1 Change log file for Exim from version 4.21
2 -------------------------------------------
7 TL/01 When running the test suite, the README says that variables such as
8 no_msglog_check are global and can be placed anywhere in a specific
9 test's script, however it was observed that placement needed to be near
10 the beginning for it to behave that way. Changed the runtest perl
11 script to read through the entire script once to detect and set these
12 variables, reset to the beginning of the script, and then run through
13 the script parsing/test process like normal.
15 TL/02 The BSD's have an arc4random API. One of the functions to induce
16 adding randomness was arc4random_stir(), but it has been removed in
17 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
18 function when detected.
20 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
21 cause callback expansion.
23 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
24 syntax errors in an expansion can be treated as a string instead of
25 logging or causing an error, due to the internal use of bool_lax
26 instead of bool when processing it.
28 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
29 server certificates when making smtp deliveries.
31 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
33 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
35 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
37 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
38 Merged patch from Sebastian Wiedenroth.
40 JH/05 Fix results-pipe from transport process. Several recipients, combined
41 with certificate use, exposed issues where response data items split
42 over buffer boundaries were not parsed properly. This eventually
43 resulted in duplicates being sent. This issue only became common enough
44 to notice due to the introduction of conection certificate information,
45 the item size being so much larger. Found and fixed by Wolfgang Breyha.
47 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
48 size buffer was used, resulting in syntax errors when an expansion
51 JH/07 Add support for directories of certificates when compiled with a GnuTLS
52 version 3.3.6 or later.
54 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
55 is EXPERIMENTAL_EVENT, the main-configuration and transport options
56 both become "event_action", the variables become $event_name, $event_data
57 and $event_defer_errno. There is a new variable $verify_mode, usable in
58 routers, transports and related events. The tls:cert event is now also
59 raised for inbound connections, if the main configuration event_action
62 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
63 server certificate names available under the smtp transport option
64 "tls_verify_cert_hostname" now do not permit multi-component wildcard
70 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
71 checkers that were complaining about end of non-void function with no
74 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
75 This was a regression intruduced in 4.83 by another bugfix.
77 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
79 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
80 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
86 TF/01 Correctly close the server side of TLS when forking for delivery.
88 When a message was received over SMTP with TLS, Exim failed to clear up
89 the incoming connection properly after forking off the child process to
90 deliver the message. In some situations the subsequent outgoing
91 delivery connection happened to have the same fd number as the incoming
92 connection previously had. Exim would try to use TLS and fail, logging
93 a "Bad file descriptor" error.
95 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
96 utilities have not been installed.
98 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
99 temporary space as the ACL may create new global variables.
101 TL/01 LDAP support uses per connection or global context settings, depending
102 upon the detected version of the libraries at build time.
104 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
105 to extract and use the src ip:port in logging and expansions as if it
106 were a direct connection from the outside internet. PPv2 support was
107 updated based on HAProxy spec change in May 2014.
109 JH/02 Add ${listextract {number}{list}{success}{fail}}.
111 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
112 Properly escape header and check for NULL return.
114 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
117 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
119 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
120 characters in header names, implemented as a verify condition.
121 Contributed by Michael Fischer v. Mollard.
123 TL/05 Rename SPF condition results err_perm and err_temp to standardized
124 results permerror and temperror. Previous values are deprecated but
125 still accepted. In a future release, err_perm and err_temp will be
126 completely removed, which will be a backward incompatibility if the
127 ACL tests for either of these two old results. Patch contributed by
128 user bes-internal on the mailing list.
130 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
132 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
133 selectors, in both main and reject logs.
135 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
138 JH/07 Add malware type "sock" for talking to simple daemon.
140 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
142 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
143 routers/transports under cutthrough routing.
145 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
146 numbers. Touch up "bool" conditional to keep the same definition.
148 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
150 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
152 JH/12 Expand items in router/transport headers_add or headers_remove lists
153 individually rather than the list as a whole. Bug 1452.
155 Required for reasonable handling of multiple headers_ options when
156 they may be empty; requires that headers_remove items with embedded
157 colons must have them doubled (or the list-separator changed).
159 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
160 view the policy declared in the DMARC record. Currently, $dmarc_status
161 is a combined value of both the record presence and the result of the
164 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
166 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
167 dnslookup router and the smtp transport (applying to the forward
170 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
171 of ldap servers used for a specific lookup. Patch provided by Heiko
174 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
175 New variable $lookup_dnssec_authenticated for observability.
177 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
178 Patch submitted by Lars Timman.
180 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
182 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
183 Requires trusted mode and valid format message id, aborts otherwise.
184 Patch contributed by Heiko Schlichting.
186 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
187 certextract with support for various fields. Bug 1358.
189 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
190 is requested by default, modifiable by smtp transport option
193 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
194 operate on certificate variables to give certificate fingerprints
195 Also new ${sha256:cert_variable}.
197 JH/23 The PRDR feature is moved from being Experimental into the mainline.
199 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
200 Christian Aistleitner.
202 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
204 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
205 file. Patch from Wolfgang Breyha.
207 JH/25 Expand the coverage of the delivery $host and $host_address to
208 client authenticators run in verify callout. Bug 1476.
210 JH/26 Port service names are now accepted for tls_on_connect_ports, to
211 align with daemon_smtp_ports. Bug 72.
213 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
214 support and error reporting did not work properly.
216 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
217 and is readable. Patch from Andrew Colin Kissa.
219 TL/14 Enhance documentation of ${run expansion and how it parses the
220 commandline after expansion, particularly in the case when an
221 unquoted variable expansion results in an empty value.
223 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
225 PP/02 Fix internal collision of T_APL on systems which support RFC3123
226 by renaming away from it. Addresses GH issue 15, reported by
229 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
231 TL/15 SECURITY: prevent double expansion in math comparison functions
232 (can expand unsanitized data). Not remotely exploitable.
239 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
241 PP/02 Make -n do something, by making it not do something.
242 When combined with -bP, the name of an option is not output.
244 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
247 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
248 $sender_host_name and config options to manage this, and basic check
251 PP/05 DSCP support for outbound connections and control modifier for inbound.
253 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
254 (Only plugin which currently uses this is kerberos4, which nobody should
255 be using, but we should make it available and other future plugins might
256 conceivably use it, even though it would break NAT; stuff *should* be
257 using channel bindings instead).
259 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
260 name; added for Sendmail compatibility; requires admin caller.
261 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
262 just ignore it); requires trusted caller.
263 Also parse but ignore: -Ac -Am -X<logfile>
266 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
268 TL/02 Add +smtp_confirmation as a default logging option.
270 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
271 Patch by Magnus Holmgren from 2007-02-20.
273 TL/04 Bugzilla 1281 - Spec typo.
274 Bugzilla 1283 - Spec typo.
275 Bugzilla 1290 - Spec grammar fixes.
277 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
279 TL/06 Add Experimental DMARC support using libopendmarc libraries.
281 TL/07 Fix an out of order global option causing a segfault. Reported to dev
282 mailing list by by Dmitry Isaikin.
284 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
286 JH/02 Support "G" suffix to numbers in ${if comparisons.
288 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
290 NM/01 Bugzilla 1197 - Spec typo
291 Bugzilla 1196 - Spec examples corrections
293 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
295 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
296 gnutls_enable_pkcs11, but renamed to more accurately indicate its
299 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
300 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
302 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
303 "acl {{name}{arg}...}", and optional args on acl condition
306 JH/05 Permit multiple router/transport headers_add/remove lines.
308 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
310 JH/07 Avoid using a waiting database for a single-message-only transport.
311 Performance patch from Paul Fisher. Bugzilla 1262.
313 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
316 JH/09 Add $headers_added variable, with content from use of ACL modifier
317 add_header (but not yet added to the message). Bugzilla 199.
319 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
320 Pulled from Bugzilla 817 by Wolfgang Breyha.
322 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
324 (nb: this is the same fix as in Exim 4.80.1)
326 JH/11 Add A= logging on delivery lines, and a client_set_id option on
329 JH/12 Add optional authenticated_sender logging to A= and a log_selector
332 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
334 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
335 advertise SMTP AUTH mechanism to us, instead of a generic
336 protocol violation error. Also, make Exim more robust to bad
337 data from the Dovecot auth socket.
339 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
341 When a queue runner is handling a message, Exim first routes the
342 recipient addresses, during which it prunes them based on the retry
343 hints database. After that it attempts to deliver the message to
344 any remaining recipients. It then updates the hints database using
347 So if a recipient address works intermittently, it can get repeatedly
348 deferred at routing time. The retry hints record remains fresh so the
349 address never reaches the final cutoff time.
351 This is a fairly common occurrence when a user is bumping up against
352 their storage quota. Exim had some logic in its local delivery code
353 to deal with this. However it did not apply to per-recipient defers
354 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
356 This change adds a proper retry rule check during routing so that the
357 final cutoff time is checked against the message's age. We only do
358 this check if there is an address retry record and there is not a
359 domain retry record; this implies that previous attempts to handle
360 the address had the retry_use_local_parts option turned on. We use
361 this as an approximation for the destination being like a local
362 delivery, as in LMTP.
364 I suspect this new check makes the old local delivery cutoff check
365 redundant, but I have not verified this so I left the code in place.
367 TF/02 Correct gecos expansion when From: is a prefix of the username.
369 Test 0254 submits a message to Exim with the header
373 When I ran the test suite under the user fanf2, Exim expanded
374 the header to contain my full name, whereas it should have added
375 a Resent-Sender: header. It erroneously treats any prefix of the
376 username as equal to the username.
378 This change corrects that bug.
380 GF/01 DCC debug and logging tidyup
381 Error conditions log to paniclog rather than rejectlog.
382 Debug lines prefixed by "DCC: " to remove any ambiguity.
384 TF/03 Avoid unnecessary rebuilds of lookup-related code.
386 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
387 Bug spotted by Jeremy Harris; was flawed since initial commit.
388 Would have resulted in OCSP responses post-SNI triggering an Exim
389 NULL dereference and crash.
391 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
393 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
394 Bug detection, analysis and fix by Samuel Thibault.
395 Bugzilla 1331, Debian bug #698092.
397 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
399 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
400 Server implementation by Todd Lyons, client by JH.
401 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
402 config variable "prdr_enable" controls whether the server
403 advertises the facility. If the client requests PRDR a new
404 acl_data_smtp_prdr ACL is called once for each recipient, after
405 the body content is received and before the acl_smtp_data ACL.
406 The client is controlled by bolth of: a hosts_try_prdr option
407 on the smtp transport, and the server advertisement.
408 Default client logging of deliveries and rejections involving
409 PRDR are flagged with the string "PRDR".
411 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
412 fclose(). Diagnosis by Todd Lyons.
414 PP/17 Update configure.default to handle IPv6 localhost better.
415 Patch by Alain Williams (plus minor tweaks).
418 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
419 This is now consistent with GnuTLS, and is now documented: the
420 previous undocumented portable approach to treating the option as
421 unset was to force an expansion failure. That still works, and
422 an empty string is now equivalent.
424 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
425 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
426 not performing validation itself.
428 PP/20 Added force_command boolean option to pipe transport.
429 Patch from Nick Koston, of cPanel Inc.
431 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
434 TF/04 Added udpsend ACL modifer and hexquote expansion operator
436 PP/21 Fix eximon continuous updating with timestamped log-files.
437 Broken in a format-string cleanup in 4.80, missed when I repaired the
438 other false fix of the same issue.
439 Report and fix from Heiko Schlichting.
442 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
443 Report from Prashanth Katuri.
445 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
446 It's SecureTransport, so affects any MacOS clients which use the
447 system-integrated TLS libraries, including email clients.
449 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
450 using a MIME ACL for non-SMTP local injection.
451 Report and assistance in diagnosis by Warren Baker.
453 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
455 JH/16 Fix comparisons for 64b. Bugzilla 1385.
457 TL/09 Add expansion variable $authenticated_fail_id to keep track of
458 last id that failed so it may be referenced in subsequent ACL's.
460 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
463 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
464 ldap library initialization, allowing self-signed CA's to be
465 used. Also properly sets require_cert option later in code by
466 using NULL (global ldap config) instead of ldap handle (per
467 session). Bug diagnosis and testing by alxgomz.
469 TL/12 Enhanced documentation in the ratelimit.pl script provided in
470 the src/util/ subdirectory.
472 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
473 renamed to Transport Post Delivery Action by Jeremy Harris, as
476 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
477 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
478 redis_servers = needs to be configured which will be used by the redis
479 lookup. Patch from Warren Baker, of The Packet Hub.
481 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
483 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
484 hostname or reverse DNS when processing a host list. Used suggestions
485 from multiple comments on this bug.
487 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
489 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
490 Missed a few lines, added it to make the runtest require no keyboard
493 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
494 contains upper case chars. Make router use caseful_local_part.
496 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
497 support when GnuTLS has been built with p11-kit.
503 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
505 This, or similar/improved, will also be change PP/11 of 4.82.
511 PP/01 Handle short writes when writing local log-files.
512 In practice, only affects FreeBSD (8 onwards).
513 Bugzilla 1053, with thanks to Dmitry Isaikin.
515 NM/01 Bugzilla 949 - Documentation tweak
517 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
520 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
522 PP/02 Implemented gsasl authenticator.
524 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
526 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
527 `pkg-config foo` for cflags/libs.
529 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
530 with rest of GSASL and with heimdal_gssapi.
532 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
533 `pkg-config foo` for cflags/libs for the TLS implementation.
535 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
536 properties get this fed in as external SSF. A number of robustness
537 and debugging improvements to the cyrus_sasl authenticator.
539 PP/08 cyrus_sasl server now expands the server_realm option.
541 PP/09 Bugzilla 1214 - Log authentication information in reject log.
542 Patch by Jeremy Harris.
544 PP/10 Added dbmjz lookup type.
546 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
548 PP/12 MAIL args handles TAB as well as SP, for better interop with
549 non-compliant senders.
550 Analysis and variant patch by Todd Lyons.
552 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
553 Bug report from Lars Müller <lars@samba.org> (via SUSE),
554 Patch from Dirk Mueller <dmueller@suse.com>
556 PP/13 tls_peerdn now print-escaped for spool files.
557 Observed some $tls_peerdn in wild which contained \n, which resulted
558 in spool file corruption.
560 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
561 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
562 or write after TLS renegotiation, which otherwise led to messages
565 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
566 as a tracking header (ie: a signed header comes before the signature).
567 Patch from Wolfgang Breyha.
569 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
570 comma-sep list; embedded commas doubled.
572 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
574 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
576 Report and patch from Dmitry Banschikov.
578 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
579 Removed SSL_clear() after SSL_new() which led to protocol negotiation
580 failures. We appear to now support TLS1.1+ with Exim.
582 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
583 lets Exim select keys and certificates based upon TLS SNI from client.
584 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
585 before an outbound SMTP session. New log_selector, +tls_sni.
587 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
588 NULL dereference. Report and patch from Alun Jones.
590 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
591 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
592 Not seeing resolver debug output on NetBSD, but suspect this is a
593 resolver implementation change.
595 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
596 Left warnings. Added "eximon gdb" invocation mode.
598 PP/21 Defaulting "accept_8bitmime" to true, not false.
600 PP/22 Added -bw for inetd wait mode support.
602 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
603 locate the relevant includes and libraries. Made this the default.
605 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
606 Bugzilla 1246, report and most of solution from Tomasz Kusy.
608 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
609 This may cause build issues on older platforms.
611 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
612 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
613 gnutls_require_mac & gnutls_require_protocols (no longer supported).
614 Added SNI support via GnuTLS too.
615 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
617 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
619 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
620 Applied second patch from Janne, implementing suggestion to default
621 multiple-strings-in-record handling to match SPF spec.
623 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
625 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
626 read-only, out of scope).
627 Patch from Wolfgang Breyha, report from Stuart Northfield.
629 PP/29 Fix three issues highlighted by clang analyser static analysis.
630 Only crash-plausible issue would require the Cambridge-specific
631 iplookup router and a misconfiguration.
632 Report from Marcin Mirosław.
634 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
636 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
637 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
638 As part of this, removing so much warning spew let me fix some minor
639 real issues in debug logging.
641 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
642 assignment on my part. Fixed.
644 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
645 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
646 Janne Snabb (who went above and beyond: thank you).
648 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
649 string otherwise requires a connection and a bunch more work and it's
650 relatively easy to get wrong. Should also expose TLS library linkage
653 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
654 64-bit ${eval} (JH/03).
656 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
657 GNU libc to support some of the 64-bit stuff, should not lead to
658 conflicts. Defined before os.h is pulled in, so if a given platform
659 needs to override this, it can.
661 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
662 protection layer was required, which is not implemented.
663 Bugzilla 1254, patch from Wolfgang Breyha.
665 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
666 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
667 tls_dhparam take prime identifiers. Also unbreak combination of
668 OpenSSL+DH_params+TLSSNI.
670 PP/39 Disable SSLv2 by default in OpenSSL support.
676 PP/01 Solaris build fix for Oracle's LDAP libraries.
677 Bugzilla 1109, patch from Stephen Usher.
679 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
681 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
684 TF/02 Fix a couple more cases where we did not log the error message
685 when unlink() failed. See also change 4.74-TF/03.
687 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
688 lock up or crash if it happened to be inside a call to libc when it
689 got a SIGUSR1 from exiwhat.
691 The SIGUSR1 handler appends the current process status to the process
692 log which is later printed by exiwhat. It used to use the general
693 purpose logging code to do this, but several functions it calls are
694 not safe for signals.
696 The new output code in the SIGUSR1 handler is specific to the process
697 log, and simple enough that it's easy to inspect for signal safety.
698 Removing some special cases also simplifies the general logging code.
699 Removing the spurious timestamps from the process log simplifies
702 TF/04 Improved ratelimit ACL condition.
704 The /noupdate option has been deprecated in favour of /readonly which
705 has clearer semantics. The /leaky, /strict, and /readonly update modes
706 are mutually exclusive. The update mode is no longer included in the
707 database key; it just determines when the database is updated. (This
708 means that when you upgrde Exim will forget old rate measurements.)
710 Exim now checks that the per_* options are used with an update mode that
711 makes sense for the current ACL. For example, when Exim is processing a
712 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
713 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
714 must specify per_mail/readonly. If you omit the update mode it defaults to
715 /leaky where that makes sense (as before) or /readonly where required.
717 The /noupdate option is now undocumented but still supported for
718 backwards compatibility. It is equivalent to /readonly except that in
719 ACLs where /readonly is required you may specify /leaky/noupdate or
720 /strict/noupdate which are treated the same as /readonly.
722 A useful new feature is the /count= option. This is a generalization
723 of the per_byte option, so that you can measure the throughput of other
724 aggregate values. For example, the per_byte option is now equivalent
725 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
727 The per_rcpt option has been generalized using the /count= mechanism
728 (though it's more complicated than the per_byte equivalence). When it is
729 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
730 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
731 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
732 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
733 Note that using per_rcpt with a non-readonly update mode in more than
734 one ACL will cause the recipients to be double-counted. (The per_mail
735 and per_byte options don't have this problem.)
737 The handling of very low rates has changed slightly. If the computed rate
738 is less than the event's count (usually one) then this event is the first
739 after a long gap. In this case the rate is set to the same as this event's
740 count, so that the first message of a spam run is counted properly.
742 The major new feature is a mechanism for counting the rate of unique
743 events. The new per_addr option counts the number of different
744 recipients that someone has sent messages to in the last time period. It
745 behaves like per_rcpt if all the recipient addresses are different, but
746 duplicate recipient addresses do not increase the measured rate. Like
747 the /count= option this is a general mechanism, so the per_addr option
748 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
749 example, measure the rate that a client uses different sender addresses
750 with the options per_mail/unique=$sender_address. There are further
751 details in the main documentation.
753 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
755 TF/06 Removed a few PCRE remnants.
757 TF/07 Automatically extract Exim's version number from tags in the git
758 repository when doing development or release builds.
760 PP/02 Raise smtp_cmd_buffer_size to 16kB.
761 Bugzilla 879. Patch from Paul Fisher.
763 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
764 Heavily based on revision 40f9a89a from Simon Arlott's tree.
767 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
769 PP/05 Variable $av_failed, true if the AV scanner deferred.
770 Bugzilla 1078. Patch from John Horne.
772 PP/06 Stop make process more reliably on build failure.
773 Bugzilla 1087. Patch from Heiko Schlittermann.
775 PP/07 Make maildir_use_size_file an _expandable_ boolean.
776 Bugzilla 1089. Patch from Heiko Schlittermann.
778 PP/08 Handle ${run} returning more data than OS pipe buffer size.
779 Bugzilla 1131. Patch from Holger Weiß.
781 PP/09 Handle IPv6 addresses with SPF.
782 Bugzilla 860. Patch from Wolfgang Breyha.
784 PP/10 GnuTLS: support TLS 1.2 & 1.1.
786 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
789 PP/11 match_* no longer expand right-hand-side by default.
790 New compile-time build option, EXPAND_LISTMATCH_RHS.
791 New expansion conditions, "inlist", "inlisti".
793 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
795 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
797 PP/14 fix log_write() format string regression from TF/03.
798 Bugzilla 1152. Patch from Dmitry Isaikin.
804 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
806 PP/02 Harmonised TLS library version reporting; only show if debugging.
807 Layout now matches that introduced for other libraries in 4.74 PP/03.
809 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
811 PP/04 New "dns_use_edns0" global option.
813 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
816 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
817 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
819 TK/01 Updated PolarSSL code to 0.14.2.
820 Bugzilla 1097. Patch from Andreas Metzler.
822 PP/07 Catch divide-by-zero in ${eval:...}.
825 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
828 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
829 format-string attack -- SECURITY: remote arbitrary code execution.
831 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
832 time unintentionally subject to list matching rules, letting the header
833 cause arbitrary Exim lookups (of items which can occur in lists, *not*
834 arbitrary string expansion). This allowed for information disclosure.
836 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
837 INT_MIN/-1 -- value coerced to INT_MAX.
843 NM/01 Workround for PCRE version dependency in version reporting
846 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
847 This fixes portability to compilers other than gcc, notably
848 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
850 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
851 makefiles for portability to HP-UX and POSIX correctness.
853 PP/01 Permit LOOKUP_foo enabling on the make command-line.
854 Also via indented variable definition in the Makefile.
855 (Debugging by Oliver Heesakkers).
857 PP/02 Restore caching of spamd results with expanded spamd_address.
858 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
860 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
861 Improves build reliability. Fix from: Frank Elsner
863 NM/02 Fix wide character breakage in the rfc2047 coding
864 Fixes bug 1064. Patch from Andrey N. Oktyabrski
866 NM/03 Allow underscore in dnslist lookups
867 Fixes bug 1026. Patch from Graeme Fowler
869 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
870 Code patches from Adam Ciarcinski of NetBSD.
872 NM/04 Fixed exiqgrep to cope with mailq missing size issue
875 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
876 is logged, to avoid truncation. Patch from John Horne.
878 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
879 Patch from Jakob Hirsch.
881 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
882 SQL string expansion failure details.
883 Patch from Andrey Oktyabrski.
885 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
886 Patch from Simon Arlott.
888 PP/09 New lookups functionality failed to compile on old gcc which rejects
889 extern declarations in function scope.
890 Patch from Oliver Fleischmann
892 PP/10 Use sig_atomic_t for flags set from signal handlers.
893 Check getgroups() return and improve debugging.
894 Fixed developed for diagnosis in bug 927 (which turned out to be
897 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
898 Patch from Mark Zealey.
900 PP/12 Bugzilla 1056: Improved spamd server selection.
901 Patch from Mark Zealey.
903 PP/13 Bugzilla 1086: Deal with maildir quota file races.
904 Based on patch from Heiko Schlittermann.
906 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
907 Patch from Uwe Doering, sign-off by Michael Haardt.
909 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
910 variable declaration deep within a block. Bug and patch from
913 PP/15 lookups-Makefile IRIX compatibilty coercion.
915 PP/16 Make DISABLE_DKIM build knob functional.
917 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
918 Patch by Simon Arlott
920 TF/03 Fix valgrind.h portability to C89 compilers that do not support
921 variable argument macros. Our copy now differs from upstream.
927 TF/01 Failure to get a lock on a hints database can have serious
928 consequences so log it to the panic log.
930 TF/02 Log LMTP confirmation messages in the same way as SMTP,
931 controlled using the smtp_confirmation log selector.
933 TF/03 Include the error message when we fail to unlink a spool file.
935 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
936 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
937 for maintaining out-of-tree patches for some time.
939 PP/01 Bugzilla 139: Documentation and portability issues.
940 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
941 Handle per-OS dynamic-module compilation flags.
943 PP/02 Let /dev/null have normal permissions.
944 The 4.73 fixes were a little too stringent and complained about the
945 permissions on /dev/null. Exempt it from some checks.
946 Reported by Andreas M. Kirchwitz.
948 PP/03 Report version information for many libraries, including
949 Exim version information for dynamically loaded libraries. Created
950 version.h, now support a version extension string for distributors
951 who patch heavily. Dynamic module ABI change.
953 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
954 privilege escalation vulnerability whereby the Exim run-time user
955 can cause root to append content of the attacker's choosing to
958 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
961 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
962 If dropping privileges for untrusted macros, we disabled normal logging
963 on the basis that it would fail; for the Exim run-time user, this is not
964 the case, and it resulted in successful deliveries going unlogged.
965 Fixed. Reported by Andreas Metzler.
971 PP/01 Date: & Message-Id: revert to normally being appended to a message,
972 only prepend for the Resent-* case. Fixes regression introduced in
973 Exim 4.70 by NM/22 for Bugzilla 607.
975 PP/02 Include check_rfc2047_length in configure.default because we're seeing
976 increasing numbers of administrators be bitten by this.
978 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
980 PP/03 Bugzilla 994: added openssl_options main configuration option.
982 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
984 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
986 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
988 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
989 without a peer certificate, leading to a segfault because of an
990 assumption that peers always have certificates. Be a little more
991 paranoid. Problem reported by Martin Tscholak.
993 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
994 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
995 NB: ClamAV planning to remove STREAM in "middle of 2010".
996 CL also introduces -bmalware, various -d+acl logging additions and
997 more caution in buffer sizes.
999 PP/09 Implemented reverse_ip expansion operator.
1001 PP/10 Bugzilla 937: provide a "debug" ACL control.
1003 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1005 PP/12 Bugzilla 973: Implement --version.
1007 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1009 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1011 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1013 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1014 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1015 ignore trailing whitespace.
1017 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1019 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1022 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1023 Notification from Dr Andrew Aitchison.
1025 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1026 ExtendedDetectionInfo response format.
1027 Notification from John Horne.
1029 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1032 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1033 XSL and documented dependency on system catalogs, with examples of how
1036 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1039 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1040 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1041 configuration file which is writeable by the Exim user or group.
1043 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1044 of configuration files to cover files specified with the -C option if
1045 they are going to be used with root privileges, not just the default
1048 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1049 option (effectively making it always true).
1051 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1052 files to be used while preserving root privileges.
1054 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1055 that rogue child processes cannot use them.
1057 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1058 run-time user, instead of root.
1060 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1061 Exim run-time user without dropping privileges.
1063 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1064 result string, instead of calling string_vformat() twice with the same
1067 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1068 for other users. Others should always drop root privileges if they use
1069 -C on the command line, even for a whitelisted configure file.
1071 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1073 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1079 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1080 $data_path, and $header_path variables; fixed documentation bugs and
1083 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1084 exipick to access non-standard spools, including the "frozen" queue
1087 NM/01 Bugzilla 965: Support mysql stored procedures.
1088 Patch from Alain Williams
1090 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1092 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1093 Patch from Andreas Metzler
1095 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1096 Patch from Kirill Miazine
1098 NM/05 Bugzilla 671: Added umask to procmail example.
1100 JJ/03 installed exipick 20100323.0, fixing doc bug
1102 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1103 directory. Notification and patch from Dan Rosenberg.
1105 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1107 TK/02 Improve log output when DKIM signing operation fails.
1109 MH/01 Treat the transport option dkim_domain as a colon separated
1110 list, not as a single string, and sign the message with each element,
1111 omitting multiple occurences of the same signer.
1113 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1114 Bugzilla 985, 986. Patch by Simon Arlott
1116 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1117 Patch by Simon Arlott
1119 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1120 MBX locking. Notification from Dan Rosenberg.
1126 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1128 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1130 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1132 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1134 NM/04 Bugzilla 915: Flag broken perl installation during build.
1140 TK/01 Added patch by Johannes Berg that expands the main option
1141 "spamd_address" if it starts with a dollar sign.
1143 TK/02 Write list of recipients to X-Envelope-Sender header when building
1144 the mbox-format spool file for content scanning (suggested by Jakob
1147 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1148 (http://www.dcc-servers.net/) support via dccifd. Activated by
1149 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1151 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1152 by Mark Daniel Reidel <mr@df.eu>.
1154 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1155 When building exim an external PCRE library is now needed -
1156 PCRE is a system library on the majority of modern systems.
1157 See entry on PCRE_LIBS in EDITME file.
1159 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1160 conversation. Added nologin parameter to request.
1161 Patch contributed by Kirill Miazine.
1163 TF/01 Do not log submission mode rewrites if they do not change the address.
1165 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1167 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1168 log files in place. Contributed by Roberto Lima.
1170 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1172 TF/03 Bugzilla 615: When checking the local_parts router precondition
1173 after a local_part_suffix or local_part_prefix option, Exim now
1174 does not use the address's named list lookup cache, since this
1175 contains cached lookups for the whole local part.
1177 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1178 Robert Millan. Documentation is in experimental-spec.txt.
1180 TF/04 Bugzilla 668: Fix parallel build (make -j).
1182 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1184 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1185 Patch provided by Jan Srzednicki.
1187 TF/05 Leading white space used to be stripped from $spam_report which
1188 wrecked the formatting. Now it is preserved.
1190 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1191 that they are available at delivery time.
1193 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1195 TF/08 TLS error reporting now respects the incoming_interface and
1196 incoming_port log selectors.
1198 TF/09 Produce a more useful error message if an SMTP transport's hosts
1199 setting expands to an empty string.
1201 NM/06 Bugzilla 744: EXPN did not work under TLS.
1202 Patch provided by Phil Pennock.
1204 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1205 Patch provided by Richard Godbee.
1207 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1208 acl_smtp_notquit, added index entry.
1210 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1211 Patch provided by Eugene Bujak.
1213 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1214 accept(). Patch provided by Maxim Dounin.
1216 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1217 Patch provided by Phil Pennock.
1219 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1221 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1222 Patch provided by Brad "anomie" Jorsch.
1224 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1225 Patch provided by Dean Brooks.
1227 TK/05 Add native DKIM support (does not depend on external libraries).
1229 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1230 Patch provided by Graeme Fowler.
1232 NM/16 Bugzilla 851: Documentation example syntax fix.
1234 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1236 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1239 NM/19 Bugzilla 745: TLS version reporting.
1240 Patch provided by Phil Pennock.
1242 NM/20 Bugzilla 167: bool: condition support.
1243 Patch provided by Phil Pennock.
1245 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1246 clients. Patch provided by Phil Pennock.
1248 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1249 Patch provided by Brad "anomie" Jorsch.
1251 NM/23 Bugzilla 687: Fix misparses in eximstats.
1252 Patch provided by Heiko Schlittermann.
1254 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1255 Patch provided by Heiko Schlittermann.
1257 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1258 plus update to original patch.
1260 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1262 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1263 Patch provided by David Brownlee.
1265 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1267 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1269 NM/30 Bugzilla 888: TLS documentation bugfixes.
1271 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1273 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1274 Unlike the original bugzilla I have changed all shell scripts in src tree.
1276 NM/33 Bugzilla 898: Transport filter timeout fix.
1277 Patch by Todd Rinaldo.
1279 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1280 Patch by Serge Demonchaux.
1282 NM/35 Bugzilla 39: Base64 decode bug fixes.
1283 Patch by Jakob Hirsch.
1285 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1287 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1289 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1291 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1297 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1298 ALT-N's libdkim that I have put here:
1299 http://duncanthrax.net/exim-experimental/
1301 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1302 were called 'true' and it seems that C99 defines that as a reserved
1303 keyword to be used with 'bool' variable types. That means you could
1304 not include C99-style headers which use bools without triggering
1305 build errors in sieve.c.
1307 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1308 as mailq or other aliases. Changed the --help handling significantly
1309 to do whats expected. exim_usage() emits usage/help information.
1311 SC/01 Added the -bylocaldomain option to eximstats.
1313 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1315 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1317 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1323 PH/01 Another patch from the Sieve maintainer.
1325 PH/02 When an IPv6 address is converted to a string for single-key lookup
1326 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1327 dots are used instead of colons so that keys in lsearch files need not
1328 contain colons. This was done some time before quoting was made available
1329 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1330 (notated using the quote facility) so as to distinguish them from IPv4
1331 keys. This meant that lookups for IP addresses in host lists did not work
1332 for iplsearch lookups.
1334 This has been fixed by arranging for IPv6 addresses to be expressed with
1335 colons if the lookup type is iplsearch. This is not incompatible, because
1336 previously such lookups could never work.
1338 The situation is now rather anomolous, since one *can* have colons in
1339 ordinary lsearch keys. However, making the change in all cases is
1340 incompatible and would probably break a number of configurations.
1342 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1345 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1346 conversion specification without a maximum field width, thereby enabling
1347 a rogue spamd server to cause a buffer overflow. While nobody in their
1348 right mind would setup Exim to query an untrusted spamd server, an
1349 attacker that gains access to a server running spamd could potentially
1350 exploit this vulnerability to run arbitrary code as the Exim user.
1352 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1353 $primary_hostname instead of what libspf2 thinks the hosts name is.
1355 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1356 a directory entry by the name of the lookup key. Previously, if a
1357 symlink pointed to a non-existing file or a file in a directory that
1358 Exim lacked permissions to read, a lookup for a key matching that
1359 symlink would fail. Now it is enough that a matching directory entry
1360 exists, symlink or not. (Bugzilla 503.)
1362 PH/03 The body_linecount and body_zerocount variables are now exported in the
1365 PH/04 Added the $dnslist_matched variable.
1367 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1368 This means they are set thereafter only if the connection becomes
1371 PH/06 Added the client_condition to authenticators so that some can be skipped
1372 by clients under certain conditions.
1374 PH/07 The error message for a badly-placed control=no_multiline_responses left
1375 "_responses" off the end of the name.
1377 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1379 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1380 (without spaces) instead of just copying the configuration text.
1382 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1384 PH/11 Added $max_received_linelength.
1386 PH/12 Added +ignore_defer and +include_defer to host lists.
1388 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1389 way in which PCRE > 7.0 is built.
1391 PH/14 Implemented queue_only_load_latch.
1393 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1394 MAIL command. The effect was to mangle the value on 64-bit systems.
1396 PH/16 Another patch from the Sieve maintainer.
1398 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1400 PH/18 If a system quota error occurred while trying to create the file for
1401 a maildir delivery, the message "Mailbox is full" was not appended to the
1402 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1403 only to a quota excession during the actual writing of the file.
1405 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1406 characters?) which causes problems in log lines. The DN values are now
1407 passed through string_printing() before being added to log lines.
1409 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1410 and InterBase are left for another time.)
1412 PH/21 Added message_body_newlines option.
1414 PH/22 Guard against possible overflow in moan_check_errorcopy().
1416 PH/23 POSIX allows open() to be a macro; guard against that.
1418 PH/24 If the recipient of an error message contained an @ in the local part
1419 (suitably quoted, of course), incorrect values were put in $domain and
1420 $local_part during the evaluation of errors_copy.
1426 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1427 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1430 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1431 issue a MAIL command.
1433 PH/02 In an ACL statement such as
1435 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1437 if a client was not listed at all, or was listed with a value other than
1438 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1439 the condition was not true (as it should be), so access was not denied.
1440 The bug was that the ! inversion was incorrectly passed on to the second
1441 item. This has been fixed.
1443 PH/03 Added additional dnslists conditions == and =& which are different from
1444 = and & when the dns lookup returns more than one IP address.
1446 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1447 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1449 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1450 FSYNC, which compiles an option called disable_fsync that allows for
1451 bypassing fsync(). The documentation is heavily laced with warnings.
1453 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1455 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1456 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1457 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1458 including adding "make clean"; (3) Added -fPIC when compiling the test
1459 dynamically loaded module, to get rid of a warning.
1461 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1462 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1463 = 0s. The bug is otherwise harmless.
1465 PH/07 There was a bug in the dovecot authenticator such that the value of
1466 $auth1 could be overwritten, and so not correctly preserved, after a
1467 successful authentication. This usually meant that the value preserved by
1468 the server_setid option was incorrect.
1470 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1472 PH/09 Installed PCRE release 7.0.
1474 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1475 run for batched SMTP input. It is now run at the start of every message
1476 in the batch. While fixing this I discovered that the process information
1477 (output by running exiwhat) was not always getting set for -bs and -bS
1478 input. This is fixed, and it now also says "batched" for BSMTP.
1480 PH/11 Added control=no_pipelining.
1482 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1483 patch, slightly modified), and move the expansion of helo_data till after
1484 the connection is made in the smtp transport (so it can use these
1487 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1489 PH/14 Added log_selector = +pid.
1491 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1493 PH/16 Add ${if forany and ${if forall.
1495 PH/17 Added dsn_from option to vary the From: line in DSNs.
1497 PH/18 Flush SMTP output before performing a callout, unless control =
1498 no_callout_flush is set.
1500 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1501 was true (the default) a successful delivery failed to delete the retry
1502 item, thus causing premature timeout of the address. The bug is now
1505 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1507 PH/21 Long custom messages for fakedefer and fakereject are now split up
1508 into multiline reponses in the same way that messages for "deny" and
1509 other ACL rejections are.
1511 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1512 with slight modification.
1514 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1515 draft, changing the syntax and factoring some duplicate code".
1517 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1518 for deliveries of the second and subsequent messages over the same SMTP
1521 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1522 ${reduce, with only minor "tidies".
1524 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1526 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1527 expansion side effects.
1529 PH/27 When a message times out after an over-quota error from an Exim-imposed
1530 quota, the bounce message says "mailbox is full". This message was not
1531 being given when it was a system quota that was exceeded. It now should
1534 MH/03 Made $recipients available in local_scan(). local_scan() already has
1535 better access to the recipient list through recipients_list[], but
1536 $recipients can be useful in postmaster-provided expansion strings.
1538 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1539 in the case of a MAIL command with additional options following the
1540 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1541 were accidentally chopped off.
1543 PH/29 SMTP synchronization checks are implemented when a command is read -
1544 there is a check that no more input is waiting when there shouldn't be
1545 any. However, for some commands, a delay in an ACL can mean that it is
1546 some time before the response is written. In this time, more input might
1547 arrive, invalidly. So now there are extra checks after an ACL has run for
1548 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1549 pipelining has not been advertised.
1551 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1553 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1554 returned in the SMTP response when smtp_return_error_details was false.
1555 This has been fixed.
1557 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1558 instead of the C I/O that was originally supplied, because problems were
1559 reported on Solaris.
1561 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1562 Exim which did not show up earlier: it was assuming that a call to
1563 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1564 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1565 was a macro that became an assignment, so it seemed to work. This has
1566 changed to a proper function call with a void return, hence the compile
1567 error. Exim's code has been fixed.
1569 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1572 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1574 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1576 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1577 because it turns out to be too big or there is a timeout) there is no
1578 "Completed" line in the log. When some messages of this type were
1579 selected by exigrep, they were listed as "not completed". Others were
1580 picked up by some special patterns. I have improved the selection
1581 criteria to be more general.
1583 PH/38 The host_find_failed option in the manualroute router can now be set
1584 to "ignore", to completely ignore a host whose IP address cannot be
1585 found. If all hosts are ignored, the behaviour is controlled by the new
1586 host_all_ignored option.
1588 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1589 homing or because of multiple MX records with /mx) generated more than
1590 one IP address, and the following item turned out to be the local host,
1591 all the secondary addresses of the first item were incorrectly removed
1592 from the list, along with the local host and any following hosts (which
1593 is what is supposed to happen).
1595 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1596 whoever called Exim into the -H file. In the case of the daemon it was
1597 behaving confusingly. When first started, it used values for whoever
1598 started the daemon, but after a SIGHUP it used the Exim user (because it
1599 calls itself on a restart). I have changed the code so that it now always
1602 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1603 message are rejected with the same error (e.g. no authentication or bad
1604 sender address), and a DATA command is nevertheless sent (as can happen
1605 with PIPELINING or a stupid MUA), the error message that was given to the
1606 RCPT commands is included in the rejection of the DATA command. This is
1607 intended to be helpful for MUAs that show only the final error to their
1610 PH/42 Another patch from the Sieve maintainer.
1612 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1613 Eximstats - Fixed some broken HTML links and added missing column headers
1615 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1616 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1618 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1620 PH/43 Yet another patch from the Sieve maintainer.
1622 PH/44 I found a way to check for a TCP/IP connection going away before sending
1623 the response to the final '.' that terminates a message, but only in the
1624 case where the client has not sent further data following the '.'
1625 (unfortunately, this is allowed). However, in many cases there won't be
1626 any further data because there won't be any more messages to send. A call
1627 to select() can be used: if it shows that the input is "ready", there is
1628 either input waiting, or the socket has been closed. An attempt to read
1629 the next input character can distinguish the two cases. Previously, Exim
1630 would have sent an OK response which the client would never have see.
1631 This could lead to message repetition. This fix should cure that, at
1632 least in a lot of common cases.
1634 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1635 advertised in response to EHLO.
1641 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1642 fixed by 4.65/MH/01 (is this a record?) are fixed:
1644 (i) An empty string was always treated as zero by the numeric comparison
1645 operators. This behaviour has been restored.
1647 (ii) It is documented that the numeric comparison operators always treat
1648 their arguments as decimal numbers. This was broken in that numbers
1649 starting with 0 were being interpreted as octal.
1651 While fixing these problems I realized that there was another issue that
1652 hadn't been noticed. Values of message_size_limit (both the global option
1653 and the transport option) were treated as octal if they started with 0.
1654 The documentation was vague. These values are now always treated as
1655 decimal, and I will make that clear in the documentation.
1661 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1662 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1665 MH/01 Don't check that the operands of numeric comparison operators are
1666 integers when their expansion is in "skipping" mode (fixes bug
1667 introduced by 4.64-PH/07).
1669 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1670 child addresses, Exim now panics and dies. Previously, because the count
1671 is held in a short int, deliveries were likely to be lost. As such a
1672 large number of recipients for a single message is ridiculous
1673 (performance will be very, very poor), I have chosen to impose a limit
1674 rather than extend the field.
1680 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1681 leftover -K file (the existence of which was triggered by #402).
1682 While we were at it, introduced process PID as part of the -K
1683 filename. This should rule out race conditions when creating
1686 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1687 processing considerably. Previous code took too long for large mails,
1688 triggering a timeout which in turn triggers #401.
1690 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1691 in the DK code in transports.c. sendfile() is not really portable,
1692 hence the _LINUX specificness.
1694 TF/01 In the add_headers option to the mail command in an Exim filter,
1695 there was a bug that Exim would claim a syntax error in any
1696 header after the first one which had an odd number of characters
1699 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1700 callout verification, Exim cached a "reject" for the entire domain. This
1701 is correct for most verifications, but it is not correct for a recipient
1702 verification with use_sender or use_postmaster set, because in that case
1703 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1704 case of MAIL FROM:<> rejection from other early rejections (e.g.
1705 rejection of HELO). When verifying a recipient using a non-null MAIL
1706 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1707 Whatever the result of the callout, the value of the domain cache is
1708 left unchanged (for any other kind of callout, getting as far as trying
1709 RCPT means that the domain itself is ok).
1711 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1714 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1715 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1716 session, the daemon ignores ECONNECT errors and logs others; it now
1717 ignores EPIPE as well.
1719 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1720 (quoted-printable decoding).
1722 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1723 later the small subsequent patch to fix an introduced bug.
1725 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1727 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1729 PH/08 An error is now given if message_size_limit is specified negative.
1731 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1732 to be given (somewhat) arbitrary names.
1734 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1737 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1738 miscellaneous code fixes
1740 PH/10 Added the log_reject_target ACL modifier to specify where to log
1743 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1744 hostname. This is wrong, because it relates to the incoming message (and
1745 probably the interface on which it is arriving) and not to the outgoing
1746 callout (which could be using a different interface). This has been
1747 changed to use the value of the helo_data option from the smtp transport
1748 instead - this is what is used when a message is actually being sent. If
1749 there is no remote transport (possible with a router that sets up host
1750 addresses), $smtp_active_hostname is used.
1752 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1753 tweaks were necessary in order to get it to work (see also 21 below):
1754 (a) The code assumed that strncpy() returns a negative number on buffer
1755 overflow, which isn't the case. Replaced with Exim's string_format()
1757 (b) There were several signed/unsigned issues. I just did the minimum
1758 hacking in of casts. There is scope for a larger refactoring.
1759 (c) The code used strcasecmp() which is not a standard C function.
1760 Replaced with Exim's strcmpic() function.
1761 (d) The code set only $1; it now sets $auth1 as well.
1762 (e) A simple test gave the error "authentication client didn't specify
1763 service in request". It would seem that Dovecot has changed its
1764 interface. Fortunately there's a specification; I followed it and
1765 changed what the client sends and it appears to be working now.
1767 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1770 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1771 address A is aliased to B and C, where B exists and C does not. Without
1772 -v the output is "A verified" because verification stops after a
1773 successful redirection if more than one address is generated. However,
1774 with -v the child addresses are also verified. Exim was outputting "A
1775 failed to verify" and then showing the successful verification for C,
1776 with its parentage. It now outputs "B failed to verify", showing B's
1777 parentage before showing the successful verification of C.
1779 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1780 look up a TXT record in a specific list after matching in a combined
1783 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1784 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1785 they consult the DNS. I had assumed they would set it the way they
1786 wanted; and indeed my experiments on Linux seem to show that in some
1787 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1788 To be on the safe side, however, I have now made the interface to
1789 host_find_byname() similar to host_find_bydns(), with an argument
1790 containing the DNS resolver options. The host_find_byname() function now
1791 sets these options at its start, just as host_find_bydns() does. The smtp
1792 transport options dns_qualify_single and dns_search_parents are passed to
1793 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
1794 of host_find_byname() use the default settings of RES_DEFNAMES
1795 (qualify_single) but not RES_DNSRCH (search_parents).
1797 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
1798 spool_read_header() do less string testing, by means of a preliminary
1799 switch on the second character of optional "-foo" lines. (This is
1800 overdue, caused by the large number of possibilities that now exist.
1801 Originally there were few.) While I was there, I also converted the
1802 str(n)cmp tests so they don't re-test the leading "-" and the first
1803 character, in the hope this might squeeze out yet more improvement.
1805 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
1806 flag allowing group syntax was set by the header_syntax check but not
1807 turned off, possible causing trouble later; (2) The flag was not being
1808 set at all for the header_verify test, causing "group"-style headers to
1809 be rejected. I have now set it in this case, and also caused header_
1810 verify to ignore an empty address taken from a group. While doing this, I
1811 came across some other cases where the code for allowing group syntax
1812 while scanning a header line wasn't quite right (mostly, not resetting
1813 the flag correctly in the right place). These bugs could have caused
1814 trouble for malformed header lines. I hope it is now all correct.
1816 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
1817 with the "reply" argument non-NULL. The code, however (which originally
1818 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
1819 but it didn't always do it. This confused somebody who was copying the
1820 code for some other use. I have removed all the tests.
1822 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
1823 feature that was used to support insecure browsers during the U.S. crypto
1824 embargo. It requires special client support, and Exim is probably the
1825 only MTA that supported it -- and would never use it because real RSA is
1826 always available. This code has been removed, because it had the bad
1827 effect of slowing Exim down by computing (never used) parameters for the
1828 RSA_EXPORT functionality.
1830 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
1831 authenticator to fail if there's a tab character in the incoming data
1832 (there should never be unless someone is messing about, as it's supposed
1833 to be base64-encoded). Also added, on Timo's advice, the "secured" option
1834 if the connection is using TLS or if the remote IP is the same as the
1835 local IP, and the "valid-client-cert option" if a client certificate has
1838 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
1839 authenticators. This can be used for authorization after authentication
1840 succeeds. (In the case of plaintext, it servers for both authentication
1843 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1844 if any retry times were supplied.
1846 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1847 connection, where there is no client IP address to check. In this
1848 situation, the verify now always succeeds.
1850 PH/25 Applied John Jetmore's -Mset patch.
1852 PH/26 Added -bem to be like -Mset, but loading a message from a file.
1854 PH/27 In a string expansion for a processed (not raw) header when multiple
1855 headers of the same name were present, leading whitespace was being
1856 removed from all of them, but trailing whitespace was being removed only
1857 from the last one. Now trailing whitespace is removed from each header
1858 before concatenation. Completely empty headers in a concatenation (as
1859 before) are ignored.
1861 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1862 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1864 PH/29 [Removed. This was a change that I later backed out, and forgot to
1865 correct the ChangeLog entry (that I had efficiently created) before
1866 committing the later change.]
1868 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1869 address errors (4xx response to RCPT) over the same connection as other
1870 messages routed to the same hosts. Such deliveries are always "forced",
1871 so retry times are not inspected. This resulted in far too many retries
1872 for the affected addresses. The effect occurred only when there were more
1873 hosts than the hosts_max_try setting in the smtp transport when it had
1874 the 4xx errors. Those hosts that it had tried were not added to the list
1875 of hosts for which the message was waiting, so if all were tried, there
1876 was no problem. Two fixes have been applied:
1878 (i) If there are any address or message errors in an SMTP delivery, none
1879 of the hosts (tried or untried) are now added to the list of hosts
1880 for which the message is waiting, so the message should not be a
1881 candidate for sending over the same connection that was used for a
1882 successful delivery of some other message. This seems entirely
1883 reasonable: after all the message is NOT "waiting for some host".
1884 This is so "obvious" that I'm not sure why it wasn't done
1885 previously. Hope I haven't missed anything, but it can't do any
1886 harm, as the worst effect is to miss an optimization.
1888 (ii) If, despite (i), such a delivery is accidentally attempted, the
1889 routing retry time is respected, so at least it doesn't keep
1890 hammering the server.
1892 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1893 in ${readsocket because some servers need this prod.
1895 PH/32 Added some extra debug output when updating a wait-xxx database.
1897 PH/33 The hint "could be header name not terminated by colon", which has been
1898 given for certain expansion errors for a long time, was not being given
1899 for the ${if def:h_colon_omitted{... case.
1901 PH/34 The spec says: "With one important exception, whenever a domain list is
1902 being scanned, $domain contains the subject domain." There was at least
1903 one case where this was not true.
1905 PH/35 The error "getsockname() failed: connection reset by peer" was being
1906 written to the panic log as well as the main log, but it isn't really
1907 panic-worthy as it just means the connection died rather early on. I have
1908 removed the panic log writing for the ECONNRESET error when getsockname()
1911 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1912 runs only) independently of the message's sender address. This meant
1913 that, if the 4xx error was in fact related to the sender, a different
1914 message to the same recipient with a different sender could confuse
1915 things. In particualar, this can happen when sending to a greylisting
1916 server, but other circumstances could also provoke similar problems.
1917 I have changed the default so that the retry time for these errors is now
1918 based a combination of the sender and recipient addresses. This change
1919 can be overridden by setting address_retry_include_sender=false in the
1922 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1923 remote server are returned as part of bounce messages. This was not
1924 happening for LMTP over a pipe (the lmtp transport), but now it is the
1925 same for both kinds of LMTP.
1927 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1928 in header lines that were in fact CNAMEs. This is no longer the case.
1930 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1931 and queue runs started by the daemon processed all messages. This has
1932 been fixed so that -R and -S can now usefully be given with -q<time>.
1934 PH/40 Import PCRE release 6.7 (fixes some bugs).
1936 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1938 PH/42 Give an error if -q is specified more than once.
1940 PH/43 Renamed the variables $interface_address and $interface_port as
1941 $received_ip_address and $received_port, to make it clear that these
1942 values apply to message reception, and not to the outgoing interface when
1943 a message is delivered. (The old names remain recognized, of course.)
1945 PH/44 There was no timeout on the connect() call when using a Unix domain
1946 socket in the ${readsocket expansion. There now is.
1948 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1949 be meaningful with "accept".
1951 SC/01 Eximstats V1.43
1952 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1954 SC/02 Eximstats V1.44
1955 Use a glob alias rather than an array ref in the generated
1956 parser. This improves both readability and performance.
1958 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1959 Collect SpamAssassin and rejection statistics.
1960 Don't display local sender or destination tables unless
1961 there is data to show.
1962 Added average volumes into the top table text output.
1964 SC/04 Eximstats V1.46
1965 Collect data on the number of addresses (recipients)
1966 as well as the number of messages.
1968 SC/05 Eximstats V1.47
1969 Added 'Message too big' to the list of mail rejection
1970 reasons (thanks to Marco Gaiarin).
1972 SC/06 Eximstats V1.48
1973 Mainlog lines which have GMT offsets and are too short to
1974 have a flag are now skipped.
1976 SC/07 Eximstats V1.49 (Alain Williams)
1977 Added the -emptyok flag.
1979 SC/08 Eximstats V1.50
1980 Fixes for obtaining the IP address from reject messages.
1982 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1983 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1984 whitesspace changes from 4.64-PH/27
1986 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1989 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1990 are found, allow negative numbers in numeric criteria)
1992 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1994 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1995 to match changes made in 4.64-PH/43
1997 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1999 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2000 "domain" is actually an IP address, and force a failure. This locks out
2001 those revolvers/nameservers that support "A-for-A" lookups, in
2002 contravention of the specifications.
2004 PH/48 When a host name was looked up from an IP address, and the subsequent
2005 forward lookup of the name timed out, the host name was left in
2006 $sender_host_name, contrary to the specification.
2008 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2009 restricted to single-key lookups, Exim was not diagnosing an error if
2010 * or *@ was used with a query-style lookup.
2012 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2014 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2015 long ago, but noone interested enough thought of it. Let's just say that
2016 the "1.1" means that there are some new functions that weren't there at
2017 some point in the past.
2019 PH/51 Error processing for expansion failure of helo_data from an smtp
2020 transport during callout processing was broken.
2022 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2023 tested/used via the -bh/-bhc/-bs options.
2025 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2026 bug, fixed in subsequent PCRE releases).
2028 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2029 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2031 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2037 SC/01 Use a glob alias rather than an array ref in eximstats generated
2038 parser. This improves both readability and performance.
2040 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2041 Don't display local sender or destination tables in eximstats unless
2042 there is data to show.
2043 Added average volumes into the eximstats top table text output.
2045 SC/03 Collect data on the number of addresses (recipients) as well
2046 as the number of messages in eximstats.
2048 TF/01 Correct an error in the documentation for the redirect router. Exim
2049 does (usually) call initgroups() when daemonizing.
2051 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2052 with consistent privilege compared to when running as a daemon.
2054 TF/03 Note in the spec that $authenticated_id is not set for local
2055 submissions from trusted users.
2057 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2058 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2060 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2061 by adding some example configuration directives to the default
2062 configuration file. A little bit of work is required to uncomment the
2063 directives and define how usernames and passwords are checked, but
2064 there is now a framework to start from.
2066 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2067 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2068 without this. I don't know how relevant this is to other LDAP libraries.
2070 PH/02 Add the verb name to the "unknown ACL verb" error.
2072 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2074 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2076 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2077 directory not expanded when it should be if an expanded home directory
2078 was set for the address (which is overridden by the transport).
2080 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2083 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2084 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2085 because it is too late at that time, and has no effect.
2087 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2088 security issue with \' (bugzilla #107). I could not use the
2089 PQescapeStringConn() function, because it needs a PGconn value as one of
2092 PH/08 When testing addresses using -bt, indicate those final addresses that
2093 are duplicates that would not cause an additional delivery. At least one
2094 person was confused, thinking that -bt output corresponded to deliveries.
2095 (Suppressing duplicates isn't a good idea as you lose the information
2096 about possibly different redirections that led to the duplicates.)
2098 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2099 systems where poll() doesn't work, in particular OS X.
2101 PH/10 Added more information to debugging output for retry time not reached.
2103 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2104 operations in malware.c.
2106 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2109 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2110 syslog_duplication set false, log lines that would normally be written
2111 both the the main log and to the reject log were not written to syslog at
2114 PH/14 In the default configuration, change the use of "message" in ACL warn
2115 statements to "add_header".
2117 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2118 not followed by a command (e.g. "seen endif").
2120 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2121 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2124 PH/17 Added extra conditions to the default value of delay_warning_condition
2128 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2129 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2130 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2133 The Auto-Submitted: and various List- headers are standardised, whereas I
2134 don't think Precedence: ever was.
2136 PH/18 Refactored debugging code in route_finduser() to show more information,
2137 in particular, the error code if getpwnam() issues one.
2139 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2140 This is apparently needed in addition to the PH/07 change above to avoid
2141 any possible encoding problems.
2143 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2144 but not after initializing Perl.
2146 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2147 output them only if debugging. By default they are written stderr,
2148 apparently, which is not desirable.
2150 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2153 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2156 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2158 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2159 authenticated or an ident call has been made. Suppress the default
2160 values for $authenticated_id and $authenticated_sender (but permit -oMai
2161 and -oMas) when testing with -bh.
2163 PH/24 Re-jigged the order of the tests in the default configuration so that the
2164 tests for valid domains and recipients precede the DNS black list and CSA
2165 tests, on the grounds that those ones are more expensive.
2167 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2168 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2169 This bug exists in every version of Exim that I still have, right back to
2172 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2173 However, an attempt to turn on case-sensitivity in a regex key by
2174 including (?-i) didn't work because the subject string was already
2175 lowercased, and the effects were non-intuitive. It turns out that a
2176 one-line patch can be used to allow (?-i) to work as expected.
2182 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2183 other effects) broke the use of negated acl sub-conditions.
2185 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2188 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2189 "Deny" causes Exim to reject the incoming connection with a 554 error.
2190 Unfortunately, if there is a major crisis, such as a disk failure,
2191 tcp-wrappers gives "deny", whereas what one would like would be some
2192 kind of temporary error. A kludge has been added to help with this.
2193 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2194 554 error is used if errno is still zero or contains ENOENT (which occurs
2195 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2198 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2200 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2201 errors. Otherwise a message that provokes a temporary error (when other
2202 messages do not) can cause a whole host to time out.
2204 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2205 addresses were routed directly to files or pipes from a redirect router.
2206 File deliveries just didn't batch; pipe deliveries might have suffered
2209 PH/06 A failure to get a lock for a hints database would erroneously always say
2210 "Failed to get write lock", even when it was really a read lock.
2212 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2213 of 0600. This has been changed to use the value of the lockfile_mode
2214 option (which defaults to 0600).
2216 PH/08 Applied small patch from the Sieve maintainer.
2218 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2219 folder from quota calculations, a direct delivery into this folder messed
2220 up the contents of the maildirsize file. This was because the regex was
2221 used only to exclude .Trash (or whatever) when the size of the mailbox
2222 was calculated. There was no check that a delivery was happening into an
2223 excluded directory. This bug has been fixed by ignoring all quota
2224 processing for deliveries into excluded directories.
2226 PH/10 Added the maildirfolder_create_regex option to appendfile.
2232 PH/01 The code for finding all the local interface addresses on a FreeBSD
2233 system running IPv6 was broken. This may well have applied to all BSD
2234 systems, as well as to others that have similar system calls. The broken
2235 code found IPv4 interfaces correctly, but gave incorrect values for the
2236 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2237 that it would not match correctly against @[] and not recognize the IPv6
2240 PH/02 The ipliteral router was not recognizing addresses of the form user@
2241 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2243 PH/03 Added disable_ipv6.
2245 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2246 decoded form, because it is most often used to construct To: headers
2247 lines in autoreplies, and the decoded form may well be syntactically
2248 invalid. However, $reply_address has leading white space removed, and all
2249 newlines turned into spaces so that the autoreply transport does not
2252 PH/05 If group was specified without a user on a router, and no group or user
2253 was specified on a transport, the group from the router was ignored.
2255 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2256 for visible compile-time settings that can be used to change these
2257 numbers, for those that want even more. Backwards compatibility with old
2258 spool files has been maintained. However, going back to a previous Exim
2259 release will lost any variables that are in spool files.
2261 PH/07 Two small changes when running in the test harness: increase delay when
2262 passing a TCP/IP connection to a new process, in case the original
2263 process has to generate a bounce, and remove special handling of
2264 127.0.0.2 (sic), which is no longer necessary.
2266 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2267 be the same on different OS.
2269 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2272 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2273 whether --show-vars was specified or not
2275 JJ/02 exipick: Added support for new ACL variable spool format introduced
2278 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2279 syntactically invalid From: or Reply-to: line, and a filter used this to
2280 generate an autoreply, and therefore failed to obtain an address for the
2281 autoreply, Exim could try to deliver to a non-existent relative file
2282 name, causing unrelated and misleading errors. What now happens is that
2283 it logs this as a hard delivery error, but does not attempt to create a
2286 PH/11 The exinext utility has a -C option for testing purposes, but although
2287 the given file was scanned by exinext itself; it wasn't being passed on
2288 when Exim was called.
2290 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2291 an end-of-file indication when reading a command response.
2293 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2294 compiled. In many other places in Exim, IPv6 addresses are always
2295 recognized, so I have changed this. It also means that IPv4 domain
2296 literals of the form [IPV4:n.n.n.n] are now always recognized.
2298 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2299 used if the router is not running as root, for example, when verifying at
2300 ACL time, or when using -bh. The debugging output from this situation was
2301 non-existent - all you got was a failure to exec. I have made two
2304 (a) Failures to set uid/gid, the current directory, or a process leader
2305 in a subprocess such as that created by queryprogram now generate
2306 suitable debugging ouput when -d is set.
2308 (b) The queryprogram router detects when it is not running as root,
2309 outputs suitable debugging information if -d is set, and then runs
2310 the subprocess without attempting to change uid/gid.
2312 PH/15 Minor change to Makefile for building test_host (undocumented testing
2315 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2316 additional section of a DNS packet that returns MX or SRV records.
2317 Instead, it always explicitly searches for A/AAAA records. This avoids
2318 major problems that occur when a DNS server includes only records of one
2319 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2320 fixed another bug: if SRV records were looked up and the corresponding
2321 address records were *not* found in the additional section, the port
2322 values from the SRV records were lost.
2324 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2325 using the correct key (the original address) when searching the retry
2326 rules in order to find which one to use for generating the retry hint.
2328 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2329 adding the default one. Similarly, if it contains a Reply-To: header, the
2330 errors_reply_to option, if set, is not used.
2332 PH/19 When calculating a retry time, Exim used to measure the "time since
2333 failure" by looking at the "first failed" field in the retry record. Now
2334 it does not use this if it is later than than the arrival time of the
2335 message. Instead it uses the arrival time. This makes for better
2336 behaviour in cases where some deliveries succeed, thus re-setting the
2337 "first failed" field. An example is a quota failure for a huge message
2338 when small messages continue to be delivered. Without this change, the
2339 "time since failure" will always be short, possible causing more frequent
2340 delivery attempts for the huge message than are intended.
2341 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2343 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2344 $1, $2, $3) because the numerical variables can be reset during some
2345 expansion items (e.g. "match"), thereby losing the authentication data.
2347 PH/21 Make -bV show the size of off_t variables so that the test suite can
2348 decide whether to run tests for quotas > 2G.
2350 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2351 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2352 filecount value is greater than 2G or if a quota value is greater than 2G
2353 on a system where the size of off_t is not greater than 4, a panic error
2356 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2357 never match. The debug and -bh output now contains an explicit error
2358 message indicating a malformed IPv4 address or mask.
2360 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2361 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2362 PH/23 above applies.
2364 PH/25 Do not write to syslog when running in the test harness. The only
2365 occasion when this arises is a failure to open the main or panic logs
2366 (for which there is an explicit test).
2368 PH/26 Added the /no_tell option to "control=freeze".
2370 PH/27 If a host name lookup failed very early in a connection, for example, if
2371 the IP address matched host_lookup and the reverse lookup yielded a name
2372 that did not have a forward lookup, an error message of the form "no IP
2373 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2374 could be logged. Now it outputs the IP address instead of "NULL".
2376 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2377 allows the sender and the authenticated sender to be set when
2378 submitting a message from within Exim. Since child_open_exim() is
2379 documented for local_scan(), the new function should be too.
2381 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2382 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2383 results in an empty string is now treated as unset.
2385 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2387 PH/31 Added sender_verify_fail logging option.
2389 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2390 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2391 tidied the source and removed it altogether.
2393 PH/33 When a queue run was abandoned because the load average was too high, a
2394 log line was always written; now it is written only if the queue_run log
2395 selector is set. In addition, the log line for abandonment now contains
2396 information about the queue run such as the pid. This is always present
2397 in "start" and "stop" lines but was omitted from the "abandon" line.
2399 PH/34 Omit spaces between a header name and the colon in the error message that
2400 is given when verify = headers_syntax fails (if there are lots of them,
2401 the message gets confusing).
2403 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2404 names, as there are now some PTR records that contain slashes. This check
2405 is only to protect against broken name servers that fall over on strange
2406 characters, so the fact that it applies to all lookups doesn't matter.
2408 PH/36 Now that the new test suite is complete, we can remove some of the
2409 special code in Exim that was needed for the old test suite. For example,
2410 sorting DNS records because real resolvers return them in an arbitrary
2411 order. The new test suite's fake resolver always returns records in the
2414 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2415 bounces) except when queue_only is set, to avoid logging races between
2416 the different processes.
2418 PH/38 Panic-die if .include specifies a non-absolute path.
2420 PH/39 A tweak to the "H" retry rule from its user.
2422 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2423 a label. They prevented compilation on older perls.
2425 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2426 a warning to be raised on newish perls.
2428 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2429 on queue. Changes to match documented behaviour of showing count of
2430 messages matching specified criteria.
2432 PH/40 Changed the default ident timeout from 30s to 5s.
2434 PH/41 Added support for the use of login_cap features, on those BSD systems
2435 that have them, for controlling the resources used by pipe deliveries.
2437 PH/42 The content-scanning code uses fopen() to create files in which to put
2438 message data. Previously it was not paying any attention to the mode of
2439 the files. Exim runs with umask(0) because the rest of the code creates
2440 files with open(), and sets the required mode explicitly. Thus, these
2441 files were ending up world-writeable. This was not a big issue, because,
2442 being within the spool directory, they were not world-accessible. I have
2443 created a function called modefopen, which takes an additional mode
2444 argument. It sets umask(777), creates the file, chmods it to the required
2445 mode, then resets the umask. All the relevant calls to fopen() in the
2446 content scanning code have been changed to use this function.
2448 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2449 to 24 hours. This avoids potential overflow problems when processing G
2450 and H retry rules. I suspect nobody ever tinkers with this value.
2452 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2454 PH/45 When the plaintext authenticator is running as a client, the server's
2455 challenges are checked to ensure they are valid base64 strings. By
2456 default, the authentication attempt is cancelled if an invalid string is
2457 received. Setting client_ignore_invalid_base64 true ignores these errors.
2458 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2459 they are received. Thus, the responses can be made to depend on the
2460 challenges. If an invalid string is ignored, an empty string is placed in
2463 PH/46 Messages that are created by the autoreply transport now contains a
2464 References: header, in accordance with RFCs 2822 and 3834.
2466 PH/47 Added authenticated_sender_force to the smtp transport.
2468 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2470 PH/49 Installed latest patch from the Sieve maintainer.
2472 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2473 was also set, the appendfile transport was unnecessarily scanning a
2474 directory of message files (e.g. for maildir delivery) to find the count
2475 of files (along with the size), even though it did not need this
2476 information. It now does the scan only if it needs to find either the
2477 size of the count of files.
2479 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2481 PH/52 Two bugs concerned with error handling when the smtp transport is
2484 (i) Exim was not creating retry information for temporary errors given
2485 for individual recipients after the DATA command when the smtp transport
2486 was used in LMTP mode. This meant that they could be retried too
2487 frequently, and not timed out correctly.
2489 (ii) Exim was setting the flag that allows error details to be returned
2490 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2491 recipients that were returned after the DATA command.
2493 PH/53 This is related to PH/52, but is more general: for any failing address,
2494 when detailed error information was permitted to be returned to the
2495 sender, but the error was temporary, then after the final timeout, only
2496 "retry timeout exceeded" was returned. Now it returns the full error as
2497 well as "retry timeout exceeded".
2499 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2500 do this, and (what is worse) MTAs that accept it.
2502 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2503 will now be deprecated.
2505 PH/56 New os.c-cygwin from the Cygwin maintainer.
2507 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2508 formats (previously only available in exim formats via -bpr, -bpru,
2509 and -bpra. Now also available in native and exiqgrep formats)
2511 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2512 with very large, slow to parse queues
2514 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2516 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2518 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2519 responses to authentication challenges, though it was showing the
2520 challenges; (ii) I've removed the CR characters from the debug output for
2523 PH/58 Allow for the insertion of a newline as well as a space when a string
2524 is turned into more than one encoded-word during RFC 2047 encoding. The
2525 Sieve code now uses this.
2527 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2528 data_4xx, lost_connection, tls_required.
2530 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2531 message was being sent as an SMTP response.
2533 PH/61 Add -l and -k options to exicyclog.
2535 PH/62 When verifying, if an address was redirected to one new address, so that
2536 verification continued, and the new address failed or deferred after
2537 having set something in $address_data, the value of $address_data was not
2538 passed back to the ACL. This was different to the case when no
2539 redirection occurred. The value is now passed back in both cases.
2541 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2542 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2543 use login_cap.h, so on its own it isn't the distinguishing feature. The
2544 new name refers directly to the setclassresources() function.
2546 PH/65 Added configuration files for NetBSD3.
2548 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2550 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2551 is preferred over IPv4.
2553 PH/68 The bounce_return_message and bounce_return_body options were not being
2554 honoured for bounces generated during the reception of non-SMTP messages.
2555 In particular, this applied to messages rejected by the ACL. This bug has
2556 been fixed. However, if bounce_return_message is true and bounce_return_
2557 body is false, the headers that are returned for a non-SMTP message
2558 include only those that have been read before the error was detected.
2559 (In the case of an ACL rejection, they have all been read.)
2561 PH/69 The HTML version of the specification is now built in a directory called
2562 spec_html instead of spec.html, because the latter looks like a path with
2563 a MIME-type, and this confuses some software.
2565 PH/70 Catch two compiler warnings in sieve.c.
2567 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2568 function verify_get_ident() calls ip_connect() to connect a socket, but
2569 if the "connect()" function timed out, ip_connect() used to close the
2570 socket. However, verify_get_ident() also closes the socket later, and in
2571 between Exim writes to the log, which may get opened at this point. When
2572 the socket was closed in ip_connect(), the log could get the same file
2573 descriptor number as the socket. This naturally causes chaos. The fix is
2574 not to close the socket in ip_connect(); the socket should be closed by
2575 the function that creates it. There was only one place in the code where
2576 this was missing, in the iplookup router, which I don't think anybody now
2577 uses, but I've fixed it anyway.
2579 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2580 well as to direct DNS lookups. Otherwise the handling of names in host
2581 lists is inconsistent and therefore confusing.
2587 PH/01 Two changes to the default runtime configuration:
2589 (1) Move the checks for relay_from_hosts and authenticated clients from
2590 after to before the (commented out) DNS black list checks.
2592 (2) Add control=submission to the relay_from_hosts and authenticated
2593 clients checks, on the grounds that messages accepted by these
2594 statements are most likely to be submissions.
2596 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2598 (1) Generate an error if the third argument for the ${prvs expansion is
2601 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2604 (3) Reset the variables that are obtained from the first argument of
2605 ${prvscheck and used in the second argument before leaving the code,
2606 because their memory is reclaimed, so using them afterwards may do
2609 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2610 one (it's much easier than Tom thought :-).
2612 (5) Because of (4), we can now allow for the use of $prvscheck_result
2613 inside the third argument.
2615 PH/03 For some reason, the default setting of PATH when running a command from
2616 a pipe transport was just "/usr/bin". I have changed it to
2619 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2620 anything to be listed in the output from -bV.
2622 PH/05 When a filter generated an autoreply, the entire To: header line was
2623 quoted in the delivery log line, like this:
2625 => >A.N.Other <ano@some.domain> <original@ddress> ...
2627 This has been changed so that it extracts the operative address. There
2628 may be more than one such address. If so, they are comma-separated, like
2631 => >ano@some.domain,ona@other.domain <original@ddress> ...
2633 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2634 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2635 not being looked up in the rDNS to get a host name, Exim was showing the
2636 IP address twice in Received: lines, even though the IP addresses were
2637 identical. For example:
2639 Received: from [1.2.3.4] (helo=[1.2.3.4])
2641 However, if the real host name was known, it was omitting the HELO data
2642 if it matched the actual IP address. This has been tidied up so that it
2643 doesn't show the same IP address twice.
2645 PH/07 When both +timestamp and +memory debugging was on, the value given by
2646 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2647 called by the debug printing, thereby overwriting the timestamp buffer.
2648 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2650 PH/08 When the original message was included in an autoreply transport, it
2651 always said "this is a copy of the message, including all the headers",
2652 even if body_only or headers_only was set. It now gives an appropriate
2655 PH/09 Applied a patch from the Sieve maintainer which:
2657 o fixes some comments
2658 o adds the (disabled) notify extension core
2659 o adds some debug output for the result of if/elsif tests
2660 o points to the current vacation draft in the documentation
2661 and documents the missing references header update
2665 o fixes a bug in processing the envelope test (when testing
2666 multiple envelope elements, the last element determinted the
2669 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2670 Electronic Mail") by including:
2672 Auto-submitted: auto-generated
2674 in the messages that it generates (bounce messages and others, such as
2675 warnings). In the case of bounce messages for non-SMTP mesages, there was
2676 also a typo: it was using "Auto_submitted" (underscore instead of
2677 hyphen). Since every message generated by Exim is necessarily in response
2678 to another message, thes have all been changed to:
2680 Auto-Submitted: auto-replied
2682 in accordance with these statements in the RFC:
2684 The auto-replied keyword:
2686 - SHOULD be used on messages sent in direct response to another
2687 message by an automatic process,
2689 - MUST NOT be used on manually-generated messages,
2691 - MAY be used on Delivery Status Notifications (DSNs) and Message
2692 Disposition Notifications (MDNs),
2694 - MUST NOT be used on messages generated by automatic or periodic
2695 processes, except for messages which are automatic responses to
2698 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2699 to the default Received: header definition.
2701 PH/12 Added log selector acl_warn_skipped (default on).
2703 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2704 variables because (a) they are in the wrong storage pool and (b) even if
2705 they were copied, it wouldn't work properly because of the caching.
2707 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2708 checking when decoding. Apparently there are clients that generate
2709 overlong encoded strings. Why am I not surprised?
2711 PH/15 If the first argument of "${if match_address" was not empty, but did not
2712 contain an "@" character, Exim crashed. Now it writes a panic log message
2713 and treats the condition as false.
2715 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2717 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2718 extension "envelope-auth". The code is finished and in agreement with
2719 other implementations, but there is no documentation so far and in fact,
2720 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2721 not changing the active code.
2723 Print executed "if" and "elsif" statements when debugging is used. This
2724 helps a great deal to understand what a filter does.
2726 Document more things not specified clearly in RFC3028. I had all this
2727 sorted out, when out of a sudden new issues came to my mind. Oops."
2729 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2732 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2733 canonical form (as documented). However, after a host name lookup from
2734 the IP address, check_host() was doing a simple string comparison with
2735 addresses acquired from the DNS when checking that the found name did
2736 have the original IP as one of its addresses. Since any found IPv6
2737 addresses are likely to be in abbreviated form, the comparison could
2738 fail. Luckily, there already exists a function for doing the comparison
2739 by converting both addresses to binary, so now that is used instead of
2740 the text comparison.
2742 PH/20 There was another similar case to PH/19, when a complete host name was
2743 given in a host list; looking up its IP address could give an abbreviated
2744 form, whereas the current host's name might or might not be abbreviated.
2745 The same fix has been applied.
2751 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2752 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2755 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2756 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2758 PH/03 Typo: missing ".o" in src/pcre/Makefile.
2760 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2761 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2762 Also, for "Auto-Submitted", treat anything other than "no" as
2763 non-personal, in accordance with RFC 3834. (Previously it treated
2764 anything starting "auto-" as non-personal.)
2766 TF/01 The control=submission/name=... option had a problem with syntax
2767 errors if the name included a slash character. The /name= option
2768 now slurps the rest of the string, so it can include any characters
2769 but it must come last in the list of options (after /sender_retain
2772 PH/05 Some modifications to the interface to the fake nameserver for the new
2780 TK/01 Added the "success_on_redirect" address verification option. See
2781 NewStuff for rationale and an example.
2783 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2785 PH/02 Patch to exigrep to allow it to work on syslog lines.
2787 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2788 fread() to skip over the body file's header line, because in Cygwin the
2789 header line is locked and is inaccessible.
2791 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
2792 co-exist for some time) to make it clear that it is the Exim ID that is
2793 referenced, not the Message-ID: header line.
2795 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
2796 string_format() function, because snprintf() does not exist on all
2799 PH/06 The use of forbid_filter_existstest now also locks out the use of the
2800 ${stat: expansion item.
2802 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
2803 protocol synchronization error", to keep the pedants happy.
2805 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
2806 well as for IRIX systems, when gcc is being used. See the host.c source
2809 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
2811 PH/10 Named domain lists were not working if used in a queue_smtp_domains
2814 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
2815 transport and to the smtp transport in LMTP mode.
2817 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
2819 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
2820 run a filter in a subprocess. This could lead to confusion in subsequent
2821 lookups in the parent process. There should also be a search_tidyup() at
2822 the end of the subprocess.
2824 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
2825 only if the host matched helo_try_verify_hosts, which caused the
2826 verification to occur when the EHLO/HELO command was issued. The ACL just
2827 tested the remembered result. Now, if a previous verification attempt has
2828 not happened, "verify = helo" does it there and then.
2830 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
2832 TK/03 Fix log output including CR from clamd.
2834 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
2835 exist provoked a memory error which could cause a segfault.
2837 PH/15 Installed PCRE 6.2
2839 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2841 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2842 of the problem. Specifically, suggested +O2 rather than +O1 for the
2845 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2847 PH/20 If a delivery was routed to a non-standard port by means of an SRV
2848 record, the port was not correctly logged when the outgoing_port log
2849 selector was set (it logged the transort's default port).
2851 PH/21 Added support for host-specific ports to manualroute, queryprogram,
2852 fallback_hosts, and "hosts" in the smtp transport.
2854 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2855 host errors such as "Connection refused".
2857 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2858 authentication with radiusclient 0.4.9:
2860 - Error returned from rc_read_config was caught wrongly
2861 - Username/password not passed on to radius server due to wrong length.
2863 The presumption is that some radiusclient API changes for 4.51/PH/17
2864 were not taken care of correctly. The code is still untested by me (my
2865 Linux distribution still has 0.3.2 of radiusclient), but it was
2866 contributed by a Radius user.
2868 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2869 expanding the "port" option of the smtp transport.
2871 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2872 while reading a MIME header. Thanks to Tom Hughes for a patch.
2874 PH/24 Include config.h inside local_scan.h so that configuration settings are
2877 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2878 that in an ACL for RCPT (for example), you can examine exactly what was
2881 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2882 commands, but it was not correctly comparing the address with the actual
2883 client host address. Thus, it would show the EHLO address in Received:
2884 header lines when this was not necessary.
2886 PH/27 Added the % operator to ${eval:}.
2888 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2889 it should be ignoring failures (because with -C, for example, it has lost
2890 privilege). It wasn't ignoring creation failures other than "already
2893 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2896 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2897 because an input line was too long, either on its own, or by virtue of
2898 too many continuations, the temporary file was not being removed, and the
2899 return code was incorrect.
2901 PH/31 Missing "BOOL" in function definition in filtertest.c.
2903 PH/32 Applied Sieve patches from the maintainer.
2905 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2907 PH/33 Added "verify = not_blind".
2909 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2910 Local/Makefile (with some defaults set). These are used in built scripts
2911 such as exicyclog, but they have never been used in the exim_install
2912 script (though there are many overriding facilities there). I have
2913 arranged that the exim_install script now takes note of these two
2916 PH/35 Installed configuration files for Dragonfly.
2918 PH/36 When a locally submitted message by a trusted user did not contain a
2919 From: header, and the sender address was obtained from -f or from an SMTP
2920 MAIL command, and the trusted user did not use -F to supply a sender
2921 name, $originator_name was incorrectly used when constructing a From:
2922 header. Furthermore, $originator_name was used for submission mode
2923 messages from external hosts without From: headers in a similar way,
2924 which is clearly wrong.
2926 PH/37 Added control=suppress_local_fixups.
2928 PH/38 When log_selector = +received_sender was set, and the addition of the
2929 sender made the log line's construction buffer exactly full, or one byte
2930 less than full, an overflow happened when the terminating "\n" was
2933 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2934 when the result of a list match is failure because a DNS lookup failed.
2936 PH/40 RM_COMMAND is now used in the building process.
2938 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2939 the "build-* directories that it finds.
2941 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2942 address in a domain literal was a prefix of an interface address.
2944 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2945 when verifying a sender address, unless rewrite_headers is false.
2947 PH/44 Wrote a long comment about why errors_to addresses are verified as
2948 recipients, not senders.
2950 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2951 the ratelimit ACL was added.
2953 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2955 PH/46 Added extra information about PostgreSQL errors to the error string.
2957 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2958 avoiding the need to install special zones in a real server. This is
2959 backwards compatible; if it can't find the fake resolver, it drops back.
2960 Thus, both old and new test suites can be run.
2962 TF/02 Added util/ratelimit.pl
2964 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2965 clock is set back in time.
2967 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2968 Candler <B.Candler@pobox.com>.
2970 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2971 OK for addresses that are the result of redirections.
2973 PH/48 A number of further additions for the benefit of the new test suite,
2974 including a fake gethostbyname() that interfaces to the fake DNS resolver
2977 TF/06 The fix for widen_domains has also been applied to qualify_single and
2978 search_parents which are the other dnslookup options that can cause
2981 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2984 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2986 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2987 service" if it encountered an MX record with an empty target hostname.
2988 The message is now "an MX or SRV record indicated no SMTP service".
2990 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2991 if the DNS of the sending site is misconfigured. This is quite a
2992 common situation. This change restores the behaviour of treating a
2993 helo verification defer as a failure.
2995 PH/51 If self=fail was set on a router, the bounce message did not include the
2996 actual error message.
3002 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3004 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3005 command itself ended in error, the underlying message about the transport
3006 filter timeout was being overwritten with the pipe command error. Now the
3007 underlying error message should be appended to the second error message.
3009 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3011 PH/02 Reduce the amount of output that "make" produces by default. Full output
3012 can still be requested.
3014 PH/03 The warning log line about a condition test deferring for a "warn" verb
3015 was being output only once per connection, rather than after each
3016 occurrence (because it was using the same function as for successful
3017 "warn" verbs). This seems wrong, so I have changed it.
3019 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3020 it should not have, which might have caused a crash in the right
3021 circumstances, but probably never did.
3023 PH/04 Installed a modified version of Tony Finch's patch to make submission
3024 mode fix the return path as well as the Sender: header line, and to
3025 add a /name= option so that you can make the user's friendly name appear
3028 TF/03 Added the control = fakedefer ACL modifier.
3030 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3031 Mark Lowes for thorough testing.
3033 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3035 TK/03 Merged latest SRS patch from Miles Wilton.
3037 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3038 with the definition in sysexits.h (which is #included earlier).
3039 Fortunately, Exim does not actually use EX_OK. The code used to try to
3040 preserve the sysexits.h value, by assumimg that macro definitions were
3041 scanned for macro replacements. I have been disabused of this notion,
3042 so now the code just undefines EX_OK before #including unistd.h.
3044 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3045 in the smtp transport. When a block could not be written in a single
3046 write() function, the timeout was being re-applied to each part-write.
3047 This seems wrong - if the receiver was accepting one byte at a time it
3048 would take for ever. The timeout is now adjusted when this happens. It
3049 doesn't have to be particularly precise.
3051 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3052 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3054 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3055 without a domain if the check to <postmaster@domain> fails.
3057 SC/01 Eximstats: added -xls and the ability to specify output files
3058 (patch written by Frank Heydlauf).
3060 SC/02 Eximstats: use FileHandles for outputing results.
3062 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3064 SC/04 Eximstats: fixed display of large numbers with -nvr option
3066 SC/05 Eximstats: fixed merging of reports with empty tables.
3068 SC/06 Eximstats: added the -include_original_destination flag
3070 SC/07 Eximstats: removed tabs and trailing whitespace.
3072 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3074 TK/06 MBOX spool code: Add real "From " MBOX separator line
3075 so the .eml file is really in mbox format (even though
3076 most programs do not really care). Patch from Alex Miller.
3078 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3079 The latter is generated from $received_to and is only set if the
3080 message has one envelope recipient. SA can use these headers,
3081 obviously out-of-the-box. Patch from Alex Miller.
3083 PH/08 The ${def test on a variable was returning false if the variable's
3084 value was "0", contrary to what the specification has always said!
3085 The result should be true unless the variable is empty.
3087 PH/09 The syntax error of a character other than { following "${if
3088 def:variable_name" (after optional whitespace) was not being diagnosed.
3089 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3090 accidental colon was present, for example, could give incorrect results.
3092 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3093 result is used (not including appendfile, where other changes are about
3096 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3097 This involved changing a lot of size variables from int to off_t. It
3098 should work with maildirs and everything.
3100 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3101 spamd dying while we are connected to it.
3103 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3106 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3107 function for BDB 4.3.
3109 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3111 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3112 the sources for PCRE have been re-arranged and more files are now
3115 PH/15 The code I had for printing potentially long long variables in PH/11
3116 above was not the best (it lost precision). The length of off_t variables
3117 is now inspected at build time, and an appropriate printing format (%ld
3118 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3119 to be "long long int" or "long int". This is needed for the internal
3120 formatting function string_vformat().
3122 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3123 the configuration file to be ":syslog", then the script "guesses" where
3124 the logs files are, rather than using the compiled in default. In our
3125 case the guess is not the same as the compiled default, so the script
3126 suddenly stopped working when I started to use syslog. The patch checks
3127 to see if log_file_path is "". If so, it attempts to read it from exim
3128 with no configuration file to get the compiled in version, before it
3129 falls back to the previous guessing code."
3131 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3132 implementing BATV in an Exim configuration. See NewStuff for the gory
3135 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3136 Makefile that are specific to HP-UX.
3138 PH/18 If the "use_postmaster" option was set for a recipient callout together
3139 with the "random" option, the postmaster address was used as the MAIL
3140 FROM address for the random test, but not for the subsequent recipient
3141 test. It is now used for both.
3143 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3144 patch removes a few documentation additions to RFC 3028, because the
3145 latest draft now contains them. It adds the new en;ascii-case comparator
3146 and a new error check for 8bit text in MIME parts. Comparator and
3147 require names are now matched exactly. I enabled the subaddress
3148 extension, but it is not well tested yet (read: it works for me)."
3150 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3151 rework some of the code of TK/09 above to avoid the hardwired use of
3152 "%lld" and "long long". Replaced the call to snprintf() with a call to
3155 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3156 records point to non-existent hosts", "retry timeout exceeded", and
3157 "retry time not reached for any host after a long failure period".
3159 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3160 experimental DomainKeys support:
3162 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3163 (2) On an error such as an illegally used "control", the wrong name for
3164 the control was given.
3166 These problems did NOT occur unless DomainKeys support was compiled.
3168 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3170 PH/24 Added ${if match_ip condition.
3172 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3173 databases so that it will be absolutely obvious if a crash occurs in the
3174 DB library. This is a regular occurrence (often caused by mis-matched
3177 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3178 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3179 picked up on a user's system that detects such things. There doesn't seem
3180 to be a gcc warning option for this - only an attribute that has to be
3181 put on the function's prototype. It seems that in Fedora Core 4 they have
3182 set this on a number of new functions. No doubt there will be more in due
3185 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3186 specify a transport. However, if an address that was verified by such a
3187 router was the subject of a callout, Exim crashed because it tried to
3188 read the rcpt_include_affixes from the non-existent transport. Now it
3189 just assumes that the setting of that option is false. This bug was
3190 introduced by 4.51/PH/31.
3192 PH/28 Changed -d+all to exclude +memory, because that information is very
3193 rarely of interest, but it makes the output a lot bigger. People tend to
3194 do -d+all out of habit.
3196 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3197 code in os-type was giving problems when libc.so lives in lib64, like on
3200 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3201 aren't the modern standard, and it seems that some systems' include files
3202 don't always have them. Exim was already checking for some of the newer
3203 ones like T_AAAA, and defining it itself. I've added checks for all the
3204 record types that Exim uses.
3206 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3207 not automatically generating a new one, as it is supposed to. This
3208 prevented TLS from working. If the file did exist, but contained invalid
3209 data, a new version was generated, as expected. It was only the case of a
3210 non-existent file that was broken.
3212 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3213 with a change in libdomainkeys > 0.64.
3215 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3216 from DNS. If the selector record carries the flag, it now has
3217 precedence over the domain-wide flag.
3219 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3221 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3222 the use of a port name that isn't defined in /etc/services) occurred, the
3223 message was deferred as in a normal delivery, and thus remained on the
3224 spool, instead of being failed because of the mua_wrapper setting. This
3225 is now fixed, and I tidied up some of the mua_wrapper messages at the
3228 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3229 the messages in a hash of arrays rather than using individual hashes.
3230 This is a bit cleaner and results in dramatic memory savings, albeit
3231 at a slight CPU cost.
3233 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3234 as requested by Marc Sherman.
3236 SC/10 Eximstats: added histograms for user specified patterns as requested
3239 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3241 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3242 fopen() in the content-scanning modules that did not already have it.
3248 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3249 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3251 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3253 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3255 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3256 to test Sieve filters that use "vacation".
3258 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3259 that changes the way the GnuTLS parameters are stored in the cache file.
3260 The new format can be generated externally. For backward compatibility,
3261 if the data in the cache doesn't make sense, Exim assumes it has read an
3262 old-format file, and it generates new data and writes a new file. This
3263 means that you can't go back to an older release without removing the
3266 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3267 work if there are any delivery delays because "one_time" forces the
3268 parent to be marked "delivered", so its unseen clone is never tried
3269 again. For this reason, Exim now forbids the simultaneous setting of
3272 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3273 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3274 entry if you want to know the details. The fix, however, neglected to
3275 consider the case when local delivery batching is involved. The test for
3276 "previously delivered" was not happening when checking to see if an
3277 address could be batched with a previous (undelivered) one; under
3278 certain circumstances this could lead to multiple deliveries to the same
3281 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3282 in its include files, and this causes problems building Exim.
3284 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3285 header_syntax) but Exim was just ignoring anything given after a slash.
3286 In particular, this caused confusion with an attempt to use "verify =
3287 reverse_host_lookup/defer_ok". An error is now given when options are
3288 supplied for verify items that do not have them. (Maybe reverse_host_
3289 lookup should have a defer_ok option, but that's a different point.)
3291 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3292 defined by RFC 821) to 2048, because there were problems with some AUTH
3293 commands, and RFC 1869 says the size should be increased for extended
3294 SMTP commands that take arguments.
3296 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3299 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3300 "unknown" error; now it says that the functionality isn't in the binary.
3302 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3303 an address' error message when a string expansion fails (syntax or
3304 whatever). Otherwise the password may appear in the log. Following change
3305 PH/42 below, there is no longer a chance of it appearing in a bounce
3308 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3310 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3311 of its addresses was ever tried. (Bugzilla bug #2.)
3313 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3314 the result incorrectly in the debug output. (It correctly added a newline
3315 to what was transported.)
3317 TF/01 Added $received_time.
3319 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3320 commented out examples of how to interface to a virus scanner and to
3321 SpamAssassin. Also added commented examples of av_scanner and
3322 spamd_address settings.
3324 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3325 and controls are allowed in which ACLs. There were a couple of minor
3326 errors. Some of the entries in the conditions table (which is a table of
3327 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3328 negation of where the condition IS allowed.
3330 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3332 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3333 header file does not have a version number, so I've had to invent a new
3334 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3335 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3336 radiusclient), but it was contributed by a Radius user.
3338 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3339 files or directories, for OpenSSL.
3341 PH/19 When an Exim process that is running as root has to create an Exim log
3342 file, it does so in a subprocess that runs as exim:exim so as to get the
3343 ownership right at creation (otherwise, other Exim processes might see
3344 the file with the wrong ownership). There was no test for failure of this
3345 fork() call, which would lead to the process getting stuck as it waited
3346 for a non-existent subprocess. Forks do occasionally fail when resources
3347 run out. I reviewed all the other calls to fork(); they all seem to check
3350 PH/20 When checking for unexpected SMTP input at connect time (before writing
3351 the banner), Exim was not dealing correctly with a non-positive return
3352 from the read() function. If the client had disconnected by this time,
3353 the result was a log entry for a synchronization error with an empty
3354 string after "input=" when read() returned zero. If read() returned -1
3355 (an event I could not check), uninitialized data bytes were printed.
3356 There were reports of junk text (parts of files, etc) appearing after
3359 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3361 PH/22 Added support for macro redefinition, and (re)definition in between
3362 driver and ACL definitions.
3364 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3365 forgetting to use the resulting value; it was using the unexpanded value.
3367 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3368 hadn't been configured. The fix is from Juergen Kreileder, who
3369 understands it better than I do:
3371 "Here's what I see happening with three configured cyrus_sasl
3372 authenticators configured (plain, login, cram-md5):
3374 On startup auth_cyrus_sasl_init() gets called for each of these.
3375 This means three calls to sasl_listmech() without a specified mech_list.
3376 => SASL tests which mechs of all available mechs actually work
3377 => three warnings about OTP not working
3378 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3380 With the patch, sasl_listmech() also gets called three times. But now
3381 SASL's mech_list option is set to the server_mech specified in the the
3382 authenticator. Or in other words, the answer from sasl_listmech()
3383 gets limited to just the mech you're testing for (which is different
3385 => the return list contains just 'plain' or 'login', 'cram-md5' or
3386 nothing depending on the value of ob->server_mech.
3388 I've just tested the patch: Authentication still works fine,
3389 unavailable mechs specified in the exim configuration are still
3390 caught, and the auth.log warnings about OTP are gone."
3392 PH/25 When debugging is enabled, the contents of the command line are added
3393 to the debugging output, even when log_selector=+arguments is not
3396 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3397 answer is "GNU", and only if the return is "GNU/something" is the answer
3400 PH/27 $acl_verify_message is now set immediately after the failure of a
3401 verification in an ACL, and so is available in subsequent modifiers. In
3402 particular, the message can be preserved by coding like this:
3404 warn !verify = sender
3405 set acl_m0 = $acl_verify_message
3407 Previously, $acl_verify_message was set only while expanding "message"
3408 and "log_message" when a very denied access.
3410 PH/28 Modified OS/os.c-Linux with
3412 -#ifndef OS_LOAD_AVERAGE
3413 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3415 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3416 nomenclature these days.)
3418 PH/29 Installed patch from the Sieve maintainer that adds the options
3419 sieve_useraddress and sieve_subaddress to the redirect router.
3421 PH/30 In these circumstances:
3422 . Two addresses routed to the same list of hosts;
3423 . First host does not offer TLS;
3424 . First host accepts first address;
3425 . First host gives temporary error to second address;
3426 . Second host offers TLS and a TLS session is established;
3427 . Second host accepts second address.
3428 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3429 and peerdn, if requested) that were in fact used only for the second
3432 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3433 attention to any local part prefix or suffix that was matched by the
3434 router that accepted the address. It now behaves in the same way as it
3435 does for delivery: the affixes are removed from the local part unless
3436 rcpt_include_affixes is set on the transport.
3438 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3439 timeout during the DATA phase of an incoming message.
3441 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3442 applied a patch sanctioned by the Sieve maintainer.
3444 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3445 the uid or gid is negative. A case of a negative gid caused this to be
3446 noticed. The fix allows for either to be negative.
3448 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3449 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3452 PH/36 The change PH/12 above was broken. Fixed it.
3454 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3455 the grounds that routing the same address twice would always produce the
3456 same answer. This might have been true once, but it is certainly no
3457 longer true now. Routing a child address may depend on the previous
3458 routing that produced that child. Some complicated redirection strategies
3459 went wrong when messages had multiple recipients, and made Exim's
3460 behaviour dependent on the order in which the addresses were given.
3462 I have moved the duplicate checking until after the routing is complete.
3463 Exim scans the addresses that are assigned to local and remote
3464 transports, and removes any duplicates. This means that more work will be
3465 done, as duplicates will always all be routed, but duplicates are
3466 presumably rare, so I don't expect this is of any significance.
3468 For deliveries to pipes, files, and autoreplies, the duplicate checking
3469 still happens during the routing process, since they are not going to be
3472 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3473 It corrects a timeout issue with spamd. This is Ian's comment: "The
3474 background is that sometimes spamd either never reads data from a
3475 connection it has accepted, or it never writes response data. The exiscan
3476 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3477 blindly assumes that writes won't block so it may never time out."
3479 PH/39 Allow G after quota size as well as K and M.
3481 PH/40 The value set for $authenticated_id in an authenticator may not contain
3482 binary zeroes or newlines because the value is written to log lines and
3483 to spool files. There was no check on this. Now the value is run through
3484 the string_printing() function so that such characters are converted to
3485 printable escape sequences.
3487 PH/41 $message_linecount is a new variable that contains the total number of
3488 lines in the message. Compare $body_linecount, which is the count for the
3491 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3492 bounce and delay warning messages, except in certain special cases, which
3495 (a) An SMTP error message from a remote host;
3496 (b) A message specified in a :fail: redirection;
3497 (c) A message specified in a "fail" command in a system filter;
3498 (d) A message specified in a FAIL return from the queryprogram router;
3499 (e) A message specified by the cannot_route_message router option.
3501 In these cases only, Exim does include the error details in bounce and
3502 warning messages. There are also a few cases where bland messages such
3503 as "unrouteable address" or "local delivery error" are given.
3505 PH/43 $value is now also set for the "else" part of a ${run expansion.
3507 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3508 being worked on, but at least Exim now implements the latest version to
3511 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3512 process to complete was treated as a delivery failure, a timeout while
3513 writing the message to the pipe was logged, but erroneously treated as a
3514 successful delivery. Such timeouts include transport filter timeouts. For
3515 consistency with the overall process timeout, these timeouts are now
3516 treated as errors, giving rise to delivery failures by default. However,
3517 there is now a new Boolean option for the pipe transport called
3518 timeout_defer, which, if set TRUE, converts the failures into defers for
3519 both kinds of timeout. A transport filter timeout is now identified in
3522 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3523 systems where "make" and "gmake" are different, calling "gmake" at top
3524 level broke things. I've arranged for the value of $(MAKE) to be passed
3525 from the Makefile to this script so that it can call the same version of
3529 A note about Exim versions 4.44 and 4.50
3530 ----------------------------------------
3532 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3533 changes of various kinds. As a consequence, a big documentation update was
3534 needed. This delayed the release for rather longer than seemed good, especially
3535 in the light of a couple of (minor) security issues. Therefore, the changes
3536 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3537 release. So 4.44 and 4.50 are in effect two different branches that both start
3540 I have left the 4.50 change log unchanged; it contains all the changes since
3541 4.43. The change log for 4.44 is below; many of its items are identical to
3542 those for 4.50. This seems to be the most sensible way to preserve the
3543 historical information.
3549 1. Minor wording change to the doc/README.SIEVE file.
3551 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3552 computation of the current number of files was incorrect.
3554 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3555 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3558 4. Give more explanation in the error message when the command for a transport
3559 filter fails to execute.
3561 5. There are several places where Exim runs a non-Exim command in a
3562 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3563 was being done only for the command run by the queryprogram router. It is
3564 now done for all such subprocesses. The other cases are: ${run, transport
3565 filters, and the commands run by the lmtp and pipe transports.
3567 6. Added CONFIGURE_GROUP build-time option.
3569 7. Some older OS have a limit of 256 on the maximum number of file
3570 descriptors. Exim was using setrlimit() to set 1000 as a large value
3571 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3572 systems. I've change it so that if it can't get 1000, it tries for 256.
3574 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3575 was an oversight, and furthermore, ever since the addition of extra
3576 controls (e.g. 4.43/32), the checks on when to allow different forms of
3577 "control" were broken. There should now be diagnostics for all cases when a
3578 control that does not make sense is encountered.
3580 9. Added the /retain_sender option to "control=submission".
3582 10. $recipients is now available in the predata ACL (oversight).
3584 11. Tidy the search cache before the fork to do a delivery from a message
3585 received from the command line. Otherwise the child will trigger a lookup
3586 failure and thereby defer the delivery if it tries to use (for example) a
3587 cached ldap connection that the parent has called unbind on.
3589 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3590 of $address_data from the recipient verification was clobbered by the
3591 sender verification.
3593 13. The value of address_data from a sender verification is now available in
3594 $sender_address_data in subsequent conditions in the ACL statement.
3596 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3598 15. Added a new option "connect=<time>" to callout options, to set a different
3601 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3602 was its contents. (It was OK if the option was not defined at all.)
3604 17. A "Completed" log line is now written for messages that are removed from
3605 the spool by the -Mrm option.
3607 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3608 information about exactly what failed.
3610 19. Added -dd to debug only the daemon process.
3612 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3613 handles timeouts, both on the server side and network timeouts. Renamed the
3614 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3616 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3617 It is now set to "smtps".
3619 22. $host_address is now set to the target address during the checking of
3620 ignore_target_hosts.
3622 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3623 being passed; this would have caused $sender_host_name to have been used if
3624 matching the list had actually called for a host name (not very likely,
3625 since this list is usually IP addresses). A host name is now passed as
3628 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3629 code that specifies a non-restarting handler (typically sigaction() in
3630 modern systems) in an attempt to fix a rare and obscure crash bug.
3632 25. Narrowed the window for a race in the daemon that could cause it to ignore
3633 SIGCHLD signals. This is not a major problem, because they are used only to
3634 wake it up if nothing else does.
3636 26. A malformed maildirsize file could cause Exim to calculate negative values
3637 for the mailbox size or file count. Odd effects could occur as a result.
3638 The maildirsize information is now recalculated if the size or filecount
3641 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3642 support for a long time. Removed HAVE_SYS_VFS_H.
3644 28. Installed the latest version of exipick from John Jetmore.
3646 29. In an address list, if the pattern was not a regular expression, an empty
3647 subject address (from a bounce message) matched only if the pattern was an
3648 empty string. Non-empty patterns were not even tested. This was the wrong
3649 because it is perfectly reasonable to use an empty address as part of a
3650 database query. An empty address is now tested by patterns that are
3651 lookups. However, all the other forms of pattern expect the subject to
3652 contain a local part and a domain, and therefore, for them, an empty
3653 address still always fails if the pattern is not itself empty.
3655 30. Exim went into a mad DNS loop when attempting to do a callout where the
3656 host was specified on an smtp transport, and looking it up yielded more
3657 than one IP address.
3659 31. Re-factored the code for checking spool and log partition space into a
3660 function that finds that data and another that does the check. The former
3661 is then used to implement four new variables: $spool_space, $log_space,
3662 $spool_inodes, and $log_inodes.
3664 32. The RFC2047 encoding function was originally intended for short strings
3665 such as real names; it was not keeping to the 75-character limit for
3666 encoded words that the RFC imposes. It now respects the limit, and
3667 generates multiple encoded words if necessary. To be on the safe side, I
3668 have increased the buffer size for the ${rfc2047: expansion operator from
3671 33. It is now permitted to omit both strings after an "if" condition; if the
3672 condition is true, the result is "true". As before, when the second string
3673 is omitted, a false condition yields an empty string. This makes it less
3674 cumbersome to write custom ACL and router conditions.
3676 34. Failure to deliver a bounce message always caused it to be frozen, even if
3677 there was an errors_to setting on the router. The errors_to setting is now
3680 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3681 canonical form (fully expanded) before being placed in
3682 $sender_host_address.
3684 36. The table in the code that translates DNS record types into text (T_A to
3685 "A" for instance) was missing entries for NS and CNAME. It is just possible
3686 that this could have caused confusion if both these types were looked up
3687 for the same domain, because the text type is used as part of Exim's
3688 per-process caching. But the chance of anyone hitting this buglet seems
3691 37. The dnsdb lookup has been extended in a number of ways.
3693 (1) There is a new type, "zns", which walks up the domain tree until it
3694 finds some nameserver records. It should be used with care.
3696 (2) There is a new type, "mxh", which is like "mx" except that it returns
3697 just the host names, not the priorities.
3699 (3) It is now possible to give a list of domains (or IP addresses) to be
3700 looked up. The behaviour when one of the lookups defers can be
3701 controlled by a keyword.
3703 (4) It is now possible to specify the separator character for use when
3704 multiple records are returned.
3706 38. The dnslists ACL condition has been extended: it is now possible to supply
3707 a list of IP addresses and/or domains to be looked up in a particular DNS
3710 39. Added log_selector=+queue_time_overall.
3712 40. When running the queue in the test harness, wait just a tad after forking a
3713 delivery process, to get repeatability of debugging output.
3715 41. Include certificate and key file names in error message when GnuTLS fails
3716 to set them up, because the GnuTLS error message doesn't include the name
3717 of the failing file when there is a problem reading it.
3719 42. Allow both -bf and -bF in the same test run.
3721 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3723 44. The "Exiscan patch" is now merged into the mainline Exim source.
3725 45. Sometimes the final signoff response after QUIT could fail to get
3726 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3727 < 0 before doing a fflush(). This bug looks as though it goes back to the
3728 introduction of TLS in release 3.20, but "sometimes" must have been rare
3729 because the tests only now provoked it.
3731 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3732 (this can affect the format of dates).
3734 47. exim_tidydb, when checking for the continued existence of a message for
3735 which it has found a message-specific retry record, was not finding
3736 messages that were in split spool directories. Consequently, it was
3737 deleting retry records that should have stayed in existence.
3739 48. Steve fixed some bugs in eximstats.
3741 49. The SPA authentication driver was not abandoning authentication and moving
3742 on to the next authenticator when an expansion was forced to fail,
3743 contradicting the general specification for all authenticators. Instead it
3744 was generating a temporary error. It now behaves as specified.
3746 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3747 (the order specifies the preference for clients). The order is now AES256,
3748 AES128, 3DES, ARCFOUR128.
3750 51. Small patch to Sieve code - explicitly set From: when generating an
3753 52. Exim crashed if a remote delivery caused a very long error message to be
3754 recorded - for instance if somebody sent an entire SpamAssassin report back
3755 as a large number of 550 error lines. This bug was coincidentally fixed by
3756 increasing the size of one of Exim's internal buffers (big_buffer) that
3757 happened as part of the Exiscan merge. However, to be on the safe side, I
3758 have made the code more robust (and fixed the comments that describe what
3761 53. Now that there can be additional text after "Completed" in log lines (if
3762 the queue_time_overall log selector is set), a one-byte patch to exigrep
3763 was needed to allow it to recognize "Completed" as not the last thing in
3766 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3767 patch that reportedly fixes this has been added. I am not expert enough to
3768 create a test for it. This is what the patch creator wrote:
3770 "I found a little strange behaviour of ldap code when working with
3771 Windows 2003 AD Domain, where users was placed in more than one
3772 Organization Units. When I tried to give exim partial DN, the exit code
3773 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
3774 But simultaneously result of request was absolutely normal ldap result,
3775 so I produce this patch..."
3777 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
3778 REFERENCE, so I have modified the code to exclude the patch when that macro
3781 55. Some experimental protocols are using DNS PTR records for new purposes. The
3782 keys for these records are domain names, not reversed IP addresses. The
3783 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3784 leaves it alone. Component reversal etc. now happens only for IP addresses.
3787 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
3789 57. Double the size of the debug message buffer (to 2048) so that more of very
3790 long debug lines gets shown.
3792 58. The exicyclog utility now does better if the number of log files to keep
3793 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
3795 59. Two changes related to the smtp_active_hostname option:
3797 (1) $smtp_active_hostname is now available as a variable.
3798 (2) The default for smtp_banner uses $smtp_active_hostname instead
3799 of $primary_hostname.
3801 60. The host_aton() function is supposed to be passed a string that is known
3802 to be a valid IP address. However, in the case of IPv6 addresses, it was
3803 not checking this. This is a hostage to fortune. Exim now panics and dies
3804 if the condition is not met. A case was found where this could be provoked
3805 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3806 components; fortuitously, this particular loophole had already been fixed
3807 by change 4.50/55 above.
3809 If there are any other similar loopholes, the new check in host_aton()
3810 itself should stop them being exploited. The report I received stated that
3811 data on the command line could provoke the exploit when Exim was running as
3812 exim, but did not say which command line option was involved. All I could
3813 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3814 running as the user.
3817 61. There was a buffer overflow vulnerability in the SPA authentication code
3818 (which came originally from the Samba project). I have added a test to the
3819 spa_base64_to_bits() function which I hope fixes it.
3822 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
3823 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
3825 63. The daemon start-up calls getloadavg() while still root for those OS that
3826 need the first call to be done as root, but it missed one case: when
3827 deliver_queue_load_max is set with deliver_drop_privilege. This is
3828 necessary for the benefit of the queue runner, because there is no re-exec
3829 when deliver_drop_privilege is set.
3831 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
3832 This has been fixed.
3834 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3835 was in use, was not putting the data itself into the right store pool;
3836 consequently, it could be overwritten for a subsequent message in the same
3837 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3840 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
3842 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
3843 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
3844 the function were treating the return as a boolean value, which happened to
3845 work because 0=false and not-0=true, but is not correct code.
3847 68. The host_aton() function was not handling scoped IPv6 addresses (those
3848 with, for example, "%eth0" on the end) correctly.
3850 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
3851 negated items (that is, ~something) in unsigned ints. Some compilers
3852 apparently mutter when there is no cast.
3854 70. If an address verification called from an ACL failed, and did not produce a
3855 user-specific message (i.e. there was only a "system" message), nothing was
3856 put in $acl_verify_message. In this situation, it now puts the system
3859 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3860 session; change 4.31/43 added the unwanted input to the log line - except
3861 that it did not do this in the start of session case. It now does.
3863 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3864 This is wrong and can cause the other end to generate a synchronization
3865 error if it is another Exim or anything else that does the synchronization
3866 check. A QUIT command is no longer sent after a timeout.
3868 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3869 during host lookups.
3871 74. The defer_ok option of callout verification was not working if it was used
3872 when verifying addresses in header lines, that is, for this case:
3874 verify = header_sender/callout=defer_ok
3876 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3877 those file descriptors could be used for SMTP connections. If anything
3878 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3879 could be sent to the SMTP client, causing chaos. The daemon now opens
3880 stdin, stdout, and stderr to /dev/null when it puts itself into the
3883 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3884 log by default. The user can override this with suitable Perl magic.
3886 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3887 the log message when discard triggers, was not working for the DATA ACL or
3888 for the non-SMTP ACL.
3890 78. Error message wording change in sieve.c.
3892 79. If smtp_accept_max_per_host was set, the number of connections could be
3893 restricted to fewer than expected, because the daemon was trying to set up
3894 a new connection before checking whether the processes handling previous
3895 connections had finished. The check for completed processes is now done
3896 earlier. On busy systems, this bug wouldn't be noticed because something
3897 else would have woken the daemon, and it would have reaped the completed
3900 80. If a message was submitted locally by a user whose login name contained one
3901 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3902 It caused a spool format error. I have fixed the spool reading code. A
3903 related problem was that the "from" clause in the Received: line became
3904 illegal because of the space(s). It is now covered by ${quote_local_part.
3906 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3909 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3911 83. Patch from Sieve maintainer to fix unterminated string problem in
3912 "vacation" handling.
3914 84. Some minor changes to the Linux configuration files to help with other
3915 OS variants using glibc.
3917 85. One more patch for Sieve to update vacation handling to latest spec.
3920 ----------------------------------------------------
3921 See the note above about the 4.44 and 4.50 releases.
3922 ----------------------------------------------------
3928 1. Change 4.43/35 introduced a bug that caused file counts to be
3929 incorrectly computed when quota_filecount was set in an appendfile
3932 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3933 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3936 3. Give more explanation in the error message when the command for a transport
3937 filter fails to execute.
3939 4. There are several places where Exim runs a non-Exim command in a
3940 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3941 was being done only for the command run by the queryprogram router. It is
3942 now done for all such subprocesses. The other cases are: ${run, transport
3943 filters, and the commands run by the lmtp and pipe transports.
3945 5. Some older OS have a limit of 256 on the maximum number of file
3946 descriptors. Exim was using setrlimit() to set 1000 as a large value
3947 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3948 systems. I've change it so that if it can't get 1000, it tries for 256.
3950 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3951 was an oversight, and furthermore, ever since the addition of extra
3952 controls (e.g. 4.43/32), the checks on when to allow different forms of
3953 "control" were broken. There should now be diagnostics for all cases when a
3954 control that does not make sense is encountered.
3956 7. $recipients is now available in the predata ACL (oversight).
3958 8. Tidy the search cache before the fork to do a delivery from a message
3959 received from the command line. Otherwise the child will trigger a lookup
3960 failure and thereby defer the delivery if it tries to use (for example) a
3961 cached ldap connection that the parent has called unbind on.
3963 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3964 of $address_data from the recipient verification was clobbered by the
3965 sender verification.
3967 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3968 was its contents. (It was OK if the option was not defined at all.)
3970 11. A "Completed" log line is now written for messages that are removed from
3971 the spool by the -Mrm option.
3973 12. $host_address is now set to the target address during the checking of
3974 ignore_target_hosts.
3976 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3977 being passed; this would have caused $sender_host_name to have been used if
3978 matching the list had actually called for a host name (not very likely,
3979 since this list is usually IP addresses). A host name is now passed as
3982 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3983 code that specifies a non-restarting handler (typically sigaction() in
3984 modern systems) in an attempt to fix a rare and obscure crash bug.
3986 15. Narrowed the window for a race in the daemon that could cause it to ignore
3987 SIGCHLD signals. This is not a major problem, because they are used only to
3988 wake it up if nothing else does.
3990 16. A malformed maildirsize file could cause Exim to calculate negative values
3991 for the mailbox size or file count. Odd effects could occur as a result.
3992 The maildirsize information is now recalculated if the size or filecount
3995 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3996 support for a long time. Removed HAVE_SYS_VFS_H.
3998 18. Updated exipick to current release from John Jetmore.
4000 19. Allow an empty sender to be matched against a lookup in an address list.
4001 Previously the only cases considered were a regular expression, or an
4004 20. Exim went into a mad DNS lookup loop when doing a callout where the
4005 host was specified on the transport, if the DNS lookup yielded more than
4008 21. The RFC2047 encoding function was originally intended for short strings
4009 such as real names; it was not keeping to the 75-character limit for
4010 encoded words that the RFC imposes. It now respects the limit, and
4011 generates multiple encoded words if necessary. To be on the safe side, I
4012 have increased the buffer size for the ${rfc2047: expansion operator from
4015 22. Failure to deliver a bounce message always caused it to be frozen, even if
4016 there was an errors_to setting on the router. The errors_to setting is now
4019 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4020 canonical form (fully expanded) before being placed in
4021 $sender_host_address.
4023 24. Updated eximstats to version 1.33
4025 25. Include certificate and key file names in error message when GnuTLS fails
4026 to set them up, because the GnuTLS error message doesn't include the name
4027 of the failing file when there is a problem reading it.
4029 26. Expand error message when OpenSSL has problems setting up cert/key files.
4032 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4033 (this can affect the format of dates).
4035 28. exim_tidydb, when checking for the continued existence of a message for
4036 which it has found a message-specific retry record, was not finding
4037 messages that were in split spool directories. Consequently, it was
4038 deleting retry records that should have stayed in existence.
4040 29. eximstats updated to version 1.35
4041 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4042 1.35 - bugfix such that pie charts by volume are generated correctly
4044 30. The SPA authentication driver was not abandoning authentication and moving
4045 on to the next authenticator when an expansion was forced to fail,
4046 contradicting the general specification for all authenticators. Instead it
4047 was generating a temporary error. It now behaves as specified.
4049 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4050 (the order specifies the preference for clients). The order is now AES256,
4051 AES128, 3DES, ARCFOUR128.
4053 31. Small patch to Sieve code - explicitly set From: when generating an
4056 32. Exim crashed if a remote delivery caused a very long error message to be
4057 recorded - for instance if somebody sent an entire SpamAssassin report back
4058 as a large number of 550 error lines. This bug was coincidentally fixed by
4059 increasing the size of one of Exim's internal buffers (big_buffer) that
4060 happened as part of the Exiscan merge. However, to be on the safe side, I
4061 have made the code more robust (and fixed the comments that describe what
4064 33. Some experimental protocols are using DNS PTR records for new purposes. The
4065 keys for these records are domain names, not reversed IP addresses. The
4066 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4067 leaves it alone. Component reversal etc. now happens only for IP addresses.
4070 34. The host_aton() function is supposed to be passed a string that is known
4071 to be a valid IP address. However, in the case of IPv6 addresses, it was
4072 not checking this. This is a hostage to fortune. Exim now panics and dies
4073 if the condition is not met. A case was found where this could be provoked
4074 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4075 components; fortuitously, this particular loophole had already been fixed
4076 by change 4.50/55 or 4.44/33 above.
4078 If there are any other similar loopholes, the new check in host_aton()
4079 itself should stop them being exploited. The report I received stated that
4080 data on the command line could provoke the exploit when Exim was running as
4081 exim, but did not say which command line option was involved. All I could
4082 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4083 running as the user.
4086 35. There was a buffer overflow vulnerability in the SPA authentication code
4087 (which came originally from the Samba project). I have added a test to the
4088 spa_base64_to_bits() function which I hope fixes it.
4091 36. The daemon start-up calls getloadavg() while still root for those OS that
4092 need the first call to be done as root, but it missed one case: when
4093 deliver_queue_load_max is set with deliver_drop_privilege. This is
4094 necessary for the benefit of the queue runner, because there is no re-exec
4095 when deliver_drop_privilege is set.
4097 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4098 was in use, was not putting the data itself into the right store pool;
4099 consequently, it could be overwritten for a subsequent message in the same
4100 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4103 38. Sometimes the final signoff response after QUIT could fail to get
4104 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4105 < 0 before doing a fflush(). This bug looks as though it goes back to the
4106 introduction of TLS in release 3.20, but "sometimes" must have been rare
4107 because the tests only now provoked it.
4113 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4114 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4115 Now it returns an integer. A number of calls were still expecting a T/F
4116 return. Fortuitously, in all cases, the tests worked in OK situations,
4117 which is the norm. However, things would have gone wrong on any write
4118 failures on the smtp file descriptor. This function is used when sending
4119 messages over SMTP and also when doing verify callouts.
4121 2. When Exim is called to do synchronous delivery of a locally submitted
4122 message (the -odf or -odi options), it no longer closes stderr before doing
4125 3. Implemented the mua_wrapper option.
4127 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4129 5. Implemented the functions header_remove(), header_testname(),
4130 header_add_at_position(), and receive_remove_recipient(), and exported them
4133 6. If an ACL "warn" statement specified the addition of headers, Exim already
4134 inserted X-ACL-Warn: at the start if there was no header name. However, it
4135 was not making this test for the second and subsequent header lines if
4136 there were newlines in the string. This meant that an invalid header could
4137 be inserted if Exim was badly configured.
4139 7. Allow an ACL "warn" statement to add header lines at the start or after all
4140 the Received: headers, as well as at the end.
4142 8. Added the rcpt_4xx retry error code.
4144 9. Added postmaster_mailfrom=xxx to callout verification option.
4146 10. Added mailfrom=xxxx to the callout verification option, for verify=
4149 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4150 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4152 12. Inserted some casts to stop certain compilers warning when using pointer
4153 differences as field lengths or precisions in printf-type calls (mostly
4154 affecting debugging statements).
4156 13. Added optional readline() support for -be (dynamically loaded).
4158 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4159 same clock tick as a message's arrival, so that its received time was the
4160 same as the "first fail" time on the retry record, and that message
4161 remained on the queue past the ultimate address timeout, every queue runner
4162 would try a delivery (because it was past the ultimate address timeout) but
4163 after another failure, the ultimate address timeout, which should have then
4164 bounced the address, did not kick in. This was a "< instead of <=" error;
4165 in most cases the first failure would have been in the next clock tick
4166 after the received time, and all would be well.
4168 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4169 being recognized when the domain list was tested by the match_domain
4170 condition in an expansion string.
4172 16. Added the ${str2b64: operator.
4174 17. Exim was always calling setrlimit() to set a large limit for the number of
4175 processes, without checking whether the existing limit was already
4176 adequate. (It did check for the limit on file descriptors.) Furthermore,
4177 errors from getrlimit() and setrlimit() were being ignored. Now they are
4178 logged to the main and panic logs, but Exim does carry on, to try to do its
4179 job under whatever limits there are.
4181 18. Imported PCRE 5.0.
4183 19. Trivial typo in log message " temporarily refused connection" (the leading
4186 20. If the log selector return_path_on_delivery was set and an address was
4187 redirected to /dev/null, the delivery process crashed because it assumed
4188 that a return path would always be set for a "successful" delivery. In this
4189 case, the whole delivery is bypassed as an optimization, and therefore no
4192 21. Internal re-arrangement: the function for sending a challenge and reading
4193 a response while authentication was assuming a zero-terminated challenge
4194 string. It's now changed to take a pointer and a length, to allow for
4195 binary data in such strings.
4197 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4199 23. Exim was not respecting finduser_retries when seeking the login of the
4200 uid under which it was called; it was always trying 10 times. (The default
4201 setting of finduser_retries is zero.) Also, it was sleeping after the final
4202 failure, which is pointless.
4204 24. Implemented tls_on_connect_ports.
4206 25. Implemented acl_smtp_predata.
4208 26. If the domain in control=submission is set empty, Exim assumes that the
4209 authenticated id is a complete email address when it generates From: or
4210 Sender: header lines.
4212 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4213 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4214 chown and chgrp in /bin and hostname in /usr/bin.
4216 28. Exim was keeping the "process log" file open after each use, just as it
4217 does for the main log. This opens the possibility of it remaining open for
4218 long periods when the USR1 signal hits a daemon. Occasional processlog
4219 errors were reported, that could have been caused by this. Anyway, it seems
4220 much more sensible not to leave this file open at all, so that is what now
4223 29. The long-running daemon process does not normally write to the log once it
4224 has entered its main loop, and it closes the log before doing so. This is
4225 so that log files can straightforwardly be renamed and moved. However,
4226 there are a couple of unusual error situations where the daemon does write
4227 log entries, and I had neglected to close the log afterwards.
4229 30. The text of an SMTP error response that was received during a remote
4230 delivery was being truncated at 512 bytes. This is too short for some of
4231 the long messages that one sometimes sees. I've increased the limit to
4234 31. It is now possible to make retry rules that apply only when a message has a
4235 specific sender, in particular, an empty sender.
4237 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4238 it possible to be selective about when SMTP synchronization is enforced.
4240 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4242 32. Implemented hosts_connection_nolog.
4244 33. Added an ACL for QUIT.
4246 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4249 35. Added mailbox_size and mailbox_filecount to appendfile.
4251 36. Added control = no_multiline_responses to ACLs.
4253 37. There was a bug in the logic of the code that waits for the clock to tick
4254 in the case where the clock went backwards by a substantial amount such
4255 that the microsecond fraction of "now" was more than the microsecond
4256 fraction of "then" (but the whole seconds number was less).
4258 38. Added support for the libradius Radius client library this is found on
4259 FreeBSD (previously only the radiusclient library was supported).
4265 1. When certain lookups returned multiple values in the form name=value, the
4266 quoting of the values was not always being done properly. Specifically:
4267 (a) If the value started with a double quote, but contained no whitespace,
4269 (b) If the value contained whitespace other than a space character (i.e.
4270 tabs or newlines or carriage returns) it was not quoted.
4271 This fix has been applied to the mysql and pgsql lookups by writing a
4272 separate quoting function and calling it from the lookup code. The fix
4273 should probably also be applied to nisplus, ibase and oracle lookups, but
4274 since I cannot test any of those, I have not disturbed their existing code.
4276 2. A hit in the callout cache for a specific address caused a log line with no
4277 reason for rejecting RCPT. Now it says "Previous (cached) callout
4278 verification failure".
4280 3. There was an off-by-one bug in the queryprogram router. An over-long
4281 return line was truncated at 256 instead of 255 characters, thereby
4282 overflowing its buffer with the terminating zero. As well as fixing this, I
4283 have increased the buffer size to 1024 (and made a note to document this).
4285 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4286 when Exim is waiting for an SMTP response from a remote server, Exim
4287 restarts its select() call on the socket, thereby resetting its timeout.
4288 This is not a problem when such interrupts are rare. Somebody set up a cron
4289 job to run exiwhat every 2 minutes, which is less than the normal select()
4290 timeout (5 or 10 minutes). This meant that the select() timeout never
4291 kicked in because it was always reset. I have fixed this by comparing the
4292 time when an interrupt arrives with the time at the start of the first call
4293 to select(). If more time than the timeout has elapsed, the interrupt is
4294 treated as a timeout.
4296 5. Some internal re-factoring in preparation for the addition of Sieve
4297 extensions (by MH). In particular, the "personal" test is moved to a
4298 separate function, and given an option for scanning Cc: and Bcc: (which is
4299 not set for Exim filters).
4301 6. When Exim created an email address using the login of the caller as the
4302 local part (e.g. when creating a From: or Sender: header line), it was not
4303 quoting the local part when it contained special characters such as @.
4305 7. Installed new OpenBSD configuration files.
4307 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4308 try to make them clearer.
4310 9. Callout options, other than the timeout value, were being ignored when
4311 verifying sender addresses in header lines. For example, when using
4313 verify = header_sender/callout=no_cache
4315 the cache was (incorrectly) being used.
4317 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4318 only the Cygwin environment.
4320 11. When return_path_on_delivery was set as a log selector, if different remote
4321 addresses in the same message used different return paths and parallel
4322 remote delivery occurred, the wrong values would sometimes be logged.
4323 (Whenever a remote delivery process finished, the return path value from
4324 the most recently started remote delivery process was logged.)
4326 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4327 lines when AUTH and/or TLS are in use. This is the "received protocol"
4328 field. Exim used to use "asmtp" for authenticated SMTP, without any
4329 indication (in the protocol name) for TLS use. Now it follows the RFC and
4330 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4331 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4332 names appear in log lines as well as in Received: header lines.
4334 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4335 extensions, and comparison tests, and to fix some bugs.
4337 14. Changes to the "personal" filter test:
4339 (1) The test was buggy in that it was just doing the equivalent of
4340 "contains" tests on header lines. For example, if a user's address was
4341 anne@some.where, the "personal" test would incorrectly be true for
4343 To: susanne@some.where
4345 This test is now done by extracting each address from the header in turn,
4346 and checking the entire address. Other tests that are part of "personal"
4347 are now done using regular expressions (for example, to check local parts
4348 of addresses in From: header lines).
4350 (2) The list of non-personal local parts in From: addresses has been
4351 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4352 taken from the Sieve specification recommendations.
4354 (3) If the message contains any header line starting with "List-" it is
4355 treated as non-personal.
4357 (4) The test for "circular" in the Subject: header line has been removed
4358 because it now seems ill-conceived.
4360 15. Minor typos in src/EDITME comments corrected.
4362 16. Installed latest exipick from John Jetmore.
4364 17. If headers_add on a router specified a text string that was too long for
4365 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4366 of string_sprintf() is now avoided.
4368 18. $message_body_size was not set (it was always zero) when running the DATA
4369 ACL and the local_scan() function.
4371 19. For the "mail" command in an Exim filter, no default was being set for
4372 the once_repeat time, causing a random time value to be used if "once" was
4373 specified. (If the value happened to be <= 0, no repeat happened.) The
4374 default is now 0s, meaning "never repeat". The "vacation" command was OK
4375 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4376 (I found it when inspecting the code).
4378 20. There is now an overall timeout for performing a callout verification. It
4379 defaults to 4 times the callout timeout, which applies to individual SMTP
4380 commands during the callout. The overall timeout applies when there is more
4381 than one host that can be tried. The timeout is checked before trying the
4382 next host. This prevents very long delays if there are a large number of
4383 hosts and all are timing out (e.g. when the network connections are timing
4384 out). The value of the overall timeout can be changed by specifying an
4385 additional sub-option for "callout", called "maxwait". For example:
4387 verify = sender/callout=5s,maxwait=20s
4389 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4390 to the end before writing, but this should make it even safer).
4392 22. Exim was forgetting that it had advertised PIPELINING for the second and
4393 subsequent messages on an SMTP connection. It was also not resetting its
4394 memory on STARTTLS and an internal HELO.
4396 23. When Exim logs an SMTP synchronization error within a session, it now
4397 records whether PIPELINING has been advertised or not.
4399 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4400 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4401 rather than long int.
4403 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4405 26. Added the never_mail option to autoreply.
4411 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4412 crash if the getsockname() call failed; this can happen if a connection is
4413 closed very soon after it is established. The problem was simply in the
4414 order in which certain operations were done, causing Exim to try to write
4415 to the SMTP stream before it had set up the file descriptor. The bug has
4416 been fixed by making things happen in the correct order.
4422 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4423 before the connection was closed, thus losing the rejection response.
4425 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4426 some early Solaris releases, but causes trouble in current releases where
4427 socklen_t is defined.
4429 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4432 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4435 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4436 that releasing the top of it at the end releases what was used for sub-
4437 expansions (unless the block got too big). However, discard this block if
4438 the first thing is a variable or header, so that we can use its block when
4439 it is dynamic (useful for very large $message_headers, for example).
4441 6. Lookups now cache *every* query, not just the most recent. A new, separate
4442 store pool is used for this. It can be recovered when all lookup caches are
4443 flushed. Lookups now release memory at the end of their result strings.
4444 This has involved some general refactoring of the lookup sources.
4446 7. Some code has been added to the store_xxx() functions to reduce the amount
4447 of flapping under certain conditions.
4449 8. log_incoming_interface used to affect only the <= reception log lines. Now
4450 it causes the local interface and port to be added to several more SMTP log
4451 lines, for example "SMTP connection from", and rejection lines.
4453 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4455 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4457 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4458 could be overwritten at the end of the current message (or the start of a
4459 new message if it was set in a HELO ACL). The value is now preserved for
4460 the duration of the SMTP connection.
4462 12. If a transport had a headers_rewrite setting, and a matching header line
4463 contained an unqualified address, that address was qualified, even if it
4464 did not match any rewriting rules. The underlying bug was that the values
4465 of the flags that permit the existence of unqualified sender and recipient
4466 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4467 non-local messages, and by -bnq for local messages) were not being
4468 preserved with the message after it was received.
4470 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4471 "next input=" as part of the text comprising the host identity instead of
4472 the correct text. The code was using the same buffer for two different
4473 strings. However, depending on which order the printing function evaluated
4474 its arguments, the bug did not always show up. Under Linux, for example, my
4475 test suite worked just fine.
4477 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4478 doesn't work with some older versions of Perl. It has been changed to "my",
4479 which in any case is probably the better facility to use.
4481 15. A really picky compiler found some instances of statements for creating
4482 error messages that either had too many or two few arguments for the format
4485 16. The size of the buffer for calls to the DNS resolver has been increased
4486 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4487 for addresses that have a lot of PTR records. This alleviates a problem; it
4488 does not fully solve it.
4490 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4491 buffer now truncates the list and logs the incident, which is the same
4492 action as happens when Exim is looking up a host name and its aliases.
4493 Previously in this situation something unpredictable would happen;
4494 sometimes it was "internal error: store_reset failed".
4496 18. If a server dropped the connection unexpectedly when an Exim client was
4497 using GnuTLS and trying to read a response, the client delivery process
4498 crashed while trying to generate an error log message.
4500 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4501 string, for example:
4503 warn message = H1: something\nH2: something
4505 the text was added as a single header line from Exim's point of view
4506 though it ended up OK in the delivered message. However, searching for the
4507 second and subsequent header lines using $h_h2: did not work. This has been
4508 fixed. Similarly, if a system filter added multiple headers in this way,
4509 the routers could not see them.
4511 20. Expanded the error message when iplsearch is called with an invalid key to
4512 suggest using net-iplsearch in a host list.
4514 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4515 ACLs are no longer actually imposed (and a message to that effect is
4518 22. If a "gecos" field in a passwd entry contained escaped characters, in
4519 particular, if it contained a \" sequence, Exim got it wrong when building
4520 a From: or a Sender: header from that name. A second bug also caused
4521 incorrect handling when an unquoted " was present following a character
4522 that needed quoting.
4524 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4525 was not being matched caselessly.
4527 24. Arranged for all hyphens in the exim.8 source to be escaped with
4530 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4531 properly. Recipient callout cache records were still being keyed to include
4532 the sender, even when use_sender was set false. This led to far more
4533 callouts that were necessary. The sender is no longer included in the key
4534 when use_sender is false.
4536 26. Added "control = submission" modifier to ACLs.
4538 27. Added the ${base62d: operator to decode base 62 numbers.
4540 28. dnsdb lookups can now access SRV records.
4542 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4543 the configuration file.
4545 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4546 (-v) mode. This makes the output for a verbose queue run more intelligible.
4548 31. Added a use_postmaster feature to recipient callouts.
4550 32. Added the $body_zerocount variable, containing the number of binary zero
4551 bytes in the message body.
4553 33. The time of last modification of the "new" subdirectory is now used as the
4554 "mailbox time last read" when there is a quota error for a maildir
4557 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4559 35. Added +ignore_unknown as a special item in host lists.
4561 36. Code for decoding IPv6 addresses in host lists is now included, even if
4562 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4563 address was recognized as an IP address, but was then not correctly decoded
4564 into binary, causing unexpected and incorrect effects when compared with
4571 1. Very minor rewording of debugging text in manualroute to say "list of
4572 hosts" instead of "hostlist".
4574 2. If verify=header_syntax was set, and a header line with an unqualified
4575 address (no domain) and a large number of spaces between the end of the
4576 name and the colon was received, the reception process suffered a buffer
4577 overflow, and (when I tested it) crashed. This was caused by some obsolete
4578 code that should have been removed. The fix is to remove it!
4580 3. When running in the test harness, delay a bit after writing a bounce
4581 message to get a bit more predictability in the log output.
4583 4. Added a call to search_tidyup() just before forking a reception process. In
4584 theory, someone could use a lookup in the expansion of smtp_accept_max_
4585 per_host which, without the tidyup, could leave open a database connection.
4587 5. Added the variables $recipient_data and $sender_data which get set from a
4588 lookup success in an ACL "recipients" or "senders" condition, or a router
4589 "senders" option, similar to $domain_data and $local_part_data.
4591 6. Moved the writing of debug_print from before to after the "senders" test
4594 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4595 problems for message scanning, either using a data ACL, or using
4596 local_scan() because the Received: header was not generated till after they
4597 were called (in order to set the time as the time of reception completion).
4598 I have revised the way this works. The header is now generated after the
4599 body is received, but before the ACL or local_scan() are called. After they
4600 are run, the timestamp in the header is updated.
4606 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4607 before starting a queue runner without re-exec. This happened only when
4608 deliver_drop_privilege was set or when the Exim user was set to root. The
4609 effect of the bug was that timeouts during subsequent deliveries caused
4610 crashes instead of being properly handled. The handler is now left at its
4611 default (and expected) setting.
4613 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4614 message, again when deliver_drop_privilege is set or Exim is run as root.
4615 The bug described in (1) was not present in this case, but the tidying up
4616 of the other signals was missing. I have made the two cases consistent.
4618 3. The ignore_target_hosts setting on a manualroute router was being ignored
4619 for hosts that were looked up using the /MX notation.
4621 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4624 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4625 operated on the sender address. After changing the $sender_address to <>
4626 for the sender address verify, Exim was re-instated it as the original
4627 (before rewriting) address, but remembering that it had rewritten it, so it
4628 wasn't rewriting it again. This bug also had the effect of breaking the
4629 sender address verification caching when the sender address was rewritten.
4631 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4632 This has been changed so that if the ip literal address matches
4633 ignore_target_hosts, the router declines.
4635 7. Added expansion conditions match_domain, match_address, and match_local_
4636 part (NOT match_host).
4638 8. The placeholder for the Received: header didn't have a length field set.
4640 9. Added code to Exim itself and to exim_lock to test for a specific race
4641 condition that could lead to file corruption when using MBX delivery. The
4642 issue is with the lockfile that is created in /tmp. If this file is removed
4643 after a process has opened it but before that process has acquired a lock,
4644 there is the potential for a second process to recreate the file and also
4645 acquire a lock. This could lead to two Exim processes writing to the file
4646 at the same time. The added code performs the same test as UW imapd; it
4647 checks after acquiring the lock that its file descriptor still refers to
4648 the same named file.
4650 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4651 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4654 11. Added the smtp_active_hostname option. If used, this will typically be made
4655 to depend on the incoming interface address. Because $interface_address is
4656 not set up until the daemon has forked a reception process, error responses
4657 that can happen earlier (such as "too many connections") no longer contain
4660 12. If an expansion in a condition on a "warn" statement fails because a lookup
4661 defers, the "warn" statement is abandoned, and the next ACL statement is
4662 processed. Previously this caused the whole ACL to be aborted.
4664 13. Added the iplsearch lookup type.
4666 14. Added ident_timeout as a log selector.
4668 15. Added tls_certificate_verified as a log selector.
4670 16. Added a global option tls_require_ciphers (compare the smtp transport
4671 option of the same name). This controls incoming TLS connections.
4673 17. I finally figured out how to make tls_require_ciphers do a similar thing
4674 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4675 before starting the TLS session.
4677 18. Tabs are now shown as \t in -bP output.
4679 19. If the log selector return_path_on_delivery was set, Exim crashed when
4680 bouncing a message because it had too many Received: header lines.
4682 20. If two routers both had headers_remove settings, and the first one included
4683 a superfluous trailing colon, the final name in the first list and the
4684 first name in the second list were incorrectly joined into one item (with a
4685 colon in the middle).
4691 1. Added -C and -D options to the exinext utility, mainly to make it easier
4692 to include in the automated testing, but these could be helpful when
4693 multiple configurations are in use.
4695 2. The exinext utility was not formatting the output nicely when there was
4696 an alternate port involved in the retry record key, nor when there was a
4697 message id as well (for retries that were specific to a specific message
4698 and a specific host). It was also confused by IPv6 addresses, because of
4699 the additional colons they contain. I have fixed the IPv4 problem, and
4700 patched it up to do a reasonable job for IPv6.
4702 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4703 delivery, the log line now contains "pipelined" if PIPELINING was used.
4705 4. An SMTP transport process used to panic and die if the bind() call to set
4706 an explicit outgoing interface failed. This has been changed; it is now
4707 treated in the same way as a connect() failure.
4709 5. A reference to $sender_host_name in the part of a conditional expansion
4710 that was being skipped was still causing a DNS lookup. This no longer
4713 6. The def: expansion condition was not recognizing references to header lines
4714 that used bh_ and bheader_.
4716 7. Added the _cache feature to named lists.
4718 8. The code for checking quota_filecount in the appendfile transport was
4719 allowing one more file than it should have been.
4721 9. For compatibility with Sendmail, the command line option
4729 and sets the incoming protocol and host name (for trusted callers). The
4730 host name and its colon can be omitted when only the protocol is to be set.
4731 Note the Exim already has two private options, -pd and -ps, that refer to
4732 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4733 "s", but I don't think that's a major issue.
4735 10. A number of refactoring changes to the code, none of which should affect
4738 (a) The number of logging options was getting close to filling up the
4739 32-bit word that was used as a bit map. I have split them into two classes:
4740 those that are passed in the argument to log_write(), and those that are
4741 only ever tested independently outside of that function. These are now in
4742 separate 32-bit words, so there is plenty of room for expansion again.
4743 There is no change in the user interface or the logging behaviour.
4745 (b) When building, for example, log lines, the code previously used a
4746 macro that called string_cat() twice, in order to add two strings. This is
4747 not really sufficiently general. Furthermore, there was one instance where
4748 it was actually wrong because one of the argument was used twice, and in
4749 one call a function was used. (As it happened, calling the function twice
4750 did not affect the overall behaviour.) The macro has been replaced by a
4751 function that can join an arbitrary number of extra strings onto a growing
4754 (c) The code for expansion conditions now uses a table and a binary chop
4755 instead of a serial search (which was left over from when there were very
4756 few conditions). Also, it now recognizes conditions like "pam" even when
4757 the relevant support is not compiled in: a suitably worded error message is
4758 given if an attempt is made to use such a condition.
4760 11. Added ${time_interval:xxxxx}.
4762 12. A bug was causing one of the ddress fields not to be passed back correctly
4763 from remote delivery subprocesses. The field in question was not being
4764 subsequently used, so this caused to problems in practice.
4766 13. Added new log selectors queue_time and deliver_time.
4768 14. Might have fixed a bug in maildirsizefile handling that threw up
4769 "unexpected character" debug warnings, and recalculated the data
4770 unnecessarily. In any case, I expanded the warning message to give more
4773 15. Added the message "Restricted characters in address" to the statements in
4774 the default ACL that block characters like @ and % in local parts.
4776 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
4777 Three changes have been made:
4779 (a) There was a serious bug; a negative response to MAIL caused the whole
4780 recipient domain to be cached as invalid, thereby blocking all messages
4781 to all local parts at the same domain, from all senders. This bug has
4782 been fixed. The domain is no longer cached after a negative response to
4783 MAIL if the sender used is not empty.
4785 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
4788 (c) A new callout option, "use_sender" has been added for people who want
4789 the modified behaviour.
4795 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
4798 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
4799 indeed breaks things for older releases.
4801 3. Added additional logging to the case where there is a problem reading data
4802 from a filter that is running in a subprocess using a pipe, in order to
4803 try to track down a specific problem.
4805 4. Testing facility fudge: when running in the test harness and attempting
4806 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
4807 getting "No route to host". Convert this to a timeout.
4809 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
4812 6. Some OS don't have socklen_t but use size_t instead. This affects the
4813 fifth argument of getsockopt() amongst other things. This is now
4814 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
4815 can be set for individual OS. I have set it for SunOS5, OSF1, and
4816 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
4817 some earlier ones do not.
4819 7. Change 4.30/15 was not doing the test caselessly.
4821 8. The standard form for an IPv6 address literal was being rejected by address
4822 parsing in, for example, MAIL and RCPT commands. An example of this kind of
4823 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4824 this, as well as the form without the "IPv6" on the front (but only when
4825 address literals are enabled, of course).
4827 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
4829 10. Exim crashed if a message with an empty sender address specified by -f
4830 encountered a router with an errors_to setting. This could be provoked only
4831 by a command such as
4835 where an empty string was supplied; "<>" did not hit this bug.
4837 11. Installed PCRE release 4.5.
4839 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
4840 remained set. It is now erased.
4842 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
4843 times from message ids (which are base 36 rather than the normal 62).
4845 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
4846 were being counted as actual protocol errors, and logged if the log
4847 selector +smtp_protocol_error was set. One cannot be perfect in this test,
4848 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
4849 and DATA following a set of rejected RCPTs do not count as protocol errors.
4850 In other words, Exim assumes they were pipelined, though this may not
4851 actually be the case. Of course, in all cases the client gets an
4852 appropriate error code.
4854 15. If a lookup fails in an ACL condition, a message about the failure may
4855 be available; it is used if testing the ACL cannot continue, because most
4856 such messages specify what the cause of the deferral is. However, some
4857 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
4858 that caused an old message to be retained and used if a later statement
4859 caused a defer, replacing the real cause of the deferral.
4861 16. If an IP address had so many PTR records that the DNS lookup buffer
4862 was not large enough to hold them, Exim could crash while trying to process
4863 the truncated data. It now detects and logs this case.
4865 17. Further to 4.21/58, another change has been made: if (and only if) the
4866 first line of a message (the first header line) ends with CRLF, a bare LF
4867 in a subsequent header line has a space inserted after it, so as not to
4868 terminate the header.
4870 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
4871 unnecessarily, used atoi() instead of strtol(), and didn't check the
4872 termination when getting file sizes from file names by regex.
4874 19. Completely re-implemented the support for maildirsize files, in the light
4875 of a number of problems with the previous contributed implementation
4876 (4.30/29). In particular:
4878 . If the quota is zero, the maildirsize file is maintained, but no quota is
4881 . If the maildir directory does not exist, it is created before any attempt
4882 to write a maildirsize file.
4884 . The quota value in the file is just a cache; if the quota is changed in
4885 the transport, the new value overrides.
4887 . A regular expression is available for excluding directories from the
4890 20. The autoreply transport checks the characters in options that define the
4891 message's headers; it allows continued headers, but it was checking with
4892 isspace() after an embedded newline instead of explicitly looking for a
4895 21. If all the "regular" hosts to which an address was routed had passed their
4896 expiry times, and had not reached their retry times, the address was
4897 bounced, even if fallback hosts were defined. Now Exim should go on to try
4900 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4901 equivalent code in the SMTP transport. Some hosts send humungous responses
4902 to HELO/EHLO, more than 1024 it seems.
4904 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4905 gives compiler warnings in some environments. I've now done it "properly",
4908 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4909 (because of problems with the built-in one) was declared to return uschar *
4910 instead of char *, causing compiler failure.
4912 25. Fixed a file descriptor leak when processing alias/forward files.
4914 26. Fixed a minor format string issue in dbfn.c.
4916 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4918 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4919 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4922 29. When a sender address is verified, it is cached, to save repeating the test
4923 when there is more than one recipient in a message. However, when the
4924 verification involves a callout, it is possible for different callout
4925 options to be set for different recipients. It is too complicated to keep
4926 track of this in the cache, so now Exim always runs a verification when a
4927 callout is required, relying on the callout cache for the optimization.
4928 The overhead is duplication of the address routing, but this should not be
4931 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4932 to be verified with callout=postmaster, and the main callout worked but the
4933 postmaster check failed, the verification correctly failed. However, if a
4934 subsequent RCPT command asked for sender verification *without* the
4935 postmaster check, incorrect caching caused this verification also to fail,
4938 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4939 it was not caching the DNS options (qualify_single, search_parents) that
4940 were used when the lookup failed. A subsequent lookup with different
4941 options therefore always gave the same answer, though there were cases
4942 where it should not have. (Example: a "domains = !$mx_any" option on a
4943 dnslookup router: the "domains" option is always processed without any
4944 widening, but the router might have qualify_single set.) Now Exim uses the
4945 cached value only when the same options are set.
4947 32. Added John Jetmore's "exipick" utility to the distribution.
4949 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4950 than a timeout (e.g. a certificate is required, and is not provided), an
4951 Exim server now closes the connection immediately. Previously it waited for
4952 the client to close - but if the client is SSL, it seems that they each
4953 wait for each other, leading to a delay before one of them times out.
4955 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4956 maintained 0.8.x compatibility because I don't think many are using it, and
4957 it is clearly obsolete.
4959 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4962 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4963 client certificate was expired. A simple patch fixes this, though I don't
4964 understand the full logic of why the verify callback is called multiple
4967 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4968 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4969 which causes problems with some clients (such as the Certicom SSL Plus
4970 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4971 disables the coutermeasure allowing Eudora to connect."
4973 38. Exim was not checking that a write() to a log file succeeded. This could
4974 lead to Bad Things if a log got too big, in particular if it hit a file
4975 size limit. Exim now panics and dies if it cannot write to a log file, just
4976 as it does if it cannot open a log file.
4978 39. Modified OS/Makefile-Linux so that it now contains
4980 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4982 The two -D definitions ensure that Exim is compiled with large file
4983 support, which makes it possible to handle log files that are bigger than
4986 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4987 instance) a domain was checked against a named list that involved a lookup,
4988 causing $domain_data to be set, then another domain was checked against the
4989 same list, then the first domain was re-checked, the value of $domain_data
4990 after the final check could be wrong. In particular, if the second check
4991 failed, it could be set empty. This bug probably also applied to
4994 41. The strip_trailing_dot option was not being applied to the address given
4995 with the -f command-line option.
4997 42. The code for reading a message's header from the spool was incrementing
4998 $received_count, but never initializing it. This meant that the value was
4999 incorrect (doubled) while delivering a message in the same process in which
5000 it was received. In the most common configuration of Exim, this never
5001 happens - a fresh exec is done - but it can happen when
5002 deliver_drop_privilege is set.
5004 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5005 it now includes up to 150 characters of the unexpected data in the log
5008 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5009 and building data strings. The size of both of these buffers was 10 000
5010 bytes - far larger than anybody would *ever* want, thought I. Needless to
5011 say, somebody hit the limit. I have increased the maximum line length to
5012 20 000 and the maximum data length of concatenated lines to 100 000. I have
5013 also fixed two bugs, because there was no checking on these buffers. Tsk,
5014 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5015 buffer is too small.
5017 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5018 lsearch lookups. Now it does.
5020 46. When parsing a route_list item in a manualroute router, a fixed-length
5021 buffer was used for the list of hosts. I made this 1024 bytes long,
5022 thinking that nobody would ever have a list of hosts that long. Wrong.
5023 Somebody had a whole pile of complicated expansion conditions, and the
5024 string was silently truncated, leading to an expansion error. It turns out
5025 that it is easier to change to an unlimited length (owing to other changes
5026 that have happened since this code was originally written) than to build
5027 structure for giving a limitation error. The length of the item that
5028 expands into the list of hosts is now unlimited.
5030 47. The lsearch lookup could not handle data where the length of text line was
5031 more than 4095 characters. Such lines were truncated, leading to shortened
5032 data being returned. It should now handle lines of any length.
5034 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5035 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5038 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5039 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5040 no longer happens in comment lines. A list of replacements is now placed
5041 at the head of all of the source files, except those whose only change is
5042 to replace PERL_COMMAND in the very first #! line.
5044 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5045 messages on the queue, with a bottom-up merge sort, using code contributed
5046 by Michael Haardt. This should make operations like -bp somewhat faster on
5047 large queues. It won't affect queue runners, except when queue_run_in_order
5050 51. Installed eximstats 1.31 in the distribution.
5052 52. Added support for SRV lookups to the dnslookup router.
5054 53. If an ACL referred to $message_body or $message_body_end, the value was not
5055 reset for any messages that followed in the same SMTP session.
5057 54. The store-handling optimization for building very long strings was not
5058 differentiating between the different store pools. I don't think this
5059 actually made any difference in practice, but I've tidied it.
5061 55. While running the routers to verify a sender address, $sender_address
5062 was still set to the sender address. This is wrong, because when routing to
5063 send a bounce to the sender, it would be empty. Therefore, I have changed
5064 it so that, while verifying a sender address, $sender_address is set to <>.
5065 (There is no change to what happens when verifying a recipient address.)
5067 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5068 target A or AAAA records (if not already returned) without resetting the
5069 qualify_single or search_parents options of the DNS resolver. These are
5070 inappropriate in this case because the targets of MX and SRV records must
5071 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5072 target that, when qualified, matched something in the local domain. These
5073 two options are now turned off when doing these lookups.
5075 57. It seems that at least some releases of Reiserfs (which does not have the
5076 concept of a fixed number of inodes) returns zero and not -1 for the
5077 number of available inodes. This interacted badly with check_spool_inodes,
5078 which assumed that -1 was the "no such thing" setting. What I have done is
5079 to check that the total number of inodes is greater than zero before doing
5080 the test of how many are available.
5082 58. When a "warn" ACL statement has a log_message modifier, the message is
5083 remembered, and not repeated. This is to avoid a lot of repetition when a
5084 message has many recipients that cause the same warning to be written.
5085 Howewer, Exim was preserving the list of already written lines for an
5086 entire SMTP session, which doesn't seem right. The memory is now reset if a
5087 new message is started.
5089 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5090 debugging output unless log_rewrite was also set.
5092 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5093 of (int)(handle) when we know that handle contains (void *)(-1).
5095 61. The Exim daemon panic-logs an error return when it closes the incoming
5096 connection. However "connection reset by peer" seems to be common, and
5097 isn't really an error worthy of noting specially, so that particular error
5100 62. When Exim is trying to find all the local interfaces, it used to panic and
5101 die if the ioctl to get the interface flags failed. However, it seems that
5102 on at least one OS (Solaris 9) it is possible to have an interface that is
5103 included in the list of interfaces, but for which you get a failure error
5104 for this call. This happens when the interface is not "plumbed" into a
5105 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5106 failure of the "get flags" call assumes that the interface is down.
5108 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5109 makes life easier for people who are doing arithmetic on fields extracted
5110 from dates, where you often get leading zeros that should not be
5111 interpreted as octal.
5113 64. Added qualify_domain to the redirect router, to override the global
5116 65. If a pathologically long header line contained very many addresses (the
5117 report of this problem mentioned 10 000) and each of them was rewritten,
5118 Exim could use up a very large amount of memory. (It kept on making new
5119 copies of the header line as it rewrote, and never released the old ones.)
5120 At the expense of a bit more processing, the header rewriting function has
5121 been changed so that it no longer eats memory in this way.
5123 66. The generation of the Received: header has been moved from the time that a
5124 message starts to be received, to the time that it finishes. The timestamp
5125 in the Received: header should now be very close to that of the <= log
5126 line. There are two side-effects of this change:
5128 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5129 logged header lines no longer include the local Received: line, because
5130 it has not yet been created. The same applies to a copy of the message
5131 that is returned to a non-SMTP sender when a message is rejected.
5133 (b) When a filter file is tested using -bf, no additional Received: header
5134 is added to the test message. After some thought, I decided that this
5137 This change does not affect the value of $received_for. It is still set
5138 after address rewriting, but before local_scan() is called.
5140 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5142 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5143 gave an unhelpful panic error message, and a defer error. I have managed to
5144 change this behaviour so that it now rejects any supplied certificate,
5145 which seems right, as the list of acceptable certificates is empty.
5147 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5148 gave an unhelpful defer error. I have not managed to make this reject any
5149 supplied certificates, but the error message it gives is "no certificate
5150 supplied", which is not helpful.
5152 70. exigrep's output now also includes lines that are not associated with any
5153 message, but which match the given pattern. Implemented by a patch from
5154 Martin Sluka, which also tidied up the Perl a bit.
5156 71. Recipient callout verification, like sender verification, was using <> in
5157 the MAIL FROM command. This isn't really the right thing, since the actual
5158 sender may affect whether the remote host accepts the recipient or not. I
5159 have changed it to use the actual sender in the callout; this means that
5160 the cache record is now keyed on a recipient/sender pair, not just the
5161 recipient address. There doesn't seem to be a real danger of callout loops,
5162 since a callout by the remote host to check the sender would use <>.
5163 [SEE ABOVE: changed after hitting problems.]
5165 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5166 temporary errors. However, in the case of such a code being given after
5167 the end of a data transmission (i.e. after ".") Exim was failing to write
5168 a retry record for the message. (Yes, there was some broken host that was
5169 actually sending 8xx at this point.)
5171 73. An unknown lookup type in a host list could cause Exim to panic-die when
5172 the list was checked. (An example that provoked this was putting <; in the
5173 middle of a list instead of at the start.) If this happened during a DATA
5174 ACL check, a -D file could be left lying around. This kind of configuration
5175 error no longer causes Exim to die; instead it causes a defer errror. The
5176 incident is still logged to the main and panic logs.
5178 74. Buglet left over from Exim 3 conversion. The message "too many messages
5179 in one connection" was written to the rejectlog but not the mainlog, except
5180 when address rewriting (yes!) was being logged.
5182 75. Added write_rejectlog option.
5184 76. When a system filter was run not as root (that is, when system_filter_user
5185 was set), the values of the $n variables were not being returned to the
5186 main process; thus, they were not subsequently available in the $sn
5189 77. Added +return_path_on_delivery log selector.
5191 78. A connection timeout was being treated differently from recipients deferred
5192 when testing hosts_max_try with a message that was older than the host's
5193 retry timeout. (The host should not be counted, thus allowing all hosts to
5194 be tried at least once before bouncing.) This may have been the cause of an
5195 occasionally reported bug whereby a message would remain on the queue
5196 longer than the retry timeout, but would be bounced if a delivery was
5197 forced. I say "may" because I never totally pinned down the problem;
5198 setting up timeout/retry tests is difficult. See also the next item.
5200 79. The ultimate address timeout was not being applied to errors that involved
5201 a combination of host plus message (for example, a timeout on a MAIL
5202 command). When an address resolved to a number of possible hosts, and they
5203 were not all tried for each delivery (e.g. because of hosts_max_try), a
5204 message could remain on the queue longer than the retry timeout.
5206 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5209 81. Fixed an obscure SMTP outgoing bug which required at least the following
5210 conditions: (a) there was another message waiting for the same server;
5211 (b) the server returned 5xx to all RCPT commands in the first message so
5212 that the message was not completed; (c) the server dropped the connection
5213 or gave a negative response to the RSET that Exim sends to abort the
5214 transaction. The observed case was a dropped connection after DATA that had
5215 been sent in pipelining mode. That is, the server had advertised PIPELINING
5216 but was not implementing it correctly. The effect of the bug was incorrect
5217 behaviour, such as trying another host, and this could lead to a crash.
5223 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5224 and daemon.c were passed as pointers to ints; they should have been
5225 pointers to socklen_t variables (which are typically unsigned ints).
5227 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5230 3. Fixed a really odd bug that affected only the testing scheme; patching a
5231 certain fixed string in the binary changed the value of another string that
5232 happened to be identical to the end of the original first string.
5234 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5235 name", it returns that address as the IP address. On some operating
5236 systems (e.g. Solaris), it also passes back the IP address string as the
5237 "host name". However, on others (e.g. Linux), it passes back an empty
5238 string. Exim wasn't checking for this, and was changing the host name to an
5239 empty string, assuming it had been canonicized.
5241 5. Although rare, it is permitted to have more than one PTR record for a given
5242 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5243 all the names associated with an address, because they do in Solaris.
5244 However, it seems that they do not in Linux for data that comes from the
5245 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5246 I found this out when I moved to a new Linux workstation and tried to run
5247 the Exim test suite.
5249 To get round this problem I have changed the code so that it now does its
5250 own call to the DNS to look up PTR records when searching for a host name.
5251 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5252 addresses that are only in /etc/hosts are still found.
5254 This behaviour is, however, controlled by an option called host_lookup_
5255 order, which defaults to "bydns:byaddr". If people want to use the other
5256 order, or indeed, just use one or the other means of lookup, they can
5257 specify it in this variable.
5259 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5260 some operating systems, this comes back from gethostbyaddr() as an empty
5261 string, and this is what Exim used to test for. However, it seems that in
5262 other systems, "." is yielded. Exim now tests for this case too.
5264 7. The values of check_spool_space and check_log_space are now held internally
5265 as a number of kilobytes instead of an absolute number of bytes. If a
5266 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5267 kilobyte. This means that much larger values can be stored.
5269 8. Exim monitor: an attempt to get the action menu when not actually pointing
5270 at a message produces an empty menu entitled "No message selected". This
5271 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5272 no entries in it ("Shell widget menu has zero width and/or height"). So I
5273 have added a single, blank menu entry in this case.
5275 9. Added ${quote_local_part.
5277 10. MIME decoding is now applied to the contents of Subject: header lines when
5280 11. Now that a reference to $sender_host_address automatically causes a reverse
5281 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5282 host lookup before query-style lookups in lists that might use this
5283 variable. This has therefore been abolished, and the "net-" prefix is no
5284 longer necessary for query-style lookups.
5286 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5287 have been a typo for LFLAGS, so it has been changed.
5289 13. The install script calls Exim with "-C /dev/null" in order to find the
5290 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5291 to be output. Howeve, since Exim outputs its version number before the
5292 error, it didn't break the script. It just looked ugly. I fixed this by
5293 always allowing "-C /dev/null" if the caller is root.
5295 14. Ignore overlarge ACL variable number when reading spool file - insurance
5296 against a later release with more variables having written the file.
5298 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5299 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5300 this, as well as the form without the "IPv6" on the front.
5302 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5303 OS/Makefile-Darwin file.
5305 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5306 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5308 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5309 variable. However, some OS use ints for inodes. I've added cast to long int
5310 to get rid of the compiler warning.
5312 19. I had forgotten to lock out "/../" in configuration file names when
5313 ALT_CONFIG_PREFIX was set.
5315 20. Routers used for verification do not need to specify transports. However,
5316 if such a router generated a host list, and callout was configured, Exim
5317 crashed, because it could not find a port number from the (non-existent)
5318 transport. It now assumes port 25 in this circumstance.
5320 21. Added the -t option to exigrep.
5322 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5323 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5324 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5325 them is set, LOOKUP_LSEARCH is forced.
5327 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5329 24. Added sender and host information to the "rejected by local_scan()" log
5330 line; previously there was no indication of these.
5332 25. Added .include_if_exists.
5334 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5335 Linux. It turns out that not all file systems support this. Apparently some
5336 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5337 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5338 not supported on the file descriptor, is now ignored when Exim is trying to
5339 sync a directory. This applies only to Linux.
5341 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5343 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5344 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5347 29. Added support for maildirsize files from supplied patch (modified a bit).
5349 30. The use of :fail: followed by an empty string could lead Exim to respond to
5350 sender verification failures with (e.g.):
5352 550 Verification failed for <xxx>
5353 550 Sender verify failed
5355 where the first response line was missing the '-' that indicates it is not
5356 the final line of the response.
5358 31. The loop for finding the name of the user that called Exim had a hardwired
5359 limit of 10; it now uses the value of finduser_retries, which is used for
5360 all other user lookups.
5362 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5365 33. Exim was neglecting to zero errno before one call of strtol() when
5366 expanding a string and expecting an integer value. On some systems this
5367 resulted in spurious "integer overflow" errors. Also, it was casting the
5368 result into an int without checking.
5370 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5371 did not work. The code looks as if it has *never* worked, though it appears
5372 to have been documented since at least releast 1.62. I have made it work.
5374 35. The "timeout_DNS" error in retry rules, also documented since at least
5375 1.62, also never worked. As it isn't clear exactly what this means, and
5376 clearly it isn't a major issue, I have abolished the feature by treating it
5377 as "timeout", and writing a warning to the main and panic logs.
5379 36. The display of retry rules for -brt wasn't always showing the error code
5382 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5383 timeout_connect_A, timeout_connect_MX.
5385 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5386 to the empty sender.
5388 39. The daemon was not analysing the content of -oX till after it had closed
5389 stderr and disconnected from the controlling terminal. This meant that any
5390 syntax errors were only noted on the panic log, and the return code from
5391 the command was 0. By re-arranging the code a little, I've made the
5392 decoding happen first, so such errors now appear on stderr, and the return
5393 code is 1. However, the actual setting up of the sockets still happens in
5394 the disconnected process, so errors there are still only recorded on the
5397 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5398 connections (as happens on some IP stacks) was logged at start up time as
5399 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5400 it from "IPv6 and IPv4", which means that two separate sockets are being
5403 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5404 says whether AF_INET or AF_INET6 was passed as an argument.
5406 42. Exiwhat output was messed up when time zones were included in log
5409 43. Exiwhat now gives more information about the daemon's listening ports,
5410 and whether -tls-on-connect was used.
5412 44. The "port" option of the smtp transport is now expanded.
5414 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5415 silently ignored. Now an error message is written to the main and panic
5418 46. There's a new ACL modifier called "logwrite" which writes to a log file
5419 as soon as it is encountered.
5421 47. Added $local_user_uid and $local_user_gid at routing time.
5423 48. Exim crashed when trying to verify a sender address that was being
5426 49. Exim was recognizing only a space character after ".include". It now also
5427 recognizes a tab character.
5429 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5430 extracting the relevant information from the specification. The man page no
5431 longer contains scrambled data for the -d option, and I've added a section
5432 at the front about calling Exim under different names.
5434 51. Added "extra_headers" argument to the "mail" command in filter files.
5436 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5439 53. Installed eximstats 1.29.
5441 54. Added transport_filter_timeout as a generic transport option.
5443 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5444 Cc: header lines. This was required by RFC 822, but it not required by RFC
5447 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5448 incoming messages that did not have them. Now it does so only if the
5449 message originates locally, that is, if there is no associated remote host
5450 address. When Resent- header lines are present, this applies to the Resent-
5451 lines rather than the non-Resent- lines.
5453 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5454 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5456 58. Messages for configuration errors now include the name of the main
5457 configuration file - useful now that there may be more than one file in a
5458 list (.included file names were always shown).
5460 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5461 for those rare installations that do not start the daemon as root or run it
5462 setuid root. I've cut out the call to initgroups() if the daemon is not
5465 60. The Exim user and group can now be bound into the binary as text strings
5466 that are looked up at the start of Exim's processing.
5468 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5470 62. Added $mailstore_basename variable.
5472 63. Installed patch to sieve.c from Michael Haardt.
5474 64. When Exim failed to open the panic log after failing to open the main log,
5475 the original message it was trying to log was written to stderr and debug
5476 output, but if they were not available (the usual case in production), it
5477 was lost. Now it is written to syslog before the two lines that record the
5478 failures to open the logs.
5480 65. Users' Exim filters run in subprocesses under the user's uid. It is
5481 possible for a "deliver" command or an alias in a "personal" command to
5482 provoke an address rewrite. If logging of address rewriting is configured,
5483 this fails because the process is not running as root or exim. There may be
5484 a better way of dealing with this, but for the moment (because 4.30 needs
5485 to be released), I have disabled address rewrite logging when running a
5486 filter in a non-root, non-exim process.
5492 1. The buildconfig auxiliary program wasn't quoting the value set for
5493 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5494 not defined. This bug was masked in 4.22 by the effect that was fixed in
5497 2. Some messages that were rejected after a message id was allocated were
5498 shown as "incomplete" by exigrep. It no longer does this for messages that
5499 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5501 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5502 have allow_domain_literals set, the ID did not get logged in the <= line.
5503 Domain literals are now always recognized in Message-ID: header lines.
5505 4. The first argument for a ${extract expansion item is the key name or field
5506 number. Leading and trailing spaces in this item were not being ignored,
5507 causing some misleading effects.
5509 5. When deliver_drop_privilege was set, single queue runner processes started
5510 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5511 same command in the process it spins off) were not dropping privilege.
5513 6. When the daemon running as "exim" started a queue runner, it always
5514 re-executed Exim in the spun-off process. This is a waste of effort when
5515 deliver_drop_privilege is set. The new process now just calls the
5516 queue-runner function directly.
5522 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5525 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5526 ignored. Though the use of -oP was forcing the writing of a pid file, it
5527 was always written to the default place.
5529 3. If the message "no IP address found for host xxxx" is generated during
5530 incoming verification, it is now followed by identification of the incoming
5531 connection (so you can more easily find what provoked it).
5533 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5535 5. Added some features to "harden" Exim a bit more against certain attacks:
5537 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5538 be put in Local/Makefile. This is like the never_users runtime option,
5539 but it cannot be overridden. The default setting is "root".
5541 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5542 prefix string with which any file named in a -C command line option
5545 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5546 is retained for -C and -D only if the caller of Exim is root. Without
5547 it, the exim user may also use -C and -D and retain privilege.
5549 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5550 command line option is disabled.
5552 6. Macro names set by the -D option must start with an upper case letter, just
5553 like macro names defined in the configuration file.
5555 7. Added "dereference=" facility to LDAP.
5557 8. Two instances of the typo "uknown" in the source files are fixed.
5559 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5560 the Configure-Makefile script screwed up while processing it.
5562 10. Incorporated PCRE 4.4.
5564 11. The SMTP synchronization check was not operating right at the start of an
5565 SMTP session. For example, it could not catch a HELO sent before the client
5566 waited for the greeting. There is now a check for outstanding input at the
5567 point when the greeting is written. Because of the duplex, asynchronous
5568 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5569 way, but not yet received, when the check is performed.
5571 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5572 on TCP/IP sockets, because this apparently causes some broken clients to
5575 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5576 unchanged) from the Cygwin maintainer.
5578 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5579 was supported instead of testing for mbx. Effectively a typo.
5581 15. The spa authenticator server code was not checking that the input it
5582 received was valid base64.
5584 16. The debug output line for the "set" modifier in ACLs was not showing the
5585 name of the variable that was being set.
5587 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5589 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5590 lookup on its own. Something else was needed to trigger the lookup. For
5591 example, a match in host_lookup or the need for a host name in a host list.
5592 Now, if $sender_host_name is referenced and the host name has not yet been
5593 looked up, a lookup is performed. If the lookup fails, the variable remains
5594 empty, and $host_lookup_failed is set to "1".
5596 19. Added "eqi" as a case-independent comparison operator.
5598 20. The saslauthd authentication condition could segfault if neither service
5599 nor realm was specified.
5601 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5602 error message that was logged was misleading, and incoming SMTP
5603 connections were dropped. The message is now more accurate, and temporary
5604 errors are given to SMTP connections.
5606 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5607 (except RSET) with a 421 error, until QUIT is received. However, it was
5608 failing to send a response to QUIT.
5610 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5611 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5612 ACL is now run after the helo_try_verify_hosts code.
5614 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5615 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5616 case-independent, so other case variants are also recognized). Apparently
5617 some systems use these upper case variants.
5619 25. If more than two messages were waiting for the same host, and a transport
5620 filter was specified for the transport, Exim sent two messages over the
5621 same TCP/IP connection, and then failed with "socket operation on non-
5622 socket" when it tried to send the third.
5624 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5626 27. The extern definition of crypt16() in expand.c was not being excluded when
5627 the OS had its own crypt16() function.
5629 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5630 as a preferred synonym for return_size_limit, both as an option and as an
5633 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5635 30. Changed the default configuration ACL to relax the local part checking rule
5636 for addresses that are not in any local domains. For these addresses,
5637 slashes and pipe symbols are allowed within local parts, but the sequence
5638 /../ is explicitly forbidden.
5640 31. SPA server authentication was not clearing the challenge buffer before
5643 32. log_message in a "warn" ACL statement was writing to the reject log as
5644 well as to the main log, which contradicts the documentation and doesn't
5645 seem right (because no rejection is happening). So I have stopped it.
5647 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5648 However, I am unable to do any testing of this.
5650 34. Fixed an infelicity in the appendfile transport. When checking directories
5651 for a mailbox, to see if any needed to be created, it was accidentally
5652 using path names with one or more superfluous leading slashes; tracing
5653 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5655 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5656 contents are added to the log line that is written for every discarded
5657 recipient. (Previously a log_message setting was ignored.)
5659 36. The ${quote: operator now quotes the string if it is empty.
5661 37. The install script runs exim in order to find its version number. If for
5662 some reason other than non-existence or emptiness, which it checks, it
5663 could not run './exim', it was installing it with an empty version number,
5664 i.e. as "exim-". This error state is now caught, and the installation is
5667 38. An argument was missing from the function that creates an error message
5668 when Exim fails to connect to the socket for saslauthd authentication.
5669 This could cause Exim to crash, or give a corrupted message.
5671 39. Added isip, isip4, and isip6 to ${if conditions.
5673 40. The ACL variables $acl_xx are now saved with the message, and can be
5674 accessed later in routers, transports, and filters.
5676 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5677 strings in the file are not string-expanded.
5679 42. If a MAIL command specified a SIZE value that was too large to fit into an
5680 int variable, the check against message_size_limit failed. Such values are
5681 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5682 day this will have to be increased, but I don't think I want to be around
5683 when emails are that large.
5690 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5691 iconv() is not standard in FreeBSD.
5693 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5694 IPv6 enabled. The observed symptom was a segmentation fault on return from
5695 the function os_common_find_running_interfaces() in src/os.c.
5697 3. In the check_special_case() function in daemon.c I had used "errno" as an
5698 argument name, which causes warnings on some systems. This was basically a
5699 typo, since it was named "eno" in the comments!
5701 4. The code that waits for the clock to tick (at a resolution of some fraction
5702 of a second) so as to ensure message-id uniqueness was always waiting for
5703 at least one whole tick, when it could have waited for less. [This is
5704 almost certainly not relevant at current processor speeds, where it is
5705 unlikely to ever wait at all. But we try to future-proof.]
5707 5. The function that sleeps for a time interval that includes fractions of a
5708 second contained a race. It did not block SIGALRM between setting the
5709 timer, and suspending (a couple of lines later). If the interval was short
5710 and the sigsuspend() was delayed until after it had expired, the suspension
5711 never ended. On busy systems this could lead to processes getting stuck for
5714 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5715 process, before it forks any delivery processes. The open lookup caching
5716 mechanism meant that the open file or database connection was passed into
5717 the delivery process. The problem was that delivery processes always tidy
5718 up cached lookup data. This could cause a problem for the next delivery
5719 process started by the queue runner, because the external queue runner
5720 process does not know about the closure. So the next delivery process
5721 still has data in the lookup cache. In the case of a file lookup, there was
5722 no problem because closing a file descriptor in a subprocess doesn't affect
5723 the parent. However, if the lookup was caching a connection to a database,
5724 the connection was closed, and the second delivery process was likely to
5725 see errors such as "PGSQL: query failed: server closed the connection
5726 unexpectedly". The problem has been fixed by closing all cached lookups
5727 in a queue runner before running a delivery process.
5729 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5730 seem to have the "const" qualifier which it has on other OS. I've
5733 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5734 *of the same type* (client or server) with the same public name that an
5735 error should be diagnosed.
5737 9. When Exim looked up a host name for an IP address, but failed to find the
5738 original IP address when looking up the host name (a safety check), it
5739 output the message "<ip address> does not match any IP for NULL", which was
5740 confusing, to say the least. The bug was that the host name should have
5741 appeared instead of "NULL".
5743 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5744 user that is built into the binary, and the -C or -D options is used, root
5745 privilege is dropped before the configuration file is read. In addition,
5746 logging is switched to stderr instead of the normal log files. If the
5747 configuration then re-defines the Exim user, the unprivileged environment
5748 is probably not what is expected, so Exim logs a panic warning message (but
5751 However, if deliver_drop_privilege is set, the unprivileged state may well
5752 be exactly what is intended, so the warning has been cut out in that case,
5753 and Exim is allowed to try to write to its normal log files.
5759 1. smtp_return_error_details was not giving details for temporary sender
5760 or receiver verification errors.
5762 2. Diagnose a configuration error if two authenticators have the same public
5765 3. Exim used not to create the message log file for a message until the first
5766 delivery attempt. This could be confusing when incoming messages were held
5767 for policy or load reasons. The message log file is now created at the time
5768 the message is received, and an initial "Received" line is written to it.
5770 4. The automatically generated man page for command line options had a minor
5771 bug that caused no ill effects; however, a more serious problem was that
5772 the procedure for building the man page automatically didn't always
5773 operate. Consequently, release 4.20 contains an out-of-date version. This
5774 shouldn't happen again.
5776 5. When building Exim with embedded Perl support, the script that builds the
5777 Makefile was calling 'perl' to find its compile-time parameters, ignoring
5778 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
5780 6. The freeze_tell option was not being used for messages that were frozen on
5781 arrival, either by an ACL or by local_scan().
5783 7. Added the smtp_incomplete_transaction log selector.
5785 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
5786 was accepting AUTH without a new EHLO.
5788 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
5789 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
5792 10. Logging of TCP/IP connections (when configured) now happens in the main
5793 daemon process instead of the child process, so that the TCP/IP connection
5794 count is more accurate (but it can never be perfect).
5796 11. The use of "drop" in a nested ACL was not being handled correctly in the
5797 outer ACL. Now, if condition failure induced by the nested "drop" causes
5798 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
5799 or "require"), the connection is dropped.
5801 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
5802 that yield "discard" can now be used with an "accept" or a "discard" verb,
5803 but an error is generated for any others (because I can't see a useful way
5804 to define what should happen).
5806 13. When an ACL is read dynamically from a file (or anywhere else), the lines
5807 are now processed in the same way as lines in the Exim configuration file.
5808 In particular, continuation lines are supported.
5810 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
5812 15. Added -ti meaning -t -i.
5814 16. Check for letters, digits, hyphens, and dots in the names of dnslist
5815 domains, and warn by logging if others are found.
5817 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
5818 returned from GIFCONF when Exim is trying to find the list of interfaces on
5819 a host. The code in os.c has been modified to copy each ifreq to an aligned
5820 structure in all cases.
5822 Also, in some cases, the returned ifreq's were being copied to a 'struct
5823 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
5824 means the last couple of bytes of an IPv6 address could be chopped if the
5825 ifreq contained only a normal sockaddr (14 bytes storage).
5827 18. Named domain lists were not supported in the hosts_treat_as_local option.
5828 An entry such as +xxxx was not recognized, and was treated as a literal
5831 19. Ensure that header lines added by a DATA ACL are included in the reject log
5832 if the ACL subsequently rejects the message.
5834 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
5835 MD5 (which is deprecated).
5837 21. When testing a filter file using -bf, Exim was writing a message when it
5838 took the sender from a "From " line in the message, but it was not doing so
5839 when it took $return_path from a Return-Path: header line. It now does.
5841 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
5842 with a valid header line field name (a series of printing characters
5843 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
5845 23. Changed "disc" in the source to "disk" to conform to the documentation and
5846 the book and for uniformity.
5848 24. Ignore Sendmail's -Ooption=value command line item.
5850 25. When execve() failed while trying to run a command in a pipe transport,
5851 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
5852 could be confused with a return value of 69 from the command itself. This
5853 has been changed to 127, the value the shell returns if it is asked to run
5854 a non-existent command. The wording for the related log line suggests a
5855 non-existent command as the problem.
5857 26. If received_header_text expands to an empty string, do not add a Received:
5858 header line to the message. (Well, it adds a token one on the spool, but
5859 marks it "old" so that it doesn't get used or transmitted.)
5861 27. Installed eximstats 1.28 (addition of -nt option).
5863 28. There was no check for failure on the call to getsockname() in the daemon
5864 code. This can fail if there is a shortage of resources on the system, with
5865 ENOMEM, for example. A temporary error is now given on failure.
5867 29. Contrary to the C standard, it seems that in some environments, the
5868 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
5869 program. Exim now does this explicitly; it affects the formatting of
5870 timestamps using strftime().
5872 30. If exiqsumm was given junk data, it threw up some uninitialized variable
5873 complaints. I've now initialized all the variables, to avoid this.
5875 32. Header lines added by a system filter were not being "seen" during
5876 transport-time rewrites.
5878 33. The info_callback() function passed to OpenSSL is set up with type void
5879 (*)(SSL *, int, int), as described somewhere. However, when calling the
5880 function (actually a macro) that sets it up, the type void(*)() is
5881 expected. I've put in a cast to prevent warnings from picky compilers.
5883 34. If a DNS black list lookup found a CNAME record, but there were no A
5884 records associated with the domain it pointed at, Exim crashed.
5886 35. If a DNS black list lookup returned more than one A record, Exim ignored
5887 all but the first. It now scans all returned addresses if a particular IP
5888 value is being sought. In this situation, the contents of the
5889 $dnslist_value variable are a list of all the addresses, separated by a
5892 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5893 to accept a host name and all its aliases if the forward lookup for any of
5894 them yielded the IP address of the incoming connection. Now it accepts only
5895 those names whose forward lookup yields the correct IP address. Any other
5896 names are discarded. This closes a loophole whereby a rogue DNS
5897 administrator could create reverse DNS records to break through a
5898 wildcarded host restriction in an ACL.
5900 37. If a user filter or a system filter that ran in a subprocess used any of
5901 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5902 the wrong values were passed to the pipe command ($thisaddress had the
5903 value of $0, $0 had the value of $1, etc). This bug was introduced by
5904 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5906 38. Improved the line breaking for long SMTP error messages from ACLs.
5907 Previously, if there was no break point between 40 and 75 characters, Exim
5908 left the rest of the message alone. Two changes have been made: (a) I've
5909 reduced the minimum length to 35 characters; (b) if it can't find a break
5910 point between 35 and 75 characters, it looks ahead and uses the first one
5911 that it finds. This may give the occasional overlong line, but at least the
5912 remaining text gets split now.
5914 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5915 file descriptors might be low, and that setting 1000 would always raise it.
5916 It turns out that in some environments, the limit is already over 1000 and
5917 that lowering it causes trouble. So now Exim takes care not to decrease it.
5919 40. When delivering a message, the value of $return_path is set to $sender_
5920 address at the start of routing (routers may change the value). By an
5921 oversight, this default was not being set up when an address was tested by
5922 -bt or -bv, which affected the outcome if any router or filter referred to
5925 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5926 log only when "message" or "log_message" are set. However, if one of the
5927 conditions was an address verification, or a call to a nested ACL, the
5928 messages generated by the underlying test were being passed through. This
5929 no longer happens. The underlying message is available in $acl_verify_
5930 message for both "message" and "log_message" expansions, so it can be
5931 passed through if needed.
5933 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5934 new expansion $bh_ to give the encoded byte string without charset
5935 translation. Translation happens only if iconv() is available; HAVE_ICONV
5936 indicates this at build time. HEADERS_CHARSET gives the charset to
5937 translate to; headers_charset can change it in the configuration, and
5938 "headers charset" can change it in an individual filter file.
5940 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5941 that creates RFC 2047 encoded "words" labels them as that charset instead
5942 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5943 expansion operator; (ii) when Exim creates a From: line for a local
5944 message; (iii) when a header line is rewritten to include a "phrase" part.
5946 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5947 buggy, causing it to skip the first lines of messages whose message ID
5948 ended in 'D'. This would not have bitten before Exim release 4.14, because
5949 message IDs were unlikely to end in 'D' before then. The effect was to have
5950 incorrect size information for certain domains.
5952 45. #include "config.h" was missing at the start of the crypt16.c module. This
5953 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5956 46. If there was a timeout during a "random" callout check, Exim treated it as
5957 a failure of the random address, and carried on sending RSET and the real
5958 address. If the delay was just some slowness somewhere, the response to the
5959 original RCPT would be taken as a response to RSET and so on, causing
5960 mayhem of various kinds.
5962 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5963 when I implemented it. It didn't allow for the fact that some option values
5964 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5965 the right test for positive values.
5967 48. Domain names in DNS records are case-independent. Exim always looks them up
5968 in lower case. Some resolvers return domain names in exactly the case they
5969 appear in the zone file, that is, they may contain uppercase letters. Not
5970 all resolvers do this - some return always lower case. Exim was treating a
5971 change of case by a resolver as a change of domain, similar to a widening
5972 of a domain abbreviation. This triggered its re-routing code and so it was
5973 trying to route what was effectively the same domain again. This normally
5974 caused routing to fail (because the router wouldn't handle the domain
5975 twice). Now Exim checks for this case specially, and just changes the
5976 casing of the domain that it ultimately uses when it transmits the message
5979 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5982 50. If a filter generated a file delivery with a non-absolute name (possible if
5983 no home directory exists for the router), the forbid_file option was not
5986 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5987 the existing equality matching.
5989 52. Exim was using ints instead of ino_t variables in some places where it was
5990 dealing with inode numbers.
5992 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5993 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5994 called TMPDIR, and if it finds it is different, it changes its value.
5996 54. The smtp_printf() function is now made available to local_scan() so
5997 additional output lines can be written before returning. There is also an
5998 smtp_fflush() function to enable the detection of a dropped connection.
5999 The variables smtp_input and smtp_batched_input are exported to
6002 55. Changed the default runtime configuration: the message "Unknown user"
6003 has been removed from the ACL, and instead placed on the localuser router,
6004 using the cannot_route_message feature. This means that any verification
6005 failures that generate their own messages won't get overridden. Similarly,
6006 the "Unrouteable address" message that was in the ACL for unverifiable
6007 relay addresses has also been removed.
6009 56. Added hosts_avoid_esmtp to the smtp transport.
6011 57. The exicyclog script was not checking for the esoteric option
6012 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6013 will work only if exicyclog is run under the appropriate euid.
6015 58. Following a discussion on the list, the rules by which Exim recognises line
6016 endings on incoming messages have been changed. The -dropcr and drop_cr
6017 options are now no-ops, retained only for backwards compatibility. The
6018 following line terminators are recognized: LF CRLF CR. However, special
6019 processing applies to CR:
6021 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6022 nor a local message in the state where . is a terminator.
6024 (ii) If a bare CR is encountered in a header line, an extra space is added
6025 after the line terminator so as not to end the header. The reasoning
6026 behind this is that bare CRs in header lines are most likely either
6027 to be mistakes, or people trying to play silly games.
6029 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6030 was being incorrectly given as 18 bytes larger than it should have been.
6031 This is a VOB (very old bug).
6033 60. This may never have affected anything current, but just in case it has:
6034 When the local host is found other than at the start of a list of hosts,
6035 the local host, those with the same MX, and any that follow, are discarded.
6036 When the list in question was part of a longer list of hosts, the following
6037 hosts (not currently being processed) were also being discarded. This no
6038 longer happens. I'm not sure if this situation could ever has previously
6041 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6044 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6045 header. This is something that is recommended in a new Internet Draft, and
6046 is something that is documented as being done by Sendmail. There are two
6047 possible values. For messages generated by the autoreply transport, Exim
6050 Auto-Submitted: auto-replied
6052 whereas for all other generated messages (e.g. bounces) it adds
6054 Auto-Submitted: auto-generated
6056 63. The "personal" condition in filters now includes a test for the
6057 Auto-Submitted: header. If it contains the string "auto-" the message it
6058 not considered personal.
6060 64. Added rcpt_include_affixes as a generic transport option.
6062 65. Added queue_only_override (default true).
6064 66. Added the syslog_duplication option.
6066 67. If what should have been the first header line of a message consisted of
6067 a space followed by a colon, Exim was mis-interpreting it as a header line.
6068 It isn't of course - it is syntactically invalid and should therefore be
6069 treated as the start of the message body. The misbehaviour could have
6070 caused a number of strange effects, including loss of data in subsequent
6071 header lines, and spool format errors.
6073 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6074 client host had authenticated. This control can now be exercised by an ACL
6075 for more flexibility.
6077 69. By default, callouts do not happen when testing with -bh. There is now a
6078 variant, -bhc, which does actually run the callout code, including
6079 consulting and updating the callout cache.
6081 70. Added support for saslauthd authentication, courtesy of Alexander
6084 71. If statvfs() failed on the spool or log directories while checking their
6085 size for availability, Exim confusingly gave the error "space shortage".
6086 Furthermore, in debugging mode it crashed with a floating point exception.
6087 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6088 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6089 serious problem, Exim now writes to the main and panic logs when this
6090 happens, with details of the failure. It then refuses to accept the
6091 incoming message, giving the message "spool directory problem" or "log
6092 directory problem" with a 421 code for SMTP messages.
6094 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6095 0, 1, and 2 exist, because some OS complain for execs without them (see
6096 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6097 descriptors. However, the code omitted to check that the open succeeded,
6098 causing mysterious errors if for some reason the permissions on /dev/null
6099 got screwed. Now Exim writes a message to the main and panic logs, and
6100 bombs out if it can't open /dev/null.
6102 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6103 interact so that it is all more flexible. It is supposed to remain
6104 backwards compatible. Also added extra_local_interfaces.
6106 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6107 to bomb out with an assertion failure - to the client this appears as a
6108 connection drop. This problem occurs in the part of the code that was taken
6109 from the Samba project. Fortunately, the assertion is in a very simple
6110 function, so I have fixed this by reproducing the function inline in the
6111 one place where it is called, and arranging for authentication to fail
6112 instead of killing the process with assert().
6114 75. The SPA client code was not working when the server requested OEM rather
6115 than Unicode encoding.
6117 76. Added code to make require_files with a specific uid setting more usable in
6118 the case where statting the file as root fails - usually a non-root-mounted
6119 NFS file system. When this happens and the failure is EACCES, Exim now
6120 forks a subprocess and does the per-uid checking as the relevant uid.
6122 77. Added process_log_path.
6124 78. If log_file_path was not explicitly set, a setting of check_log_space or
6125 check_log_inodes was ignored.
6127 79. If a space check for the spool or log partitions fails, the incident is now
6128 logged. Of course, in the latter case the data may get lost...
6130 80. Added the %p formatting code to string_format() so that it can be used to
6131 print addresses in debug_print(). Adjusted all the address printing in the
6132 debugging in store.c to use %p rather than %d.
6134 81. There was a concern that a line of code in smtp_in.c could overflow a
6135 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6136 initially expressed, the concern was not well-founded, because trailing
6137 spaces are removed early. However, if the trailing spaces were followed by
6138 a NULL, they did not get removed, so the overflow was possible. Two fixes
6141 (a) I re-wrote the offending code in a cleaner fashion.
6142 (b) If an incoming SMTP command contains a NULL character, it is rejected
6145 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6146 runs initgroups(), so that if the Exim user is in any additional groups,
6147 they will be used during message reception.
6153 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob