4980 10. Exim crashed if a message with an empty sender address specified by -f
4981 encountered a router with an errors_to setting. This could be provoked only
4982 by a command such as
4986 where an empty string was supplied; "<>" did not hit this bug.
4988 11. Installed PCRE release 4.5.
4990 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
4991 remained set. It is now erased.
4993 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
4994 times from message ids (which are base 36 rather than the normal 62).
4996 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
4997 were being counted as actual protocol errors, and logged if the log
4998 selector +smtp_protocol_error was set. One cannot be perfect in this test,
4999 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5000 and DATA following a set of rejected RCPTs do not count as protocol errors.
5001 In other words, Exim assumes they were pipelined, though this may not
5002 actually be the case. Of course, in all cases the client gets an
5003 appropriate error code.
5005 15. If a lookup fails in an ACL condition, a message about the failure may
5006 be available; it is used if testing the ACL cannot continue, because most
5007 such messages specify what the cause of the deferral is. However, some
5008 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5009 that caused an old message to be retained and used if a later statement
5010 caused a defer, replacing the real cause of the deferral.
5012 16. If an IP address had so many PTR records that the DNS lookup buffer
5013 was not large enough to hold them, Exim could crash while trying to process
5014 the truncated data. It now detects and logs this case.
5016 17. Further to 4.21/58, another change has been made: if (and only if) the
5017 first line of a message (the first header line) ends with CRLF, a bare LF
5018 in a subsequent header line has a space inserted after it, so as not to
5019 terminate the header.
5021 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5022 unnecessarily, used atoi() instead of strtol(), and didn't check the
5023 termination when getting file sizes from file names by regex.
5025 19. Completely re-implemented the support for maildirsize files, in the light
5026 of a number of problems with the previous contributed implementation
5027 (4.30/29). In particular:
5029 . If the quota is zero, the maildirsize file is maintained, but no quota is
5032 . If the maildir directory does not exist, it is created before any attempt
5033 to write a maildirsize file.
5035 . The quota value in the file is just a cache; if the quota is changed in
5036 the transport, the new value overrides.
5038 . A regular expression is available for excluding directories from the
5041 20. The autoreply transport checks the characters in options that define the
5042 message's headers; it allows continued headers, but it was checking with
5043 isspace() after an embedded newline instead of explicitly looking for a
5046 21. If all the "regular" hosts to which an address was routed had passed their
5047 expiry times, and had not reached their retry times, the address was
5048 bounced, even if fallback hosts were defined. Now Exim should go on to try
5051 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5052 equivalent code in the SMTP transport. Some hosts send humungous responses
5053 to HELO/EHLO, more than 1024 it seems.
5055 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5056 gives compiler warnings in some environments. I've now done it "properly",
5059 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5060 (because of problems with the built-in one) was declared to return uschar *
5061 instead of char *, causing compiler failure.
5063 25. Fixed a file descriptor leak when processing alias/forward files.
5065 26. Fixed a minor format string issue in dbfn.c.
5067 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5069 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5070 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5073 29. When a sender address is verified, it is cached, to save repeating the test
5074 when there is more than one recipient in a message. However, when the
5075 verification involves a callout, it is possible for different callout
5076 options to be set for different recipients. It is too complicated to keep
5077 track of this in the cache, so now Exim always runs a verification when a
5078 callout is required, relying on the callout cache for the optimization.
5079 The overhead is duplication of the address routing, but this should not be
5082 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5083 to be verified with callout=postmaster, and the main callout worked but the
5084 postmaster check failed, the verification correctly failed. However, if a
5085 subsequent RCPT command asked for sender verification *without* the
5086 postmaster check, incorrect caching caused this verification also to fail,
5089 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5090 it was not caching the DNS options (qualify_single, search_parents) that
5091 were used when the lookup failed. A subsequent lookup with different
5092 options therefore always gave the same answer, though there were cases
5093 where it should not have. (Example: a "domains = !$mx_any" option on a
5094 dnslookup router: the "domains" option is always processed without any
5095 widening, but the router might have qualify_single set.) Now Exim uses the
5096 cached value only when the same options are set.
5098 32. Added John Jetmore's "exipick" utility to the distribution.
5100 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5101 than a timeout (e.g. a certificate is required, and is not provided), an
5102 Exim server now closes the connection immediately. Previously it waited for
5103 the client to close - but if the client is SSL, it seems that they each
5104 wait for each other, leading to a delay before one of them times out.
5106 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5107 maintained 0.8.x compatibility because I don't think many are using it, and
5108 it is clearly obsolete.
5110 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5113 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5114 client certificate was expired. A simple patch fixes this, though I don't
5115 understand the full logic of why the verify callback is called multiple
5118 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5119 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5120 which causes problems with some clients (such as the Certicom SSL Plus
5121 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5122 disables the coutermeasure allowing Eudora to connect."
5124 38. Exim was not checking that a write() to a log file succeeded. This could
5125 lead to Bad Things if a log got too big, in particular if it hit a file
5126 size limit. Exim now panics and dies if it cannot write to a log file, just
5127 as it does if it cannot open a log file.
5129 39. Modified OS/Makefile-Linux so that it now contains
5131 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5133 The two -D definitions ensure that Exim is compiled with large file
5134 support, which makes it possible to handle log files that are bigger than
5137 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5138 instance) a domain was checked against a named list that involved a lookup,
5139 causing $domain_data to be set, then another domain was checked against the
5140 same list, then the first domain was re-checked, the value of $domain_data
5141 after the final check could be wrong. In particular, if the second check
5142 failed, it could be set empty. This bug probably also applied to
5145 41. The strip_trailing_dot option was not being applied to the address given
5146 with the -f command-line option.
5148 42. The code for reading a message's header from the spool was incrementing
5149 $received_count, but never initializing it. This meant that the value was
5150 incorrect (doubled) while delivering a message in the same process in which
5151 it was received. In the most common configuration of Exim, this never
5152 happens - a fresh exec is done - but it can happen when
5153 deliver_drop_privilege is set.
5155 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5156 it now includes up to 150 characters of the unexpected data in the log
5159 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5160 and building data strings. The size of both of these buffers was 10 000
5161 bytes - far larger than anybody would *ever* want, thought I. Needless to
5162 say, somebody hit the limit. I have increased the maximum line length to
5163 20 000 and the maximum data length of concatenated lines to 100 000. I have
5164 also fixed two bugs, because there was no checking on these buffers. Tsk,
5165 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5166 buffer is too small.
5168 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5169 lsearch lookups. Now it does.
5171 46. When parsing a route_list item in a manualroute router, a fixed-length
5172 buffer was used for the list of hosts. I made this 1024 bytes long,
5173 thinking that nobody would ever have a list of hosts that long. Wrong.
5174 Somebody had a whole pile of complicated expansion conditions, and the
5175 string was silently truncated, leading to an expansion error. It turns out
5176 that it is easier to change to an unlimited length (owing to other changes
5177 that have happened since this code was originally written) than to build
5178 structure for giving a limitation error. The length of the item that
5179 expands into the list of hosts is now unlimited.
5181 47. The lsearch lookup could not handle data where the length of text line was
5182 more than 4095 characters. Such lines were truncated, leading to shortened
5183 data being returned. It should now handle lines of any length.
5185 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5186 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5189 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5190 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5191 no longer happens in comment lines. A list of replacements is now placed
5192 at the head of all of the source files, except those whose only change is
5193 to replace PERL_COMMAND in the very first #! line.
5195 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5196 messages on the queue, with a bottom-up merge sort, using code contributed
5197 by Michael Haardt. This should make operations like -bp somewhat faster on
5198 large queues. It won't affect queue runners, except when queue_run_in_order
5201 51. Installed eximstats 1.31 in the distribution.
5203 52. Added support for SRV lookups to the dnslookup router.
5205 53. If an ACL referred to $message_body or $message_body_end, the value was not
5206 reset for any messages that followed in the same SMTP session.
5208 54. The store-handling optimization for building very long strings was not
5209 differentiating between the different store pools. I don't think this
5210 actually made any difference in practice, but I've tidied it.
5212 55. While running the routers to verify a sender address, $sender_address
5213 was still set to the sender address. This is wrong, because when routing to
5214 send a bounce to the sender, it would be empty. Therefore, I have changed
5215 it so that, while verifying a sender address, $sender_address is set to <>.
5216 (There is no change to what happens when verifying a recipient address.)
5218 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5219 target A or AAAA records (if not already returned) without resetting the
5220 qualify_single or search_parents options of the DNS resolver. These are
5221 inappropriate in this case because the targets of MX and SRV records must
5222 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5223 target that, when qualified, matched something in the local domain. These
5224 two options are now turned off when doing these lookups.
5226 57. It seems that at least some releases of Reiserfs (which does not have the
5227 concept of a fixed number of inodes) returns zero and not -1 for the
5228 number of available inodes. This interacted badly with check_spool_inodes,
5229 which assumed that -1 was the "no such thing" setting. What I have done is
5230 to check that the total number of inodes is greater than zero before doing
5231 the test of how many are available.
5233 58. When a "warn" ACL statement has a log_message modifier, the message is
5234 remembered, and not repeated. This is to avoid a lot of repetition when a
5235 message has many recipients that cause the same warning to be written.
5236 Howewer, Exim was preserving the list of already written lines for an
5237 entire SMTP session, which doesn't seem right. The memory is now reset if a
5238 new message is started.
5240 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5241 debugging output unless log_rewrite was also set.
5243 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5244 of (int)(handle) when we know that handle contains (void *)(-1).
5246 61. The Exim daemon panic-logs an error return when it closes the incoming
5247 connection. However "connection reset by peer" seems to be common, and
5248 isn't really an error worthy of noting specially, so that particular error
5251 62. When Exim is trying to find all the local interfaces, it used to panic and
5252 die if the ioctl to get the interface flags failed. However, it seems that
5253 on at least one OS (Solaris 9) it is possible to have an interface that is
5254 included in the list of interfaces, but for which you get a failure error
5255 for this call. This happens when the interface is not "plumbed" into a
5256 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5257 failure of the "get flags" call assumes that the interface is down.
5259 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5260 makes life easier for people who are doing arithmetic on fields extracted
5261 from dates, where you often get leading zeros that should not be
5262 interpreted as octal.
5264 64. Added qualify_domain to the redirect router, to override the global
5267 65. If a pathologically long header line contained very many addresses (the
5268 report of this problem mentioned 10 000) and each of them was rewritten,
5269 Exim could use up a very large amount of memory. (It kept on making new
5270 copies of the header line as it rewrote, and never released the old ones.)
5271 At the expense of a bit more processing, the header rewriting function has
5272 been changed so that it no longer eats memory in this way.
5274 66. The generation of the Received: header has been moved from the time that a
5275 message starts to be received, to the time that it finishes. The timestamp
5276 in the Received: header should now be very close to that of the <= log
5277 line. There are two side-effects of this change:
5279 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5280 logged header lines no longer include the local Received: line, because
5281 it has not yet been created. The same applies to a copy of the message
5282 that is returned to a non-SMTP sender when a message is rejected.
5284 (b) When a filter file is tested using -bf, no additional Received: header
5285 is added to the test message. After some thought, I decided that this
5288 This change does not affect the value of $received_for. It is still set
5289 after address rewriting, but before local_scan() is called.
5291 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5293 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5294 gave an unhelpful panic error message, and a defer error. I have managed to
5295 change this behaviour so that it now rejects any supplied certificate,
5296 which seems right, as the list of acceptable certificates is empty.
5298 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5299 gave an unhelpful defer error. I have not managed to make this reject any
5300 supplied certificates, but the error message it gives is "no certificate
5301 supplied", which is not helpful.
5303 70. exigrep's output now also includes lines that are not associated with any
5304 message, but which match the given pattern. Implemented by a patch from
5305 Martin Sluka, which also tidied up the Perl a bit.
5307 71. Recipient callout verification, like sender verification, was using <> in
5308 the MAIL FROM command. This isn't really the right thing, since the actual
5309 sender may affect whether the remote host accepts the recipient or not. I
5310 have changed it to use the actual sender in the callout; this means that
5311 the cache record is now keyed on a recipient/sender pair, not just the
5312 recipient address. There doesn't seem to be a real danger of callout loops,
5313 since a callout by the remote host to check the sender would use <>.
5314 [SEE ABOVE: changed after hitting problems.]
5316 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5317 temporary errors. However, in the case of such a code being given after
5318 the end of a data transmission (i.e. after ".") Exim was failing to write
5319 a retry record for the message. (Yes, there was some broken host that was
5320 actually sending 8xx at this point.)
5322 73. An unknown lookup type in a host list could cause Exim to panic-die when
5323 the list was checked. (An example that provoked this was putting <; in the
5324 middle of a list instead of at the start.) If this happened during a DATA
5325 ACL check, a -D file could be left lying around. This kind of configuration
5326 error no longer causes Exim to die; instead it causes a defer errror. The
5327 incident is still logged to the main and panic logs.
5329 74. Buglet left over from Exim 3 conversion. The message "too many messages
5330 in one connection" was written to the rejectlog but not the mainlog, except
5331 when address rewriting (yes!) was being logged.
5333 75. Added write_rejectlog option.
5335 76. When a system filter was run not as root (that is, when system_filter_user
5336 was set), the values of the $n variables were not being returned to the
5337 main process; thus, they were not subsequently available in the $sn
5340 77. Added +return_path_on_delivery log selector.
5342 78. A connection timeout was being treated differently from recipients deferred
5343 when testing hosts_max_try with a message that was older than the host's
5344 retry timeout. (The host should not be counted, thus allowing all hosts to
5345 be tried at least once before bouncing.) This may have been the cause of an
5346 occasionally reported bug whereby a message would remain on the queue
5347 longer than the retry timeout, but would be bounced if a delivery was
5348 forced. I say "may" because I never totally pinned down the problem;
5349 setting up timeout/retry tests is difficult. See also the next item.
5351 79. The ultimate address timeout was not being applied to errors that involved
5352 a combination of host plus message (for example, a timeout on a MAIL
5353 command). When an address resolved to a number of possible hosts, and they
5354 were not all tried for each delivery (e.g. because of hosts_max_try), a
5355 message could remain on the queue longer than the retry timeout.
5357 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5360 81. Fixed an obscure SMTP outgoing bug which required at least the following
5361 conditions: (a) there was another message waiting for the same server;
5362 (b) the server returned 5xx to all RCPT commands in the first message so
5363 that the message was not completed; (c) the server dropped the connection
5364 or gave a negative response to the RSET that Exim sends to abort the
5365 transaction. The observed case was a dropped connection after DATA that had
5366 been sent in pipelining mode. That is, the server had advertised PIPELINING
5367 but was not implementing it correctly. The effect of the bug was incorrect
5368 behaviour, such as trying another host, and this could lead to a crash.
5374 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5375 and daemon.c were passed as pointers to ints; they should have been
5376 pointers to socklen_t variables (which are typically unsigned ints).
5378 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5381 3. Fixed a really odd bug that affected only the testing scheme; patching a
5382 certain fixed string in the binary changed the value of another string that
5383 happened to be identical to the end of the original first string.
5385 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5386 name", it returns that address as the IP address. On some operating
5387 systems (e.g. Solaris), it also passes back the IP address string as the
5388 "host name". However, on others (e.g. Linux), it passes back an empty
5389 string. Exim wasn't checking for this, and was changing the host name to an
5390 empty string, assuming it had been canonicized.
5392 5. Although rare, it is permitted to have more than one PTR record for a given
5393 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5394 all the names associated with an address, because they do in Solaris.
5395 However, it seems that they do not in Linux for data that comes from the
5396 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5397 I found this out when I moved to a new Linux workstation and tried to run
5398 the Exim test suite.
5400 To get round this problem I have changed the code so that it now does its
5401 own call to the DNS to look up PTR records when searching for a host name.
5402 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5403 addresses that are only in /etc/hosts are still found.
5405 This behaviour is, however, controlled by an option called host_lookup_
5406 order, which defaults to "bydns:byaddr". If people want to use the other
5407 order, or indeed, just use one or the other means of lookup, they can
5408 specify it in this variable.
5410 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5411 some operating systems, this comes back from gethostbyaddr() as an empty
5412 string, and this is what Exim used to test for. However, it seems that in
5413 other systems, "." is yielded. Exim now tests for this case too.
5415 7. The values of check_spool_space and check_log_space are now held internally
5416 as a number of kilobytes instead of an absolute number of bytes. If a
5417 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5418 kilobyte. This means that much larger values can be stored.
5420 8. Exim monitor: an attempt to get the action menu when not actually pointing
5421 at a message produces an empty menu entitled "No message selected". This
5422 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5423 no entries in it ("Shell widget menu has zero width and/or height"). So I
5424 have added a single, blank menu entry in this case.
5426 9. Added ${quote_local_part.
5428 10. MIME decoding is now applied to the contents of Subject: header lines when
5431 11. Now that a reference to $sender_host_address automatically causes a reverse
5432 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5433 host lookup before query-style lookups in lists that might use this
5434 variable. This has therefore been abolished, and the "net-" prefix is no
5435 longer necessary for query-style lookups.
5437 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5438 have been a typo for LFLAGS, so it has been changed.
5440 13. The install script calls Exim with "-C /dev/null" in order to find the
5441 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5442 to be output. Howeve, since Exim outputs its version number before the
5443 error, it didn't break the script. It just looked ugly. I fixed this by
5444 always allowing "-C /dev/null" if the caller is root.
5446 14. Ignore overlarge ACL variable number when reading spool file - insurance
5447 against a later release with more variables having written the file.
5449 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5450 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5451 this, as well as the form without the "IPv6" on the front.
5453 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5454 OS/Makefile-Darwin file.
5456 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5457 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5459 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5460 variable. However, some OS use ints for inodes. I've added cast to long int
5461 to get rid of the compiler warning.
5463 19. I had forgotten to lock out "/../" in configuration file names when
5464 ALT_CONFIG_PREFIX was set.
5466 20. Routers used for verification do not need to specify transports. However,
5467 if such a router generated a host list, and callout was configured, Exim
5468 crashed, because it could not find a port number from the (non-existent)
5469 transport. It now assumes port 25 in this circumstance.
5471 21. Added the -t option to exigrep.
5473 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5474 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5475 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5476 them is set, LOOKUP_LSEARCH is forced.
5478 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5480 24. Added sender and host information to the "rejected by local_scan()" log
5481 line; previously there was no indication of these.
5483 25. Added .include_if_exists.
5485 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5486 Linux. It turns out that not all file systems support this. Apparently some
5487 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5488 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5489 not supported on the file descriptor, is now ignored when Exim is trying to
5490 sync a directory. This applies only to Linux.
5492 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5494 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5495 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5498 29. Added support for maildirsize files from supplied patch (modified a bit).
5500 30. The use of :fail: followed by an empty string could lead Exim to respond to
5501 sender verification failures with (e.g.):
5503 550 Verification failed for <xxx>
5504 550 Sender verify failed
5506 where the first response line was missing the '-' that indicates it is not
5507 the final line of the response.
5509 31. The loop for finding the name of the user that called Exim had a hardwired
5510 limit of 10; it now uses the value of finduser_retries, which is used for
5511 all other user lookups.
5513 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5516 33. Exim was neglecting to zero errno before one call of strtol() when
5517 expanding a string and expecting an integer value. On some systems this
5518 resulted in spurious "integer overflow" errors. Also, it was casting the
5519 result into an int without checking.
5521 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5522 did not work. The code looks as if it has *never* worked, though it appears
5523 to have been documented since at least releast 1.62. I have made it work.
5525 35. The "timeout_DNS" error in retry rules, also documented since at least
5526 1.62, also never worked. As it isn't clear exactly what this means, and
5527 clearly it isn't a major issue, I have abolished the feature by treating it
5528 as "timeout", and writing a warning to the main and panic logs.
5530 36. The display of retry rules for -brt wasn't always showing the error code
5533 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5534 timeout_connect_A, timeout_connect_MX.
5536 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5537 to the empty sender.
5539 39. The daemon was not analysing the content of -oX till after it had closed
5540 stderr and disconnected from the controlling terminal. This meant that any
5541 syntax errors were only noted on the panic log, and the return code from
5542 the command was 0. By re-arranging the code a little, I've made the
5543 decoding happen first, so such errors now appear on stderr, and the return
5544 code is 1. However, the actual setting up of the sockets still happens in
5545 the disconnected process, so errors there are still only recorded on the
5548 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5549 connections (as happens on some IP stacks) was logged at start up time as
5550 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5551 it from "IPv6 and IPv4", which means that two separate sockets are being
5554 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5555 says whether AF_INET or AF_INET6 was passed as an argument.
5557 42. Exiwhat output was messed up when time zones were included in log
5560 43. Exiwhat now gives more information about the daemon's listening ports,
5561 and whether -tls-on-connect was used.
5563 44. The "port" option of the smtp transport is now expanded.
5565 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5566 silently ignored. Now an error message is written to the main and panic
5569 46. There's a new ACL modifier called "logwrite" which writes to a log file
5570 as soon as it is encountered.
5572 47. Added $local_user_uid and $local_user_gid at routing time.
5574 48. Exim crashed when trying to verify a sender address that was being
5577 49. Exim was recognizing only a space character after ".include". It now also
5578 recognizes a tab character.
5580 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5581 extracting the relevant information from the specification. The man page no
5582 longer contains scrambled data for the -d option, and I've added a section
5583 at the front about calling Exim under different names.
5585 51. Added "extra_headers" argument to the "mail" command in filter files.
5587 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5590 53. Installed eximstats 1.29.
5592 54. Added transport_filter_timeout as a generic transport option.
5594 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5595 Cc: header lines. This was required by RFC 822, but it not required by RFC
5598 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5599 incoming messages that did not have them. Now it does so only if the
5600 message originates locally, that is, if there is no associated remote host
5601 address. When Resent- header lines are present, this applies to the Resent-
5602 lines rather than the non-Resent- lines.
5604 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5605 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5607 58. Messages for configuration errors now include the name of the main
5608 configuration file - useful now that there may be more than one file in a
5609 list (.included file names were always shown).
5611 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5612 for those rare installations that do not start the daemon as root or run it
5613 setuid root. I've cut out the call to initgroups() if the daemon is not
5616 60. The Exim user and group can now be bound into the binary as text strings
5617 that are looked up at the start of Exim's processing.
5619 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5621 62. Added $mailstore_basename variable.
5623 63. Installed patch to sieve.c from Michael Haardt.
5625 64. When Exim failed to open the panic log after failing to open the main log,
5626 the original message it was trying to log was written to stderr and debug
5627 output, but if they were not available (the usual case in production), it
5628 was lost. Now it is written to syslog before the two lines that record the
5629 failures to open the logs.
5631 65. Users' Exim filters run in subprocesses under the user's uid. It is
5632 possible for a "deliver" command or an alias in a "personal" command to
5633 provoke an address rewrite. If logging of address rewriting is configured,
5634 this fails because the process is not running as root or exim. There may be
5635 a better way of dealing with this, but for the moment (because 4.30 needs
5636 to be released), I have disabled address rewrite logging when running a
5637 filter in a non-root, non-exim process.
5643 1. The buildconfig auxiliary program wasn't quoting the value set for
5644 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5645 not defined. This bug was masked in 4.22 by the effect that was fixed in
5648 2. Some messages that were rejected after a message id was allocated were
5649 shown as "incomplete" by exigrep. It no longer does this for messages that
5650 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5652 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5653 have allow_domain_literals set, the ID did not get logged in the <= line.
5654 Domain literals are now always recognized in Message-ID: header lines.
5656 4. The first argument for a ${extract expansion item is the key name or field
5657 number. Leading and trailing spaces in this item were not being ignored,
5658 causing some misleading effects.
5660 5. When deliver_drop_privilege was set, single queue runner processes started
5661 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5662 same command in the process it spins off) were not dropping privilege.
5664 6. When the daemon running as "exim" started a queue runner, it always
5665 re-executed Exim in the spun-off process. This is a waste of effort when
5666 deliver_drop_privilege is set. The new process now just calls the
5667 queue-runner function directly.
5673 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5676 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5677 ignored. Though the use of -oP was forcing the writing of a pid file, it
5678 was always written to the default place.
5680 3. If the message "no IP address found for host xxxx" is generated during
5681 incoming verification, it is now followed by identification of the incoming
5682 connection (so you can more easily find what provoked it).
5684 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5686 5. Added some features to "harden" Exim a bit more against certain attacks:
5688 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5689 be put in Local/Makefile. This is like the never_users runtime option,
5690 but it cannot be overridden. The default setting is "root".
5692 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5693 prefix string with which any file named in a -C command line option
5696 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5697 is retained for -C and -D only if the caller of Exim is root. Without
5698 it, the exim user may also use -C and -D and retain privilege.
5700 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5701 command line option is disabled.
5703 6. Macro names set by the -D option must start with an upper case letter, just
5704 like macro names defined in the configuration file.
5706 7. Added "dereference=" facility to LDAP.
5708 8. Two instances of the typo "uknown" in the source files are fixed.
5710 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5711 the Configure-Makefile script screwed up while processing it.
5713 10. Incorporated PCRE 4.4.
5715 11. The SMTP synchronization check was not operating right at the start of an
5716 SMTP session. For example, it could not catch a HELO sent before the client
5717 waited for the greeting. There is now a check for outstanding input at the
5718 point when the greeting is written. Because of the duplex, asynchronous
5719 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5720 way, but not yet received, when the check is performed.
5722 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5723 on TCP/IP sockets, because this apparently causes some broken clients to
5726 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5727 unchanged) from the Cygwin maintainer.
5729 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5730 was supported instead of testing for mbx. Effectively a typo.
5732 15. The spa authenticator server code was not checking that the input it
5733 received was valid base64.
5735 16. The debug output line for the "set" modifier in ACLs was not showing the
5736 name of the variable that was being set.
5738 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5740 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5741 lookup on its own. Something else was needed to trigger the lookup. For
5742 example, a match in host_lookup or the need for a host name in a host list.
5743 Now, if $sender_host_name is referenced and the host name has not yet been
5744 looked up, a lookup is performed. If the lookup fails, the variable remains
5745 empty, and $host_lookup_failed is set to "1".
5747 19. Added "eqi" as a case-independent comparison operator.
5749 20. The saslauthd authentication condition could segfault if neither service
5750 nor realm was specified.
5752 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5753 error message that was logged was misleading, and incoming SMTP
5754 connections were dropped. The message is now more accurate, and temporary
5755 errors are given to SMTP connections.
5757 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5758 (except RSET) with a 421 error, until QUIT is received. However, it was
5759 failing to send a response to QUIT.
5761 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5762 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5763 ACL is now run after the helo_try_verify_hosts code.
5765 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5766 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5767 case-independent, so other case variants are also recognized). Apparently
5768 some systems use these upper case variants.
5770 25. If more than two messages were waiting for the same host, and a transport
5771 filter was specified for the transport, Exim sent two messages over the
5772 same TCP/IP connection, and then failed with "socket operation on non-
5773 socket" when it tried to send the third.
5775 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5777 27. The extern definition of crypt16() in expand.c was not being excluded when
5778 the OS had its own crypt16() function.
5780 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5781 as a preferred synonym for return_size_limit, both as an option and as an
5784 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5786 30. Changed the default configuration ACL to relax the local part checking rule
5787 for addresses that are not in any local domains. For these addresses,
5788 slashes and pipe symbols are allowed within local parts, but the sequence
5789 /../ is explicitly forbidden.
5791 31. SPA server authentication was not clearing the challenge buffer before
5794 32. log_message in a "warn" ACL statement was writing to the reject log as
5795 well as to the main log, which contradicts the documentation and doesn't
5796 seem right (because no rejection is happening). So I have stopped it.
5798 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5799 However, I am unable to do any testing of this.
5801 34. Fixed an infelicity in the appendfile transport. When checking directories
5802 for a mailbox, to see if any needed to be created, it was accidentally
5803 using path names with one or more superfluous leading slashes; tracing
5804 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5806 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5807 contents are added to the log line that is written for every discarded
5808 recipient. (Previously a log_message setting was ignored.)
5810 36. The ${quote: operator now quotes the string if it is empty.
5812 37. The install script runs exim in order to find its version number. If for
5813 some reason other than non-existence or emptiness, which it checks, it
5814 could not run './exim', it was installing it with an empty version number,
5815 i.e. as "exim-". This error state is now caught, and the installation is
5818 38. An argument was missing from the function that creates an error message
5819 when Exim fails to connect to the socket for saslauthd authentication.
5820 This could cause Exim to crash, or give a corrupted message.
5822 39. Added isip, isip4, and isip6 to ${if conditions.
5824 40. The ACL variables $acl_xx are now saved with the message, and can be
5825 accessed later in routers, transports, and filters.
5827 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5828 strings in the file are not string-expanded.
5830 42. If a MAIL command specified a SIZE value that was too large to fit into an
5831 int variable, the check against message_size_limit failed. Such values are
5832 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5833 day this will have to be increased, but I don't think I want to be around
5834 when emails are that large.
5841 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5842 iconv() is not standard in FreeBSD.
5844 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5845 IPv6 enabled. The observed symptom was a segmentation fault on return from
5846 the function os_common_find_running_interfaces() in src/os.c.
5848 3. In the check_special_case() function in daemon.c I had used "errno" as an
5849 argument name, which causes warnings on some systems. This was basically a
5850 typo, since it was named "eno" in the comments!
5852 4. The code that waits for the clock to tick (at a resolution of some fraction
5853 of a second) so as to ensure message-id uniqueness was always waiting for
5854 at least one whole tick, when it could have waited for less. [This is
5855 almost certainly not relevant at current processor speeds, where it is
5856 unlikely to ever wait at all. But we try to future-proof.]
5858 5. The function that sleeps for a time interval that includes fractions of a
5859 second contained a race. It did not block SIGALRM between setting the
5860 timer, and suspending (a couple of lines later). If the interval was short
5861 and the sigsuspend() was delayed until after it had expired, the suspension
5862 never ended. On busy systems this could lead to processes getting stuck for
5865 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5866 process, before it forks any delivery processes. The open lookup caching
5867 mechanism meant that the open file or database connection was passed into
5868 the delivery process. The problem was that delivery processes always tidy
5869 up cached lookup data. This could cause a problem for the next delivery
5870 process started by the queue runner, because the external queue runner
5871 process does not know about the closure. So the next delivery process
5872 still has data in the lookup cache. In the case of a file lookup, there was
5873 no problem because closing a file descriptor in a subprocess doesn't affect
5874 the parent. However, if the lookup was caching a connection to a database,
5875 the connection was closed, and the second delivery process was likely to
5876 see errors such as "PGSQL: query failed: server closed the connection
5877 unexpectedly". The problem has been fixed by closing all cached lookups
5878 in a queue runner before running a delivery process.
5880 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5881 seem to have the "const" qualifier which it has on other OS. I've
5884 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5885 *of the same type* (client or server) with the same public name that an
5886 error should be diagnosed.
5888 9. When Exim looked up a host name for an IP address, but failed to find the
5889 original IP address when looking up the host name (a safety check), it
5890 output the message "<ip address> does not match any IP for NULL", which was
5891 confusing, to say the least. The bug was that the host name should have
5892 appeared instead of "NULL".
5894 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5895 user that is built into the binary, and the -C or -D options is used, root
5896 privilege is dropped before the configuration file is read. In addition,
5897 logging is switched to stderr instead of the normal log files. If the
5898 configuration then re-defines the Exim user, the unprivileged environment
5899 is probably not what is expected, so Exim logs a panic warning message (but
5902 However, if deliver_drop_privilege is set, the unprivileged state may well
5903 be exactly what is intended, so the warning has been cut out in that case,
5904 and Exim is allowed to try to write to its normal log files.
5910 1. smtp_return_error_details was not giving details for temporary sender
5911 or receiver verification errors.
5913 2. Diagnose a configuration error if two authenticators have the same public
5916 3. Exim used not to create the message log file for a message until the first
5917 delivery attempt. This could be confusing when incoming messages were held
5918 for policy or load reasons. The message log file is now created at the time
5919 the message is received, and an initial "Received" line is written to it.
5921 4. The automatically generated man page for command line options had a minor
5922 bug that caused no ill effects; however, a more serious problem was that
5923 the procedure for building the man page automatically didn't always
5924 operate. Consequently, release 4.20 contains an out-of-date version. This
5925 shouldn't happen again.
5927 5. When building Exim with embedded Perl support, the script that builds the
5928 Makefile was calling 'perl' to find its compile-time parameters, ignoring
5929 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
5931 6. The freeze_tell option was not being used for messages that were frozen on
5932 arrival, either by an ACL or by local_scan().
5934 7. Added the smtp_incomplete_transaction log selector.
5936 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
5937 was accepting AUTH without a new EHLO.
5939 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
5940 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
5943 10. Logging of TCP/IP connections (when configured) now happens in the main
5944 daemon process instead of the child process, so that the TCP/IP connection
5945 count is more accurate (but it can never be perfect).
5947 11. The use of "drop" in a nested ACL was not being handled correctly in the
5948 outer ACL. Now, if condition failure induced by the nested "drop" causes
5949 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
5950 or "require"), the connection is dropped.
5952 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
5953 that yield "discard" can now be used with an "accept" or a "discard" verb,
5954 but an error is generated for any others (because I can't see a useful way
5955 to define what should happen).
5957 13. When an ACL is read dynamically from a file (or anywhere else), the lines
5958 are now processed in the same way as lines in the Exim configuration file.
5959 In particular, continuation lines are supported.
5961 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
5963 15. Added -ti meaning -t -i.
5965 16. Check for letters, digits, hyphens, and dots in the names of dnslist
5966 domains, and warn by logging if others are found.
5968 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
5969 returned from GIFCONF when Exim is trying to find the list of interfaces on
5970 a host. The code in os.c has been modified to copy each ifreq to an aligned
5971 structure in all cases.
5973 Also, in some cases, the returned ifreq's were being copied to a 'struct
5974 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
5975 means the last couple of bytes of an IPv6 address could be chopped if the
5976 ifreq contained only a normal sockaddr (14 bytes storage).
5978 18. Named domain lists were not supported in the hosts_treat_as_local option.
5979 An entry such as +xxxx was not recognized, and was treated as a literal
5982 19. Ensure that header lines added by a DATA ACL are included in the reject log
5983 if the ACL subsequently rejects the message.
5985 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
5986 MD5 (which is deprecated).
5988 21. When testing a filter file using -bf, Exim was writing a message when it
5989 took the sender from a "From " line in the message, but it was not doing so
5990 when it took $return_path from a Return-Path: header line. It now does.
5992 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
5993 with a valid header line field name (a series of printing characters
5994 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
5996 23. Changed "disc" in the source to "disk" to conform to the documentation and
5997 the book and for uniformity.
5999 24. Ignore Sendmail's -Ooption=value command line item.
6001 25. When execve() failed while trying to run a command in a pipe transport,
6002 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6003 could be confused with a return value of 69 from the command itself. This
6004 has been changed to 127, the value the shell returns if it is asked to run
6005 a non-existent command. The wording for the related log line suggests a
6006 non-existent command as the problem.
6008 26. If received_header_text expands to an empty string, do not add a Received:
6009 header line to the message. (Well, it adds a token one on the spool, but
6010 marks it "old" so that it doesn't get used or transmitted.)
6012 27. Installed eximstats 1.28 (addition of -nt option).
6014 28. There was no check for failure on the call to getsockname() in the daemon
6015 code. This can fail if there is a shortage of resources on the system, with
6016 ENOMEM, for example. A temporary error is now given on failure.
6018 29. Contrary to the C standard, it seems that in some environments, the
6019 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6020 program. Exim now does this explicitly; it affects the formatting of
6021 timestamps using strftime().
6023 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6024 complaints. I've now initialized all the variables, to avoid this.
6026 32. Header lines added by a system filter were not being "seen" during
6027 transport-time rewrites.
6029 33. The info_callback() function passed to OpenSSL is set up with type void
6030 (*)(SSL *, int, int), as described somewhere. However, when calling the
6031 function (actually a macro) that sets it up, the type void(*)() is
6032 expected. I've put in a cast to prevent warnings from picky compilers.
6034 34. If a DNS black list lookup found a CNAME record, but there were no A
6035 records associated with the domain it pointed at, Exim crashed.
6037 35. If a DNS black list lookup returned more than one A record, Exim ignored
6038 all but the first. It now scans all returned addresses if a particular IP
6039 value is being sought. In this situation, the contents of the
6040 $dnslist_value variable are a list of all the addresses, separated by a
6043 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6044 to accept a host name and all its aliases if the forward lookup for any of
6045 them yielded the IP address of the incoming connection. Now it accepts only
6046 those names whose forward lookup yields the correct IP address. Any other
6047 names are discarded. This closes a loophole whereby a rogue DNS
6048 administrator could create reverse DNS records to break through a
6049 wildcarded host restriction in an ACL.
6051 37. If a user filter or a system filter that ran in a subprocess used any of
6052 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6053 the wrong values were passed to the pipe command ($thisaddress had the
6054 value of $0, $0 had the value of $1, etc). This bug was introduced by
6055 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6057 38. Improved the line breaking for long SMTP error messages from ACLs.
6058 Previously, if there was no break point between 40 and 75 characters, Exim
6059 left the rest of the message alone. Two changes have been made: (a) I've
6060 reduced the minimum length to 35 characters; (b) if it can't find a break
6061 point between 35 and 75 characters, it looks ahead and uses the first one
6062 that it finds. This may give the occasional overlong line, but at least the
6063 remaining text gets split now.
6065 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6066 file descriptors might be low, and that setting 1000 would always raise it.
6067 It turns out that in some environments, the limit is already over 1000 and
6068 that lowering it causes trouble. So now Exim takes care not to decrease it.
6070 40. When delivering a message, the value of $return_path is set to $sender_
6071 address at the start of routing (routers may change the value). By an
6072 oversight, this default was not being set up when an address was tested by
6073 -bt or -bv, which affected the outcome if any router or filter referred to
6076 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6077 log only when "message" or "log_message" are set. However, if one of the
6078 conditions was an address verification, or a call to a nested ACL, the
6079 messages generated by the underlying test were being passed through. This
6080 no longer happens. The underlying message is available in $acl_verify_
6081 message for both "message" and "log_message" expansions, so it can be
6082 passed through if needed.
6084 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6085 new expansion $bh_ to give the encoded byte string without charset
6086 translation. Translation happens only if iconv() is available; HAVE_ICONV
6087 indicates this at build time. HEADERS_CHARSET gives the charset to
6088 translate to; headers_charset can change it in the configuration, and
6089 "headers charset" can change it in an individual filter file.
6091 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6092 that creates RFC 2047 encoded "words" labels them as that charset instead
6093 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6094 expansion operator; (ii) when Exim creates a From: line for a local
6095 message; (iii) when a header line is rewritten to include a "phrase" part.
6097 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6098 buggy, causing it to skip the first lines of messages whose message ID
6099 ended in 'D'. This would not have bitten before Exim release 4.14, because
6100 message IDs were unlikely to end in 'D' before then. The effect was to have
6101 incorrect size information for certain domains.
6103 45. #include "config.h" was missing at the start of the crypt16.c module. This
6104 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6107 46. If there was a timeout during a "random" callout check, Exim treated it as
6108 a failure of the random address, and carried on sending RSET and the real
6109 address. If the delay was just some slowness somewhere, the response to the
6110 original RCPT would be taken as a response to RSET and so on, causing
6111 mayhem of various kinds.
6113 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6114 when I implemented it. It didn't allow for the fact that some option values
6115 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6116 the right test for positive values.
6118 48. Domain names in DNS records are case-independent. Exim always looks them up
6119 in lower case. Some resolvers return domain names in exactly the case they
6120 appear in the zone file, that is, they may contain uppercase letters. Not
6121 all resolvers do this - some return always lower case. Exim was treating a
6122 change of case by a resolver as a change of domain, similar to a widening
6123 of a domain abbreviation. This triggered its re-routing code and so it was
6124 trying to route what was effectively the same domain again. This normally
6125 caused routing to fail (because the router wouldn't handle the domain
6126 twice). Now Exim checks for this case specially, and just changes the
6127 casing of the domain that it ultimately uses when it transmits the message
6130 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6133 50. If a filter generated a file delivery with a non-absolute name (possible if
6134 no home directory exists for the router), the forbid_file option was not
6137 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6138 the existing equality matching.
6140 52. Exim was using ints instead of ino_t variables in some places where it was
6141 dealing with inode numbers.
6143 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6144 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6145 called TMPDIR, and if it finds it is different, it changes its value.
6147 54. The smtp_printf() function is now made available to local_scan() so
6148 additional output lines can be written before returning. There is also an
6149 smtp_fflush() function to enable the detection of a dropped connection.
6150 The variables smtp_input and smtp_batched_input are exported to
6153 55. Changed the default runtime configuration: the message "Unknown user"
6154 has been removed from the ACL, and instead placed on the localuser router,
6155 using the cannot_route_message feature. This means that any verification
6156 failures that generate their own messages won't get overridden. Similarly,
6157 the "Unrouteable address" message that was in the ACL for unverifiable
6158 relay addresses has also been removed.
6160 56. Added hosts_avoid_esmtp to the smtp transport.
6162 57. The exicyclog script was not checking for the esoteric option
6163 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6164 will work only if exicyclog is run under the appropriate euid.
6166 58. Following a discussion on the list, the rules by which Exim recognises line
6167 endings on incoming messages have been changed. The -dropcr and drop_cr
6168 options are now no-ops, retained only for backwards compatibility. The
6169 following line terminators are recognized: LF CRLF CR. However, special
6170 processing applies to CR:
6172 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6173 nor a local message in the state where . is a terminator.
6175 (ii) If a bare CR is encountered in a header line, an extra space is added
6176 after the line terminator so as not to end the header. The reasoning
6177 behind this is that bare CRs in header lines are most likely either
6178 to be mistakes, or people trying to play silly games.
6180 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6181 was being incorrectly given as 18 bytes larger than it should have been.
6182 This is a VOB (very old bug).
6184 60. This may never have affected anything current, but just in case it has:
6185 When the local host is found other than at the start of a list of hosts,
6186 the local host, those with the same MX, and any that follow, are discarded.
6187 When the list in question was part of a longer list of hosts, the following
6188 hosts (not currently being processed) were also being discarded. This no
6189 longer happens. I'm not sure if this situation could ever has previously
6192 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6195 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6196 header. This is something that is recommended in a new Internet Draft, and
6197 is something that is documented as being done by Sendmail. There are two
6198 possible values. For messages generated by the autoreply transport, Exim
6201 Auto-Submitted: auto-replied
6203 whereas for all other generated messages (e.g. bounces) it adds
6205 Auto-Submitted: auto-generated
6207 63. The "personal" condition in filters now includes a test for the
6208 Auto-Submitted: header. If it contains the string "auto-" the message it
6209 not considered personal.
6211 64. Added rcpt_include_affixes as a generic transport option.
6213 65. Added queue_only_override (default true).
6215 66. Added the syslog_duplication option.
6217 67. If what should have been the first header line of a message consisted of
6218 a space followed by a colon, Exim was mis-interpreting it as a header line.
6219 It isn't of course - it is syntactically invalid and should therefore be
6220 treated as the start of the message body. The misbehaviour could have
6221 caused a number of strange effects, including loss of data in subsequent
6222 header lines, and spool format errors.
6224 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6225 client host had authenticated. This control can now be exercised by an ACL
6226 for more flexibility.
6228 69. By default, callouts do not happen when testing with -bh. There is now a
6229 variant, -bhc, which does actually run the callout code, including
6230 consulting and updating the callout cache.
6232 70. Added support for saslauthd authentication, courtesy of Alexander
6235 71. If statvfs() failed on the spool or log directories while checking their
6236 size for availability, Exim confusingly gave the error "space shortage".
6237 Furthermore, in debugging mode it crashed with a floating point exception.
6238 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6239 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6240 serious problem, Exim now writes to the main and panic logs when this
6241 happens, with details of the failure. It then refuses to accept the
6242 incoming message, giving the message "spool directory problem" or "log
6243 directory problem" with a 421 code for SMTP messages.
6245 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6246 0, 1, and 2 exist, because some OS complain for execs without them (see
6247 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6248 descriptors. However, the code omitted to check that the open succeeded,
6249 causing mysterious errors if for some reason the permissions on /dev/null
6250 got screwed. Now Exim writes a message to the main and panic logs, and
6251 bombs out if it can't open /dev/null.
6253 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6254 interact so that it is all more flexible. It is supposed to remain
6255 backwards compatible. Also added extra_local_interfaces.
6257 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6258 to bomb out with an assertion failure - to the client this appears as a
6259 connection drop. This problem occurs in the part of the code that was taken
6260 from the Samba project. Fortunately, the assertion is in a very simple
6261 function, so I have fixed this by reproducing the function inline in the
6262 one place where it is called, and arranging for authentication to fail
6263 instead of killing the process with assert().
6265 75. The SPA client code was not working when the server requested OEM rather
6266 than Unicode encoding.
6268 76. Added code to make require_files with a specific uid setting more usable in
6269 the case where statting the file as root fails - usually a non-root-mounted
6270 NFS file system. When this happens and the failure is EACCES, Exim now
6271 forks a subprocess and does the per-uid checking as the relevant uid.
6273 77. Added process_log_path.
6275 78. If log_file_path was not explicitly set, a setting of check_log_space or
6276 check_log_inodes was ignored.
6278 79. If a space check for the spool or log partitions fails, the incident is now
6279 logged. Of course, in the latter case the data may get lost...
6281 80. Added the %p formatting code to string_format() so that it can be used to
6282 print addresses in debug_print(). Adjusted all the address printing in the
6283 debugging in store.c to use %p rather than %d.
6285 81. There was a concern that a line of code in smtp_in.c could overflow a
6286 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6287 initially expressed, the concern was not well-founded, because trailing
6288 spaces are removed early. However, if the trailing spaces were followed by
6289 a NULL, they did not get removed, so the overflow was possible. Two fixes
6292 (a) I re-wrote the offending code in a cleaner fashion.
6293 (b) If an incoming SMTP command contains a NULL character, it is rejected
6296 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6297 runs initgroups(), so that if the Exim user is in any additional groups,
6298 they will be used during message reception.
6304 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob