1 Change log file for Exim from version 4.21
2 -------------------------------------------
7 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
8 and 3.4.4 - once the server is enabled to respond to an OCSP request
9 it does even when not requested, resulting in a stapling non-aware
10 client dropping the TLS connection.
12 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
13 support variable-length bit vectors. No functional change.
15 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
16 The I= interface field on outgoing lines is now after the H= remote
17 host field, same as incoming lines. There is a separate
18 outgoing_interface log selector which allows you to disable the
21 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
22 If not running log_selector +smtp_connection the mainlog would be held
23 open indefinitely after a "too many connections" event, including to a
24 deleted file after a log rotate. Leave the per net connection logging
25 leaving it open for efficiency as that will be quickly detected by the
26 check on the next write.
28 HS/01 Bug 1671: Fix post transport crash.
29 Processing the wait-<transport> messages could crash the delivery
30 process if the message IDs didn't exist for some reason. When
31 using 'split_spool_directory=yes' the construction of the spool
32 file name failed already, exposing the same netto behaviour.
34 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
35 mime_regex ACL conditions.
37 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
38 to DSN fail messages (bounces): remote IP, remote greeting, remote response
39 to HELO, local diagnostic string.
41 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
42 log line to debug. Even when configured with a tls authenticator many
43 client connections are expected to not authenticate in this way, so
44 an authenticate fail is not an error.
46 HS/02 Add the Exim version string to the process info. This way exiwhat
47 gives some more detail about the running daemon.
49 JH/06 Bug 1395: time-limit cacheing of DNS lookups, to the TTL value. This may
50 matter for fast-change records such as DNSBLs.
52 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
53 retry record, even if constant. There may be multiple transports with
54 different interface settings and the retry behaviour needs to be kept
57 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
59 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
61 JH/10 Bug 840: fix log_defer_output option of pipe transport
63 JH/11 Bug 830: use same host for all RCPTS of a message, even under
64 hosts_randomize. This matters a lot when combined with mua_wrapper.
66 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
67 ${quote_pgsql:<string>} operator.
69 JH/13 Bug 1708: avoid misaligned access in cached lookup.
71 JH/14 Change header file name for freeradius-client. Relevant if compiling
72 with Radius support; from the Gentoo tree and checked under Fedora.
74 JH/15 Bug 1712: Introduce $prdr_requested flag variable
76 JH/16 Bug 1714: Permit an empty string as expansion result for transport
77 option transport_filter, meaning no filtering.
79 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
81 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
82 defaults to "*" (all hosts). The variable is now available when not built
83 with TLS, default unset, mainly to enable keeping the testuite sane.
84 If a server certificate is not supplied (via tls_certificate) an error is
85 logged, and clients will find TLS connections fail on startup. Presumably
86 they will retry in-clear.
87 Packagers of Exim are strongly encouraged to create a server certificate
90 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
91 with the $config_file variable.
93 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
94 in transport context, after the attempt, and per-recipient. The latter type
95 is per host attempted. The event data is the error message, and the errno
96 information encodes the lookup type (A vs. MX) used for the (first) host,
97 and the trailing two digits of the smtp 4xx reponse.
99 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
100 to write to mainlog (or rejectlog, paniclog) in the window between file
101 creation and permissions/ownership being changed. Particularly affects
102 installations where exicyclog is run as root, rather than exim user;
103 result is that the running daemon panics and dies.
105 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
107 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
108 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
109 "pri" and "weight". Note that the previous implicit priority given by the
110 list order is no longer honoured.
112 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalisation
115 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
116 by defining SUPPORT_SOCKS.
118 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
119 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
120 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
121 variables are renamed to proxy_{local,external}_{address,port}.
123 JH/27 Move Internationalisation support from Experimental to mainline, enabled
124 for a build by defining SUPPORT_I18N
126 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
127 of the query string, and make ${quote_redis:} do that quoting.
129 JH/29 Move Events support from Experimental to mainline, enabled by default
130 and removable for a build by defining DISABLE_EVENT.
132 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
134 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
135 cached by the daemon.
137 JH/32 Move Redis support from Experimental to mainline, enabled for a build
138 by defining LOOKUP_REDIS. The libhiredis library is required.
140 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
141 keys are given for lookup.
143 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
144 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
145 only supported when built with TLS support. The PolarSSL SHA routines
146 are still used when the TLS library is too old for convenient support.
148 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
149 openssl_options), for security. OpenSSL forces this from version 1.1.0
150 server-side so match that on older versions.
152 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
153 allocation for $value could be released as the expansion processing
154 concluded, but leaving the global pointer active for it.
156 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
157 and to use the domains and local_parts ACL conditions.
159 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
160 incorrectly not doubled on cutthrough transmission, hence seen as a
161 body-termination at the receiving system - resulting in truncated mails.
162 Commonly the sender saw a TCP-level error, and retransmitted the message
163 via the normal store-and-forward channel. This could result in duplicates
164 received - but deduplicating mailstores were liable to retain only the
165 initial truncated version.
167 JH/39 Bug 1781: Fix use of private-keys having trailing '=' in the base-64.
169 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
175 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
178 JH/02 The smtp transport option "multi_domain" is now expanded.
180 JH/03 The smtp transport now requests PRDR by default, if the server offers
183 JH/04 Certificate name checking on server certificates, when exim is a client,
184 is now done by default. The transport option tls_verify_cert_hostnames
185 can be used to disable this per-host. The build option
186 EXPERIMENTAL_CERTNAMES is withdrawn.
188 JH/05 The value of the tls_verify_certificates smtp transport and main options
189 default to the word "system" to access the system default CA bundle.
190 For GnuTLS, only version 3.0.20 or later.
192 JH/06 Verification of the server certificate for a TLS connection is now tried
193 (but not required) by default. The verification status is now logged by
194 default, for both outbound TLS and client-certificate supplying inbound
197 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
200 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
201 Status Notification (bounce) messages are now MIME format per RFC 3464.
202 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
203 under the control of the dsn_advertise_hosts option, and routers may
204 have a dsn_lasthop option.
206 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
207 default, modifiable by a malware= option. The list separator for
208 the options can now be changed in the usual way. Bug 68.
210 JH/10 The smtp_receive_timeout main option is now expanded before use.
212 JH/11 The incoming_interface log option now also enables logging of the
213 local interface on delivery outgoing connections.
215 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
216 if the interface and destination host and port all match.
218 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
221 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
222 Patch from Andrew Lewis.
224 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
225 now supports optional time-restrictions, weighting, and priority
226 modifiers per server. Patch originally by <rommer@active.by>.
228 JH/16 The spamd_address main option now supports a mixed list of local
229 and remote servers. Remote servers can be IPv6 addresses, and
230 specify a port-range.
232 JH/17 Bug 68: The spamd_address main option now supports an optional
233 timeout value per server.
235 JH/18 Bug 1581: Router and transport options headers_add/remove can
236 now have the list separator specified.
238 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
241 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
244 JH/21 Support for the A6 type of dns record is withdrawn.
246 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
247 rather than the verbs used.
249 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
250 from 255 to 1024 chars.
252 JH/24 Verification callouts now attempt to use TLS by default.
254 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
255 are generic router options now. The defaults didn't change.
257 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
258 Original patch from Alexander Shikoff, worked over by JH.
260 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
261 files if ZCAT_COMMAND is not executable.
263 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
265 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
267 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
268 Normally benign, it bites when the pair was led to by a CNAME;
269 modern usage is to not canoicalize the domain to a CNAME target
270 (and we were inconsistent anyway for A-only vs AAAA+A).
272 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
274 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
275 when evaluating $sender_host_dnssec.
277 JH/31 Check the HELO verification lookup for DNSSEC, adding new
278 $sender_helo_dnssec variable.
280 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
282 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
284 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
286 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
287 documented as working, but never had. Support all but $spam_report.
289 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
290 added for tls authenticator.
295 TL/01 When running the test suite, the README says that variables such as
296 no_msglog_check are global and can be placed anywhere in a specific
297 test's script, however it was observed that placement needed to be near
298 the beginning for it to behave that way. Changed the runtest perl
299 script to read through the entire script once to detect and set these
300 variables, reset to the beginning of the script, and then run through
301 the script parsing/test process like normal.
303 TL/02 The BSD's have an arc4random API. One of the functions to induce
304 adding randomness was arc4random_stir(), but it has been removed in
305 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
306 function when detected.
308 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
309 cause callback expansion.
311 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
312 syntax errors in an expansion can be treated as a string instead of
313 logging or causing an error, due to the internal use of bool_lax
314 instead of bool when processing it.
316 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
317 server certificates when making smtp deliveries.
319 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
321 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
323 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
325 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
326 Merged patch from Sebastian Wiedenroth.
328 JH/05 Fix results-pipe from transport process. Several recipients, combined
329 with certificate use, exposed issues where response data items split
330 over buffer boundaries were not parsed properly. This eventually
331 resulted in duplicates being sent. This issue only became common enough
332 to notice due to the introduction of conection certificate information,
333 the item size being so much larger. Found and fixed by Wolfgang Breyha.
335 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
336 size buffer was used, resulting in syntax errors when an expansion
339 JH/07 Add support for directories of certificates when compiled with a GnuTLS
340 version 3.3.6 or later.
342 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
343 is EXPERIMENTAL_EVENT, the main-configuration and transport options
344 both become "event_action", the variables become $event_name, $event_data
345 and $event_defer_errno. There is a new variable $verify_mode, usable in
346 routers, transports and related events. The tls:cert event is now also
347 raised for inbound connections, if the main configuration event_action
350 TL/06 In test suite, disable OCSP for old versions of openssl which contained
351 early OCSP support, but no stapling (appears to be less than 1.0.0).
353 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
354 server certificate names available under the smtp transport option
355 "tls_verify_cert_hostname" now do not permit multi-component wildcard
358 JH/10 Time-related extraction expansions from certificates now use the main
359 option "timezone" setting for output formatting, and are consistent
360 between OpenSSL and GnuTLS compilations. Bug 1541.
362 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
363 encoded parameter in the incoming message. Bug 1558.
365 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
366 include certificate info, eximon was claiming there were spoolfile
369 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
371 JH/14 Log delivery-related information more consistently, using the sequence
372 "H=<name> [<ip>]" wherever possible.
374 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
375 are problematic for Debian distribution, omit them from the release
378 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
380 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
383 JH/17 Fix a null-indirection in certextract expansions when a nondefault
384 output list separator was used.
389 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
390 checkers that were complaining about end of non-void function with no
393 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
394 This was a regression intruduced in 4.83 by another bugfix.
396 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
398 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
399 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
405 TF/01 Correctly close the server side of TLS when forking for delivery.
407 When a message was received over SMTP with TLS, Exim failed to clear up
408 the incoming connection properly after forking off the child process to
409 deliver the message. In some situations the subsequent outgoing
410 delivery connection happened to have the same fd number as the incoming
411 connection previously had. Exim would try to use TLS and fail, logging
412 a "Bad file descriptor" error.
414 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
415 utilities have not been installed.
417 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
418 temporary space as the ACL may create new global variables.
420 TL/01 LDAP support uses per connection or global context settings, depending
421 upon the detected version of the libraries at build time.
423 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
424 to extract and use the src ip:port in logging and expansions as if it
425 were a direct connection from the outside internet. PPv2 support was
426 updated based on HAProxy spec change in May 2014.
428 JH/02 Add ${listextract {number}{list}{success}{fail}}.
430 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
431 Properly escape header and check for NULL return.
433 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
436 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
438 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
439 characters in header names, implemented as a verify condition.
440 Contributed by Michael Fischer v. Mollard.
442 TL/05 Rename SPF condition results err_perm and err_temp to standardized
443 results permerror and temperror. Previous values are deprecated but
444 still accepted. In a future release, err_perm and err_temp will be
445 completely removed, which will be a backward incompatibility if the
446 ACL tests for either of these two old results. Patch contributed by
447 user bes-internal on the mailing list.
449 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
451 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
452 selectors, in both main and reject logs.
454 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
457 JH/07 Add malware type "sock" for talking to simple daemon.
459 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
461 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
462 routers/transports under cutthrough routing.
464 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
465 numbers. Touch up "bool" conditional to keep the same definition.
467 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
469 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
471 JH/12 Expand items in router/transport headers_add or headers_remove lists
472 individually rather than the list as a whole. Bug 1452.
474 Required for reasonable handling of multiple headers_ options when
475 they may be empty; requires that headers_remove items with embedded
476 colons must have them doubled (or the list-separator changed).
478 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
479 view the policy declared in the DMARC record. Currently, $dmarc_status
480 is a combined value of both the record presence and the result of the
483 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
485 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
486 dnslookup router and the smtp transport (applying to the forward
489 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
490 of ldap servers used for a specific lookup. Patch provided by Heiko
493 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
494 New variable $lookup_dnssec_authenticated for observability.
496 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
497 Patch submitted by Lars Timman.
499 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
501 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
502 Requires trusted mode and valid format message id, aborts otherwise.
503 Patch contributed by Heiko Schlichting.
505 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
506 certextract with support for various fields. Bug 1358.
508 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
509 is requested by default, modifiable by smtp transport option
512 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
513 operate on certificate variables to give certificate fingerprints
514 Also new ${sha256:cert_variable}.
516 JH/23 The PRDR feature is moved from being Experimental into the mainline.
518 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
519 Christian Aistleitner.
521 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
523 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
524 file. Patch from Wolfgang Breyha.
526 JH/25 Expand the coverage of the delivery $host and $host_address to
527 client authenticators run in verify callout. Bug 1476.
529 JH/26 Port service names are now accepted for tls_on_connect_ports, to
530 align with daemon_smtp_ports. Bug 72.
532 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
533 support and error reporting did not work properly.
535 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
536 and is readable. Patch from Andrew Colin Kissa.
538 TL/14 Enhance documentation of ${run expansion and how it parses the
539 commandline after expansion, particularly in the case when an
540 unquoted variable expansion results in an empty value.
542 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
544 PP/02 Fix internal collision of T_APL on systems which support RFC3123
545 by renaming away from it. Addresses GH issue 15, reported by
548 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
550 TL/15 SECURITY: prevent double expansion in math comparison functions
551 (can expand unsanitized data). Not remotely exploitable.
558 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
560 PP/02 Make -n do something, by making it not do something.
561 When combined with -bP, the name of an option is not output.
563 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
566 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
567 $sender_host_name and config options to manage this, and basic check
570 PP/05 DSCP support for outbound connections and control modifier for inbound.
572 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
573 (Only plugin which currently uses this is kerberos4, which nobody should
574 be using, but we should make it available and other future plugins might
575 conceivably use it, even though it would break NAT; stuff *should* be
576 using channel bindings instead).
578 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
579 name; added for Sendmail compatibility; requires admin caller.
580 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
581 just ignore it); requires trusted caller.
582 Also parse but ignore: -Ac -Am -X<logfile>
585 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
587 TL/02 Add +smtp_confirmation as a default logging option.
589 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
590 Patch by Magnus Holmgren from 2007-02-20.
592 TL/04 Bugzilla 1281 - Spec typo.
593 Bugzilla 1283 - Spec typo.
594 Bugzilla 1290 - Spec grammar fixes.
596 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
598 TL/06 Add Experimental DMARC support using libopendmarc libraries.
600 TL/07 Fix an out of order global option causing a segfault. Reported to dev
601 mailing list by by Dmitry Isaikin.
603 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
605 JH/02 Support "G" suffix to numbers in ${if comparisons.
607 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
609 NM/01 Bugzilla 1197 - Spec typo
610 Bugzilla 1196 - Spec examples corrections
612 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
614 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
615 gnutls_enable_pkcs11, but renamed to more accurately indicate its
618 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
619 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
621 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
622 "acl {{name}{arg}...}", and optional args on acl condition
625 JH/05 Permit multiple router/transport headers_add/remove lines.
627 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
629 JH/07 Avoid using a waiting database for a single-message-only transport.
630 Performance patch from Paul Fisher. Bugzilla 1262.
632 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
635 JH/09 Add $headers_added variable, with content from use of ACL modifier
636 add_header (but not yet added to the message). Bugzilla 199.
638 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
639 Pulled from Bugzilla 817 by Wolfgang Breyha.
641 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
643 (nb: this is the same fix as in Exim 4.80.1)
645 JH/11 Add A= logging on delivery lines, and a client_set_id option on
648 JH/12 Add optional authenticated_sender logging to A= and a log_selector
651 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
653 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
654 advertise SMTP AUTH mechanism to us, instead of a generic
655 protocol violation error. Also, make Exim more robust to bad
656 data from the Dovecot auth socket.
658 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
660 When a queue runner is handling a message, Exim first routes the
661 recipient addresses, during which it prunes them based on the retry
662 hints database. After that it attempts to deliver the message to
663 any remaining recipients. It then updates the hints database using
666 So if a recipient address works intermittently, it can get repeatedly
667 deferred at routing time. The retry hints record remains fresh so the
668 address never reaches the final cutoff time.
670 This is a fairly common occurrence when a user is bumping up against
671 their storage quota. Exim had some logic in its local delivery code
672 to deal with this. However it did not apply to per-recipient defers
673 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
675 This change adds a proper retry rule check during routing so that the
676 final cutoff time is checked against the message's age. We only do
677 this check if there is an address retry record and there is not a
678 domain retry record; this implies that previous attempts to handle
679 the address had the retry_use_local_parts option turned on. We use
680 this as an approximation for the destination being like a local
681 delivery, as in LMTP.
683 I suspect this new check makes the old local delivery cutoff check
684 redundant, but I have not verified this so I left the code in place.
686 TF/02 Correct gecos expansion when From: is a prefix of the username.
688 Test 0254 submits a message to Exim with the header
692 When I ran the test suite under the user fanf2, Exim expanded
693 the header to contain my full name, whereas it should have added
694 a Resent-Sender: header. It erroneously treats any prefix of the
695 username as equal to the username.
697 This change corrects that bug.
699 GF/01 DCC debug and logging tidyup
700 Error conditions log to paniclog rather than rejectlog.
701 Debug lines prefixed by "DCC: " to remove any ambiguity.
703 TF/03 Avoid unnecessary rebuilds of lookup-related code.
705 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
706 Bug spotted by Jeremy Harris; was flawed since initial commit.
707 Would have resulted in OCSP responses post-SNI triggering an Exim
708 NULL dereference and crash.
710 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
712 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
713 Bug detection, analysis and fix by Samuel Thibault.
714 Bugzilla 1331, Debian bug #698092.
716 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
718 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
719 Server implementation by Todd Lyons, client by JH.
720 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
721 config variable "prdr_enable" controls whether the server
722 advertises the facility. If the client requests PRDR a new
723 acl_data_smtp_prdr ACL is called once for each recipient, after
724 the body content is received and before the acl_smtp_data ACL.
725 The client is controlled by bolth of: a hosts_try_prdr option
726 on the smtp transport, and the server advertisement.
727 Default client logging of deliveries and rejections involving
728 PRDR are flagged with the string "PRDR".
730 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
731 fclose(). Diagnosis by Todd Lyons.
733 PP/17 Update configure.default to handle IPv6 localhost better.
734 Patch by Alain Williams (plus minor tweaks).
737 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
738 This is now consistent with GnuTLS, and is now documented: the
739 previous undocumented portable approach to treating the option as
740 unset was to force an expansion failure. That still works, and
741 an empty string is now equivalent.
743 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
744 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
745 not performing validation itself.
747 PP/20 Added force_command boolean option to pipe transport.
748 Patch from Nick Koston, of cPanel Inc.
750 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
753 TF/04 Added udpsend ACL modifer and hexquote expansion operator
755 PP/21 Fix eximon continuous updating with timestamped log-files.
756 Broken in a format-string cleanup in 4.80, missed when I repaired the
757 other false fix of the same issue.
758 Report and fix from Heiko Schlichting.
761 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
762 Report from Prashanth Katuri.
764 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
765 It's SecureTransport, so affects any MacOS clients which use the
766 system-integrated TLS libraries, including email clients.
768 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
769 using a MIME ACL for non-SMTP local injection.
770 Report and assistance in diagnosis by Warren Baker.
772 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
774 JH/16 Fix comparisons for 64b. Bugzilla 1385.
776 TL/09 Add expansion variable $authenticated_fail_id to keep track of
777 last id that failed so it may be referenced in subsequent ACL's.
779 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
782 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
783 ldap library initialization, allowing self-signed CA's to be
784 used. Also properly sets require_cert option later in code by
785 using NULL (global ldap config) instead of ldap handle (per
786 session). Bug diagnosis and testing by alxgomz.
788 TL/12 Enhanced documentation in the ratelimit.pl script provided in
789 the src/util/ subdirectory.
791 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
792 renamed to Transport Post Delivery Action by Jeremy Harris, as
795 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
796 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
797 redis_servers = needs to be configured which will be used by the redis
798 lookup. Patch from Warren Baker, of The Packet Hub.
800 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
802 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
803 hostname or reverse DNS when processing a host list. Used suggestions
804 from multiple comments on this bug.
806 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
808 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
809 Missed a few lines, added it to make the runtest require no keyboard
812 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
813 contains upper case chars. Make router use caseful_local_part.
815 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
816 support when GnuTLS has been built with p11-kit.
822 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
824 This, or similar/improved, will also be change PP/11 of 4.82.
830 PP/01 Handle short writes when writing local log-files.
831 In practice, only affects FreeBSD (8 onwards).
832 Bugzilla 1053, with thanks to Dmitry Isaikin.
834 NM/01 Bugzilla 949 - Documentation tweak
836 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
839 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
841 PP/02 Implemented gsasl authenticator.
843 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
845 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
846 `pkg-config foo` for cflags/libs.
848 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
849 with rest of GSASL and with heimdal_gssapi.
851 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
852 `pkg-config foo` for cflags/libs for the TLS implementation.
854 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
855 properties get this fed in as external SSF. A number of robustness
856 and debugging improvements to the cyrus_sasl authenticator.
858 PP/08 cyrus_sasl server now expands the server_realm option.
860 PP/09 Bugzilla 1214 - Log authentication information in reject log.
861 Patch by Jeremy Harris.
863 PP/10 Added dbmjz lookup type.
865 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
867 PP/12 MAIL args handles TAB as well as SP, for better interop with
868 non-compliant senders.
869 Analysis and variant patch by Todd Lyons.
871 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
872 Bug report from Lars Müller <lars@samba.org> (via SUSE),
873 Patch from Dirk Mueller <dmueller@suse.com>
875 PP/13 tls_peerdn now print-escaped for spool files.
876 Observed some $tls_peerdn in wild which contained \n, which resulted
877 in spool file corruption.
879 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
880 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
881 or write after TLS renegotiation, which otherwise led to messages
884 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
885 as a tracking header (ie: a signed header comes before the signature).
886 Patch from Wolfgang Breyha.
888 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
889 comma-sep list; embedded commas doubled.
891 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
893 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
895 Report and patch from Dmitry Banschikov.
897 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
898 Removed SSL_clear() after SSL_new() which led to protocol negotiation
899 failures. We appear to now support TLS1.1+ with Exim.
901 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
902 lets Exim select keys and certificates based upon TLS SNI from client.
903 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
904 before an outbound SMTP session. New log_selector, +tls_sni.
906 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
907 NULL dereference. Report and patch from Alun Jones.
909 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
910 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
911 Not seeing resolver debug output on NetBSD, but suspect this is a
912 resolver implementation change.
914 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
915 Left warnings. Added "eximon gdb" invocation mode.
917 PP/21 Defaulting "accept_8bitmime" to true, not false.
919 PP/22 Added -bw for inetd wait mode support.
921 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
922 locate the relevant includes and libraries. Made this the default.
924 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
925 Bugzilla 1246, report and most of solution from Tomasz Kusy.
927 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
928 This may cause build issues on older platforms.
930 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
931 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
932 gnutls_require_mac & gnutls_require_protocols (no longer supported).
933 Added SNI support via GnuTLS too.
934 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
936 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
938 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
939 Applied second patch from Janne, implementing suggestion to default
940 multiple-strings-in-record handling to match SPF spec.
942 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
944 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
945 read-only, out of scope).
946 Patch from Wolfgang Breyha, report from Stuart Northfield.
948 PP/29 Fix three issues highlighted by clang analyser static analysis.
949 Only crash-plausible issue would require the Cambridge-specific
950 iplookup router and a misconfiguration.
951 Report from Marcin Mirosław.
953 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
955 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
956 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
957 As part of this, removing so much warning spew let me fix some minor
958 real issues in debug logging.
960 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
961 assignment on my part. Fixed.
963 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
964 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
965 Janne Snabb (who went above and beyond: thank you).
967 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
968 string otherwise requires a connection and a bunch more work and it's
969 relatively easy to get wrong. Should also expose TLS library linkage
972 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
973 64-bit ${eval} (JH/03).
975 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
976 GNU libc to support some of the 64-bit stuff, should not lead to
977 conflicts. Defined before os.h is pulled in, so if a given platform
978 needs to override this, it can.
980 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
981 protection layer was required, which is not implemented.
982 Bugzilla 1254, patch from Wolfgang Breyha.
984 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
985 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
986 tls_dhparam take prime identifiers. Also unbreak combination of
987 OpenSSL+DH_params+TLSSNI.
989 PP/39 Disable SSLv2 by default in OpenSSL support.
995 PP/01 Solaris build fix for Oracle's LDAP libraries.
996 Bugzilla 1109, patch from Stephen Usher.
998 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1000 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1003 TF/02 Fix a couple more cases where we did not log the error message
1004 when unlink() failed. See also change 4.74-TF/03.
1006 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1007 lock up or crash if it happened to be inside a call to libc when it
1008 got a SIGUSR1 from exiwhat.
1010 The SIGUSR1 handler appends the current process status to the process
1011 log which is later printed by exiwhat. It used to use the general
1012 purpose logging code to do this, but several functions it calls are
1013 not safe for signals.
1015 The new output code in the SIGUSR1 handler is specific to the process
1016 log, and simple enough that it's easy to inspect for signal safety.
1017 Removing some special cases also simplifies the general logging code.
1018 Removing the spurious timestamps from the process log simplifies
1021 TF/04 Improved ratelimit ACL condition.
1023 The /noupdate option has been deprecated in favour of /readonly which
1024 has clearer semantics. The /leaky, /strict, and /readonly update modes
1025 are mutually exclusive. The update mode is no longer included in the
1026 database key; it just determines when the database is updated. (This
1027 means that when you upgrde Exim will forget old rate measurements.)
1029 Exim now checks that the per_* options are used with an update mode that
1030 makes sense for the current ACL. For example, when Exim is processing a
1031 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1032 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1033 must specify per_mail/readonly. If you omit the update mode it defaults to
1034 /leaky where that makes sense (as before) or /readonly where required.
1036 The /noupdate option is now undocumented but still supported for
1037 backwards compatibility. It is equivalent to /readonly except that in
1038 ACLs where /readonly is required you may specify /leaky/noupdate or
1039 /strict/noupdate which are treated the same as /readonly.
1041 A useful new feature is the /count= option. This is a generalization
1042 of the per_byte option, so that you can measure the throughput of other
1043 aggregate values. For example, the per_byte option is now equivalent
1044 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1046 The per_rcpt option has been generalized using the /count= mechanism
1047 (though it's more complicated than the per_byte equivalence). When it is
1048 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1049 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1050 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1051 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1052 Note that using per_rcpt with a non-readonly update mode in more than
1053 one ACL will cause the recipients to be double-counted. (The per_mail
1054 and per_byte options don't have this problem.)
1056 The handling of very low rates has changed slightly. If the computed rate
1057 is less than the event's count (usually one) then this event is the first
1058 after a long gap. In this case the rate is set to the same as this event's
1059 count, so that the first message of a spam run is counted properly.
1061 The major new feature is a mechanism for counting the rate of unique
1062 events. The new per_addr option counts the number of different
1063 recipients that someone has sent messages to in the last time period. It
1064 behaves like per_rcpt if all the recipient addresses are different, but
1065 duplicate recipient addresses do not increase the measured rate. Like
1066 the /count= option this is a general mechanism, so the per_addr option
1067 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1068 example, measure the rate that a client uses different sender addresses
1069 with the options per_mail/unique=$sender_address. There are further
1070 details in the main documentation.
1072 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1074 TF/06 Removed a few PCRE remnants.
1076 TF/07 Automatically extract Exim's version number from tags in the git
1077 repository when doing development or release builds.
1079 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1080 Bugzilla 879. Patch from Paul Fisher.
1082 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1083 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1086 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1088 PP/05 Variable $av_failed, true if the AV scanner deferred.
1089 Bugzilla 1078. Patch from John Horne.
1091 PP/06 Stop make process more reliably on build failure.
1092 Bugzilla 1087. Patch from Heiko Schlittermann.
1094 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1095 Bugzilla 1089. Patch from Heiko Schlittermann.
1097 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1098 Bugzilla 1131. Patch from Holger Weiß.
1100 PP/09 Handle IPv6 addresses with SPF.
1101 Bugzilla 860. Patch from Wolfgang Breyha.
1103 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1105 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1108 PP/11 match_* no longer expand right-hand-side by default.
1109 New compile-time build option, EXPAND_LISTMATCH_RHS.
1110 New expansion conditions, "inlist", "inlisti".
1112 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1114 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1116 PP/14 fix log_write() format string regression from TF/03.
1117 Bugzilla 1152. Patch from Dmitry Isaikin.
1123 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1125 PP/02 Harmonised TLS library version reporting; only show if debugging.
1126 Layout now matches that introduced for other libraries in 4.74 PP/03.
1128 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1130 PP/04 New "dns_use_edns0" global option.
1132 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1135 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1136 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1138 TK/01 Updated PolarSSL code to 0.14.2.
1139 Bugzilla 1097. Patch from Andreas Metzler.
1141 PP/07 Catch divide-by-zero in ${eval:...}.
1142 Fixes bugzilla 1102.
1144 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1147 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1148 format-string attack -- SECURITY: remote arbitrary code execution.
1150 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1151 time unintentionally subject to list matching rules, letting the header
1152 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1153 arbitrary string expansion). This allowed for information disclosure.
1155 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1156 INT_MIN/-1 -- value coerced to INT_MAX.
1162 NM/01 Workround for PCRE version dependency in version reporting
1165 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1166 This fixes portability to compilers other than gcc, notably
1167 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1169 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1170 makefiles for portability to HP-UX and POSIX correctness.
1172 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1173 Also via indented variable definition in the Makefile.
1174 (Debugging by Oliver Heesakkers).
1176 PP/02 Restore caching of spamd results with expanded spamd_address.
1177 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1179 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1180 Improves build reliability. Fix from: Frank Elsner
1182 NM/02 Fix wide character breakage in the rfc2047 coding
1183 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1185 NM/03 Allow underscore in dnslist lookups
1186 Fixes bug 1026. Patch from Graeme Fowler
1188 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1189 Code patches from Adam Ciarcinski of NetBSD.
1191 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1194 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1195 is logged, to avoid truncation. Patch from John Horne.
1197 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1198 Patch from Jakob Hirsch.
1200 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1201 SQL string expansion failure details.
1202 Patch from Andrey Oktyabrski.
1204 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1205 Patch from Simon Arlott.
1207 PP/09 New lookups functionality failed to compile on old gcc which rejects
1208 extern declarations in function scope.
1209 Patch from Oliver Fleischmann
1211 PP/10 Use sig_atomic_t for flags set from signal handlers.
1212 Check getgroups() return and improve debugging.
1213 Fixed developed for diagnosis in bug 927 (which turned out to be
1216 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1217 Patch from Mark Zealey.
1219 PP/12 Bugzilla 1056: Improved spamd server selection.
1220 Patch from Mark Zealey.
1222 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1223 Based on patch from Heiko Schlittermann.
1225 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1226 Patch from Uwe Doering, sign-off by Michael Haardt.
1228 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1229 variable declaration deep within a block. Bug and patch from
1232 PP/15 lookups-Makefile IRIX compatibilty coercion.
1234 PP/16 Make DISABLE_DKIM build knob functional.
1236 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1237 Patch by Simon Arlott
1239 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1240 variable argument macros. Our copy now differs from upstream.
1246 TF/01 Failure to get a lock on a hints database can have serious
1247 consequences so log it to the panic log.
1249 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1250 controlled using the smtp_confirmation log selector.
1252 TF/03 Include the error message when we fail to unlink a spool file.
1254 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1255 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1256 for maintaining out-of-tree patches for some time.
1258 PP/01 Bugzilla 139: Documentation and portability issues.
1259 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1260 Handle per-OS dynamic-module compilation flags.
1262 PP/02 Let /dev/null have normal permissions.
1263 The 4.73 fixes were a little too stringent and complained about the
1264 permissions on /dev/null. Exempt it from some checks.
1265 Reported by Andreas M. Kirchwitz.
1267 PP/03 Report version information for many libraries, including
1268 Exim version information for dynamically loaded libraries. Created
1269 version.h, now support a version extension string for distributors
1270 who patch heavily. Dynamic module ABI change.
1272 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1273 privilege escalation vulnerability whereby the Exim run-time user
1274 can cause root to append content of the attacker's choosing to
1277 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1280 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1281 If dropping privileges for untrusted macros, we disabled normal logging
1282 on the basis that it would fail; for the Exim run-time user, this is not
1283 the case, and it resulted in successful deliveries going unlogged.
1284 Fixed. Reported by Andreas Metzler.
1290 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1291 only prepend for the Resent-* case. Fixes regression introduced in
1292 Exim 4.70 by NM/22 for Bugzilla 607.
1294 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1295 increasing numbers of administrators be bitten by this.
1297 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1299 PP/03 Bugzilla 994: added openssl_options main configuration option.
1301 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1303 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1305 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1307 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1308 without a peer certificate, leading to a segfault because of an
1309 assumption that peers always have certificates. Be a little more
1310 paranoid. Problem reported by Martin Tscholak.
1312 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1313 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1314 NB: ClamAV planning to remove STREAM in "middle of 2010".
1315 CL also introduces -bmalware, various -d+acl logging additions and
1316 more caution in buffer sizes.
1318 PP/09 Implemented reverse_ip expansion operator.
1320 PP/10 Bugzilla 937: provide a "debug" ACL control.
1322 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1324 PP/12 Bugzilla 973: Implement --version.
1326 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1328 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1330 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1332 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1333 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1334 ignore trailing whitespace.
1336 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1338 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1341 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1342 Notification from Dr Andrew Aitchison.
1344 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1345 ExtendedDetectionInfo response format.
1346 Notification from John Horne.
1348 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1351 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1352 XSL and documented dependency on system catalogs, with examples of how
1355 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1358 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1359 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1360 configuration file which is writeable by the Exim user or group.
1362 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1363 of configuration files to cover files specified with the -C option if
1364 they are going to be used with root privileges, not just the default
1367 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1368 option (effectively making it always true).
1370 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1371 files to be used while preserving root privileges.
1373 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1374 that rogue child processes cannot use them.
1376 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1377 run-time user, instead of root.
1379 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1380 Exim run-time user without dropping privileges.
1382 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1383 result string, instead of calling string_vformat() twice with the same
1386 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1387 for other users. Others should always drop root privileges if they use
1388 -C on the command line, even for a whitelisted configure file.
1390 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1392 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1398 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1399 $data_path, and $header_path variables; fixed documentation bugs and
1402 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1403 exipick to access non-standard spools, including the "frozen" queue
1406 NM/01 Bugzilla 965: Support mysql stored procedures.
1407 Patch from Alain Williams
1409 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1411 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1412 Patch from Andreas Metzler
1414 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1415 Patch from Kirill Miazine
1417 NM/05 Bugzilla 671: Added umask to procmail example.
1419 JJ/03 installed exipick 20100323.0, fixing doc bug
1421 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1422 directory. Notification and patch from Dan Rosenberg.
1424 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1426 TK/02 Improve log output when DKIM signing operation fails.
1428 MH/01 Treat the transport option dkim_domain as a colon separated
1429 list, not as a single string, and sign the message with each element,
1430 omitting multiple occurences of the same signer.
1432 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1433 Bugzilla 985, 986. Patch by Simon Arlott
1435 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1436 Patch by Simon Arlott
1438 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1439 MBX locking. Notification from Dan Rosenberg.
1445 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1447 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1449 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1451 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1453 NM/04 Bugzilla 915: Flag broken perl installation during build.
1459 TK/01 Added patch by Johannes Berg that expands the main option
1460 "spamd_address" if it starts with a dollar sign.
1462 TK/02 Write list of recipients to X-Envelope-Sender header when building
1463 the mbox-format spool file for content scanning (suggested by Jakob
1466 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1467 (http://www.dcc-servers.net/) support via dccifd. Activated by
1468 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1470 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1471 by Mark Daniel Reidel <mr@df.eu>.
1473 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1474 When building exim an external PCRE library is now needed -
1475 PCRE is a system library on the majority of modern systems.
1476 See entry on PCRE_LIBS in EDITME file.
1478 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1479 conversation. Added nologin parameter to request.
1480 Patch contributed by Kirill Miazine.
1482 TF/01 Do not log submission mode rewrites if they do not change the address.
1484 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1486 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1487 log files in place. Contributed by Roberto Lima.
1489 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1491 TF/03 Bugzilla 615: When checking the local_parts router precondition
1492 after a local_part_suffix or local_part_prefix option, Exim now
1493 does not use the address's named list lookup cache, since this
1494 contains cached lookups for the whole local part.
1496 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1497 Robert Millan. Documentation is in experimental-spec.txt.
1499 TF/04 Bugzilla 668: Fix parallel build (make -j).
1501 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1503 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1504 Patch provided by Jan Srzednicki.
1506 TF/05 Leading white space used to be stripped from $spam_report which
1507 wrecked the formatting. Now it is preserved.
1509 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1510 that they are available at delivery time.
1512 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1514 TF/08 TLS error reporting now respects the incoming_interface and
1515 incoming_port log selectors.
1517 TF/09 Produce a more useful error message if an SMTP transport's hosts
1518 setting expands to an empty string.
1520 NM/06 Bugzilla 744: EXPN did not work under TLS.
1521 Patch provided by Phil Pennock.
1523 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1524 Patch provided by Richard Godbee.
1526 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1527 acl_smtp_notquit, added index entry.
1529 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1530 Patch provided by Eugene Bujak.
1532 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1533 accept(). Patch provided by Maxim Dounin.
1535 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1536 Patch provided by Phil Pennock.
1538 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1540 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1541 Patch provided by Brad "anomie" Jorsch.
1543 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1544 Patch provided by Dean Brooks.
1546 TK/05 Add native DKIM support (does not depend on external libraries).
1548 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1549 Patch provided by Graeme Fowler.
1551 NM/16 Bugzilla 851: Documentation example syntax fix.
1553 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1555 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1558 NM/19 Bugzilla 745: TLS version reporting.
1559 Patch provided by Phil Pennock.
1561 NM/20 Bugzilla 167: bool: condition support.
1562 Patch provided by Phil Pennock.
1564 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1565 clients. Patch provided by Phil Pennock.
1567 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1568 Patch provided by Brad "anomie" Jorsch.
1570 NM/23 Bugzilla 687: Fix misparses in eximstats.
1571 Patch provided by Heiko Schlittermann.
1573 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1574 Patch provided by Heiko Schlittermann.
1576 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1577 plus update to original patch.
1579 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1581 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1582 Patch provided by David Brownlee.
1584 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1586 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1588 NM/30 Bugzilla 888: TLS documentation bugfixes.
1590 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1592 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1593 Unlike the original bugzilla I have changed all shell scripts in src tree.
1595 NM/33 Bugzilla 898: Transport filter timeout fix.
1596 Patch by Todd Rinaldo.
1598 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1599 Patch by Serge Demonchaux.
1601 NM/35 Bugzilla 39: Base64 decode bug fixes.
1602 Patch by Jakob Hirsch.
1604 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1606 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1608 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1610 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1616 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1617 ALT-N's libdkim that I have put here:
1618 http://duncanthrax.net/exim-experimental/
1620 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1621 were called 'true' and it seems that C99 defines that as a reserved
1622 keyword to be used with 'bool' variable types. That means you could
1623 not include C99-style headers which use bools without triggering
1624 build errors in sieve.c.
1626 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1627 as mailq or other aliases. Changed the --help handling significantly
1628 to do whats expected. exim_usage() emits usage/help information.
1630 SC/01 Added the -bylocaldomain option to eximstats.
1632 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1634 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1636 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1642 PH/01 Another patch from the Sieve maintainer.
1644 PH/02 When an IPv6 address is converted to a string for single-key lookup
1645 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1646 dots are used instead of colons so that keys in lsearch files need not
1647 contain colons. This was done some time before quoting was made available
1648 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1649 (notated using the quote facility) so as to distinguish them from IPv4
1650 keys. This meant that lookups for IP addresses in host lists did not work
1651 for iplsearch lookups.
1653 This has been fixed by arranging for IPv6 addresses to be expressed with
1654 colons if the lookup type is iplsearch. This is not incompatible, because
1655 previously such lookups could never work.
1657 The situation is now rather anomolous, since one *can* have colons in
1658 ordinary lsearch keys. However, making the change in all cases is
1659 incompatible and would probably break a number of configurations.
1661 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1664 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1665 conversion specification without a maximum field width, thereby enabling
1666 a rogue spamd server to cause a buffer overflow. While nobody in their
1667 right mind would setup Exim to query an untrusted spamd server, an
1668 attacker that gains access to a server running spamd could potentially
1669 exploit this vulnerability to run arbitrary code as the Exim user.
1671 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1672 $primary_hostname instead of what libspf2 thinks the hosts name is.
1674 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1675 a directory entry by the name of the lookup key. Previously, if a
1676 symlink pointed to a non-existing file or a file in a directory that
1677 Exim lacked permissions to read, a lookup for a key matching that
1678 symlink would fail. Now it is enough that a matching directory entry
1679 exists, symlink or not. (Bugzilla 503.)
1681 PH/03 The body_linecount and body_zerocount variables are now exported in the
1684 PH/04 Added the $dnslist_matched variable.
1686 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1687 This means they are set thereafter only if the connection becomes
1690 PH/06 Added the client_condition to authenticators so that some can be skipped
1691 by clients under certain conditions.
1693 PH/07 The error message for a badly-placed control=no_multiline_responses left
1694 "_responses" off the end of the name.
1696 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1698 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1699 (without spaces) instead of just copying the configuration text.
1701 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1703 PH/11 Added $max_received_linelength.
1705 PH/12 Added +ignore_defer and +include_defer to host lists.
1707 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1708 way in which PCRE > 7.0 is built.
1710 PH/14 Implemented queue_only_load_latch.
1712 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1713 MAIL command. The effect was to mangle the value on 64-bit systems.
1715 PH/16 Another patch from the Sieve maintainer.
1717 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1719 PH/18 If a system quota error occurred while trying to create the file for
1720 a maildir delivery, the message "Mailbox is full" was not appended to the
1721 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1722 only to a quota excession during the actual writing of the file.
1724 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1725 characters?) which causes problems in log lines. The DN values are now
1726 passed through string_printing() before being added to log lines.
1728 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1729 and InterBase are left for another time.)
1731 PH/21 Added message_body_newlines option.
1733 PH/22 Guard against possible overflow in moan_check_errorcopy().
1735 PH/23 POSIX allows open() to be a macro; guard against that.
1737 PH/24 If the recipient of an error message contained an @ in the local part
1738 (suitably quoted, of course), incorrect values were put in $domain and
1739 $local_part during the evaluation of errors_copy.
1745 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1746 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1749 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1750 issue a MAIL command.
1752 PH/02 In an ACL statement such as
1754 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1756 if a client was not listed at all, or was listed with a value other than
1757 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1758 the condition was not true (as it should be), so access was not denied.
1759 The bug was that the ! inversion was incorrectly passed on to the second
1760 item. This has been fixed.
1762 PH/03 Added additional dnslists conditions == and =& which are different from
1763 = and & when the dns lookup returns more than one IP address.
1765 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1766 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1768 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1769 FSYNC, which compiles an option called disable_fsync that allows for
1770 bypassing fsync(). The documentation is heavily laced with warnings.
1772 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1774 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1775 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1776 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1777 including adding "make clean"; (3) Added -fPIC when compiling the test
1778 dynamically loaded module, to get rid of a warning.
1780 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1781 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1782 = 0s. The bug is otherwise harmless.
1784 PH/07 There was a bug in the dovecot authenticator such that the value of
1785 $auth1 could be overwritten, and so not correctly preserved, after a
1786 successful authentication. This usually meant that the value preserved by
1787 the server_setid option was incorrect.
1789 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1791 PH/09 Installed PCRE release 7.0.
1793 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1794 run for batched SMTP input. It is now run at the start of every message
1795 in the batch. While fixing this I discovered that the process information
1796 (output by running exiwhat) was not always getting set for -bs and -bS
1797 input. This is fixed, and it now also says "batched" for BSMTP.
1799 PH/11 Added control=no_pipelining.
1801 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1802 patch, slightly modified), and move the expansion of helo_data till after
1803 the connection is made in the smtp transport (so it can use these
1806 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1808 PH/14 Added log_selector = +pid.
1810 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1812 PH/16 Add ${if forany and ${if forall.
1814 PH/17 Added dsn_from option to vary the From: line in DSNs.
1816 PH/18 Flush SMTP output before performing a callout, unless control =
1817 no_callout_flush is set.
1819 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1820 was true (the default) a successful delivery failed to delete the retry
1821 item, thus causing premature timeout of the address. The bug is now
1824 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1826 PH/21 Long custom messages for fakedefer and fakereject are now split up
1827 into multiline reponses in the same way that messages for "deny" and
1828 other ACL rejections are.
1830 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1831 with slight modification.
1833 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1834 draft, changing the syntax and factoring some duplicate code".
1836 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1837 for deliveries of the second and subsequent messages over the same SMTP
1840 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1841 ${reduce, with only minor "tidies".
1843 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1845 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1846 expansion side effects.
1848 PH/27 When a message times out after an over-quota error from an Exim-imposed
1849 quota, the bounce message says "mailbox is full". This message was not
1850 being given when it was a system quota that was exceeded. It now should
1853 MH/03 Made $recipients available in local_scan(). local_scan() already has
1854 better access to the recipient list through recipients_list[], but
1855 $recipients can be useful in postmaster-provided expansion strings.
1857 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1858 in the case of a MAIL command with additional options following the
1859 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1860 were accidentally chopped off.
1862 PH/29 SMTP synchronization checks are implemented when a command is read -
1863 there is a check that no more input is waiting when there shouldn't be
1864 any. However, for some commands, a delay in an ACL can mean that it is
1865 some time before the response is written. In this time, more input might
1866 arrive, invalidly. So now there are extra checks after an ACL has run for
1867 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1868 pipelining has not been advertised.
1870 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1872 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1873 returned in the SMTP response when smtp_return_error_details was false.
1874 This has been fixed.
1876 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1877 instead of the C I/O that was originally supplied, because problems were
1878 reported on Solaris.
1880 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1881 Exim which did not show up earlier: it was assuming that a call to
1882 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1883 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1884 was a macro that became an assignment, so it seemed to work. This has
1885 changed to a proper function call with a void return, hence the compile
1886 error. Exim's code has been fixed.
1888 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1891 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1893 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1895 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1896 because it turns out to be too big or there is a timeout) there is no
1897 "Completed" line in the log. When some messages of this type were
1898 selected by exigrep, they were listed as "not completed". Others were
1899 picked up by some special patterns. I have improved the selection
1900 criteria to be more general.
1902 PH/38 The host_find_failed option in the manualroute router can now be set
1903 to "ignore", to completely ignore a host whose IP address cannot be
1904 found. If all hosts are ignored, the behaviour is controlled by the new
1905 host_all_ignored option.
1907 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1908 homing or because of multiple MX records with /mx) generated more than
1909 one IP address, and the following item turned out to be the local host,
1910 all the secondary addresses of the first item were incorrectly removed
1911 from the list, along with the local host and any following hosts (which
1912 is what is supposed to happen).
1914 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1915 whoever called Exim into the -H file. In the case of the daemon it was
1916 behaving confusingly. When first started, it used values for whoever
1917 started the daemon, but after a SIGHUP it used the Exim user (because it
1918 calls itself on a restart). I have changed the code so that it now always
1921 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1922 message are rejected with the same error (e.g. no authentication or bad
1923 sender address), and a DATA command is nevertheless sent (as can happen
1924 with PIPELINING or a stupid MUA), the error message that was given to the
1925 RCPT commands is included in the rejection of the DATA command. This is
1926 intended to be helpful for MUAs that show only the final error to their
1929 PH/42 Another patch from the Sieve maintainer.
1931 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1932 Eximstats - Fixed some broken HTML links and added missing column headers
1934 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1935 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1937 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1939 PH/43 Yet another patch from the Sieve maintainer.
1941 PH/44 I found a way to check for a TCP/IP connection going away before sending
1942 the response to the final '.' that terminates a message, but only in the
1943 case where the client has not sent further data following the '.'
1944 (unfortunately, this is allowed). However, in many cases there won't be
1945 any further data because there won't be any more messages to send. A call
1946 to select() can be used: if it shows that the input is "ready", there is
1947 either input waiting, or the socket has been closed. An attempt to read
1948 the next input character can distinguish the two cases. Previously, Exim
1949 would have sent an OK response which the client would never have see.
1950 This could lead to message repetition. This fix should cure that, at
1951 least in a lot of common cases.
1953 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1954 advertised in response to EHLO.
1960 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1961 fixed by 4.65/MH/01 (is this a record?) are fixed:
1963 (i) An empty string was always treated as zero by the numeric comparison
1964 operators. This behaviour has been restored.
1966 (ii) It is documented that the numeric comparison operators always treat
1967 their arguments as decimal numbers. This was broken in that numbers
1968 starting with 0 were being interpreted as octal.
1970 While fixing these problems I realized that there was another issue that
1971 hadn't been noticed. Values of message_size_limit (both the global option
1972 and the transport option) were treated as octal if they started with 0.
1973 The documentation was vague. These values are now always treated as
1974 decimal, and I will make that clear in the documentation.
1980 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1981 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1984 MH/01 Don't check that the operands of numeric comparison operators are
1985 integers when their expansion is in "skipping" mode (fixes bug
1986 introduced by 4.64-PH/07).
1988 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1989 child addresses, Exim now panics and dies. Previously, because the count
1990 is held in a short int, deliveries were likely to be lost. As such a
1991 large number of recipients for a single message is ridiculous
1992 (performance will be very, very poor), I have chosen to impose a limit
1993 rather than extend the field.
1999 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2000 leftover -K file (the existence of which was triggered by #402).
2001 While we were at it, introduced process PID as part of the -K
2002 filename. This should rule out race conditions when creating
2005 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2006 processing considerably. Previous code took too long for large mails,
2007 triggering a timeout which in turn triggers #401.
2009 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2010 in the DK code in transports.c. sendfile() is not really portable,
2011 hence the _LINUX specificness.
2013 TF/01 In the add_headers option to the mail command in an Exim filter,
2014 there was a bug that Exim would claim a syntax error in any
2015 header after the first one which had an odd number of characters
2018 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2019 callout verification, Exim cached a "reject" for the entire domain. This
2020 is correct for most verifications, but it is not correct for a recipient
2021 verification with use_sender or use_postmaster set, because in that case
2022 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2023 case of MAIL FROM:<> rejection from other early rejections (e.g.
2024 rejection of HELO). When verifying a recipient using a non-null MAIL
2025 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2026 Whatever the result of the callout, the value of the domain cache is
2027 left unchanged (for any other kind of callout, getting as far as trying
2028 RCPT means that the domain itself is ok).
2030 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2033 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2034 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2035 session, the daemon ignores ECONNECT errors and logs others; it now
2036 ignores EPIPE as well.
2038 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2039 (quoted-printable decoding).
2041 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2042 later the small subsequent patch to fix an introduced bug.
2044 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2046 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2048 PH/08 An error is now given if message_size_limit is specified negative.
2050 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2051 to be given (somewhat) arbitrary names.
2053 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2056 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2057 miscellaneous code fixes
2059 PH/10 Added the log_reject_target ACL modifier to specify where to log
2062 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2063 hostname. This is wrong, because it relates to the incoming message (and
2064 probably the interface on which it is arriving) and not to the outgoing
2065 callout (which could be using a different interface). This has been
2066 changed to use the value of the helo_data option from the smtp transport
2067 instead - this is what is used when a message is actually being sent. If
2068 there is no remote transport (possible with a router that sets up host
2069 addresses), $smtp_active_hostname is used.
2071 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2072 tweaks were necessary in order to get it to work (see also 21 below):
2073 (a) The code assumed that strncpy() returns a negative number on buffer
2074 overflow, which isn't the case. Replaced with Exim's string_format()
2076 (b) There were several signed/unsigned issues. I just did the minimum
2077 hacking in of casts. There is scope for a larger refactoring.
2078 (c) The code used strcasecmp() which is not a standard C function.
2079 Replaced with Exim's strcmpic() function.
2080 (d) The code set only $1; it now sets $auth1 as well.
2081 (e) A simple test gave the error "authentication client didn't specify
2082 service in request". It would seem that Dovecot has changed its
2083 interface. Fortunately there's a specification; I followed it and
2084 changed what the client sends and it appears to be working now.
2086 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2089 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2090 address A is aliased to B and C, where B exists and C does not. Without
2091 -v the output is "A verified" because verification stops after a
2092 successful redirection if more than one address is generated. However,
2093 with -v the child addresses are also verified. Exim was outputting "A
2094 failed to verify" and then showing the successful verification for C,
2095 with its parentage. It now outputs "B failed to verify", showing B's
2096 parentage before showing the successful verification of C.
2098 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2099 look up a TXT record in a specific list after matching in a combined
2102 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2103 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2104 they consult the DNS. I had assumed they would set it the way they
2105 wanted; and indeed my experiments on Linux seem to show that in some
2106 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2107 To be on the safe side, however, I have now made the interface to
2108 host_find_byname() similar to host_find_bydns(), with an argument
2109 containing the DNS resolver options. The host_find_byname() function now
2110 sets these options at its start, just as host_find_bydns() does. The smtp
2111 transport options dns_qualify_single and dns_search_parents are passed to
2112 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2113 of host_find_byname() use the default settings of RES_DEFNAMES
2114 (qualify_single) but not RES_DNSRCH (search_parents).
2116 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2117 spool_read_header() do less string testing, by means of a preliminary
2118 switch on the second character of optional "-foo" lines. (This is
2119 overdue, caused by the large number of possibilities that now exist.
2120 Originally there were few.) While I was there, I also converted the
2121 str(n)cmp tests so they don't re-test the leading "-" and the first
2122 character, in the hope this might squeeze out yet more improvement.
2124 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2125 flag allowing group syntax was set by the header_syntax check but not
2126 turned off, possible causing trouble later; (2) The flag was not being
2127 set at all for the header_verify test, causing "group"-style headers to
2128 be rejected. I have now set it in this case, and also caused header_
2129 verify to ignore an empty address taken from a group. While doing this, I
2130 came across some other cases where the code for allowing group syntax
2131 while scanning a header line wasn't quite right (mostly, not resetting
2132 the flag correctly in the right place). These bugs could have caused
2133 trouble for malformed header lines. I hope it is now all correct.
2135 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2136 with the "reply" argument non-NULL. The code, however (which originally
2137 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2138 but it didn't always do it. This confused somebody who was copying the
2139 code for some other use. I have removed all the tests.
2141 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2142 feature that was used to support insecure browsers during the U.S. crypto
2143 embargo. It requires special client support, and Exim is probably the
2144 only MTA that supported it -- and would never use it because real RSA is
2145 always available. This code has been removed, because it had the bad
2146 effect of slowing Exim down by computing (never used) parameters for the
2147 RSA_EXPORT functionality.
2149 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2150 authenticator to fail if there's a tab character in the incoming data
2151 (there should never be unless someone is messing about, as it's supposed
2152 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2153 if the connection is using TLS or if the remote IP is the same as the
2154 local IP, and the "valid-client-cert option" if a client certificate has
2157 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2158 authenticators. This can be used for authorization after authentication
2159 succeeds. (In the case of plaintext, it servers for both authentication
2162 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2163 if any retry times were supplied.
2165 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2166 connection, where there is no client IP address to check. In this
2167 situation, the verify now always succeeds.
2169 PH/25 Applied John Jetmore's -Mset patch.
2171 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2173 PH/27 In a string expansion for a processed (not raw) header when multiple
2174 headers of the same name were present, leading whitespace was being
2175 removed from all of them, but trailing whitespace was being removed only
2176 from the last one. Now trailing whitespace is removed from each header
2177 before concatenation. Completely empty headers in a concatenation (as
2178 before) are ignored.
2180 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2181 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2183 PH/29 [Removed. This was a change that I later backed out, and forgot to
2184 correct the ChangeLog entry (that I had efficiently created) before
2185 committing the later change.]
2187 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2188 address errors (4xx response to RCPT) over the same connection as other
2189 messages routed to the same hosts. Such deliveries are always "forced",
2190 so retry times are not inspected. This resulted in far too many retries
2191 for the affected addresses. The effect occurred only when there were more
2192 hosts than the hosts_max_try setting in the smtp transport when it had
2193 the 4xx errors. Those hosts that it had tried were not added to the list
2194 of hosts for which the message was waiting, so if all were tried, there
2195 was no problem. Two fixes have been applied:
2197 (i) If there are any address or message errors in an SMTP delivery, none
2198 of the hosts (tried or untried) are now added to the list of hosts
2199 for which the message is waiting, so the message should not be a
2200 candidate for sending over the same connection that was used for a
2201 successful delivery of some other message. This seems entirely
2202 reasonable: after all the message is NOT "waiting for some host".
2203 This is so "obvious" that I'm not sure why it wasn't done
2204 previously. Hope I haven't missed anything, but it can't do any
2205 harm, as the worst effect is to miss an optimization.
2207 (ii) If, despite (i), such a delivery is accidentally attempted, the
2208 routing retry time is respected, so at least it doesn't keep
2209 hammering the server.
2211 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2212 in ${readsocket because some servers need this prod.
2214 PH/32 Added some extra debug output when updating a wait-xxx database.
2216 PH/33 The hint "could be header name not terminated by colon", which has been
2217 given for certain expansion errors for a long time, was not being given
2218 for the ${if def:h_colon_omitted{... case.
2220 PH/34 The spec says: "With one important exception, whenever a domain list is
2221 being scanned, $domain contains the subject domain." There was at least
2222 one case where this was not true.
2224 PH/35 The error "getsockname() failed: connection reset by peer" was being
2225 written to the panic log as well as the main log, but it isn't really
2226 panic-worthy as it just means the connection died rather early on. I have
2227 removed the panic log writing for the ECONNRESET error when getsockname()
2230 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2231 runs only) independently of the message's sender address. This meant
2232 that, if the 4xx error was in fact related to the sender, a different
2233 message to the same recipient with a different sender could confuse
2234 things. In particualar, this can happen when sending to a greylisting
2235 server, but other circumstances could also provoke similar problems.
2236 I have changed the default so that the retry time for these errors is now
2237 based a combination of the sender and recipient addresses. This change
2238 can be overridden by setting address_retry_include_sender=false in the
2241 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2242 remote server are returned as part of bounce messages. This was not
2243 happening for LMTP over a pipe (the lmtp transport), but now it is the
2244 same for both kinds of LMTP.
2246 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2247 in header lines that were in fact CNAMEs. This is no longer the case.
2249 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2250 and queue runs started by the daemon processed all messages. This has
2251 been fixed so that -R and -S can now usefully be given with -q<time>.
2253 PH/40 Import PCRE release 6.7 (fixes some bugs).
2255 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2257 PH/42 Give an error if -q is specified more than once.
2259 PH/43 Renamed the variables $interface_address and $interface_port as
2260 $received_ip_address and $received_port, to make it clear that these
2261 values apply to message reception, and not to the outgoing interface when
2262 a message is delivered. (The old names remain recognized, of course.)
2264 PH/44 There was no timeout on the connect() call when using a Unix domain
2265 socket in the ${readsocket expansion. There now is.
2267 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2268 be meaningful with "accept".
2270 SC/01 Eximstats V1.43
2271 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2273 SC/02 Eximstats V1.44
2274 Use a glob alias rather than an array ref in the generated
2275 parser. This improves both readability and performance.
2277 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2278 Collect SpamAssassin and rejection statistics.
2279 Don't display local sender or destination tables unless
2280 there is data to show.
2281 Added average volumes into the top table text output.
2283 SC/04 Eximstats V1.46
2284 Collect data on the number of addresses (recipients)
2285 as well as the number of messages.
2287 SC/05 Eximstats V1.47
2288 Added 'Message too big' to the list of mail rejection
2289 reasons (thanks to Marco Gaiarin).
2291 SC/06 Eximstats V1.48
2292 Mainlog lines which have GMT offsets and are too short to
2293 have a flag are now skipped.
2295 SC/07 Eximstats V1.49 (Alain Williams)
2296 Added the -emptyok flag.
2298 SC/08 Eximstats V1.50
2299 Fixes for obtaining the IP address from reject messages.
2301 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2302 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2303 whitesspace changes from 4.64-PH/27
2305 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2308 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2309 are found, allow negative numbers in numeric criteria)
2311 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2313 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2314 to match changes made in 4.64-PH/43
2316 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2318 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2319 "domain" is actually an IP address, and force a failure. This locks out
2320 those revolvers/nameservers that support "A-for-A" lookups, in
2321 contravention of the specifications.
2323 PH/48 When a host name was looked up from an IP address, and the subsequent
2324 forward lookup of the name timed out, the host name was left in
2325 $sender_host_name, contrary to the specification.
2327 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2328 restricted to single-key lookups, Exim was not diagnosing an error if
2329 * or *@ was used with a query-style lookup.
2331 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2333 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2334 long ago, but noone interested enough thought of it. Let's just say that
2335 the "1.1" means that there are some new functions that weren't there at
2336 some point in the past.
2338 PH/51 Error processing for expansion failure of helo_data from an smtp
2339 transport during callout processing was broken.
2341 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2342 tested/used via the -bh/-bhc/-bs options.
2344 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2345 bug, fixed in subsequent PCRE releases).
2347 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2348 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2350 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2356 SC/01 Use a glob alias rather than an array ref in eximstats generated
2357 parser. This improves both readability and performance.
2359 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2360 Don't display local sender or destination tables in eximstats unless
2361 there is data to show.
2362 Added average volumes into the eximstats top table text output.
2364 SC/03 Collect data on the number of addresses (recipients) as well
2365 as the number of messages in eximstats.
2367 TF/01 Correct an error in the documentation for the redirect router. Exim
2368 does (usually) call initgroups() when daemonizing.
2370 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2371 with consistent privilege compared to when running as a daemon.
2373 TF/03 Note in the spec that $authenticated_id is not set for local
2374 submissions from trusted users.
2376 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2377 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2379 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2380 by adding some example configuration directives to the default
2381 configuration file. A little bit of work is required to uncomment the
2382 directives and define how usernames and passwords are checked, but
2383 there is now a framework to start from.
2385 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2386 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2387 without this. I don't know how relevant this is to other LDAP libraries.
2389 PH/02 Add the verb name to the "unknown ACL verb" error.
2391 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2393 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2395 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2396 directory not expanded when it should be if an expanded home directory
2397 was set for the address (which is overridden by the transport).
2399 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2402 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2403 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2404 because it is too late at that time, and has no effect.
2406 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2407 security issue with \' (bugzilla #107). I could not use the
2408 PQescapeStringConn() function, because it needs a PGconn value as one of
2411 PH/08 When testing addresses using -bt, indicate those final addresses that
2412 are duplicates that would not cause an additional delivery. At least one
2413 person was confused, thinking that -bt output corresponded to deliveries.
2414 (Suppressing duplicates isn't a good idea as you lose the information
2415 about possibly different redirections that led to the duplicates.)
2417 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2418 systems where poll() doesn't work, in particular OS X.
2420 PH/10 Added more information to debugging output for retry time not reached.
2422 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2423 operations in malware.c.
2425 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2428 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2429 syslog_duplication set false, log lines that would normally be written
2430 both the the main log and to the reject log were not written to syslog at
2433 PH/14 In the default configuration, change the use of "message" in ACL warn
2434 statements to "add_header".
2436 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2437 not followed by a command (e.g. "seen endif").
2439 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2440 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2443 PH/17 Added extra conditions to the default value of delay_warning_condition
2447 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2448 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2449 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2452 The Auto-Submitted: and various List- headers are standardised, whereas I
2453 don't think Precedence: ever was.
2455 PH/18 Refactored debugging code in route_finduser() to show more information,
2456 in particular, the error code if getpwnam() issues one.
2458 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2459 This is apparently needed in addition to the PH/07 change above to avoid
2460 any possible encoding problems.
2462 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2463 but not after initializing Perl.
2465 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2466 output them only if debugging. By default they are written stderr,
2467 apparently, which is not desirable.
2469 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2472 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2475 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2477 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2478 authenticated or an ident call has been made. Suppress the default
2479 values for $authenticated_id and $authenticated_sender (but permit -oMai
2480 and -oMas) when testing with -bh.
2482 PH/24 Re-jigged the order of the tests in the default configuration so that the
2483 tests for valid domains and recipients precede the DNS black list and CSA
2484 tests, on the grounds that those ones are more expensive.
2486 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2487 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2488 This bug exists in every version of Exim that I still have, right back to
2491 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2492 However, an attempt to turn on case-sensitivity in a regex key by
2493 including (?-i) didn't work because the subject string was already
2494 lowercased, and the effects were non-intuitive. It turns out that a
2495 one-line patch can be used to allow (?-i) to work as expected.
2501 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2502 other effects) broke the use of negated acl sub-conditions.
2504 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2507 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2508 "Deny" causes Exim to reject the incoming connection with a 554 error.
2509 Unfortunately, if there is a major crisis, such as a disk failure,
2510 tcp-wrappers gives "deny", whereas what one would like would be some
2511 kind of temporary error. A kludge has been added to help with this.
2512 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2513 554 error is used if errno is still zero or contains ENOENT (which occurs
2514 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2517 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2519 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2520 errors. Otherwise a message that provokes a temporary error (when other
2521 messages do not) can cause a whole host to time out.
2523 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2524 addresses were routed directly to files or pipes from a redirect router.
2525 File deliveries just didn't batch; pipe deliveries might have suffered
2528 PH/06 A failure to get a lock for a hints database would erroneously always say
2529 "Failed to get write lock", even when it was really a read lock.
2531 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2532 of 0600. This has been changed to use the value of the lockfile_mode
2533 option (which defaults to 0600).
2535 PH/08 Applied small patch from the Sieve maintainer.
2537 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2538 folder from quota calculations, a direct delivery into this folder messed
2539 up the contents of the maildirsize file. This was because the regex was
2540 used only to exclude .Trash (or whatever) when the size of the mailbox
2541 was calculated. There was no check that a delivery was happening into an
2542 excluded directory. This bug has been fixed by ignoring all quota
2543 processing for deliveries into excluded directories.
2545 PH/10 Added the maildirfolder_create_regex option to appendfile.
2551 PH/01 The code for finding all the local interface addresses on a FreeBSD
2552 system running IPv6 was broken. This may well have applied to all BSD
2553 systems, as well as to others that have similar system calls. The broken
2554 code found IPv4 interfaces correctly, but gave incorrect values for the
2555 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2556 that it would not match correctly against @[] and not recognize the IPv6
2559 PH/02 The ipliteral router was not recognizing addresses of the form user@
2560 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2562 PH/03 Added disable_ipv6.
2564 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2565 decoded form, because it is most often used to construct To: headers
2566 lines in autoreplies, and the decoded form may well be syntactically
2567 invalid. However, $reply_address has leading white space removed, and all
2568 newlines turned into spaces so that the autoreply transport does not
2571 PH/05 If group was specified without a user on a router, and no group or user
2572 was specified on a transport, the group from the router was ignored.
2574 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2575 for visible compile-time settings that can be used to change these
2576 numbers, for those that want even more. Backwards compatibility with old
2577 spool files has been maintained. However, going back to a previous Exim
2578 release will lost any variables that are in spool files.
2580 PH/07 Two small changes when running in the test harness: increase delay when
2581 passing a TCP/IP connection to a new process, in case the original
2582 process has to generate a bounce, and remove special handling of
2583 127.0.0.2 (sic), which is no longer necessary.
2585 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2586 be the same on different OS.
2588 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2591 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2592 whether --show-vars was specified or not
2594 JJ/02 exipick: Added support for new ACL variable spool format introduced
2597 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2598 syntactically invalid From: or Reply-to: line, and a filter used this to
2599 generate an autoreply, and therefore failed to obtain an address for the
2600 autoreply, Exim could try to deliver to a non-existent relative file
2601 name, causing unrelated and misleading errors. What now happens is that
2602 it logs this as a hard delivery error, but does not attempt to create a
2605 PH/11 The exinext utility has a -C option for testing purposes, but although
2606 the given file was scanned by exinext itself; it wasn't being passed on
2607 when Exim was called.
2609 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2610 an end-of-file indication when reading a command response.
2612 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2613 compiled. In many other places in Exim, IPv6 addresses are always
2614 recognized, so I have changed this. It also means that IPv4 domain
2615 literals of the form [IPV4:n.n.n.n] are now always recognized.
2617 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2618 used if the router is not running as root, for example, when verifying at
2619 ACL time, or when using -bh. The debugging output from this situation was
2620 non-existent - all you got was a failure to exec. I have made two
2623 (a) Failures to set uid/gid, the current directory, or a process leader
2624 in a subprocess such as that created by queryprogram now generate
2625 suitable debugging ouput when -d is set.
2627 (b) The queryprogram router detects when it is not running as root,
2628 outputs suitable debugging information if -d is set, and then runs
2629 the subprocess without attempting to change uid/gid.
2631 PH/15 Minor change to Makefile for building test_host (undocumented testing
2634 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2635 additional section of a DNS packet that returns MX or SRV records.
2636 Instead, it always explicitly searches for A/AAAA records. This avoids
2637 major problems that occur when a DNS server includes only records of one
2638 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2639 fixed another bug: if SRV records were looked up and the corresponding
2640 address records were *not* found in the additional section, the port
2641 values from the SRV records were lost.
2643 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2644 using the correct key (the original address) when searching the retry
2645 rules in order to find which one to use for generating the retry hint.
2647 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2648 adding the default one. Similarly, if it contains a Reply-To: header, the
2649 errors_reply_to option, if set, is not used.
2651 PH/19 When calculating a retry time, Exim used to measure the "time since
2652 failure" by looking at the "first failed" field in the retry record. Now
2653 it does not use this if it is later than than the arrival time of the
2654 message. Instead it uses the arrival time. This makes for better
2655 behaviour in cases where some deliveries succeed, thus re-setting the
2656 "first failed" field. An example is a quota failure for a huge message
2657 when small messages continue to be delivered. Without this change, the
2658 "time since failure" will always be short, possible causing more frequent
2659 delivery attempts for the huge message than are intended.
2660 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2662 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2663 $1, $2, $3) because the numerical variables can be reset during some
2664 expansion items (e.g. "match"), thereby losing the authentication data.
2666 PH/21 Make -bV show the size of off_t variables so that the test suite can
2667 decide whether to run tests for quotas > 2G.
2669 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2670 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2671 filecount value is greater than 2G or if a quota value is greater than 2G
2672 on a system where the size of off_t is not greater than 4, a panic error
2675 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2676 never match. The debug and -bh output now contains an explicit error
2677 message indicating a malformed IPv4 address or mask.
2679 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2680 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2681 PH/23 above applies.
2683 PH/25 Do not write to syslog when running in the test harness. The only
2684 occasion when this arises is a failure to open the main or panic logs
2685 (for which there is an explicit test).
2687 PH/26 Added the /no_tell option to "control=freeze".
2689 PH/27 If a host name lookup failed very early in a connection, for example, if
2690 the IP address matched host_lookup and the reverse lookup yielded a name
2691 that did not have a forward lookup, an error message of the form "no IP
2692 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2693 could be logged. Now it outputs the IP address instead of "NULL".
2695 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2696 allows the sender and the authenticated sender to be set when
2697 submitting a message from within Exim. Since child_open_exim() is
2698 documented for local_scan(), the new function should be too.
2700 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2701 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2702 results in an empty string is now treated as unset.
2704 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2706 PH/31 Added sender_verify_fail logging option.
2708 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2709 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2710 tidied the source and removed it altogether.
2712 PH/33 When a queue run was abandoned because the load average was too high, a
2713 log line was always written; now it is written only if the queue_run log
2714 selector is set. In addition, the log line for abandonment now contains
2715 information about the queue run such as the pid. This is always present
2716 in "start" and "stop" lines but was omitted from the "abandon" line.
2718 PH/34 Omit spaces between a header name and the colon in the error message that
2719 is given when verify = headers_syntax fails (if there are lots of them,
2720 the message gets confusing).
2722 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2723 names, as there are now some PTR records that contain slashes. This check
2724 is only to protect against broken name servers that fall over on strange
2725 characters, so the fact that it applies to all lookups doesn't matter.
2727 PH/36 Now that the new test suite is complete, we can remove some of the
2728 special code in Exim that was needed for the old test suite. For example,
2729 sorting DNS records because real resolvers return them in an arbitrary
2730 order. The new test suite's fake resolver always returns records in the
2733 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2734 bounces) except when queue_only is set, to avoid logging races between
2735 the different processes.
2737 PH/38 Panic-die if .include specifies a non-absolute path.
2739 PH/39 A tweak to the "H" retry rule from its user.
2741 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2742 a label. They prevented compilation on older perls.
2744 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2745 a warning to be raised on newish perls.
2747 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2748 on queue. Changes to match documented behaviour of showing count of
2749 messages matching specified criteria.
2751 PH/40 Changed the default ident timeout from 30s to 5s.
2753 PH/41 Added support for the use of login_cap features, on those BSD systems
2754 that have them, for controlling the resources used by pipe deliveries.
2756 PH/42 The content-scanning code uses fopen() to create files in which to put
2757 message data. Previously it was not paying any attention to the mode of
2758 the files. Exim runs with umask(0) because the rest of the code creates
2759 files with open(), and sets the required mode explicitly. Thus, these
2760 files were ending up world-writeable. This was not a big issue, because,
2761 being within the spool directory, they were not world-accessible. I have
2762 created a function called modefopen, which takes an additional mode
2763 argument. It sets umask(777), creates the file, chmods it to the required
2764 mode, then resets the umask. All the relevant calls to fopen() in the
2765 content scanning code have been changed to use this function.
2767 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2768 to 24 hours. This avoids potential overflow problems when processing G
2769 and H retry rules. I suspect nobody ever tinkers with this value.
2771 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2773 PH/45 When the plaintext authenticator is running as a client, the server's
2774 challenges are checked to ensure they are valid base64 strings. By
2775 default, the authentication attempt is cancelled if an invalid string is
2776 received. Setting client_ignore_invalid_base64 true ignores these errors.
2777 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2778 they are received. Thus, the responses can be made to depend on the
2779 challenges. If an invalid string is ignored, an empty string is placed in
2782 PH/46 Messages that are created by the autoreply transport now contains a
2783 References: header, in accordance with RFCs 2822 and 3834.
2785 PH/47 Added authenticated_sender_force to the smtp transport.
2787 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2789 PH/49 Installed latest patch from the Sieve maintainer.
2791 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2792 was also set, the appendfile transport was unnecessarily scanning a
2793 directory of message files (e.g. for maildir delivery) to find the count
2794 of files (along with the size), even though it did not need this
2795 information. It now does the scan only if it needs to find either the
2796 size of the count of files.
2798 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2800 PH/52 Two bugs concerned with error handling when the smtp transport is
2803 (i) Exim was not creating retry information for temporary errors given
2804 for individual recipients after the DATA command when the smtp transport
2805 was used in LMTP mode. This meant that they could be retried too
2806 frequently, and not timed out correctly.
2808 (ii) Exim was setting the flag that allows error details to be returned
2809 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2810 recipients that were returned after the DATA command.
2812 PH/53 This is related to PH/52, but is more general: for any failing address,
2813 when detailed error information was permitted to be returned to the
2814 sender, but the error was temporary, then after the final timeout, only
2815 "retry timeout exceeded" was returned. Now it returns the full error as
2816 well as "retry timeout exceeded".
2818 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2819 do this, and (what is worse) MTAs that accept it.
2821 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2822 will now be deprecated.
2824 PH/56 New os.c-cygwin from the Cygwin maintainer.
2826 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2827 formats (previously only available in exim formats via -bpr, -bpru,
2828 and -bpra. Now also available in native and exiqgrep formats)
2830 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2831 with very large, slow to parse queues
2833 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2835 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2837 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2838 responses to authentication challenges, though it was showing the
2839 challenges; (ii) I've removed the CR characters from the debug output for
2842 PH/58 Allow for the insertion of a newline as well as a space when a string
2843 is turned into more than one encoded-word during RFC 2047 encoding. The
2844 Sieve code now uses this.
2846 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2847 data_4xx, lost_connection, tls_required.
2849 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2850 message was being sent as an SMTP response.
2852 PH/61 Add -l and -k options to exicyclog.
2854 PH/62 When verifying, if an address was redirected to one new address, so that
2855 verification continued, and the new address failed or deferred after
2856 having set something in $address_data, the value of $address_data was not
2857 passed back to the ACL. This was different to the case when no
2858 redirection occurred. The value is now passed back in both cases.
2860 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2861 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2862 use login_cap.h, so on its own it isn't the distinguishing feature. The
2863 new name refers directly to the setclassresources() function.
2865 PH/65 Added configuration files for NetBSD3.
2867 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2869 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2870 is preferred over IPv4.
2872 PH/68 The bounce_return_message and bounce_return_body options were not being
2873 honoured for bounces generated during the reception of non-SMTP messages.
2874 In particular, this applied to messages rejected by the ACL. This bug has
2875 been fixed. However, if bounce_return_message is true and bounce_return_
2876 body is false, the headers that are returned for a non-SMTP message
2877 include only those that have been read before the error was detected.
2878 (In the case of an ACL rejection, they have all been read.)
2880 PH/69 The HTML version of the specification is now built in a directory called
2881 spec_html instead of spec.html, because the latter looks like a path with
2882 a MIME-type, and this confuses some software.
2884 PH/70 Catch two compiler warnings in sieve.c.
2886 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2887 function verify_get_ident() calls ip_connect() to connect a socket, but
2888 if the "connect()" function timed out, ip_connect() used to close the
2889 socket. However, verify_get_ident() also closes the socket later, and in
2890 between Exim writes to the log, which may get opened at this point. When
2891 the socket was closed in ip_connect(), the log could get the same file
2892 descriptor number as the socket. This naturally causes chaos. The fix is
2893 not to close the socket in ip_connect(); the socket should be closed by
2894 the function that creates it. There was only one place in the code where
2895 this was missing, in the iplookup router, which I don't think anybody now
2896 uses, but I've fixed it anyway.
2898 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2899 well as to direct DNS lookups. Otherwise the handling of names in host
2900 lists is inconsistent and therefore confusing.
2906 PH/01 Two changes to the default runtime configuration:
2908 (1) Move the checks for relay_from_hosts and authenticated clients from
2909 after to before the (commented out) DNS black list checks.
2911 (2) Add control=submission to the relay_from_hosts and authenticated
2912 clients checks, on the grounds that messages accepted by these
2913 statements are most likely to be submissions.
2915 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2917 (1) Generate an error if the third argument for the ${prvs expansion is
2920 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2923 (3) Reset the variables that are obtained from the first argument of
2924 ${prvscheck and used in the second argument before leaving the code,
2925 because their memory is reclaimed, so using them afterwards may do
2928 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2929 one (it's much easier than Tom thought :-).
2931 (5) Because of (4), we can now allow for the use of $prvscheck_result
2932 inside the third argument.
2934 PH/03 For some reason, the default setting of PATH when running a command from
2935 a pipe transport was just "/usr/bin". I have changed it to
2938 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2939 anything to be listed in the output from -bV.
2941 PH/05 When a filter generated an autoreply, the entire To: header line was
2942 quoted in the delivery log line, like this:
2944 => >A.N.Other <ano@some.domain> <original@ddress> ...
2946 This has been changed so that it extracts the operative address. There
2947 may be more than one such address. If so, they are comma-separated, like
2950 => >ano@some.domain,ona@other.domain <original@ddress> ...
2952 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2953 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2954 not being looked up in the rDNS to get a host name, Exim was showing the
2955 IP address twice in Received: lines, even though the IP addresses were
2956 identical. For example:
2958 Received: from [1.2.3.4] (helo=[1.2.3.4])
2960 However, if the real host name was known, it was omitting the HELO data
2961 if it matched the actual IP address. This has been tidied up so that it
2962 doesn't show the same IP address twice.
2964 PH/07 When both +timestamp and +memory debugging was on, the value given by
2965 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2966 called by the debug printing, thereby overwriting the timestamp buffer.
2967 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2969 PH/08 When the original message was included in an autoreply transport, it
2970 always said "this is a copy of the message, including all the headers",
2971 even if body_only or headers_only was set. It now gives an appropriate
2974 PH/09 Applied a patch from the Sieve maintainer which:
2976 o fixes some comments
2977 o adds the (disabled) notify extension core
2978 o adds some debug output for the result of if/elsif tests
2979 o points to the current vacation draft in the documentation
2980 and documents the missing references header update
2984 o fixes a bug in processing the envelope test (when testing
2985 multiple envelope elements, the last element determinted the
2988 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2989 Electronic Mail") by including:
2991 Auto-submitted: auto-generated
2993 in the messages that it generates (bounce messages and others, such as
2994 warnings). In the case of bounce messages for non-SMTP mesages, there was
2995 also a typo: it was using "Auto_submitted" (underscore instead of
2996 hyphen). Since every message generated by Exim is necessarily in response
2997 to another message, thes have all been changed to:
2999 Auto-Submitted: auto-replied
3001 in accordance with these statements in the RFC:
3003 The auto-replied keyword:
3005 - SHOULD be used on messages sent in direct response to another
3006 message by an automatic process,
3008 - MUST NOT be used on manually-generated messages,
3010 - MAY be used on Delivery Status Notifications (DSNs) and Message
3011 Disposition Notifications (MDNs),
3013 - MUST NOT be used on messages generated by automatic or periodic
3014 processes, except for messages which are automatic responses to
3017 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3018 to the default Received: header definition.
3020 PH/12 Added log selector acl_warn_skipped (default on).
3022 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3023 variables because (a) they are in the wrong storage pool and (b) even if
3024 they were copied, it wouldn't work properly because of the caching.
3026 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3027 checking when decoding. Apparently there are clients that generate
3028 overlong encoded strings. Why am I not surprised?
3030 PH/15 If the first argument of "${if match_address" was not empty, but did not
3031 contain an "@" character, Exim crashed. Now it writes a panic log message
3032 and treats the condition as false.
3034 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3036 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3037 extension "envelope-auth". The code is finished and in agreement with
3038 other implementations, but there is no documentation so far and in fact,
3039 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3040 not changing the active code.
3042 Print executed "if" and "elsif" statements when debugging is used. This
3043 helps a great deal to understand what a filter does.
3045 Document more things not specified clearly in RFC3028. I had all this
3046 sorted out, when out of a sudden new issues came to my mind. Oops."
3048 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3051 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3052 canonical form (as documented). However, after a host name lookup from
3053 the IP address, check_host() was doing a simple string comparison with
3054 addresses acquired from the DNS when checking that the found name did
3055 have the original IP as one of its addresses. Since any found IPv6
3056 addresses are likely to be in abbreviated form, the comparison could
3057 fail. Luckily, there already exists a function for doing the comparison
3058 by converting both addresses to binary, so now that is used instead of
3059 the text comparison.
3061 PH/20 There was another similar case to PH/19, when a complete host name was
3062 given in a host list; looking up its IP address could give an abbreviated
3063 form, whereas the current host's name might or might not be abbreviated.
3064 The same fix has been applied.
3070 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3071 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3074 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3075 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3077 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3079 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3080 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3081 Also, for "Auto-Submitted", treat anything other than "no" as
3082 non-personal, in accordance with RFC 3834. (Previously it treated
3083 anything starting "auto-" as non-personal.)
3085 TF/01 The control=submission/name=... option had a problem with syntax
3086 errors if the name included a slash character. The /name= option
3087 now slurps the rest of the string, so it can include any characters
3088 but it must come last in the list of options (after /sender_retain
3091 PH/05 Some modifications to the interface to the fake nameserver for the new
3099 TK/01 Added the "success_on_redirect" address verification option. See
3100 NewStuff for rationale and an example.
3102 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3104 PH/02 Patch to exigrep to allow it to work on syslog lines.
3106 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3107 fread() to skip over the body file's header line, because in Cygwin the
3108 header line is locked and is inaccessible.
3110 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3111 co-exist for some time) to make it clear that it is the Exim ID that is
3112 referenced, not the Message-ID: header line.
3114 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3115 string_format() function, because snprintf() does not exist on all
3118 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3119 ${stat: expansion item.
3121 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3122 protocol synchronization error", to keep the pedants happy.
3124 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3125 well as for IRIX systems, when gcc is being used. See the host.c source
3128 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3130 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3133 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3134 transport and to the smtp transport in LMTP mode.
3136 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3138 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3139 run a filter in a subprocess. This could lead to confusion in subsequent
3140 lookups in the parent process. There should also be a search_tidyup() at
3141 the end of the subprocess.
3143 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3144 only if the host matched helo_try_verify_hosts, which caused the
3145 verification to occur when the EHLO/HELO command was issued. The ACL just
3146 tested the remembered result. Now, if a previous verification attempt has
3147 not happened, "verify = helo" does it there and then.
3149 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3151 TK/03 Fix log output including CR from clamd.
3153 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3154 exist provoked a memory error which could cause a segfault.
3156 PH/15 Installed PCRE 6.2
3158 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3160 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3161 of the problem. Specifically, suggested +O2 rather than +O1 for the
3164 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3166 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3167 record, the port was not correctly logged when the outgoing_port log
3168 selector was set (it logged the transort's default port).
3170 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3171 fallback_hosts, and "hosts" in the smtp transport.
3173 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3174 host errors such as "Connection refused".
3176 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3177 authentication with radiusclient 0.4.9:
3179 - Error returned from rc_read_config was caught wrongly
3180 - Username/password not passed on to radius server due to wrong length.
3182 The presumption is that some radiusclient API changes for 4.51/PH/17
3183 were not taken care of correctly. The code is still untested by me (my
3184 Linux distribution still has 0.3.2 of radiusclient), but it was
3185 contributed by a Radius user.
3187 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3188 expanding the "port" option of the smtp transport.
3190 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3191 while reading a MIME header. Thanks to Tom Hughes for a patch.
3193 PH/24 Include config.h inside local_scan.h so that configuration settings are
3196 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3197 that in an ACL for RCPT (for example), you can examine exactly what was
3200 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3201 commands, but it was not correctly comparing the address with the actual
3202 client host address. Thus, it would show the EHLO address in Received:
3203 header lines when this was not necessary.
3205 PH/27 Added the % operator to ${eval:}.
3207 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3208 it should be ignoring failures (because with -C, for example, it has lost
3209 privilege). It wasn't ignoring creation failures other than "already
3212 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3215 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3216 because an input line was too long, either on its own, or by virtue of
3217 too many continuations, the temporary file was not being removed, and the
3218 return code was incorrect.
3220 PH/31 Missing "BOOL" in function definition in filtertest.c.
3222 PH/32 Applied Sieve patches from the maintainer.
3224 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3226 PH/33 Added "verify = not_blind".
3228 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3229 Local/Makefile (with some defaults set). These are used in built scripts
3230 such as exicyclog, but they have never been used in the exim_install
3231 script (though there are many overriding facilities there). I have
3232 arranged that the exim_install script now takes note of these two
3235 PH/35 Installed configuration files for Dragonfly.
3237 PH/36 When a locally submitted message by a trusted user did not contain a
3238 From: header, and the sender address was obtained from -f or from an SMTP
3239 MAIL command, and the trusted user did not use -F to supply a sender
3240 name, $originator_name was incorrectly used when constructing a From:
3241 header. Furthermore, $originator_name was used for submission mode
3242 messages from external hosts without From: headers in a similar way,
3243 which is clearly wrong.
3245 PH/37 Added control=suppress_local_fixups.
3247 PH/38 When log_selector = +received_sender was set, and the addition of the
3248 sender made the log line's construction buffer exactly full, or one byte
3249 less than full, an overflow happened when the terminating "\n" was
3252 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3253 when the result of a list match is failure because a DNS lookup failed.
3255 PH/40 RM_COMMAND is now used in the building process.
3257 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3258 the "build-* directories that it finds.
3260 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3261 address in a domain literal was a prefix of an interface address.
3263 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3264 when verifying a sender address, unless rewrite_headers is false.
3266 PH/44 Wrote a long comment about why errors_to addresses are verified as
3267 recipients, not senders.
3269 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3270 the ratelimit ACL was added.
3272 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3274 PH/46 Added extra information about PostgreSQL errors to the error string.
3276 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3277 avoiding the need to install special zones in a real server. This is
3278 backwards compatible; if it can't find the fake resolver, it drops back.
3279 Thus, both old and new test suites can be run.
3281 TF/02 Added util/ratelimit.pl
3283 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3284 clock is set back in time.
3286 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3287 Candler <B.Candler@pobox.com>.
3289 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3290 OK for addresses that are the result of redirections.
3292 PH/48 A number of further additions for the benefit of the new test suite,
3293 including a fake gethostbyname() that interfaces to the fake DNS resolver
3296 TF/06 The fix for widen_domains has also been applied to qualify_single and
3297 search_parents which are the other dnslookup options that can cause
3300 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3303 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3305 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3306 service" if it encountered an MX record with an empty target hostname.
3307 The message is now "an MX or SRV record indicated no SMTP service".
3309 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3310 if the DNS of the sending site is misconfigured. This is quite a
3311 common situation. This change restores the behaviour of treating a
3312 helo verification defer as a failure.
3314 PH/51 If self=fail was set on a router, the bounce message did not include the
3315 actual error message.
3321 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3323 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3324 command itself ended in error, the underlying message about the transport
3325 filter timeout was being overwritten with the pipe command error. Now the
3326 underlying error message should be appended to the second error message.
3328 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3330 PH/02 Reduce the amount of output that "make" produces by default. Full output
3331 can still be requested.
3333 PH/03 The warning log line about a condition test deferring for a "warn" verb
3334 was being output only once per connection, rather than after each
3335 occurrence (because it was using the same function as for successful
3336 "warn" verbs). This seems wrong, so I have changed it.
3338 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3339 it should not have, which might have caused a crash in the right
3340 circumstances, but probably never did.
3342 PH/04 Installed a modified version of Tony Finch's patch to make submission
3343 mode fix the return path as well as the Sender: header line, and to
3344 add a /name= option so that you can make the user's friendly name appear
3347 TF/03 Added the control = fakedefer ACL modifier.
3349 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3350 Mark Lowes for thorough testing.
3352 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3354 TK/03 Merged latest SRS patch from Miles Wilton.
3356 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3357 with the definition in sysexits.h (which is #included earlier).
3358 Fortunately, Exim does not actually use EX_OK. The code used to try to
3359 preserve the sysexits.h value, by assumimg that macro definitions were
3360 scanned for macro replacements. I have been disabused of this notion,
3361 so now the code just undefines EX_OK before #including unistd.h.
3363 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3364 in the smtp transport. When a block could not be written in a single
3365 write() function, the timeout was being re-applied to each part-write.
3366 This seems wrong - if the receiver was accepting one byte at a time it
3367 would take for ever. The timeout is now adjusted when this happens. It
3368 doesn't have to be particularly precise.
3370 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3371 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3373 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3374 without a domain if the check to <postmaster@domain> fails.
3376 SC/01 Eximstats: added -xls and the ability to specify output files
3377 (patch written by Frank Heydlauf).
3379 SC/02 Eximstats: use FileHandles for outputing results.
3381 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3383 SC/04 Eximstats: fixed display of large numbers with -nvr option
3385 SC/05 Eximstats: fixed merging of reports with empty tables.
3387 SC/06 Eximstats: added the -include_original_destination flag
3389 SC/07 Eximstats: removed tabs and trailing whitespace.
3391 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3393 TK/06 MBOX spool code: Add real "From " MBOX separator line
3394 so the .eml file is really in mbox format (even though
3395 most programs do not really care). Patch from Alex Miller.
3397 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3398 The latter is generated from $received_to and is only set if the
3399 message has one envelope recipient. SA can use these headers,
3400 obviously out-of-the-box. Patch from Alex Miller.
3402 PH/08 The ${def test on a variable was returning false if the variable's
3403 value was "0", contrary to what the specification has always said!
3404 The result should be true unless the variable is empty.
3406 PH/09 The syntax error of a character other than { following "${if
3407 def:variable_name" (after optional whitespace) was not being diagnosed.
3408 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3409 accidental colon was present, for example, could give incorrect results.
3411 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3412 result is used (not including appendfile, where other changes are about
3415 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3416 This involved changing a lot of size variables from int to off_t. It
3417 should work with maildirs and everything.
3419 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3420 spamd dying while we are connected to it.
3422 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3425 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3426 function for BDB 4.3.
3428 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3430 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3431 the sources for PCRE have been re-arranged and more files are now
3434 PH/15 The code I had for printing potentially long long variables in PH/11
3435 above was not the best (it lost precision). The length of off_t variables
3436 is now inspected at build time, and an appropriate printing format (%ld
3437 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3438 to be "long long int" or "long int". This is needed for the internal
3439 formatting function string_vformat().
3441 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3442 the configuration file to be ":syslog", then the script "guesses" where
3443 the logs files are, rather than using the compiled in default. In our
3444 case the guess is not the same as the compiled default, so the script
3445 suddenly stopped working when I started to use syslog. The patch checks
3446 to see if log_file_path is "". If so, it attempts to read it from exim
3447 with no configuration file to get the compiled in version, before it
3448 falls back to the previous guessing code."
3450 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3451 implementing BATV in an Exim configuration. See NewStuff for the gory
3454 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3455 Makefile that are specific to HP-UX.
3457 PH/18 If the "use_postmaster" option was set for a recipient callout together
3458 with the "random" option, the postmaster address was used as the MAIL
3459 FROM address for the random test, but not for the subsequent recipient
3460 test. It is now used for both.
3462 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3463 patch removes a few documentation additions to RFC 3028, because the
3464 latest draft now contains them. It adds the new en;ascii-case comparator
3465 and a new error check for 8bit text in MIME parts. Comparator and
3466 require names are now matched exactly. I enabled the subaddress
3467 extension, but it is not well tested yet (read: it works for me)."
3469 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3470 rework some of the code of TK/09 above to avoid the hardwired use of
3471 "%lld" and "long long". Replaced the call to snprintf() with a call to
3474 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3475 records point to non-existent hosts", "retry timeout exceeded", and
3476 "retry time not reached for any host after a long failure period".
3478 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3479 experimental DomainKeys support:
3481 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3482 (2) On an error such as an illegally used "control", the wrong name for
3483 the control was given.
3485 These problems did NOT occur unless DomainKeys support was compiled.
3487 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3489 PH/24 Added ${if match_ip condition.
3491 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3492 databases so that it will be absolutely obvious if a crash occurs in the
3493 DB library. This is a regular occurrence (often caused by mis-matched
3496 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3497 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3498 picked up on a user's system that detects such things. There doesn't seem
3499 to be a gcc warning option for this - only an attribute that has to be
3500 put on the function's prototype. It seems that in Fedora Core 4 they have
3501 set this on a number of new functions. No doubt there will be more in due
3504 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3505 specify a transport. However, if an address that was verified by such a
3506 router was the subject of a callout, Exim crashed because it tried to
3507 read the rcpt_include_affixes from the non-existent transport. Now it
3508 just assumes that the setting of that option is false. This bug was
3509 introduced by 4.51/PH/31.
3511 PH/28 Changed -d+all to exclude +memory, because that information is very
3512 rarely of interest, but it makes the output a lot bigger. People tend to
3513 do -d+all out of habit.
3515 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3516 code in os-type was giving problems when libc.so lives in lib64, like on
3519 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3520 aren't the modern standard, and it seems that some systems' include files
3521 don't always have them. Exim was already checking for some of the newer
3522 ones like T_AAAA, and defining it itself. I've added checks for all the
3523 record types that Exim uses.
3525 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3526 not automatically generating a new one, as it is supposed to. This
3527 prevented TLS from working. If the file did exist, but contained invalid
3528 data, a new version was generated, as expected. It was only the case of a
3529 non-existent file that was broken.
3531 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3532 with a change in libdomainkeys > 0.64.
3534 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3535 from DNS. If the selector record carries the flag, it now has
3536 precedence over the domain-wide flag.
3538 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3540 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3541 the use of a port name that isn't defined in /etc/services) occurred, the
3542 message was deferred as in a normal delivery, and thus remained on the
3543 spool, instead of being failed because of the mua_wrapper setting. This
3544 is now fixed, and I tidied up some of the mua_wrapper messages at the
3547 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3548 the messages in a hash of arrays rather than using individual hashes.
3549 This is a bit cleaner and results in dramatic memory savings, albeit
3550 at a slight CPU cost.
3552 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3553 as requested by Marc Sherman.
3555 SC/10 Eximstats: added histograms for user specified patterns as requested
3558 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3560 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3561 fopen() in the content-scanning modules that did not already have it.
3567 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3568 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3570 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3572 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3574 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3575 to test Sieve filters that use "vacation".
3577 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3578 that changes the way the GnuTLS parameters are stored in the cache file.
3579 The new format can be generated externally. For backward compatibility,
3580 if the data in the cache doesn't make sense, Exim assumes it has read an
3581 old-format file, and it generates new data and writes a new file. This
3582 means that you can't go back to an older release without removing the
3585 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3586 work if there are any delivery delays because "one_time" forces the
3587 parent to be marked "delivered", so its unseen clone is never tried
3588 again. For this reason, Exim now forbids the simultaneous setting of
3591 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3592 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3593 entry if you want to know the details. The fix, however, neglected to
3594 consider the case when local delivery batching is involved. The test for
3595 "previously delivered" was not happening when checking to see if an
3596 address could be batched with a previous (undelivered) one; under
3597 certain circumstances this could lead to multiple deliveries to the same
3600 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3601 in its include files, and this causes problems building Exim.
3603 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3604 header_syntax) but Exim was just ignoring anything given after a slash.
3605 In particular, this caused confusion with an attempt to use "verify =
3606 reverse_host_lookup/defer_ok". An error is now given when options are
3607 supplied for verify items that do not have them. (Maybe reverse_host_
3608 lookup should have a defer_ok option, but that's a different point.)
3610 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3611 defined by RFC 821) to 2048, because there were problems with some AUTH
3612 commands, and RFC 1869 says the size should be increased for extended
3613 SMTP commands that take arguments.
3615 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3618 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3619 "unknown" error; now it says that the functionality isn't in the binary.
3621 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3622 an address' error message when a string expansion fails (syntax or
3623 whatever). Otherwise the password may appear in the log. Following change
3624 PH/42 below, there is no longer a chance of it appearing in a bounce
3627 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3629 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3630 of its addresses was ever tried. (Bugzilla bug #2.)
3632 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3633 the result incorrectly in the debug output. (It correctly added a newline
3634 to what was transported.)
3636 TF/01 Added $received_time.
3638 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3639 commented out examples of how to interface to a virus scanner and to
3640 SpamAssassin. Also added commented examples of av_scanner and
3641 spamd_address settings.
3643 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3644 and controls are allowed in which ACLs. There were a couple of minor
3645 errors. Some of the entries in the conditions table (which is a table of
3646 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3647 negation of where the condition IS allowed.
3649 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3651 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3652 header file does not have a version number, so I've had to invent a new
3653 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3654 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3655 radiusclient), but it was contributed by a Radius user.
3657 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3658 files or directories, for OpenSSL.
3660 PH/19 When an Exim process that is running as root has to create an Exim log
3661 file, it does so in a subprocess that runs as exim:exim so as to get the
3662 ownership right at creation (otherwise, other Exim processes might see
3663 the file with the wrong ownership). There was no test for failure of this
3664 fork() call, which would lead to the process getting stuck as it waited
3665 for a non-existent subprocess. Forks do occasionally fail when resources
3666 run out. I reviewed all the other calls to fork(); they all seem to check
3669 PH/20 When checking for unexpected SMTP input at connect time (before writing
3670 the banner), Exim was not dealing correctly with a non-positive return
3671 from the read() function. If the client had disconnected by this time,
3672 the result was a log entry for a synchronization error with an empty
3673 string after "input=" when read() returned zero. If read() returned -1
3674 (an event I could not check), uninitialized data bytes were printed.
3675 There were reports of junk text (parts of files, etc) appearing after
3678 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3680 PH/22 Added support for macro redefinition, and (re)definition in between
3681 driver and ACL definitions.
3683 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3684 forgetting to use the resulting value; it was using the unexpanded value.
3686 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3687 hadn't been configured. The fix is from Juergen Kreileder, who
3688 understands it better than I do:
3690 "Here's what I see happening with three configured cyrus_sasl
3691 authenticators configured (plain, login, cram-md5):
3693 On startup auth_cyrus_sasl_init() gets called for each of these.
3694 This means three calls to sasl_listmech() without a specified mech_list.
3695 => SASL tests which mechs of all available mechs actually work
3696 => three warnings about OTP not working
3697 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3699 With the patch, sasl_listmech() also gets called three times. But now
3700 SASL's mech_list option is set to the server_mech specified in the the
3701 authenticator. Or in other words, the answer from sasl_listmech()
3702 gets limited to just the mech you're testing for (which is different
3704 => the return list contains just 'plain' or 'login', 'cram-md5' or
3705 nothing depending on the value of ob->server_mech.
3707 I've just tested the patch: Authentication still works fine,
3708 unavailable mechs specified in the exim configuration are still
3709 caught, and the auth.log warnings about OTP are gone."
3711 PH/25 When debugging is enabled, the contents of the command line are added
3712 to the debugging output, even when log_selector=+arguments is not
3715 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3716 answer is "GNU", and only if the return is "GNU/something" is the answer
3719 PH/27 $acl_verify_message is now set immediately after the failure of a
3720 verification in an ACL, and so is available in subsequent modifiers. In
3721 particular, the message can be preserved by coding like this:
3723 warn !verify = sender
3724 set acl_m0 = $acl_verify_message
3726 Previously, $acl_verify_message was set only while expanding "message"
3727 and "log_message" when a very denied access.
3729 PH/28 Modified OS/os.c-Linux with
3731 -#ifndef OS_LOAD_AVERAGE
3732 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3734 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3735 nomenclature these days.)
3737 PH/29 Installed patch from the Sieve maintainer that adds the options
3738 sieve_useraddress and sieve_subaddress to the redirect router.
3740 PH/30 In these circumstances:
3741 . Two addresses routed to the same list of hosts;
3742 . First host does not offer TLS;
3743 . First host accepts first address;
3744 . First host gives temporary error to second address;
3745 . Second host offers TLS and a TLS session is established;
3746 . Second host accepts second address.
3747 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3748 and peerdn, if requested) that were in fact used only for the second
3751 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3752 attention to any local part prefix or suffix that was matched by the
3753 router that accepted the address. It now behaves in the same way as it
3754 does for delivery: the affixes are removed from the local part unless
3755 rcpt_include_affixes is set on the transport.
3757 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3758 timeout during the DATA phase of an incoming message.
3760 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3761 applied a patch sanctioned by the Sieve maintainer.
3763 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3764 the uid or gid is negative. A case of a negative gid caused this to be
3765 noticed. The fix allows for either to be negative.
3767 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3768 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3771 PH/36 The change PH/12 above was broken. Fixed it.
3773 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3774 the grounds that routing the same address twice would always produce the
3775 same answer. This might have been true once, but it is certainly no
3776 longer true now. Routing a child address may depend on the previous
3777 routing that produced that child. Some complicated redirection strategies
3778 went wrong when messages had multiple recipients, and made Exim's
3779 behaviour dependent on the order in which the addresses were given.
3781 I have moved the duplicate checking until after the routing is complete.
3782 Exim scans the addresses that are assigned to local and remote
3783 transports, and removes any duplicates. This means that more work will be
3784 done, as duplicates will always all be routed, but duplicates are
3785 presumably rare, so I don't expect this is of any significance.
3787 For deliveries to pipes, files, and autoreplies, the duplicate checking
3788 still happens during the routing process, since they are not going to be
3791 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3792 It corrects a timeout issue with spamd. This is Ian's comment: "The
3793 background is that sometimes spamd either never reads data from a
3794 connection it has accepted, or it never writes response data. The exiscan
3795 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3796 blindly assumes that writes won't block so it may never time out."
3798 PH/39 Allow G after quota size as well as K and M.
3800 PH/40 The value set for $authenticated_id in an authenticator may not contain
3801 binary zeroes or newlines because the value is written to log lines and
3802 to spool files. There was no check on this. Now the value is run through
3803 the string_printing() function so that such characters are converted to
3804 printable escape sequences.
3806 PH/41 $message_linecount is a new variable that contains the total number of
3807 lines in the message. Compare $body_linecount, which is the count for the
3810 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3811 bounce and delay warning messages, except in certain special cases, which
3814 (a) An SMTP error message from a remote host;
3815 (b) A message specified in a :fail: redirection;
3816 (c) A message specified in a "fail" command in a system filter;
3817 (d) A message specified in a FAIL return from the queryprogram router;
3818 (e) A message specified by the cannot_route_message router option.
3820 In these cases only, Exim does include the error details in bounce and
3821 warning messages. There are also a few cases where bland messages such
3822 as "unrouteable address" or "local delivery error" are given.
3824 PH/43 $value is now also set for the "else" part of a ${run expansion.
3826 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3827 being worked on, but at least Exim now implements the latest version to
3830 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3831 process to complete was treated as a delivery failure, a timeout while
3832 writing the message to the pipe was logged, but erroneously treated as a
3833 successful delivery. Such timeouts include transport filter timeouts. For
3834 consistency with the overall process timeout, these timeouts are now
3835 treated as errors, giving rise to delivery failures by default. However,
3836 there is now a new Boolean option for the pipe transport called
3837 timeout_defer, which, if set TRUE, converts the failures into defers for
3838 both kinds of timeout. A transport filter timeout is now identified in
3841 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3842 systems where "make" and "gmake" are different, calling "gmake" at top
3843 level broke things. I've arranged for the value of $(MAKE) to be passed
3844 from the Makefile to this script so that it can call the same version of
3848 A note about Exim versions 4.44 and 4.50
3849 ----------------------------------------
3851 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3852 changes of various kinds. As a consequence, a big documentation update was
3853 needed. This delayed the release for rather longer than seemed good, especially
3854 in the light of a couple of (minor) security issues. Therefore, the changes
3855 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3856 release. So 4.44 and 4.50 are in effect two different branches that both start
3859 I have left the 4.50 change log unchanged; it contains all the changes since
3860 4.43. The change log for 4.44 is below; many of its items are identical to
3861 those for 4.50. This seems to be the most sensible way to preserve the
3862 historical information.
3868 1. Minor wording change to the doc/README.SIEVE file.
3870 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3871 computation of the current number of files was incorrect.
3873 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3874 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3877 4. Give more explanation in the error message when the command for a transport
3878 filter fails to execute.
3880 5. There are several places where Exim runs a non-Exim command in a
3881 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3882 was being done only for the command run by the queryprogram router. It is
3883 now done for all such subprocesses. The other cases are: ${run, transport
3884 filters, and the commands run by the lmtp and pipe transports.
3886 6. Added CONFIGURE_GROUP build-time option.
3888 7. Some older OS have a limit of 256 on the maximum number of file
3889 descriptors. Exim was using setrlimit() to set 1000 as a large value
3890 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3891 systems. I've change it so that if it can't get 1000, it tries for 256.
3893 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3894 was an oversight, and furthermore, ever since the addition of extra
3895 controls (e.g. 4.43/32), the checks on when to allow different forms of
3896 "control" were broken. There should now be diagnostics for all cases when a
3897 control that does not make sense is encountered.
3899 9. Added the /retain_sender option to "control=submission".
3901 10. $recipients is now available in the predata ACL (oversight).
3903 11. Tidy the search cache before the fork to do a delivery from a message
3904 received from the command line. Otherwise the child will trigger a lookup
3905 failure and thereby defer the delivery if it tries to use (for example) a
3906 cached ldap connection that the parent has called unbind on.
3908 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3909 of $address_data from the recipient verification was clobbered by the
3910 sender verification.
3912 13. The value of address_data from a sender verification is now available in
3913 $sender_address_data in subsequent conditions in the ACL statement.
3915 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3917 15. Added a new option "connect=<time>" to callout options, to set a different
3920 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3921 was its contents. (It was OK if the option was not defined at all.)
3923 17. A "Completed" log line is now written for messages that are removed from
3924 the spool by the -Mrm option.
3926 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3927 information about exactly what failed.
3929 19. Added -dd to debug only the daemon process.
3931 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3932 handles timeouts, both on the server side and network timeouts. Renamed the
3933 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3935 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3936 It is now set to "smtps".
3938 22. $host_address is now set to the target address during the checking of
3939 ignore_target_hosts.
3941 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3942 being passed; this would have caused $sender_host_name to have been used if
3943 matching the list had actually called for a host name (not very likely,
3944 since this list is usually IP addresses). A host name is now passed as
3947 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3948 code that specifies a non-restarting handler (typically sigaction() in
3949 modern systems) in an attempt to fix a rare and obscure crash bug.
3951 25. Narrowed the window for a race in the daemon that could cause it to ignore
3952 SIGCHLD signals. This is not a major problem, because they are used only to
3953 wake it up if nothing else does.
3955 26. A malformed maildirsize file could cause Exim to calculate negative values
3956 for the mailbox size or file count. Odd effects could occur as a result.
3957 The maildirsize information is now recalculated if the size or filecount
3960 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3961 support for a long time. Removed HAVE_SYS_VFS_H.
3963 28. Installed the latest version of exipick from John Jetmore.
3965 29. In an address list, if the pattern was not a regular expression, an empty
3966 subject address (from a bounce message) matched only if the pattern was an
3967 empty string. Non-empty patterns were not even tested. This was the wrong
3968 because it is perfectly reasonable to use an empty address as part of a
3969 database query. An empty address is now tested by patterns that are
3970 lookups. However, all the other forms of pattern expect the subject to
3971 contain a local part and a domain, and therefore, for them, an empty
3972 address still always fails if the pattern is not itself empty.
3974 30. Exim went into a mad DNS loop when attempting to do a callout where the
3975 host was specified on an smtp transport, and looking it up yielded more
3976 than one IP address.
3978 31. Re-factored the code for checking spool and log partition space into a
3979 function that finds that data and another that does the check. The former
3980 is then used to implement four new variables: $spool_space, $log_space,
3981 $spool_inodes, and $log_inodes.
3983 32. The RFC2047 encoding function was originally intended for short strings
3984 such as real names; it was not keeping to the 75-character limit for
3985 encoded words that the RFC imposes. It now respects the limit, and
3986 generates multiple encoded words if necessary. To be on the safe side, I
3987 have increased the buffer size for the ${rfc2047: expansion operator from
3990 33. It is now permitted to omit both strings after an "if" condition; if the
3991 condition is true, the result is "true". As before, when the second string
3992 is omitted, a false condition yields an empty string. This makes it less
3993 cumbersome to write custom ACL and router conditions.
3995 34. Failure to deliver a bounce message always caused it to be frozen, even if
3996 there was an errors_to setting on the router. The errors_to setting is now
3999 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4000 canonical form (fully expanded) before being placed in
4001 $sender_host_address.
4003 36. The table in the code that translates DNS record types into text (T_A to
4004 "A" for instance) was missing entries for NS and CNAME. It is just possible
4005 that this could have caused confusion if both these types were looked up
4006 for the same domain, because the text type is used as part of Exim's
4007 per-process caching. But the chance of anyone hitting this buglet seems
4010 37. The dnsdb lookup has been extended in a number of ways.
4012 (1) There is a new type, "zns", which walks up the domain tree until it
4013 finds some nameserver records. It should be used with care.
4015 (2) There is a new type, "mxh", which is like "mx" except that it returns
4016 just the host names, not the priorities.
4018 (3) It is now possible to give a list of domains (or IP addresses) to be
4019 looked up. The behaviour when one of the lookups defers can be
4020 controlled by a keyword.
4022 (4) It is now possible to specify the separator character for use when
4023 multiple records are returned.
4025 38. The dnslists ACL condition has been extended: it is now possible to supply
4026 a list of IP addresses and/or domains to be looked up in a particular DNS
4029 39. Added log_selector=+queue_time_overall.
4031 40. When running the queue in the test harness, wait just a tad after forking a
4032 delivery process, to get repeatability of debugging output.
4034 41. Include certificate and key file names in error message when GnuTLS fails
4035 to set them up, because the GnuTLS error message doesn't include the name
4036 of the failing file when there is a problem reading it.
4038 42. Allow both -bf and -bF in the same test run.
4040 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4042 44. The "Exiscan patch" is now merged into the mainline Exim source.
4044 45. Sometimes the final signoff response after QUIT could fail to get
4045 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4046 < 0 before doing a fflush(). This bug looks as though it goes back to the
4047 introduction of TLS in release 3.20, but "sometimes" must have been rare
4048 because the tests only now provoked it.
4050 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4051 (this can affect the format of dates).
4053 47. exim_tidydb, when checking for the continued existence of a message for
4054 which it has found a message-specific retry record, was not finding
4055 messages that were in split spool directories. Consequently, it was
4056 deleting retry records that should have stayed in existence.
4058 48. Steve fixed some bugs in eximstats.
4060 49. The SPA authentication driver was not abandoning authentication and moving
4061 on to the next authenticator when an expansion was forced to fail,
4062 contradicting the general specification for all authenticators. Instead it
4063 was generating a temporary error. It now behaves as specified.
4065 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4066 (the order specifies the preference for clients). The order is now AES256,
4067 AES128, 3DES, ARCFOUR128.
4069 51. Small patch to Sieve code - explicitly set From: when generating an
4072 52. Exim crashed if a remote delivery caused a very long error message to be
4073 recorded - for instance if somebody sent an entire SpamAssassin report back
4074 as a large number of 550 error lines. This bug was coincidentally fixed by
4075 increasing the size of one of Exim's internal buffers (big_buffer) that
4076 happened as part of the Exiscan merge. However, to be on the safe side, I
4077 have made the code more robust (and fixed the comments that describe what
4080 53. Now that there can be additional text after "Completed" in log lines (if
4081 the queue_time_overall log selector is set), a one-byte patch to exigrep
4082 was needed to allow it to recognize "Completed" as not the last thing in
4085 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4086 patch that reportedly fixes this has been added. I am not expert enough to
4087 create a test for it. This is what the patch creator wrote:
4089 "I found a little strange behaviour of ldap code when working with
4090 Windows 2003 AD Domain, where users was placed in more than one
4091 Organization Units. When I tried to give exim partial DN, the exit code
4092 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4093 But simultaneously result of request was absolutely normal ldap result,
4094 so I produce this patch..."
4096 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4097 REFERENCE, so I have modified the code to exclude the patch when that macro
4100 55. Some experimental protocols are using DNS PTR records for new purposes. The
4101 keys for these records are domain names, not reversed IP addresses. The
4102 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4103 leaves it alone. Component reversal etc. now happens only for IP addresses.
4106 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4108 57. Double the size of the debug message buffer (to 2048) so that more of very
4109 long debug lines gets shown.
4111 58. The exicyclog utility now does better if the number of log files to keep
4112 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4114 59. Two changes related to the smtp_active_hostname option:
4116 (1) $smtp_active_hostname is now available as a variable.
4117 (2) The default for smtp_banner uses $smtp_active_hostname instead
4118 of $primary_hostname.
4120 60. The host_aton() function is supposed to be passed a string that is known
4121 to be a valid IP address. However, in the case of IPv6 addresses, it was
4122 not checking this. This is a hostage to fortune. Exim now panics and dies
4123 if the condition is not met. A case was found where this could be provoked
4124 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4125 components; fortuitously, this particular loophole had already been fixed
4126 by change 4.50/55 above.
4128 If there are any other similar loopholes, the new check in host_aton()
4129 itself should stop them being exploited. The report I received stated that
4130 data on the command line could provoke the exploit when Exim was running as
4131 exim, but did not say which command line option was involved. All I could
4132 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4133 running as the user.
4136 61. There was a buffer overflow vulnerability in the SPA authentication code
4137 (which came originally from the Samba project). I have added a test to the
4138 spa_base64_to_bits() function which I hope fixes it.
4141 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4142 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4144 63. The daemon start-up calls getloadavg() while still root for those OS that
4145 need the first call to be done as root, but it missed one case: when
4146 deliver_queue_load_max is set with deliver_drop_privilege. This is
4147 necessary for the benefit of the queue runner, because there is no re-exec
4148 when deliver_drop_privilege is set.
4150 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4151 This has been fixed.
4153 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4154 was in use, was not putting the data itself into the right store pool;
4155 consequently, it could be overwritten for a subsequent message in the same
4156 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4159 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4161 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4162 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4163 the function were treating the return as a boolean value, which happened to
4164 work because 0=false and not-0=true, but is not correct code.
4166 68. The host_aton() function was not handling scoped IPv6 addresses (those
4167 with, for example, "%eth0" on the end) correctly.
4169 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4170 negated items (that is, ~something) in unsigned ints. Some compilers
4171 apparently mutter when there is no cast.
4173 70. If an address verification called from an ACL failed, and did not produce a
4174 user-specific message (i.e. there was only a "system" message), nothing was
4175 put in $acl_verify_message. In this situation, it now puts the system
4178 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4179 session; change 4.31/43 added the unwanted input to the log line - except
4180 that it did not do this in the start of session case. It now does.
4182 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4183 This is wrong and can cause the other end to generate a synchronization
4184 error if it is another Exim or anything else that does the synchronization
4185 check. A QUIT command is no longer sent after a timeout.
4187 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4188 during host lookups.
4190 74. The defer_ok option of callout verification was not working if it was used
4191 when verifying addresses in header lines, that is, for this case:
4193 verify = header_sender/callout=defer_ok
4195 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4196 those file descriptors could be used for SMTP connections. If anything
4197 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4198 could be sent to the SMTP client, causing chaos. The daemon now opens
4199 stdin, stdout, and stderr to /dev/null when it puts itself into the
4202 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4203 log by default. The user can override this with suitable Perl magic.
4205 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4206 the log message when discard triggers, was not working for the DATA ACL or
4207 for the non-SMTP ACL.
4209 78. Error message wording change in sieve.c.
4211 79. If smtp_accept_max_per_host was set, the number of connections could be
4212 restricted to fewer than expected, because the daemon was trying to set up
4213 a new connection before checking whether the processes handling previous
4214 connections had finished. The check for completed processes is now done
4215 earlier. On busy systems, this bug wouldn't be noticed because something
4216 else would have woken the daemon, and it would have reaped the completed
4219 80. If a message was submitted locally by a user whose login name contained one
4220 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4221 It caused a spool format error. I have fixed the spool reading code. A
4222 related problem was that the "from" clause in the Received: line became
4223 illegal because of the space(s). It is now covered by ${quote_local_part.
4225 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4228 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4230 83. Patch from Sieve maintainer to fix unterminated string problem in
4231 "vacation" handling.
4233 84. Some minor changes to the Linux configuration files to help with other
4234 OS variants using glibc.
4236 85. One more patch for Sieve to update vacation handling to latest spec.
4239 ----------------------------------------------------
4240 See the note above about the 4.44 and 4.50 releases.
4241 ----------------------------------------------------
4247 1. Change 4.43/35 introduced a bug that caused file counts to be
4248 incorrectly computed when quota_filecount was set in an appendfile
4251 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4252 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4255 3. Give more explanation in the error message when the command for a transport
4256 filter fails to execute.
4258 4. There are several places where Exim runs a non-Exim command in a
4259 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4260 was being done only for the command run by the queryprogram router. It is
4261 now done for all such subprocesses. The other cases are: ${run, transport
4262 filters, and the commands run by the lmtp and pipe transports.
4264 5. Some older OS have a limit of 256 on the maximum number of file
4265 descriptors. Exim was using setrlimit() to set 1000 as a large value
4266 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4267 systems. I've change it so that if it can't get 1000, it tries for 256.
4269 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4270 was an oversight, and furthermore, ever since the addition of extra
4271 controls (e.g. 4.43/32), the checks on when to allow different forms of
4272 "control" were broken. There should now be diagnostics for all cases when a
4273 control that does not make sense is encountered.
4275 7. $recipients is now available in the predata ACL (oversight).
4277 8. Tidy the search cache before the fork to do a delivery from a message
4278 received from the command line. Otherwise the child will trigger a lookup
4279 failure and thereby defer the delivery if it tries to use (for example) a
4280 cached ldap connection that the parent has called unbind on.
4282 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4283 of $address_data from the recipient verification was clobbered by the
4284 sender verification.
4286 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4287 was its contents. (It was OK if the option was not defined at all.)
4289 11. A "Completed" log line is now written for messages that are removed from
4290 the spool by the -Mrm option.
4292 12. $host_address is now set to the target address during the checking of
4293 ignore_target_hosts.
4295 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4296 being passed; this would have caused $sender_host_name to have been used if
4297 matching the list had actually called for a host name (not very likely,
4298 since this list is usually IP addresses). A host name is now passed as
4301 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4302 code that specifies a non-restarting handler (typically sigaction() in
4303 modern systems) in an attempt to fix a rare and obscure crash bug.
4305 15. Narrowed the window for a race in the daemon that could cause it to ignore
4306 SIGCHLD signals. This is not a major problem, because they are used only to
4307 wake it up if nothing else does.
4309 16. A malformed maildirsize file could cause Exim to calculate negative values
4310 for the mailbox size or file count. Odd effects could occur as a result.
4311 The maildirsize information is now recalculated if the size or filecount
4314 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4315 support for a long time. Removed HAVE_SYS_VFS_H.
4317 18. Updated exipick to current release from John Jetmore.
4319 19. Allow an empty sender to be matched against a lookup in an address list.
4320 Previously the only cases considered were a regular expression, or an
4323 20. Exim went into a mad DNS lookup loop when doing a callout where the
4324 host was specified on the transport, if the DNS lookup yielded more than
4327 21. The RFC2047 encoding function was originally intended for short strings
4328 such as real names; it was not keeping to the 75-character limit for
4329 encoded words that the RFC imposes. It now respects the limit, and
4330 generates multiple encoded words if necessary. To be on the safe side, I
4331 have increased the buffer size for the ${rfc2047: expansion operator from
4334 22. Failure to deliver a bounce message always caused it to be frozen, even if
4335 there was an errors_to setting on the router. The errors_to setting is now
4338 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4339 canonical form (fully expanded) before being placed in
4340 $sender_host_address.
4342 24. Updated eximstats to version 1.33
4344 25. Include certificate and key file names in error message when GnuTLS fails
4345 to set them up, because the GnuTLS error message doesn't include the name
4346 of the failing file when there is a problem reading it.
4348 26. Expand error message when OpenSSL has problems setting up cert/key files.
4351 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4352 (this can affect the format of dates).
4354 28. exim_tidydb, when checking for the continued existence of a message for
4355 which it has found a message-specific retry record, was not finding
4356 messages that were in split spool directories. Consequently, it was
4357 deleting retry records that should have stayed in existence.
4359 29. eximstats updated to version 1.35
4360 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4361 1.35 - bugfix such that pie charts by volume are generated correctly
4363 30. The SPA authentication driver was not abandoning authentication and moving
4364 on to the next authenticator when an expansion was forced to fail,
4365 contradicting the general specification for all authenticators. Instead it
4366 was generating a temporary error. It now behaves as specified.
4368 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4369 (the order specifies the preference for clients). The order is now AES256,
4370 AES128, 3DES, ARCFOUR128.
4372 31. Small patch to Sieve code - explicitly set From: when generating an
4375 32. Exim crashed if a remote delivery caused a very long error message to be
4376 recorded - for instance if somebody sent an entire SpamAssassin report back
4377 as a large number of 550 error lines. This bug was coincidentally fixed by
4378 increasing the size of one of Exim's internal buffers (big_buffer) that
4379 happened as part of the Exiscan merge. However, to be on the safe side, I
4380 have made the code more robust (and fixed the comments that describe what
4383 33. Some experimental protocols are using DNS PTR records for new purposes. The
4384 keys for these records are domain names, not reversed IP addresses. The
4385 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4386 leaves it alone. Component reversal etc. now happens only for IP addresses.
4389 34. The host_aton() function is supposed to be passed a string that is known
4390 to be a valid IP address. However, in the case of IPv6 addresses, it was
4391 not checking this. This is a hostage to fortune. Exim now panics and dies
4392 if the condition is not met. A case was found where this could be provoked
4393 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4394 components; fortuitously, this particular loophole had already been fixed
4395 by change 4.50/55 or 4.44/33 above.
4397 If there are any other similar loopholes, the new check in host_aton()
4398 itself should stop them being exploited. The report I received stated that
4399 data on the command line could provoke the exploit when Exim was running as
4400 exim, but did not say which command line option was involved. All I could
4401 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4402 running as the user.
4405 35. There was a buffer overflow vulnerability in the SPA authentication code
4406 (which came originally from the Samba project). I have added a test to the
4407 spa_base64_to_bits() function which I hope fixes it.
4410 36. The daemon start-up calls getloadavg() while still root for those OS that
4411 need the first call to be done as root, but it missed one case: when
4412 deliver_queue_load_max is set with deliver_drop_privilege. This is
4413 necessary for the benefit of the queue runner, because there is no re-exec
4414 when deliver_drop_privilege is set.
4416 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4417 was in use, was not putting the data itself into the right store pool;
4418 consequently, it could be overwritten for a subsequent message in the same
4419 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4422 38. Sometimes the final signoff response after QUIT could fail to get
4423 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4424 < 0 before doing a fflush(). This bug looks as though it goes back to the
4425 introduction of TLS in release 3.20, but "sometimes" must have been rare
4426 because the tests only now provoked it.
4432 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4433 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4434 Now it returns an integer. A number of calls were still expecting a T/F
4435 return. Fortuitously, in all cases, the tests worked in OK situations,
4436 which is the norm. However, things would have gone wrong on any write
4437 failures on the smtp file descriptor. This function is used when sending
4438 messages over SMTP and also when doing verify callouts.
4440 2. When Exim is called to do synchronous delivery of a locally submitted
4441 message (the -odf or -odi options), it no longer closes stderr before doing
4444 3. Implemented the mua_wrapper option.
4446 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4448 5. Implemented the functions header_remove(), header_testname(),
4449 header_add_at_position(), and receive_remove_recipient(), and exported them
4452 6. If an ACL "warn" statement specified the addition of headers, Exim already
4453 inserted X-ACL-Warn: at the start if there was no header name. However, it
4454 was not making this test for the second and subsequent header lines if
4455 there were newlines in the string. This meant that an invalid header could
4456 be inserted if Exim was badly configured.
4458 7. Allow an ACL "warn" statement to add header lines at the start or after all
4459 the Received: headers, as well as at the end.
4461 8. Added the rcpt_4xx retry error code.
4463 9. Added postmaster_mailfrom=xxx to callout verification option.
4465 10. Added mailfrom=xxxx to the callout verification option, for verify=
4468 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4469 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4471 12. Inserted some casts to stop certain compilers warning when using pointer
4472 differences as field lengths or precisions in printf-type calls (mostly
4473 affecting debugging statements).
4475 13. Added optional readline() support for -be (dynamically loaded).
4477 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4478 same clock tick as a message's arrival, so that its received time was the
4479 same as the "first fail" time on the retry record, and that message
4480 remained on the queue past the ultimate address timeout, every queue runner
4481 would try a delivery (because it was past the ultimate address timeout) but
4482 after another failure, the ultimate address timeout, which should have then
4483 bounced the address, did not kick in. This was a "< instead of <=" error;
4484 in most cases the first failure would have been in the next clock tick
4485 after the received time, and all would be well.
4487 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4488 being recognized when the domain list was tested by the match_domain
4489 condition in an expansion string.
4491 16. Added the ${str2b64: operator.
4493 17. Exim was always calling setrlimit() to set a large limit for the number of
4494 processes, without checking whether the existing limit was already
4495 adequate. (It did check for the limit on file descriptors.) Furthermore,
4496 errors from getrlimit() and setrlimit() were being ignored. Now they are
4497 logged to the main and panic logs, but Exim does carry on, to try to do its
4498 job under whatever limits there are.
4500 18. Imported PCRE 5.0.
4502 19. Trivial typo in log message " temporarily refused connection" (the leading
4505 20. If the log selector return_path_on_delivery was set and an address was
4506 redirected to /dev/null, the delivery process crashed because it assumed
4507 that a return path would always be set for a "successful" delivery. In this
4508 case, the whole delivery is bypassed as an optimization, and therefore no
4511 21. Internal re-arrangement: the function for sending a challenge and reading
4512 a response while authentication was assuming a zero-terminated challenge
4513 string. It's now changed to take a pointer and a length, to allow for
4514 binary data in such strings.
4516 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4518 23. Exim was not respecting finduser_retries when seeking the login of the
4519 uid under which it was called; it was always trying 10 times. (The default
4520 setting of finduser_retries is zero.) Also, it was sleeping after the final
4521 failure, which is pointless.
4523 24. Implemented tls_on_connect_ports.
4525 25. Implemented acl_smtp_predata.
4527 26. If the domain in control=submission is set empty, Exim assumes that the
4528 authenticated id is a complete email address when it generates From: or
4529 Sender: header lines.
4531 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4532 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4533 chown and chgrp in /bin and hostname in /usr/bin.
4535 28. Exim was keeping the "process log" file open after each use, just as it
4536 does for the main log. This opens the possibility of it remaining open for
4537 long periods when the USR1 signal hits a daemon. Occasional processlog
4538 errors were reported, that could have been caused by this. Anyway, it seems
4539 much more sensible not to leave this file open at all, so that is what now
4542 29. The long-running daemon process does not normally write to the log once it
4543 has entered its main loop, and it closes the log before doing so. This is
4544 so that log files can straightforwardly be renamed and moved. However,
4545 there are a couple of unusual error situations where the daemon does write
4546 log entries, and I had neglected to close the log afterwards.
4548 30. The text of an SMTP error response that was received during a remote
4549 delivery was being truncated at 512 bytes. This is too short for some of
4550 the long messages that one sometimes sees. I've increased the limit to
4553 31. It is now possible to make retry rules that apply only when a message has a
4554 specific sender, in particular, an empty sender.
4556 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4557 it possible to be selective about when SMTP synchronization is enforced.
4559 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4561 32. Implemented hosts_connection_nolog.
4563 33. Added an ACL for QUIT.
4565 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4568 35. Added mailbox_size and mailbox_filecount to appendfile.
4570 36. Added control = no_multiline_responses to ACLs.
4572 37. There was a bug in the logic of the code that waits for the clock to tick
4573 in the case where the clock went backwards by a substantial amount such
4574 that the microsecond fraction of "now" was more than the microsecond
4575 fraction of "then" (but the whole seconds number was less).
4577 38. Added support for the libradius Radius client library this is found on
4578 FreeBSD (previously only the radiusclient library was supported).
4584 1. When certain lookups returned multiple values in the form name=value, the
4585 quoting of the values was not always being done properly. Specifically:
4586 (a) If the value started with a double quote, but contained no whitespace,
4588 (b) If the value contained whitespace other than a space character (i.e.
4589 tabs or newlines or carriage returns) it was not quoted.
4590 This fix has been applied to the mysql and pgsql lookups by writing a
4591 separate quoting function and calling it from the lookup code. The fix
4592 should probably also be applied to nisplus, ibase and oracle lookups, but
4593 since I cannot test any of those, I have not disturbed their existing code.
4595 2. A hit in the callout cache for a specific address caused a log line with no
4596 reason for rejecting RCPT. Now it says "Previous (cached) callout
4597 verification failure".
4599 3. There was an off-by-one bug in the queryprogram router. An over-long
4600 return line was truncated at 256 instead of 255 characters, thereby
4601 overflowing its buffer with the terminating zero. As well as fixing this, I
4602 have increased the buffer size to 1024 (and made a note to document this).
4604 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4605 when Exim is waiting for an SMTP response from a remote server, Exim
4606 restarts its select() call on the socket, thereby resetting its timeout.
4607 This is not a problem when such interrupts are rare. Somebody set up a cron
4608 job to run exiwhat every 2 minutes, which is less than the normal select()
4609 timeout (5 or 10 minutes). This meant that the select() timeout never
4610 kicked in because it was always reset. I have fixed this by comparing the
4611 time when an interrupt arrives with the time at the start of the first call
4612 to select(). If more time than the timeout has elapsed, the interrupt is
4613 treated as a timeout.
4615 5. Some internal re-factoring in preparation for the addition of Sieve
4616 extensions (by MH). In particular, the "personal" test is moved to a
4617 separate function, and given an option for scanning Cc: and Bcc: (which is
4618 not set for Exim filters).
4620 6. When Exim created an email address using the login of the caller as the
4621 local part (e.g. when creating a From: or Sender: header line), it was not
4622 quoting the local part when it contained special characters such as @.
4624 7. Installed new OpenBSD configuration files.
4626 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4627 try to make them clearer.
4629 9. Callout options, other than the timeout value, were being ignored when
4630 verifying sender addresses in header lines. For example, when using
4632 verify = header_sender/callout=no_cache
4634 the cache was (incorrectly) being used.
4636 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4637 only the Cygwin environment.
4639 11. When return_path_on_delivery was set as a log selector, if different remote
4640 addresses in the same message used different return paths and parallel
4641 remote delivery occurred, the wrong values would sometimes be logged.
4642 (Whenever a remote delivery process finished, the return path value from
4643 the most recently started remote delivery process was logged.)
4645 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4646 lines when AUTH and/or TLS are in use. This is the "received protocol"
4647 field. Exim used to use "asmtp" for authenticated SMTP, without any
4648 indication (in the protocol name) for TLS use. Now it follows the RFC and
4649 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4650 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4651 names appear in log lines as well as in Received: header lines.
4653 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4654 extensions, and comparison tests, and to fix some bugs.
4656 14. Changes to the "personal" filter test:
4658 (1) The test was buggy in that it was just doing the equivalent of
4659 "contains" tests on header lines. For example, if a user's address was
4660 anne@some.where, the "personal" test would incorrectly be true for
4662 To: susanne@some.where
4664 This test is now done by extracting each address from the header in turn,
4665 and checking the entire address. Other tests that are part of "personal"
4666 are now done using regular expressions (for example, to check local parts
4667 of addresses in From: header lines).
4669 (2) The list of non-personal local parts in From: addresses has been
4670 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4671 taken from the Sieve specification recommendations.
4673 (3) If the message contains any header line starting with "List-" it is
4674 treated as non-personal.
4676 (4) The test for "circular" in the Subject: header line has been removed
4677 because it now seems ill-conceived.
4679 15. Minor typos in src/EDITME comments corrected.
4681 16. Installed latest exipick from John Jetmore.
4683 17. If headers_add on a router specified a text string that was too long for
4684 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4685 of string_sprintf() is now avoided.
4687 18. $message_body_size was not set (it was always zero) when running the DATA
4688 ACL and the local_scan() function.
4690 19. For the "mail" command in an Exim filter, no default was being set for
4691 the once_repeat time, causing a random time value to be used if "once" was
4692 specified. (If the value happened to be <= 0, no repeat happened.) The
4693 default is now 0s, meaning "never repeat". The "vacation" command was OK
4694 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4695 (I found it when inspecting the code).
4697 20. There is now an overall timeout for performing a callout verification. It
4698 defaults to 4 times the callout timeout, which applies to individual SMTP
4699 commands during the callout. The overall timeout applies when there is more
4700 than one host that can be tried. The timeout is checked before trying the
4701 next host. This prevents very long delays if there are a large number of
4702 hosts and all are timing out (e.g. when the network connections are timing
4703 out). The value of the overall timeout can be changed by specifying an
4704 additional sub-option for "callout", called "maxwait". For example:
4706 verify = sender/callout=5s,maxwait=20s
4708 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4709 to the end before writing, but this should make it even safer).
4711 22. Exim was forgetting that it had advertised PIPELINING for the second and
4712 subsequent messages on an SMTP connection. It was also not resetting its
4713 memory on STARTTLS and an internal HELO.
4715 23. When Exim logs an SMTP synchronization error within a session, it now
4716 records whether PIPELINING has been advertised or not.
4718 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4719 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4720 rather than long int.
4722 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4724 26. Added the never_mail option to autoreply.
4730 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4731 crash if the getsockname() call failed; this can happen if a connection is
4732 closed very soon after it is established. The problem was simply in the
4733 order in which certain operations were done, causing Exim to try to write
4734 to the SMTP stream before it had set up the file descriptor. The bug has
4735 been fixed by making things happen in the correct order.
4741 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4742 before the connection was closed, thus losing the rejection response.
4744 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4745 some early Solaris releases, but causes trouble in current releases where
4746 socklen_t is defined.
4748 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4751 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4754 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4755 that releasing the top of it at the end releases what was used for sub-
4756 expansions (unless the block got too big). However, discard this block if
4757 the first thing is a variable or header, so that we can use its block when
4758 it is dynamic (useful for very large $message_headers, for example).
4760 6. Lookups now cache *every* query, not just the most recent. A new, separate
4761 store pool is used for this. It can be recovered when all lookup caches are
4762 flushed. Lookups now release memory at the end of their result strings.
4763 This has involved some general refactoring of the lookup sources.
4765 7. Some code has been added to the store_xxx() functions to reduce the amount
4766 of flapping under certain conditions.
4768 8. log_incoming_interface used to affect only the <= reception log lines. Now
4769 it causes the local interface and port to be added to several more SMTP log
4770 lines, for example "SMTP connection from", and rejection lines.
4772 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4774 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4776 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4777 could be overwritten at the end of the current message (or the start of a
4778 new message if it was set in a HELO ACL). The value is now preserved for
4779 the duration of the SMTP connection.
4781 12. If a transport had a headers_rewrite setting, and a matching header line
4782 contained an unqualified address, that address was qualified, even if it
4783 did not match any rewriting rules. The underlying bug was that the values
4784 of the flags that permit the existence of unqualified sender and recipient
4785 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4786 non-local messages, and by -bnq for local messages) were not being
4787 preserved with the message after it was received.
4789 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4790 "next input=" as part of the text comprising the host identity instead of
4791 the correct text. The code was using the same buffer for two different
4792 strings. However, depending on which order the printing function evaluated
4793 its arguments, the bug did not always show up. Under Linux, for example, my
4794 test suite worked just fine.
4796 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4797 doesn't work with some older versions of Perl. It has been changed to "my",
4798 which in any case is probably the better facility to use.
4800 15. A really picky compiler found some instances of statements for creating
4801 error messages that either had too many or two few arguments for the format
4804 16. The size of the buffer for calls to the DNS resolver has been increased
4805 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4806 for addresses that have a lot of PTR records. This alleviates a problem; it
4807 does not fully solve it.
4809 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4810 buffer now truncates the list and logs the incident, which is the same
4811 action as happens when Exim is looking up a host name and its aliases.
4812 Previously in this situation something unpredictable would happen;
4813 sometimes it was "internal error: store_reset failed".
4815 18. If a server dropped the connection unexpectedly when an Exim client was
4816 using GnuTLS and trying to read a response, the client delivery process
4817 crashed while trying to generate an error log message.
4819 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4820 string, for example:
4822 warn message = H1: something\nH2: something
4824 the text was added as a single header line from Exim's point of view
4825 though it ended up OK in the delivered message. However, searching for the
4826 second and subsequent header lines using $h_h2: did not work. This has been
4827 fixed. Similarly, if a system filter added multiple headers in this way,
4828 the routers could not see them.
4830 20. Expanded the error message when iplsearch is called with an invalid key to
4831 suggest using net-iplsearch in a host list.
4833 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4834 ACLs are no longer actually imposed (and a message to that effect is
4837 22. If a "gecos" field in a passwd entry contained escaped characters, in
4838 particular, if it contained a \" sequence, Exim got it wrong when building
4839 a From: or a Sender: header from that name. A second bug also caused
4840 incorrect handling when an unquoted " was present following a character
4841 that needed quoting.
4843 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4844 was not being matched caselessly.
4846 24. Arranged for all hyphens in the exim.8 source to be escaped with
4849 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4850 properly. Recipient callout cache records were still being keyed to include
4851 the sender, even when use_sender was set false. This led to far more
4852 callouts that were necessary. The sender is no longer included in the key
4853 when use_sender is false.
4855 26. Added "control = submission" modifier to ACLs.
4857 27. Added the ${base62d: operator to decode base 62 numbers.
4859 28. dnsdb lookups can now access SRV records.
4861 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4862 the configuration file.
4864 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4865 (-v) mode. This makes the output for a verbose queue run more intelligible.
4867 31. Added a use_postmaster feature to recipient callouts.
4869 32. Added the $body_zerocount variable, containing the number of binary zero
4870 bytes in the message body.
4872 33. The time of last modification of the "new" subdirectory is now used as the
4873 "mailbox time last read" when there is a quota error for a maildir
4876 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4878 35. Added +ignore_unknown as a special item in host lists.
4880 36. Code for decoding IPv6 addresses in host lists is now included, even if
4881 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4882 address was recognized as an IP address, but was then not correctly decoded
4883 into binary, causing unexpected and incorrect effects when compared with
4890 1. Very minor rewording of debugging text in manualroute to say "list of
4891 hosts" instead of "hostlist".
4893 2. If verify=header_syntax was set, and a header line with an unqualified
4894 address (no domain) and a large number of spaces between the end of the
4895 name and the colon was received, the reception process suffered a buffer
4896 overflow, and (when I tested it) crashed. This was caused by some obsolete
4897 code that should have been removed. The fix is to remove it!
4899 3. When running in the test harness, delay a bit after writing a bounce
4900 message to get a bit more predictability in the log output.
4902 4. Added a call to search_tidyup() just before forking a reception process. In
4903 theory, someone could use a lookup in the expansion of smtp_accept_max_
4904 per_host which, without the tidyup, could leave open a database connection.
4906 5. Added the variables $recipient_data and $sender_data which get set from a
4907 lookup success in an ACL "recipients" or "senders" condition, or a router
4908 "senders" option, similar to $domain_data and $local_part_data.
4910 6. Moved the writing of debug_print from before to after the "senders" test
4913 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4914 problems for message scanning, either using a data ACL, or using
4915 local_scan() because the Received: header was not generated till after they
4916 were called (in order to set the time as the time of reception completion).
4917 I have revised the way this works. The header is now generated after the
4918 body is received, but before the ACL or local_scan() are called. After they
4919 are run, the timestamp in the header is updated.
4925 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4926 before starting a queue runner without re-exec. This happened only when
4927 deliver_drop_privilege was set or when the Exim user was set to root. The
4928 effect of the bug was that timeouts during subsequent deliveries caused
4929 crashes instead of being properly handled. The handler is now left at its
4930 default (and expected) setting.
4932 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4933 message, again when deliver_drop_privilege is set or Exim is run as root.
4934 The bug described in (1) was not present in this case, but the tidying up
4935 of the other signals was missing. I have made the two cases consistent.
4937 3. The ignore_target_hosts setting on a manualroute router was being ignored
4938 for hosts that were looked up using the /MX notation.
4940 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4943 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4944 operated on the sender address. After changing the $sender_address to <>
4945 for the sender address verify, Exim was re-instated it as the original
4946 (before rewriting) address, but remembering that it had rewritten it, so it
4947 wasn't rewriting it again. This bug also had the effect of breaking the
4948 sender address verification caching when the sender address was rewritten.
4950 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4951 This has been changed so that if the ip literal address matches
4952 ignore_target_hosts, the router declines.
4954 7. Added expansion conditions match_domain, match_address, and match_local_
4955 part (NOT match_host).
4957 8. The placeholder for the Received: header didn't have a length field set.
4959 9. Added code to Exim itself and to exim_lock to test for a specific race
4960 condition that could lead to file corruption when using MBX delivery. The
4961 issue is with the lockfile that is created in /tmp. If this file is removed
4962 after a process has opened it but before that process has acquired a lock,
4963 there is the potential for a second process to recreate the file and also
4964 acquire a lock. This could lead to two Exim processes writing to the file
4965 at the same time. The added code performs the same test as UW imapd; it
4966 checks after acquiring the lock that its file descriptor still refers to
4967 the same named file.
4969 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4970 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4973 11. Added the smtp_active_hostname option. If used, this will typically be made
4974 to depend on the incoming interface address. Because $interface_address is
4975 not set up until the daemon has forked a reception process, error responses
4976 that can happen earlier (such as "too many connections") no longer contain
4979 12. If an expansion in a condition on a "warn" statement fails because a lookup
4980 defers, the "warn" statement is abandoned, and the next ACL statement is
4981 processed. Previously this caused the whole ACL to be aborted.
4983 13. Added the iplsearch lookup type.
4985 14. Added ident_timeout as a log selector.
4987 15. Added tls_certificate_verified as a log selector.
4989 16. Added a global option tls_require_ciphers (compare the smtp transport
4990 option of the same name). This controls incoming TLS connections.
4992 17. I finally figured out how to make tls_require_ciphers do a similar thing
4993 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4994 before starting the TLS session.
4996 18. Tabs are now shown as \t in -bP output.
4998 19. If the log selector return_path_on_delivery was set, Exim crashed when
4999 bouncing a message because it had too many Received: header lines.
5001 20. If two routers both had headers_remove settings, and the first one included
5002 a superfluous trailing colon, the final name in the first list and the
5003 first name in the second list were incorrectly joined into one item (with a
5004 colon in the middle).
5010 1. Added -C and -D options to the exinext utility, mainly to make it easier
5011 to include in the automated testing, but these could be helpful when
5012 multiple configurations are in use.
5014 2. The exinext utility was not formatting the output nicely when there was
5015 an alternate port involved in the retry record key, nor when there was a
5016 message id as well (for retries that were specific to a specific message
5017 and a specific host). It was also confused by IPv6 addresses, because of
5018 the additional colons they contain. I have fixed the IPv4 problem, and
5019 patched it up to do a reasonable job for IPv6.
5021 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5022 delivery, the log line now contains "pipelined" if PIPELINING was used.
5024 4. An SMTP transport process used to panic and die if the bind() call to set
5025 an explicit outgoing interface failed. This has been changed; it is now
5026 treated in the same way as a connect() failure.
5028 5. A reference to $sender_host_name in the part of a conditional expansion
5029 that was being skipped was still causing a DNS lookup. This no longer
5032 6. The def: expansion condition was not recognizing references to header lines
5033 that used bh_ and bheader_.
5035 7. Added the _cache feature to named lists.
5037 8. The code for checking quota_filecount in the appendfile transport was
5038 allowing one more file than it should have been.
5040 9. For compatibility with Sendmail, the command line option
5048 and sets the incoming protocol and host name (for trusted callers). The
5049 host name and its colon can be omitted when only the protocol is to be set.
5050 Note the Exim already has two private options, -pd and -ps, that refer to
5051 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5052 "s", but I don't think that's a major issue.
5054 10. A number of refactoring changes to the code, none of which should affect
5057 (a) The number of logging options was getting close to filling up the
5058 32-bit word that was used as a bit map. I have split them into two classes:
5059 those that are passed in the argument to log_write(), and those that are
5060 only ever tested independently outside of that function. These are now in
5061 separate 32-bit words, so there is plenty of room for expansion again.
5062 There is no change in the user interface or the logging behaviour.
5064 (b) When building, for example, log lines, the code previously used a
5065 macro that called string_cat() twice, in order to add two strings. This is
5066 not really sufficiently general. Furthermore, there was one instance where
5067 it was actually wrong because one of the argument was used twice, and in
5068 one call a function was used. (As it happened, calling the function twice
5069 did not affect the overall behaviour.) The macro has been replaced by a
5070 function that can join an arbitrary number of extra strings onto a growing
5073 (c) The code for expansion conditions now uses a table and a binary chop
5074 instead of a serial search (which was left over from when there were very
5075 few conditions). Also, it now recognizes conditions like "pam" even when
5076 the relevant support is not compiled in: a suitably worded error message is
5077 given if an attempt is made to use such a condition.
5079 11. Added ${time_interval:xxxxx}.
5081 12. A bug was causing one of the ddress fields not to be passed back correctly
5082 from remote delivery subprocesses. The field in question was not being
5083 subsequently used, so this caused to problems in practice.
5085 13. Added new log selectors queue_time and deliver_time.
5087 14. Might have fixed a bug in maildirsizefile handling that threw up
5088 "unexpected character" debug warnings, and recalculated the data
5089 unnecessarily. In any case, I expanded the warning message to give more
5092 15. Added the message "Restricted characters in address" to the statements in
5093 the default ACL that block characters like @ and % in local parts.
5095 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5096 Three changes have been made:
5098 (a) There was a serious bug; a negative response to MAIL caused the whole
5099 recipient domain to be cached as invalid, thereby blocking all messages
5100 to all local parts at the same domain, from all senders. This bug has
5101 been fixed. The domain is no longer cached after a negative response to
5102 MAIL if the sender used is not empty.
5104 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5107 (c) A new callout option, "use_sender" has been added for people who want
5108 the modified behaviour.
5114 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5117 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5118 indeed breaks things for older releases.
5120 3. Added additional logging to the case where there is a problem reading data
5121 from a filter that is running in a subprocess using a pipe, in order to
5122 try to track down a specific problem.
5124 4. Testing facility fudge: when running in the test harness and attempting
5125 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5126 getting "No route to host". Convert this to a timeout.
5128 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5131 6. Some OS don't have socklen_t but use size_t instead. This affects the
5132 fifth argument of getsockopt() amongst other things. This is now
5133 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5134 can be set for individual OS. I have set it for SunOS5, OSF1, and
5135 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5136 some earlier ones do not.
5138 7. Change 4.30/15 was not doing the test caselessly.
5140 8. The standard form for an IPv6 address literal was being rejected by address
5141 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5142 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5143 this, as well as the form without the "IPv6" on the front (but only when
5144 address literals are enabled, of course).
5146 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5148 10. Exim crashed if a message with an empty sender address specified by -f
5149 encountered a router with an errors_to setting. This could be provoked only
5150 by a command such as
5154 where an empty string was supplied; "<>" did not hit this bug.
5156 11. Installed PCRE release 4.5.
5158 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5159 remained set. It is now erased.
5161 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5162 times from message ids (which are base 36 rather than the normal 62).
5164 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5165 were being counted as actual protocol errors, and logged if the log
5166 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5167 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5168 and DATA following a set of rejected RCPTs do not count as protocol errors.
5169 In other words, Exim assumes they were pipelined, though this may not
5170 actually be the case. Of course, in all cases the client gets an
5171 appropriate error code.
5173 15. If a lookup fails in an ACL condition, a message about the failure may
5174 be available; it is used if testing the ACL cannot continue, because most
5175 such messages specify what the cause of the deferral is. However, some
5176 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5177 that caused an old message to be retained and used if a later statement
5178 caused a defer, replacing the real cause of the deferral.
5180 16. If an IP address had so many PTR records that the DNS lookup buffer
5181 was not large enough to hold them, Exim could crash while trying to process
5182 the truncated data. It now detects and logs this case.
5184 17. Further to 4.21/58, another change has been made: if (and only if) the
5185 first line of a message (the first header line) ends with CRLF, a bare LF
5186 in a subsequent header line has a space inserted after it, so as not to
5187 terminate the header.
5189 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5190 unnecessarily, used atoi() instead of strtol(), and didn't check the
5191 termination when getting file sizes from file names by regex.
5193 19. Completely re-implemented the support for maildirsize files, in the light
5194 of a number of problems with the previous contributed implementation
5195 (4.30/29). In particular:
5197 . If the quota is zero, the maildirsize file is maintained, but no quota is
5200 . If the maildir directory does not exist, it is created before any attempt
5201 to write a maildirsize file.
5203 . The quota value in the file is just a cache; if the quota is changed in
5204 the transport, the new value overrides.
5206 . A regular expression is available for excluding directories from the
5209 20. The autoreply transport checks the characters in options that define the
5210 message's headers; it allows continued headers, but it was checking with
5211 isspace() after an embedded newline instead of explicitly looking for a
5214 21. If all the "regular" hosts to which an address was routed had passed their
5215 expiry times, and had not reached their retry times, the address was
5216 bounced, even if fallback hosts were defined. Now Exim should go on to try
5219 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5220 equivalent code in the SMTP transport. Some hosts send humungous responses
5221 to HELO/EHLO, more than 1024 it seems.
5223 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5224 gives compiler warnings in some environments. I've now done it "properly",
5227 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5228 (because of problems with the built-in one) was declared to return uschar *
5229 instead of char *, causing compiler failure.
5231 25. Fixed a file descriptor leak when processing alias/forward files.
5233 26. Fixed a minor format string issue in dbfn.c.
5235 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5237 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5238 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5241 29. When a sender address is verified, it is cached, to save repeating the test
5242 when there is more than one recipient in a message. However, when the
5243 verification involves a callout, it is possible for different callout
5244 options to be set for different recipients. It is too complicated to keep
5245 track of this in the cache, so now Exim always runs a verification when a
5246 callout is required, relying on the callout cache for the optimization.
5247 The overhead is duplication of the address routing, but this should not be
5250 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5251 to be verified with callout=postmaster, and the main callout worked but the
5252 postmaster check failed, the verification correctly failed. However, if a
5253 subsequent RCPT command asked for sender verification *without* the
5254 postmaster check, incorrect caching caused this verification also to fail,
5257 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5258 it was not caching the DNS options (qualify_single, search_parents) that
5259 were used when the lookup failed. A subsequent lookup with different
5260 options therefore always gave the same answer, though there were cases
5261 where it should not have. (Example: a "domains = !$mx_any" option on a
5262 dnslookup router: the "domains" option is always processed without any
5263 widening, but the router might have qualify_single set.) Now Exim uses the
5264 cached value only when the same options are set.
5266 32. Added John Jetmore's "exipick" utility to the distribution.
5268 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5269 than a timeout (e.g. a certificate is required, and is not provided), an
5270 Exim server now closes the connection immediately. Previously it waited for
5271 the client to close - but if the client is SSL, it seems that they each
5272 wait for each other, leading to a delay before one of them times out.
5274 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5275 maintained 0.8.x compatibility because I don't think many are using it, and
5276 it is clearly obsolete.
5278 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5281 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5282 client certificate was expired. A simple patch fixes this, though I don't
5283 understand the full logic of why the verify callback is called multiple
5286 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5287 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5288 which causes problems with some clients (such as the Certicom SSL Plus
5289 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5290 disables the coutermeasure allowing Eudora to connect."
5292 38. Exim was not checking that a write() to a log file succeeded. This could
5293 lead to Bad Things if a log got too big, in particular if it hit a file
5294 size limit. Exim now panics and dies if it cannot write to a log file, just
5295 as it does if it cannot open a log file.
5297 39. Modified OS/Makefile-Linux so that it now contains
5299 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5301 The two -D definitions ensure that Exim is compiled with large file
5302 support, which makes it possible to handle log files that are bigger than
5305 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5306 instance) a domain was checked against a named list that involved a lookup,
5307 causing $domain_data to be set, then another domain was checked against the
5308 same list, then the first domain was re-checked, the value of $domain_data
5309 after the final check could be wrong. In particular, if the second check
5310 failed, it could be set empty. This bug probably also applied to
5313 41. The strip_trailing_dot option was not being applied to the address given
5314 with the -f command-line option.
5316 42. The code for reading a message's header from the spool was incrementing
5317 $received_count, but never initializing it. This meant that the value was
5318 incorrect (doubled) while delivering a message in the same process in which
5319 it was received. In the most common configuration of Exim, this never
5320 happens - a fresh exec is done - but it can happen when
5321 deliver_drop_privilege is set.
5323 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5324 it now includes up to 150 characters of the unexpected data in the log
5327 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5328 and building data strings. The size of both of these buffers was 10 000
5329 bytes - far larger than anybody would *ever* want, thought I. Needless to
5330 say, somebody hit the limit. I have increased the maximum line length to
5331 20 000 and the maximum data length of concatenated lines to 100 000. I have
5332 also fixed two bugs, because there was no checking on these buffers. Tsk,
5333 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5334 buffer is too small.
5336 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5337 lsearch lookups. Now it does.
5339 46. When parsing a route_list item in a manualroute router, a fixed-length
5340 buffer was used for the list of hosts. I made this 1024 bytes long,
5341 thinking that nobody would ever have a list of hosts that long. Wrong.
5342 Somebody had a whole pile of complicated expansion conditions, and the
5343 string was silently truncated, leading to an expansion error. It turns out
5344 that it is easier to change to an unlimited length (owing to other changes
5345 that have happened since this code was originally written) than to build
5346 structure for giving a limitation error. The length of the item that
5347 expands into the list of hosts is now unlimited.
5349 47. The lsearch lookup could not handle data where the length of text line was
5350 more than 4095 characters. Such lines were truncated, leading to shortened
5351 data being returned. It should now handle lines of any length.
5353 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5354 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5357 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5358 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5359 no longer happens in comment lines. A list of replacements is now placed
5360 at the head of all of the source files, except those whose only change is
5361 to replace PERL_COMMAND in the very first #! line.
5363 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5364 messages on the queue, with a bottom-up merge sort, using code contributed
5365 by Michael Haardt. This should make operations like -bp somewhat faster on
5366 large queues. It won't affect queue runners, except when queue_run_in_order
5369 51. Installed eximstats 1.31 in the distribution.
5371 52. Added support for SRV lookups to the dnslookup router.
5373 53. If an ACL referred to $message_body or $message_body_end, the value was not
5374 reset for any messages that followed in the same SMTP session.
5376 54. The store-handling optimization for building very long strings was not
5377 differentiating between the different store pools. I don't think this
5378 actually made any difference in practice, but I've tidied it.
5380 55. While running the routers to verify a sender address, $sender_address
5381 was still set to the sender address. This is wrong, because when routing to
5382 send a bounce to the sender, it would be empty. Therefore, I have changed
5383 it so that, while verifying a sender address, $sender_address is set to <>.
5384 (There is no change to what happens when verifying a recipient address.)
5386 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5387 target A or AAAA records (if not already returned) without resetting the
5388 qualify_single or search_parents options of the DNS resolver. These are
5389 inappropriate in this case because the targets of MX and SRV records must
5390 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5391 target that, when qualified, matched something in the local domain. These
5392 two options are now turned off when doing these lookups.
5394 57. It seems that at least some releases of Reiserfs (which does not have the
5395 concept of a fixed number of inodes) returns zero and not -1 for the
5396 number of available inodes. This interacted badly with check_spool_inodes,
5397 which assumed that -1 was the "no such thing" setting. What I have done is
5398 to check that the total number of inodes is greater than zero before doing
5399 the test of how many are available.
5401 58. When a "warn" ACL statement has a log_message modifier, the message is
5402 remembered, and not repeated. This is to avoid a lot of repetition when a
5403 message has many recipients that cause the same warning to be written.
5404 Howewer, Exim was preserving the list of already written lines for an
5405 entire SMTP session, which doesn't seem right. The memory is now reset if a
5406 new message is started.
5408 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5409 debugging output unless log_rewrite was also set.
5411 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5412 of (int)(handle) when we know that handle contains (void *)(-1).
5414 61. The Exim daemon panic-logs an error return when it closes the incoming
5415 connection. However "connection reset by peer" seems to be common, and
5416 isn't really an error worthy of noting specially, so that particular error
5419 62. When Exim is trying to find all the local interfaces, it used to panic and
5420 die if the ioctl to get the interface flags failed. However, it seems that
5421 on at least one OS (Solaris 9) it is possible to have an interface that is
5422 included in the list of interfaces, but for which you get a failure error
5423 for this call. This happens when the interface is not "plumbed" into a
5424 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5425 failure of the "get flags" call assumes that the interface is down.
5427 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5428 makes life easier for people who are doing arithmetic on fields extracted
5429 from dates, where you often get leading zeros that should not be
5430 interpreted as octal.
5432 64. Added qualify_domain to the redirect router, to override the global
5435 65. If a pathologically long header line contained very many addresses (the
5436 report of this problem mentioned 10 000) and each of them was rewritten,
5437 Exim could use up a very large amount of memory. (It kept on making new
5438 copies of the header line as it rewrote, and never released the old ones.)
5439 At the expense of a bit more processing, the header rewriting function has
5440 been changed so that it no longer eats memory in this way.
5442 66. The generation of the Received: header has been moved from the time that a
5443 message starts to be received, to the time that it finishes. The timestamp
5444 in the Received: header should now be very close to that of the <= log
5445 line. There are two side-effects of this change:
5447 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5448 logged header lines no longer include the local Received: line, because
5449 it has not yet been created. The same applies to a copy of the message
5450 that is returned to a non-SMTP sender when a message is rejected.
5452 (b) When a filter file is tested using -bf, no additional Received: header
5453 is added to the test message. After some thought, I decided that this
5456 This change does not affect the value of $received_for. It is still set
5457 after address rewriting, but before local_scan() is called.
5459 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5461 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5462 gave an unhelpful panic error message, and a defer error. I have managed to
5463 change this behaviour so that it now rejects any supplied certificate,
5464 which seems right, as the list of acceptable certificates is empty.
5466 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5467 gave an unhelpful defer error. I have not managed to make this reject any
5468 supplied certificates, but the error message it gives is "no certificate
5469 supplied", which is not helpful.
5471 70. exigrep's output now also includes lines that are not associated with any
5472 message, but which match the given pattern. Implemented by a patch from
5473 Martin Sluka, which also tidied up the Perl a bit.
5475 71. Recipient callout verification, like sender verification, was using <> in
5476 the MAIL FROM command. This isn't really the right thing, since the actual
5477 sender may affect whether the remote host accepts the recipient or not. I
5478 have changed it to use the actual sender in the callout; this means that
5479 the cache record is now keyed on a recipient/sender pair, not just the
5480 recipient address. There doesn't seem to be a real danger of callout loops,
5481 since a callout by the remote host to check the sender would use <>.
5482 [SEE ABOVE: changed after hitting problems.]
5484 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5485 temporary errors. However, in the case of such a code being given after
5486 the end of a data transmission (i.e. after ".") Exim was failing to write
5487 a retry record for the message. (Yes, there was some broken host that was
5488 actually sending 8xx at this point.)
5490 73. An unknown lookup type in a host list could cause Exim to panic-die when
5491 the list was checked. (An example that provoked this was putting <; in the
5492 middle of a list instead of at the start.) If this happened during a DATA
5493 ACL check, a -D file could be left lying around. This kind of configuration
5494 error no longer causes Exim to die; instead it causes a defer errror. The
5495 incident is still logged to the main and panic logs.
5497 74. Buglet left over from Exim 3 conversion. The message "too many messages
5498 in one connection" was written to the rejectlog but not the mainlog, except
5499 when address rewriting (yes!) was being logged.
5501 75. Added write_rejectlog option.
5503 76. When a system filter was run not as root (that is, when system_filter_user
5504 was set), the values of the $n variables were not being returned to the
5505 main process; thus, they were not subsequently available in the $sn
5508 77. Added +return_path_on_delivery log selector.
5510 78. A connection timeout was being treated differently from recipients deferred
5511 when testing hosts_max_try with a message that was older than the host's
5512 retry timeout. (The host should not be counted, thus allowing all hosts to
5513 be tried at least once before bouncing.) This may have been the cause of an
5514 occasionally reported bug whereby a message would remain on the queue
5515 longer than the retry timeout, but would be bounced if a delivery was
5516 forced. I say "may" because I never totally pinned down the problem;
5517 setting up timeout/retry tests is difficult. See also the next item.
5519 79. The ultimate address timeout was not being applied to errors that involved
5520 a combination of host plus message (for example, a timeout on a MAIL
5521 command). When an address resolved to a number of possible hosts, and they
5522 were not all tried for each delivery (e.g. because of hosts_max_try), a
5523 message could remain on the queue longer than the retry timeout.
5525 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5528 81. Fixed an obscure SMTP outgoing bug which required at least the following
5529 conditions: (a) there was another message waiting for the same server;
5530 (b) the server returned 5xx to all RCPT commands in the first message so
5531 that the message was not completed; (c) the server dropped the connection
5532 or gave a negative response to the RSET that Exim sends to abort the
5533 transaction. The observed case was a dropped connection after DATA that had
5534 been sent in pipelining mode. That is, the server had advertised PIPELINING
5535 but was not implementing it correctly. The effect of the bug was incorrect
5536 behaviour, such as trying another host, and this could lead to a crash.
5542 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5543 and daemon.c were passed as pointers to ints; they should have been
5544 pointers to socklen_t variables (which are typically unsigned ints).
5546 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5549 3. Fixed a really odd bug that affected only the testing scheme; patching a
5550 certain fixed string in the binary changed the value of another string that
5551 happened to be identical to the end of the original first string.
5553 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5554 name", it returns that address as the IP address. On some operating
5555 systems (e.g. Solaris), it also passes back the IP address string as the
5556 "host name". However, on others (e.g. Linux), it passes back an empty
5557 string. Exim wasn't checking for this, and was changing the host name to an
5558 empty string, assuming it had been canonicized.
5560 5. Although rare, it is permitted to have more than one PTR record for a given
5561 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5562 all the names associated with an address, because they do in Solaris.
5563 However, it seems that they do not in Linux for data that comes from the
5564 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5565 I found this out when I moved to a new Linux workstation and tried to run
5566 the Exim test suite.
5568 To get round this problem I have changed the code so that it now does its
5569 own call to the DNS to look up PTR records when searching for a host name.
5570 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5571 addresses that are only in /etc/hosts are still found.
5573 This behaviour is, however, controlled by an option called host_lookup_
5574 order, which defaults to "bydns:byaddr". If people want to use the other
5575 order, or indeed, just use one or the other means of lookup, they can
5576 specify it in this variable.
5578 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5579 some operating systems, this comes back from gethostbyaddr() as an empty
5580 string, and this is what Exim used to test for. However, it seems that in
5581 other systems, "." is yielded. Exim now tests for this case too.
5583 7. The values of check_spool_space and check_log_space are now held internally
5584 as a number of kilobytes instead of an absolute number of bytes. If a
5585 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5586 kilobyte. This means that much larger values can be stored.
5588 8. Exim monitor: an attempt to get the action menu when not actually pointing
5589 at a message produces an empty menu entitled "No message selected". This
5590 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5591 no entries in it ("Shell widget menu has zero width and/or height"). So I
5592 have added a single, blank menu entry in this case.
5594 9. Added ${quote_local_part.
5596 10. MIME decoding is now applied to the contents of Subject: header lines when
5599 11. Now that a reference to $sender_host_address automatically causes a reverse
5600 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5601 host lookup before query-style lookups in lists that might use this
5602 variable. This has therefore been abolished, and the "net-" prefix is no
5603 longer necessary for query-style lookups.
5605 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5606 have been a typo for LFLAGS, so it has been changed.
5608 13. The install script calls Exim with "-C /dev/null" in order to find the
5609 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5610 to be output. Howeve, since Exim outputs its version number before the
5611 error, it didn't break the script. It just looked ugly. I fixed this by
5612 always allowing "-C /dev/null" if the caller is root.
5614 14. Ignore overlarge ACL variable number when reading spool file - insurance
5615 against a later release with more variables having written the file.
5617 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5618 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5619 this, as well as the form without the "IPv6" on the front.
5621 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5622 OS/Makefile-Darwin file.
5624 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5625 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5627 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5628 variable. However, some OS use ints for inodes. I've added cast to long int
5629 to get rid of the compiler warning.
5631 19. I had forgotten to lock out "/../" in configuration file names when
5632 ALT_CONFIG_PREFIX was set.
5634 20. Routers used for verification do not need to specify transports. However,
5635 if such a router generated a host list, and callout was configured, Exim
5636 crashed, because it could not find a port number from the (non-existent)
5637 transport. It now assumes port 25 in this circumstance.
5639 21. Added the -t option to exigrep.
5641 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5642 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5643 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5644 them is set, LOOKUP_LSEARCH is forced.
5646 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5648 24. Added sender and host information to the "rejected by local_scan()" log
5649 line; previously there was no indication of these.
5651 25. Added .include_if_exists.
5653 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5654 Linux. It turns out that not all file systems support this. Apparently some
5655 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5656 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5657 not supported on the file descriptor, is now ignored when Exim is trying to
5658 sync a directory. This applies only to Linux.
5660 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5662 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5663 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5666 29. Added support for maildirsize files from supplied patch (modified a bit).
5668 30. The use of :fail: followed by an empty string could lead Exim to respond to
5669 sender verification failures with (e.g.):
5671 550 Verification failed for <xxx>
5672 550 Sender verify failed
5674 where the first response line was missing the '-' that indicates it is not
5675 the final line of the response.
5677 31. The loop for finding the name of the user that called Exim had a hardwired
5678 limit of 10; it now uses the value of finduser_retries, which is used for
5679 all other user lookups.
5681 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5684 33. Exim was neglecting to zero errno before one call of strtol() when
5685 expanding a string and expecting an integer value. On some systems this
5686 resulted in spurious "integer overflow" errors. Also, it was casting the
5687 result into an int without checking.
5689 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5690 did not work. The code looks as if it has *never* worked, though it appears
5691 to have been documented since at least releast 1.62. I have made it work.
5693 35. The "timeout_DNS" error in retry rules, also documented since at least
5694 1.62, also never worked. As it isn't clear exactly what this means, and
5695 clearly it isn't a major issue, I have abolished the feature by treating it
5696 as "timeout", and writing a warning to the main and panic logs.
5698 36. The display of retry rules for -brt wasn't always showing the error code
5701 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5702 timeout_connect_A, timeout_connect_MX.
5704 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5705 to the empty sender.
5707 39. The daemon was not analysing the content of -oX till after it had closed
5708 stderr and disconnected from the controlling terminal. This meant that any
5709 syntax errors were only noted on the panic log, and the return code from
5710 the command was 0. By re-arranging the code a little, I've made the
5711 decoding happen first, so such errors now appear on stderr, and the return
5712 code is 1. However, the actual setting up of the sockets still happens in
5713 the disconnected process, so errors there are still only recorded on the
5716 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5717 connections (as happens on some IP stacks) was logged at start up time as
5718 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5719 it from "IPv6 and IPv4", which means that two separate sockets are being
5722 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5723 says whether AF_INET or AF_INET6 was passed as an argument.
5725 42. Exiwhat output was messed up when time zones were included in log
5728 43. Exiwhat now gives more information about the daemon's listening ports,
5729 and whether -tls-on-connect was used.
5731 44. The "port" option of the smtp transport is now expanded.
5733 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5734 silently ignored. Now an error message is written to the main and panic
5737 46. There's a new ACL modifier called "logwrite" which writes to a log file
5738 as soon as it is encountered.
5740 47. Added $local_user_uid and $local_user_gid at routing time.
5742 48. Exim crashed when trying to verify a sender address that was being
5745 49. Exim was recognizing only a space character after ".include". It now also
5746 recognizes a tab character.
5748 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5749 extracting the relevant information from the specification. The man page no
5750 longer contains scrambled data for the -d option, and I've added a section
5751 at the front about calling Exim under different names.
5753 51. Added "extra_headers" argument to the "mail" command in filter files.
5755 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5758 53. Installed eximstats 1.29.
5760 54. Added transport_filter_timeout as a generic transport option.
5762 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5763 Cc: header lines. This was required by RFC 822, but it not required by RFC
5766 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5767 incoming messages that did not have them. Now it does so only if the
5768 message originates locally, that is, if there is no associated remote host
5769 address. When Resent- header lines are present, this applies to the Resent-
5770 lines rather than the non-Resent- lines.
5772 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5773 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5775 58. Messages for configuration errors now include the name of the main
5776 configuration file - useful now that there may be more than one file in a
5777 list (.included file names were always shown).
5779 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5780 for those rare installations that do not start the daemon as root or run it
5781 setuid root. I've cut out the call to initgroups() if the daemon is not
5784 60. The Exim user and group can now be bound into the binary as text strings
5785 that are looked up at the start of Exim's processing.
5787 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5789 62. Added $mailstore_basename variable.
5791 63. Installed patch to sieve.c from Michael Haardt.
5793 64. When Exim failed to open the panic log after failing to open the main log,
5794 the original message it was trying to log was written to stderr and debug
5795 output, but if they were not available (the usual case in production), it
5796 was lost. Now it is written to syslog before the two lines that record the
5797 failures to open the logs.
5799 65. Users' Exim filters run in subprocesses under the user's uid. It is
5800 possible for a "deliver" command or an alias in a "personal" command to
5801 provoke an address rewrite. If logging of address rewriting is configured,
5802 this fails because the process is not running as root or exim. There may be
5803 a better way of dealing with this, but for the moment (because 4.30 needs
5804 to be released), I have disabled address rewrite logging when running a
5805 filter in a non-root, non-exim process.
5811 1. The buildconfig auxiliary program wasn't quoting the value set for
5812 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5813 not defined. This bug was masked in 4.22 by the effect that was fixed in
5816 2. Some messages that were rejected after a message id was allocated were
5817 shown as "incomplete" by exigrep. It no longer does this for messages that
5818 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5820 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5821 have allow_domain_literals set, the ID did not get logged in the <= line.
5822 Domain literals are now always recognized in Message-ID: header lines.
5824 4. The first argument for a ${extract expansion item is the key name or field
5825 number. Leading and trailing spaces in this item were not being ignored,
5826 causing some misleading effects.
5828 5. When deliver_drop_privilege was set, single queue runner processes started
5829 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5830 same command in the process it spins off) were not dropping privilege.
5832 6. When the daemon running as "exim" started a queue runner, it always
5833 re-executed Exim in the spun-off process. This is a waste of effort when
5834 deliver_drop_privilege is set. The new process now just calls the
5835 queue-runner function directly.
5841 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5844 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5845 ignored. Though the use of -oP was forcing the writing of a pid file, it
5846 was always written to the default place.
5848 3. If the message "no IP address found for host xxxx" is generated during
5849 incoming verification, it is now followed by identification of the incoming
5850 connection (so you can more easily find what provoked it).
5852 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5854 5. Added some features to "harden" Exim a bit more against certain attacks:
5856 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5857 be put in Local/Makefile. This is like the never_users runtime option,
5858 but it cannot be overridden. The default setting is "root".
5860 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5861 prefix string with which any file named in a -C command line option
5864 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5865 is retained for -C and -D only if the caller of Exim is root. Without
5866 it, the exim user may also use -C and -D and retain privilege.
5868 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5869 command line option is disabled.
5871 6. Macro names set by the -D option must start with an upper case letter, just
5872 like macro names defined in the configuration file.
5874 7. Added "dereference=" facility to LDAP.
5876 8. Two instances of the typo "uknown" in the source files are fixed.
5878 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5879 the Configure-Makefile script screwed up while processing it.
5881 10. Incorporated PCRE 4.4.
5883 11. The SMTP synchronization check was not operating right at the start of an
5884 SMTP session. For example, it could not catch a HELO sent before the client
5885 waited for the greeting. There is now a check for outstanding input at the
5886 point when the greeting is written. Because of the duplex, asynchronous
5887 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5888 way, but not yet received, when the check is performed.
5890 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5891 on TCP/IP sockets, because this apparently causes some broken clients to
5894 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5895 unchanged) from the Cygwin maintainer.
5897 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5898 was supported instead of testing for mbx. Effectively a typo.
5900 15. The spa authenticator server code was not checking that the input it
5901 received was valid base64.
5903 16. The debug output line for the "set" modifier in ACLs was not showing the
5904 name of the variable that was being set.
5906 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5908 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5909 lookup on its own. Something else was needed to trigger the lookup. For
5910 example, a match in host_lookup or the need for a host name in a host list.
5911 Now, if $sender_host_name is referenced and the host name has not yet been
5912 looked up, a lookup is performed. If the lookup fails, the variable remains
5913 empty, and $host_lookup_failed is set to "1".
5915 19. Added "eqi" as a case-independent comparison operator.
5917 20. The saslauthd authentication condition could segfault if neither service
5918 nor realm was specified.
5920 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5921 error message that was logged was misleading, and incoming SMTP
5922 connections were dropped. The message is now more accurate, and temporary
5923 errors are given to SMTP connections.
5925 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5926 (except RSET) with a 421 error, until QUIT is received. However, it was
5927 failing to send a response to QUIT.
5929 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5930 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5931 ACL is now run after the helo_try_verify_hosts code.
5933 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5934 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5935 case-independent, so other case variants are also recognized). Apparently
5936 some systems use these upper case variants.
5938 25. If more than two messages were waiting for the same host, and a transport
5939 filter was specified for the transport, Exim sent two messages over the
5940 same TCP/IP connection, and then failed with "socket operation on non-
5941 socket" when it tried to send the third.
5943 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5945 27. The extern definition of crypt16() in expand.c was not being excluded when
5946 the OS had its own crypt16() function.
5948 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5949 as a preferred synonym for return_size_limit, both as an option and as an
5952 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5954 30. Changed the default configuration ACL to relax the local part checking rule
5955 for addresses that are not in any local domains. For these addresses,
5956 slashes and pipe symbols are allowed within local parts, but the sequence
5957 /../ is explicitly forbidden.
5959 31. SPA server authentication was not clearing the challenge buffer before
5962 32. log_message in a "warn" ACL statement was writing to the reject log as
5963 well as to the main log, which contradicts the documentation and doesn't
5964 seem right (because no rejection is happening). So I have stopped it.
5966 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5967 However, I am unable to do any testing of this.
5969 34. Fixed an infelicity in the appendfile transport. When checking directories
5970 for a mailbox, to see if any needed to be created, it was accidentally
5971 using path names with one or more superfluous leading slashes; tracing
5972 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5974 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5975 contents are added to the log line that is written for every discarded
5976 recipient. (Previously a log_message setting was ignored.)
5978 36. The ${quote: operator now quotes the string if it is empty.
5980 37. The install script runs exim in order to find its version number. If for
5981 some reason other than non-existence or emptiness, which it checks, it
5982 could not run './exim', it was installing it with an empty version number,
5983 i.e. as "exim-". This error state is now caught, and the installation is
5986 38. An argument was missing from the function that creates an error message
5987 when Exim fails to connect to the socket for saslauthd authentication.
5988 This could cause Exim to crash, or give a corrupted message.
5990 39. Added isip, isip4, and isip6 to ${if conditions.
5992 40. The ACL variables $acl_xx are now saved with the message, and can be
5993 accessed later in routers, transports, and filters.
5995 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5996 strings in the file are not string-expanded.
5998 42. If a MAIL command specified a SIZE value that was too large to fit into an
5999 int variable, the check against message_size_limit failed. Such values are
6000 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6001 day this will have to be increased, but I don't think I want to be around
6002 when emails are that large.
6009 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6010 iconv() is not standard in FreeBSD.
6012 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6013 IPv6 enabled. The observed symptom was a segmentation fault on return from
6014 the function os_common_find_running_interfaces() in src/os.c.
6016 3. In the check_special_case() function in daemon.c I had used "errno" as an
6017 argument name, which causes warnings on some systems. This was basically a
6018 typo, since it was named "eno" in the comments!
6020 4. The code that waits for the clock to tick (at a resolution of some fraction
6021 of a second) so as to ensure message-id uniqueness was always waiting for
6022 at least one whole tick, when it could have waited for less. [This is
6023 almost certainly not relevant at current processor speeds, where it is
6024 unlikely to ever wait at all. But we try to future-proof.]
6026 5. The function that sleeps for a time interval that includes fractions of a
6027 second contained a race. It did not block SIGALRM between setting the
6028 timer, and suspending (a couple of lines later). If the interval was short
6029 and the sigsuspend() was delayed until after it had expired, the suspension
6030 never ended. On busy systems this could lead to processes getting stuck for
6033 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6034 process, before it forks any delivery processes. The open lookup caching
6035 mechanism meant that the open file or database connection was passed into
6036 the delivery process. The problem was that delivery processes always tidy
6037 up cached lookup data. This could cause a problem for the next delivery
6038 process started by the queue runner, because the external queue runner
6039 process does not know about the closure. So the next delivery process
6040 still has data in the lookup cache. In the case of a file lookup, there was
6041 no problem because closing a file descriptor in a subprocess doesn't affect
6042 the parent. However, if the lookup was caching a connection to a database,
6043 the connection was closed, and the second delivery process was likely to
6044 see errors such as "PGSQL: query failed: server closed the connection
6045 unexpectedly". The problem has been fixed by closing all cached lookups
6046 in a queue runner before running a delivery process.
6048 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6049 seem to have the "const" qualifier which it has on other OS. I've
6052 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6053 *of the same type* (client or server) with the same public name that an
6054 error should be diagnosed.
6056 9. When Exim looked up a host name for an IP address, but failed to find the
6057 original IP address when looking up the host name (a safety check), it
6058 output the message "<ip address> does not match any IP for NULL", which was
6059 confusing, to say the least. The bug was that the host name should have
6060 appeared instead of "NULL".
6062 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6063 user that is built into the binary, and the -C or -D options is used, root
6064 privilege is dropped before the configuration file is read. In addition,
6065 logging is switched to stderr instead of the normal log files. If the
6066 configuration then re-defines the Exim user, the unprivileged environment
6067 is probably not what is expected, so Exim logs a panic warning message (but
6070 However, if deliver_drop_privilege is set, the unprivileged state may well
6071 be exactly what is intended, so the warning has been cut out in that case,
6072 and Exim is allowed to try to write to its normal log files.
6078 1. smtp_return_error_details was not giving details for temporary sender
6079 or receiver verification errors.
6081 2. Diagnose a configuration error if two authenticators have the same public
6084 3. Exim used not to create the message log file for a message until the first
6085 delivery attempt. This could be confusing when incoming messages were held
6086 for policy or load reasons. The message log file is now created at the time
6087 the message is received, and an initial "Received" line is written to it.
6089 4. The automatically generated man page for command line options had a minor
6090 bug that caused no ill effects; however, a more serious problem was that
6091 the procedure for building the man page automatically didn't always
6092 operate. Consequently, release 4.20 contains an out-of-date version. This
6093 shouldn't happen again.
6095 5. When building Exim with embedded Perl support, the script that builds the
6096 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6097 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6099 6. The freeze_tell option was not being used for messages that were frozen on
6100 arrival, either by an ACL or by local_scan().
6102 7. Added the smtp_incomplete_transaction log selector.
6104 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6105 was accepting AUTH without a new EHLO.
6107 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6108 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6111 10. Logging of TCP/IP connections (when configured) now happens in the main
6112 daemon process instead of the child process, so that the TCP/IP connection
6113 count is more accurate (but it can never be perfect).
6115 11. The use of "drop" in a nested ACL was not being handled correctly in the
6116 outer ACL. Now, if condition failure induced by the nested "drop" causes
6117 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6118 or "require"), the connection is dropped.
6120 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6121 that yield "discard" can now be used with an "accept" or a "discard" verb,
6122 but an error is generated for any others (because I can't see a useful way
6123 to define what should happen).
6125 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6126 are now processed in the same way as lines in the Exim configuration file.
6127 In particular, continuation lines are supported.
6129 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6131 15. Added -ti meaning -t -i.
6133 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6134 domains, and warn by logging if others are found.
6136 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6137 returned from GIFCONF when Exim is trying to find the list of interfaces on
6138 a host. The code in os.c has been modified to copy each ifreq to an aligned
6139 structure in all cases.
6141 Also, in some cases, the returned ifreq's were being copied to a 'struct
6142 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6143 means the last couple of bytes of an IPv6 address could be chopped if the
6144 ifreq contained only a normal sockaddr (14 bytes storage).
6146 18. Named domain lists were not supported in the hosts_treat_as_local option.
6147 An entry such as +xxxx was not recognized, and was treated as a literal
6150 19. Ensure that header lines added by a DATA ACL are included in the reject log
6151 if the ACL subsequently rejects the message.
6153 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6154 MD5 (which is deprecated).
6156 21. When testing a filter file using -bf, Exim was writing a message when it
6157 took the sender from a "From " line in the message, but it was not doing so
6158 when it took $return_path from a Return-Path: header line. It now does.
6160 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6161 with a valid header line field name (a series of printing characters
6162 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6164 23. Changed "disc" in the source to "disk" to conform to the documentation and
6165 the book and for uniformity.
6167 24. Ignore Sendmail's -Ooption=value command line item.
6169 25. When execve() failed while trying to run a command in a pipe transport,
6170 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6171 could be confused with a return value of 69 from the command itself. This
6172 has been changed to 127, the value the shell returns if it is asked to run
6173 a non-existent command. The wording for the related log line suggests a
6174 non-existent command as the problem.
6176 26. If received_header_text expands to an empty string, do not add a Received:
6177 header line to the message. (Well, it adds a token one on the spool, but
6178 marks it "old" so that it doesn't get used or transmitted.)
6180 27. Installed eximstats 1.28 (addition of -nt option).
6182 28. There was no check for failure on the call to getsockname() in the daemon
6183 code. This can fail if there is a shortage of resources on the system, with
6184 ENOMEM, for example. A temporary error is now given on failure.
6186 29. Contrary to the C standard, it seems that in some environments, the
6187 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6188 program. Exim now does this explicitly; it affects the formatting of
6189 timestamps using strftime().
6191 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6192 complaints. I've now initialized all the variables, to avoid this.
6194 32. Header lines added by a system filter were not being "seen" during
6195 transport-time rewrites.
6197 33. The info_callback() function passed to OpenSSL is set up with type void
6198 (*)(SSL *, int, int), as described somewhere. However, when calling the
6199 function (actually a macro) that sets it up, the type void(*)() is
6200 expected. I've put in a cast to prevent warnings from picky compilers.
6202 34. If a DNS black list lookup found a CNAME record, but there were no A
6203 records associated with the domain it pointed at, Exim crashed.
6205 35. If a DNS black list lookup returned more than one A record, Exim ignored
6206 all but the first. It now scans all returned addresses if a particular IP
6207 value is being sought. In this situation, the contents of the
6208 $dnslist_value variable are a list of all the addresses, separated by a
6211 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6212 to accept a host name and all its aliases if the forward lookup for any of
6213 them yielded the IP address of the incoming connection. Now it accepts only
6214 those names whose forward lookup yields the correct IP address. Any other
6215 names are discarded. This closes a loophole whereby a rogue DNS
6216 administrator could create reverse DNS records to break through a
6217 wildcarded host restriction in an ACL.
6219 37. If a user filter or a system filter that ran in a subprocess used any of
6220 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6221 the wrong values were passed to the pipe command ($thisaddress had the
6222 value of $0, $0 had the value of $1, etc). This bug was introduced by
6223 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6225 38. Improved the line breaking for long SMTP error messages from ACLs.
6226 Previously, if there was no break point between 40 and 75 characters, Exim
6227 left the rest of the message alone. Two changes have been made: (a) I've
6228 reduced the minimum length to 35 characters; (b) if it can't find a break
6229 point between 35 and 75 characters, it looks ahead and uses the first one
6230 that it finds. This may give the occasional overlong line, but at least the
6231 remaining text gets split now.
6233 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6234 file descriptors might be low, and that setting 1000 would always raise it.
6235 It turns out that in some environments, the limit is already over 1000 and
6236 that lowering it causes trouble. So now Exim takes care not to decrease it.
6238 40. When delivering a message, the value of $return_path is set to $sender_
6239 address at the start of routing (routers may change the value). By an
6240 oversight, this default was not being set up when an address was tested by
6241 -bt or -bv, which affected the outcome if any router or filter referred to
6244 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6245 log only when "message" or "log_message" are set. However, if one of the
6246 conditions was an address verification, or a call to a nested ACL, the
6247 messages generated by the underlying test were being passed through. This
6248 no longer happens. The underlying message is available in $acl_verify_
6249 message for both "message" and "log_message" expansions, so it can be
6250 passed through if needed.
6252 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6253 new expansion $bh_ to give the encoded byte string without charset
6254 translation. Translation happens only if iconv() is available; HAVE_ICONV
6255 indicates this at build time. HEADERS_CHARSET gives the charset to
6256 translate to; headers_charset can change it in the configuration, and
6257 "headers charset" can change it in an individual filter file.
6259 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6260 that creates RFC 2047 encoded "words" labels them as that charset instead
6261 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6262 expansion operator; (ii) when Exim creates a From: line for a local
6263 message; (iii) when a header line is rewritten to include a "phrase" part.
6265 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6266 buggy, causing it to skip the first lines of messages whose message ID
6267 ended in 'D'. This would not have bitten before Exim release 4.14, because
6268 message IDs were unlikely to end in 'D' before then. The effect was to have
6269 incorrect size information for certain domains.
6271 45. #include "config.h" was missing at the start of the crypt16.c module. This
6272 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6275 46. If there was a timeout during a "random" callout check, Exim treated it as
6276 a failure of the random address, and carried on sending RSET and the real
6277 address. If the delay was just some slowness somewhere, the response to the
6278 original RCPT would be taken as a response to RSET and so on, causing
6279 mayhem of various kinds.
6281 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6282 when I implemented it. It didn't allow for the fact that some option values
6283 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6284 the right test for positive values.
6286 48. Domain names in DNS records are case-independent. Exim always looks them up
6287 in lower case. Some resolvers return domain names in exactly the case they
6288 appear in the zone file, that is, they may contain uppercase letters. Not
6289 all resolvers do this - some return always lower case. Exim was treating a
6290 change of case by a resolver as a change of domain, similar to a widening
6291 of a domain abbreviation. This triggered its re-routing code and so it was
6292 trying to route what was effectively the same domain again. This normally
6293 caused routing to fail (because the router wouldn't handle the domain
6294 twice). Now Exim checks for this case specially, and just changes the
6295 casing of the domain that it ultimately uses when it transmits the message
6298 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6301 50. If a filter generated a file delivery with a non-absolute name (possible if
6302 no home directory exists for the router), the forbid_file option was not
6305 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6306 the existing equality matching.
6308 52. Exim was using ints instead of ino_t variables in some places where it was
6309 dealing with inode numbers.
6311 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6312 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6313 called TMPDIR, and if it finds it is different, it changes its value.
6315 54. The smtp_printf() function is now made available to local_scan() so
6316 additional output lines can be written before returning. There is also an
6317 smtp_fflush() function to enable the detection of a dropped connection.
6318 The variables smtp_input and smtp_batched_input are exported to
6321 55. Changed the default runtime configuration: the message "Unknown user"
6322 has been removed from the ACL, and instead placed on the localuser router,
6323 using the cannot_route_message feature. This means that any verification
6324 failures that generate their own messages won't get overridden. Similarly,
6325 the "Unrouteable address" message that was in the ACL for unverifiable
6326 relay addresses has also been removed.
6328 56. Added hosts_avoid_esmtp to the smtp transport.
6330 57. The exicyclog script was not checking for the esoteric option
6331 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6332 will work only if exicyclog is run under the appropriate euid.
6334 58. Following a discussion on the list, the rules by which Exim recognises line
6335 endings on incoming messages have been changed. The -dropcr and drop_cr
6336 options are now no-ops, retained only for backwards compatibility. The
6337 following line terminators are recognized: LF CRLF CR. However, special
6338 processing applies to CR:
6340 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6341 nor a local message in the state where . is a terminator.
6343 (ii) If a bare CR is encountered in a header line, an extra space is added
6344 after the line terminator so as not to end the header. The reasoning
6345 behind this is that bare CRs in header lines are most likely either
6346 to be mistakes, or people trying to play silly games.
6348 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6349 was being incorrectly given as 18 bytes larger than it should have been.
6350 This is a VOB (very old bug).
6352 60. This may never have affected anything current, but just in case it has:
6353 When the local host is found other than at the start of a list of hosts,
6354 the local host, those with the same MX, and any that follow, are discarded.
6355 When the list in question was part of a longer list of hosts, the following
6356 hosts (not currently being processed) were also being discarded. This no
6357 longer happens. I'm not sure if this situation could ever has previously
6360 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6363 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6364 header. This is something that is recommended in a new Internet Draft, and
6365 is something that is documented as being done by Sendmail. There are two
6366 possible values. For messages generated by the autoreply transport, Exim
6369 Auto-Submitted: auto-replied
6371 whereas for all other generated messages (e.g. bounces) it adds
6373 Auto-Submitted: auto-generated
6375 63. The "personal" condition in filters now includes a test for the
6376 Auto-Submitted: header. If it contains the string "auto-" the message it
6377 not considered personal.
6379 64. Added rcpt_include_affixes as a generic transport option.
6381 65. Added queue_only_override (default true).
6383 66. Added the syslog_duplication option.
6385 67. If what should have been the first header line of a message consisted of
6386 a space followed by a colon, Exim was mis-interpreting it as a header line.
6387 It isn't of course - it is syntactically invalid and should therefore be
6388 treated as the start of the message body. The misbehaviour could have
6389 caused a number of strange effects, including loss of data in subsequent
6390 header lines, and spool format errors.
6392 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6393 client host had authenticated. This control can now be exercised by an ACL
6394 for more flexibility.
6396 69. By default, callouts do not happen when testing with -bh. There is now a
6397 variant, -bhc, which does actually run the callout code, including
6398 consulting and updating the callout cache.
6400 70. Added support for saslauthd authentication, courtesy of Alexander
6403 71. If statvfs() failed on the spool or log directories while checking their
6404 size for availability, Exim confusingly gave the error "space shortage".
6405 Furthermore, in debugging mode it crashed with a floating point exception.
6406 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6407 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6408 serious problem, Exim now writes to the main and panic logs when this
6409 happens, with details of the failure. It then refuses to accept the
6410 incoming message, giving the message "spool directory problem" or "log
6411 directory problem" with a 421 code for SMTP messages.
6413 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6414 0, 1, and 2 exist, because some OS complain for execs without them (see
6415 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6416 descriptors. However, the code omitted to check that the open succeeded,
6417 causing mysterious errors if for some reason the permissions on /dev/null
6418 got screwed. Now Exim writes a message to the main and panic logs, and
6419 bombs out if it can't open /dev/null.
6421 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6422 interact so that it is all more flexible. It is supposed to remain
6423 backwards compatible. Also added extra_local_interfaces.
6425 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6426 to bomb out with an assertion failure - to the client this appears as a
6427 connection drop. This problem occurs in the part of the code that was taken
6428 from the Samba project. Fortunately, the assertion is in a very simple
6429 function, so I have fixed this by reproducing the function inline in the
6430 one place where it is called, and arranging for authentication to fail
6431 instead of killing the process with assert().
6433 75. The SPA client code was not working when the server requested OEM rather
6434 than Unicode encoding.
6436 76. Added code to make require_files with a specific uid setting more usable in
6437 the case where statting the file as root fails - usually a non-root-mounted
6438 NFS file system. When this happens and the failure is EACCES, Exim now
6439 forks a subprocess and does the per-uid checking as the relevant uid.
6441 77. Added process_log_path.
6443 78. If log_file_path was not explicitly set, a setting of check_log_space or
6444 check_log_inodes was ignored.
6446 79. If a space check for the spool or log partitions fails, the incident is now
6447 logged. Of course, in the latter case the data may get lost...
6449 80. Added the %p formatting code to string_format() so that it can be used to
6450 print addresses in debug_print(). Adjusted all the address printing in the
6451 debugging in store.c to use %p rather than %d.
6453 81. There was a concern that a line of code in smtp_in.c could overflow a
6454 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6455 initially expressed, the concern was not well-founded, because trailing
6456 spaces are removed early. However, if the trailing spaces were followed by
6457 a NULL, they did not get removed, so the overflow was possible. Two fixes
6460 (a) I re-wrote the offending code in a cleaner fashion.
6461 (b) If an incoming SMTP command contains a NULL character, it is rejected
6464 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6465 runs initgroups(), so that if the Exim user is in any additional groups,
6466 they will be used during message reception.
6472 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob