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 DKIM private-keys having trailing '=' in the base-64.
169 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
171 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
172 we're in there, support oversigning also; bug 1309.
178 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
181 JH/02 The smtp transport option "multi_domain" is now expanded.
183 JH/03 The smtp transport now requests PRDR by default, if the server offers
186 JH/04 Certificate name checking on server certificates, when exim is a client,
187 is now done by default. The transport option tls_verify_cert_hostnames
188 can be used to disable this per-host. The build option
189 EXPERIMENTAL_CERTNAMES is withdrawn.
191 JH/05 The value of the tls_verify_certificates smtp transport and main options
192 default to the word "system" to access the system default CA bundle.
193 For GnuTLS, only version 3.0.20 or later.
195 JH/06 Verification of the server certificate for a TLS connection is now tried
196 (but not required) by default. The verification status is now logged by
197 default, for both outbound TLS and client-certificate supplying inbound
200 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
203 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
204 Status Notification (bounce) messages are now MIME format per RFC 3464.
205 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
206 under the control of the dsn_advertise_hosts option, and routers may
207 have a dsn_lasthop option.
209 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
210 default, modifiable by a malware= option. The list separator for
211 the options can now be changed in the usual way. Bug 68.
213 JH/10 The smtp_receive_timeout main option is now expanded before use.
215 JH/11 The incoming_interface log option now also enables logging of the
216 local interface on delivery outgoing connections.
218 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
219 if the interface and destination host and port all match.
221 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
224 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
225 Patch from Andrew Lewis.
227 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
228 now supports optional time-restrictions, weighting, and priority
229 modifiers per server. Patch originally by <rommer@active.by>.
231 JH/16 The spamd_address main option now supports a mixed list of local
232 and remote servers. Remote servers can be IPv6 addresses, and
233 specify a port-range.
235 JH/17 Bug 68: The spamd_address main option now supports an optional
236 timeout value per server.
238 JH/18 Bug 1581: Router and transport options headers_add/remove can
239 now have the list separator specified.
241 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
244 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
247 JH/21 Support for the A6 type of dns record is withdrawn.
249 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
250 rather than the verbs used.
252 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
253 from 255 to 1024 chars.
255 JH/24 Verification callouts now attempt to use TLS by default.
257 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
258 are generic router options now. The defaults didn't change.
260 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
261 Original patch from Alexander Shikoff, worked over by JH.
263 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
264 files if ZCAT_COMMAND is not executable.
266 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
268 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
270 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
271 Normally benign, it bites when the pair was led to by a CNAME;
272 modern usage is to not canoicalize the domain to a CNAME target
273 (and we were inconsistent anyway for A-only vs AAAA+A).
275 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
277 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
278 when evaluating $sender_host_dnssec.
280 JH/31 Check the HELO verification lookup for DNSSEC, adding new
281 $sender_helo_dnssec variable.
283 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
285 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
287 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
289 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
290 documented as working, but never had. Support all but $spam_report.
292 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
293 added for tls authenticator.
298 TL/01 When running the test suite, the README says that variables such as
299 no_msglog_check are global and can be placed anywhere in a specific
300 test's script, however it was observed that placement needed to be near
301 the beginning for it to behave that way. Changed the runtest perl
302 script to read through the entire script once to detect and set these
303 variables, reset to the beginning of the script, and then run through
304 the script parsing/test process like normal.
306 TL/02 The BSD's have an arc4random API. One of the functions to induce
307 adding randomness was arc4random_stir(), but it has been removed in
308 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
309 function when detected.
311 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
312 cause callback expansion.
314 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
315 syntax errors in an expansion can be treated as a string instead of
316 logging or causing an error, due to the internal use of bool_lax
317 instead of bool when processing it.
319 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
320 server certificates when making smtp deliveries.
322 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
324 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
326 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
328 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
329 Merged patch from Sebastian Wiedenroth.
331 JH/05 Fix results-pipe from transport process. Several recipients, combined
332 with certificate use, exposed issues where response data items split
333 over buffer boundaries were not parsed properly. This eventually
334 resulted in duplicates being sent. This issue only became common enough
335 to notice due to the introduction of conection certificate information,
336 the item size being so much larger. Found and fixed by Wolfgang Breyha.
338 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
339 size buffer was used, resulting in syntax errors when an expansion
342 JH/07 Add support for directories of certificates when compiled with a GnuTLS
343 version 3.3.6 or later.
345 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
346 is EXPERIMENTAL_EVENT, the main-configuration and transport options
347 both become "event_action", the variables become $event_name, $event_data
348 and $event_defer_errno. There is a new variable $verify_mode, usable in
349 routers, transports and related events. The tls:cert event is now also
350 raised for inbound connections, if the main configuration event_action
353 TL/06 In test suite, disable OCSP for old versions of openssl which contained
354 early OCSP support, but no stapling (appears to be less than 1.0.0).
356 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
357 server certificate names available under the smtp transport option
358 "tls_verify_cert_hostname" now do not permit multi-component wildcard
361 JH/10 Time-related extraction expansions from certificates now use the main
362 option "timezone" setting for output formatting, and are consistent
363 between OpenSSL and GnuTLS compilations. Bug 1541.
365 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
366 encoded parameter in the incoming message. Bug 1558.
368 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
369 include certificate info, eximon was claiming there were spoolfile
372 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
374 JH/14 Log delivery-related information more consistently, using the sequence
375 "H=<name> [<ip>]" wherever possible.
377 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
378 are problematic for Debian distribution, omit them from the release
381 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
383 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
386 JH/17 Fix a null-indirection in certextract expansions when a nondefault
387 output list separator was used.
392 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
393 checkers that were complaining about end of non-void function with no
396 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
397 This was a regression intruduced in 4.83 by another bugfix.
399 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
401 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
402 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
408 TF/01 Correctly close the server side of TLS when forking for delivery.
410 When a message was received over SMTP with TLS, Exim failed to clear up
411 the incoming connection properly after forking off the child process to
412 deliver the message. In some situations the subsequent outgoing
413 delivery connection happened to have the same fd number as the incoming
414 connection previously had. Exim would try to use TLS and fail, logging
415 a "Bad file descriptor" error.
417 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
418 utilities have not been installed.
420 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
421 temporary space as the ACL may create new global variables.
423 TL/01 LDAP support uses per connection or global context settings, depending
424 upon the detected version of the libraries at build time.
426 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
427 to extract and use the src ip:port in logging and expansions as if it
428 were a direct connection from the outside internet. PPv2 support was
429 updated based on HAProxy spec change in May 2014.
431 JH/02 Add ${listextract {number}{list}{success}{fail}}.
433 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
434 Properly escape header and check for NULL return.
436 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
439 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
441 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
442 characters in header names, implemented as a verify condition.
443 Contributed by Michael Fischer v. Mollard.
445 TL/05 Rename SPF condition results err_perm and err_temp to standardized
446 results permerror and temperror. Previous values are deprecated but
447 still accepted. In a future release, err_perm and err_temp will be
448 completely removed, which will be a backward incompatibility if the
449 ACL tests for either of these two old results. Patch contributed by
450 user bes-internal on the mailing list.
452 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
454 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
455 selectors, in both main and reject logs.
457 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
460 JH/07 Add malware type "sock" for talking to simple daemon.
462 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
464 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
465 routers/transports under cutthrough routing.
467 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
468 numbers. Touch up "bool" conditional to keep the same definition.
470 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
472 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
474 JH/12 Expand items in router/transport headers_add or headers_remove lists
475 individually rather than the list as a whole. Bug 1452.
477 Required for reasonable handling of multiple headers_ options when
478 they may be empty; requires that headers_remove items with embedded
479 colons must have them doubled (or the list-separator changed).
481 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
482 view the policy declared in the DMARC record. Currently, $dmarc_status
483 is a combined value of both the record presence and the result of the
486 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
488 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
489 dnslookup router and the smtp transport (applying to the forward
492 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
493 of ldap servers used for a specific lookup. Patch provided by Heiko
496 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
497 New variable $lookup_dnssec_authenticated for observability.
499 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
500 Patch submitted by Lars Timman.
502 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
504 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
505 Requires trusted mode and valid format message id, aborts otherwise.
506 Patch contributed by Heiko Schlichting.
508 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
509 certextract with support for various fields. Bug 1358.
511 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
512 is requested by default, modifiable by smtp transport option
515 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
516 operate on certificate variables to give certificate fingerprints
517 Also new ${sha256:cert_variable}.
519 JH/23 The PRDR feature is moved from being Experimental into the mainline.
521 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
522 Christian Aistleitner.
524 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
526 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
527 file. Patch from Wolfgang Breyha.
529 JH/25 Expand the coverage of the delivery $host and $host_address to
530 client authenticators run in verify callout. Bug 1476.
532 JH/26 Port service names are now accepted for tls_on_connect_ports, to
533 align with daemon_smtp_ports. Bug 72.
535 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
536 support and error reporting did not work properly.
538 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
539 and is readable. Patch from Andrew Colin Kissa.
541 TL/14 Enhance documentation of ${run expansion and how it parses the
542 commandline after expansion, particularly in the case when an
543 unquoted variable expansion results in an empty value.
545 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
547 PP/02 Fix internal collision of T_APL on systems which support RFC3123
548 by renaming away from it. Addresses GH issue 15, reported by
551 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
553 TL/15 SECURITY: prevent double expansion in math comparison functions
554 (can expand unsanitized data). Not remotely exploitable.
561 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
563 PP/02 Make -n do something, by making it not do something.
564 When combined with -bP, the name of an option is not output.
566 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
569 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
570 $sender_host_name and config options to manage this, and basic check
573 PP/05 DSCP support for outbound connections and control modifier for inbound.
575 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
576 (Only plugin which currently uses this is kerberos4, which nobody should
577 be using, but we should make it available and other future plugins might
578 conceivably use it, even though it would break NAT; stuff *should* be
579 using channel bindings instead).
581 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
582 name; added for Sendmail compatibility; requires admin caller.
583 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
584 just ignore it); requires trusted caller.
585 Also parse but ignore: -Ac -Am -X<logfile>
588 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
590 TL/02 Add +smtp_confirmation as a default logging option.
592 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
593 Patch by Magnus Holmgren from 2007-02-20.
595 TL/04 Bugzilla 1281 - Spec typo.
596 Bugzilla 1283 - Spec typo.
597 Bugzilla 1290 - Spec grammar fixes.
599 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
601 TL/06 Add Experimental DMARC support using libopendmarc libraries.
603 TL/07 Fix an out of order global option causing a segfault. Reported to dev
604 mailing list by by Dmitry Isaikin.
606 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
608 JH/02 Support "G" suffix to numbers in ${if comparisons.
610 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
612 NM/01 Bugzilla 1197 - Spec typo
613 Bugzilla 1196 - Spec examples corrections
615 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
617 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
618 gnutls_enable_pkcs11, but renamed to more accurately indicate its
621 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
622 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
624 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
625 "acl {{name}{arg}...}", and optional args on acl condition
628 JH/05 Permit multiple router/transport headers_add/remove lines.
630 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
632 JH/07 Avoid using a waiting database for a single-message-only transport.
633 Performance patch from Paul Fisher. Bugzilla 1262.
635 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
638 JH/09 Add $headers_added variable, with content from use of ACL modifier
639 add_header (but not yet added to the message). Bugzilla 199.
641 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
642 Pulled from Bugzilla 817 by Wolfgang Breyha.
644 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
646 (nb: this is the same fix as in Exim 4.80.1)
648 JH/11 Add A= logging on delivery lines, and a client_set_id option on
651 JH/12 Add optional authenticated_sender logging to A= and a log_selector
654 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
656 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
657 advertise SMTP AUTH mechanism to us, instead of a generic
658 protocol violation error. Also, make Exim more robust to bad
659 data from the Dovecot auth socket.
661 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
663 When a queue runner is handling a message, Exim first routes the
664 recipient addresses, during which it prunes them based on the retry
665 hints database. After that it attempts to deliver the message to
666 any remaining recipients. It then updates the hints database using
669 So if a recipient address works intermittently, it can get repeatedly
670 deferred at routing time. The retry hints record remains fresh so the
671 address never reaches the final cutoff time.
673 This is a fairly common occurrence when a user is bumping up against
674 their storage quota. Exim had some logic in its local delivery code
675 to deal with this. However it did not apply to per-recipient defers
676 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
678 This change adds a proper retry rule check during routing so that the
679 final cutoff time is checked against the message's age. We only do
680 this check if there is an address retry record and there is not a
681 domain retry record; this implies that previous attempts to handle
682 the address had the retry_use_local_parts option turned on. We use
683 this as an approximation for the destination being like a local
684 delivery, as in LMTP.
686 I suspect this new check makes the old local delivery cutoff check
687 redundant, but I have not verified this so I left the code in place.
689 TF/02 Correct gecos expansion when From: is a prefix of the username.
691 Test 0254 submits a message to Exim with the header
695 When I ran the test suite under the user fanf2, Exim expanded
696 the header to contain my full name, whereas it should have added
697 a Resent-Sender: header. It erroneously treats any prefix of the
698 username as equal to the username.
700 This change corrects that bug.
702 GF/01 DCC debug and logging tidyup
703 Error conditions log to paniclog rather than rejectlog.
704 Debug lines prefixed by "DCC: " to remove any ambiguity.
706 TF/03 Avoid unnecessary rebuilds of lookup-related code.
708 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
709 Bug spotted by Jeremy Harris; was flawed since initial commit.
710 Would have resulted in OCSP responses post-SNI triggering an Exim
711 NULL dereference and crash.
713 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
715 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
716 Bug detection, analysis and fix by Samuel Thibault.
717 Bugzilla 1331, Debian bug #698092.
719 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
721 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
722 Server implementation by Todd Lyons, client by JH.
723 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
724 config variable "prdr_enable" controls whether the server
725 advertises the facility. If the client requests PRDR a new
726 acl_data_smtp_prdr ACL is called once for each recipient, after
727 the body content is received and before the acl_smtp_data ACL.
728 The client is controlled by bolth of: a hosts_try_prdr option
729 on the smtp transport, and the server advertisement.
730 Default client logging of deliveries and rejections involving
731 PRDR are flagged with the string "PRDR".
733 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
734 fclose(). Diagnosis by Todd Lyons.
736 PP/17 Update configure.default to handle IPv6 localhost better.
737 Patch by Alain Williams (plus minor tweaks).
740 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
741 This is now consistent with GnuTLS, and is now documented: the
742 previous undocumented portable approach to treating the option as
743 unset was to force an expansion failure. That still works, and
744 an empty string is now equivalent.
746 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
747 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
748 not performing validation itself.
750 PP/20 Added force_command boolean option to pipe transport.
751 Patch from Nick Koston, of cPanel Inc.
753 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
756 TF/04 Added udpsend ACL modifer and hexquote expansion operator
758 PP/21 Fix eximon continuous updating with timestamped log-files.
759 Broken in a format-string cleanup in 4.80, missed when I repaired the
760 other false fix of the same issue.
761 Report and fix from Heiko Schlichting.
764 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
765 Report from Prashanth Katuri.
767 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
768 It's SecureTransport, so affects any MacOS clients which use the
769 system-integrated TLS libraries, including email clients.
771 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
772 using a MIME ACL for non-SMTP local injection.
773 Report and assistance in diagnosis by Warren Baker.
775 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
777 JH/16 Fix comparisons for 64b. Bugzilla 1385.
779 TL/09 Add expansion variable $authenticated_fail_id to keep track of
780 last id that failed so it may be referenced in subsequent ACL's.
782 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
785 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
786 ldap library initialization, allowing self-signed CA's to be
787 used. Also properly sets require_cert option later in code by
788 using NULL (global ldap config) instead of ldap handle (per
789 session). Bug diagnosis and testing by alxgomz.
791 TL/12 Enhanced documentation in the ratelimit.pl script provided in
792 the src/util/ subdirectory.
794 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
795 renamed to Transport Post Delivery Action by Jeremy Harris, as
798 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
799 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
800 redis_servers = needs to be configured which will be used by the redis
801 lookup. Patch from Warren Baker, of The Packet Hub.
803 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
805 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
806 hostname or reverse DNS when processing a host list. Used suggestions
807 from multiple comments on this bug.
809 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
811 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
812 Missed a few lines, added it to make the runtest require no keyboard
815 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
816 contains upper case chars. Make router use caseful_local_part.
818 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
819 support when GnuTLS has been built with p11-kit.
825 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
827 This, or similar/improved, will also be change PP/11 of 4.82.
833 PP/01 Handle short writes when writing local log-files.
834 In practice, only affects FreeBSD (8 onwards).
835 Bugzilla 1053, with thanks to Dmitry Isaikin.
837 NM/01 Bugzilla 949 - Documentation tweak
839 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
842 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
844 PP/02 Implemented gsasl authenticator.
846 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
848 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
849 `pkg-config foo` for cflags/libs.
851 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
852 with rest of GSASL and with heimdal_gssapi.
854 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
855 `pkg-config foo` for cflags/libs for the TLS implementation.
857 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
858 properties get this fed in as external SSF. A number of robustness
859 and debugging improvements to the cyrus_sasl authenticator.
861 PP/08 cyrus_sasl server now expands the server_realm option.
863 PP/09 Bugzilla 1214 - Log authentication information in reject log.
864 Patch by Jeremy Harris.
866 PP/10 Added dbmjz lookup type.
868 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
870 PP/12 MAIL args handles TAB as well as SP, for better interop with
871 non-compliant senders.
872 Analysis and variant patch by Todd Lyons.
874 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
875 Bug report from Lars Müller <lars@samba.org> (via SUSE),
876 Patch from Dirk Mueller <dmueller@suse.com>
878 PP/13 tls_peerdn now print-escaped for spool files.
879 Observed some $tls_peerdn in wild which contained \n, which resulted
880 in spool file corruption.
882 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
883 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
884 or write after TLS renegotiation, which otherwise led to messages
887 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
888 as a tracking header (ie: a signed header comes before the signature).
889 Patch from Wolfgang Breyha.
891 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
892 comma-sep list; embedded commas doubled.
894 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
896 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
898 Report and patch from Dmitry Banschikov.
900 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
901 Removed SSL_clear() after SSL_new() which led to protocol negotiation
902 failures. We appear to now support TLS1.1+ with Exim.
904 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
905 lets Exim select keys and certificates based upon TLS SNI from client.
906 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
907 before an outbound SMTP session. New log_selector, +tls_sni.
909 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
910 NULL dereference. Report and patch from Alun Jones.
912 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
913 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
914 Not seeing resolver debug output on NetBSD, but suspect this is a
915 resolver implementation change.
917 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
918 Left warnings. Added "eximon gdb" invocation mode.
920 PP/21 Defaulting "accept_8bitmime" to true, not false.
922 PP/22 Added -bw for inetd wait mode support.
924 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
925 locate the relevant includes and libraries. Made this the default.
927 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
928 Bugzilla 1246, report and most of solution from Tomasz Kusy.
930 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
931 This may cause build issues on older platforms.
933 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
934 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
935 gnutls_require_mac & gnutls_require_protocols (no longer supported).
936 Added SNI support via GnuTLS too.
937 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
939 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
941 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
942 Applied second patch from Janne, implementing suggestion to default
943 multiple-strings-in-record handling to match SPF spec.
945 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
947 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
948 read-only, out of scope).
949 Patch from Wolfgang Breyha, report from Stuart Northfield.
951 PP/29 Fix three issues highlighted by clang analyser static analysis.
952 Only crash-plausible issue would require the Cambridge-specific
953 iplookup router and a misconfiguration.
954 Report from Marcin Mirosław.
956 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
958 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
959 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
960 As part of this, removing so much warning spew let me fix some minor
961 real issues in debug logging.
963 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
964 assignment on my part. Fixed.
966 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
967 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
968 Janne Snabb (who went above and beyond: thank you).
970 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
971 string otherwise requires a connection and a bunch more work and it's
972 relatively easy to get wrong. Should also expose TLS library linkage
975 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
976 64-bit ${eval} (JH/03).
978 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
979 GNU libc to support some of the 64-bit stuff, should not lead to
980 conflicts. Defined before os.h is pulled in, so if a given platform
981 needs to override this, it can.
983 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
984 protection layer was required, which is not implemented.
985 Bugzilla 1254, patch from Wolfgang Breyha.
987 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
988 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
989 tls_dhparam take prime identifiers. Also unbreak combination of
990 OpenSSL+DH_params+TLSSNI.
992 PP/39 Disable SSLv2 by default in OpenSSL support.
998 PP/01 Solaris build fix for Oracle's LDAP libraries.
999 Bugzilla 1109, patch from Stephen Usher.
1001 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1003 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1006 TF/02 Fix a couple more cases where we did not log the error message
1007 when unlink() failed. See also change 4.74-TF/03.
1009 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1010 lock up or crash if it happened to be inside a call to libc when it
1011 got a SIGUSR1 from exiwhat.
1013 The SIGUSR1 handler appends the current process status to the process
1014 log which is later printed by exiwhat. It used to use the general
1015 purpose logging code to do this, but several functions it calls are
1016 not safe for signals.
1018 The new output code in the SIGUSR1 handler is specific to the process
1019 log, and simple enough that it's easy to inspect for signal safety.
1020 Removing some special cases also simplifies the general logging code.
1021 Removing the spurious timestamps from the process log simplifies
1024 TF/04 Improved ratelimit ACL condition.
1026 The /noupdate option has been deprecated in favour of /readonly which
1027 has clearer semantics. The /leaky, /strict, and /readonly update modes
1028 are mutually exclusive. The update mode is no longer included in the
1029 database key; it just determines when the database is updated. (This
1030 means that when you upgrde Exim will forget old rate measurements.)
1032 Exim now checks that the per_* options are used with an update mode that
1033 makes sense for the current ACL. For example, when Exim is processing a
1034 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1035 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1036 must specify per_mail/readonly. If you omit the update mode it defaults to
1037 /leaky where that makes sense (as before) or /readonly where required.
1039 The /noupdate option is now undocumented but still supported for
1040 backwards compatibility. It is equivalent to /readonly except that in
1041 ACLs where /readonly is required you may specify /leaky/noupdate or
1042 /strict/noupdate which are treated the same as /readonly.
1044 A useful new feature is the /count= option. This is a generalization
1045 of the per_byte option, so that you can measure the throughput of other
1046 aggregate values. For example, the per_byte option is now equivalent
1047 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1049 The per_rcpt option has been generalized using the /count= mechanism
1050 (though it's more complicated than the per_byte equivalence). When it is
1051 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1052 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1053 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1054 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1055 Note that using per_rcpt with a non-readonly update mode in more than
1056 one ACL will cause the recipients to be double-counted. (The per_mail
1057 and per_byte options don't have this problem.)
1059 The handling of very low rates has changed slightly. If the computed rate
1060 is less than the event's count (usually one) then this event is the first
1061 after a long gap. In this case the rate is set to the same as this event's
1062 count, so that the first message of a spam run is counted properly.
1064 The major new feature is a mechanism for counting the rate of unique
1065 events. The new per_addr option counts the number of different
1066 recipients that someone has sent messages to in the last time period. It
1067 behaves like per_rcpt if all the recipient addresses are different, but
1068 duplicate recipient addresses do not increase the measured rate. Like
1069 the /count= option this is a general mechanism, so the per_addr option
1070 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1071 example, measure the rate that a client uses different sender addresses
1072 with the options per_mail/unique=$sender_address. There are further
1073 details in the main documentation.
1075 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1077 TF/06 Removed a few PCRE remnants.
1079 TF/07 Automatically extract Exim's version number from tags in the git
1080 repository when doing development or release builds.
1082 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1083 Bugzilla 879. Patch from Paul Fisher.
1085 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1086 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1089 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1091 PP/05 Variable $av_failed, true if the AV scanner deferred.
1092 Bugzilla 1078. Patch from John Horne.
1094 PP/06 Stop make process more reliably on build failure.
1095 Bugzilla 1087. Patch from Heiko Schlittermann.
1097 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1098 Bugzilla 1089. Patch from Heiko Schlittermann.
1100 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1101 Bugzilla 1131. Patch from Holger Weiß.
1103 PP/09 Handle IPv6 addresses with SPF.
1104 Bugzilla 860. Patch from Wolfgang Breyha.
1106 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1108 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1111 PP/11 match_* no longer expand right-hand-side by default.
1112 New compile-time build option, EXPAND_LISTMATCH_RHS.
1113 New expansion conditions, "inlist", "inlisti".
1115 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1117 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1119 PP/14 fix log_write() format string regression from TF/03.
1120 Bugzilla 1152. Patch from Dmitry Isaikin.
1126 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1128 PP/02 Harmonised TLS library version reporting; only show if debugging.
1129 Layout now matches that introduced for other libraries in 4.74 PP/03.
1131 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1133 PP/04 New "dns_use_edns0" global option.
1135 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1138 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1139 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1141 TK/01 Updated PolarSSL code to 0.14.2.
1142 Bugzilla 1097. Patch from Andreas Metzler.
1144 PP/07 Catch divide-by-zero in ${eval:...}.
1145 Fixes bugzilla 1102.
1147 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1150 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1151 format-string attack -- SECURITY: remote arbitrary code execution.
1153 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1154 time unintentionally subject to list matching rules, letting the header
1155 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1156 arbitrary string expansion). This allowed for information disclosure.
1158 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1159 INT_MIN/-1 -- value coerced to INT_MAX.
1165 NM/01 Workround for PCRE version dependency in version reporting
1168 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1169 This fixes portability to compilers other than gcc, notably
1170 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1172 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1173 makefiles for portability to HP-UX and POSIX correctness.
1175 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1176 Also via indented variable definition in the Makefile.
1177 (Debugging by Oliver Heesakkers).
1179 PP/02 Restore caching of spamd results with expanded spamd_address.
1180 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1182 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1183 Improves build reliability. Fix from: Frank Elsner
1185 NM/02 Fix wide character breakage in the rfc2047 coding
1186 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1188 NM/03 Allow underscore in dnslist lookups
1189 Fixes bug 1026. Patch from Graeme Fowler
1191 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1192 Code patches from Adam Ciarcinski of NetBSD.
1194 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1197 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1198 is logged, to avoid truncation. Patch from John Horne.
1200 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1201 Patch from Jakob Hirsch.
1203 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1204 SQL string expansion failure details.
1205 Patch from Andrey Oktyabrski.
1207 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1208 Patch from Simon Arlott.
1210 PP/09 New lookups functionality failed to compile on old gcc which rejects
1211 extern declarations in function scope.
1212 Patch from Oliver Fleischmann
1214 PP/10 Use sig_atomic_t for flags set from signal handlers.
1215 Check getgroups() return and improve debugging.
1216 Fixed developed for diagnosis in bug 927 (which turned out to be
1219 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1220 Patch from Mark Zealey.
1222 PP/12 Bugzilla 1056: Improved spamd server selection.
1223 Patch from Mark Zealey.
1225 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1226 Based on patch from Heiko Schlittermann.
1228 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1229 Patch from Uwe Doering, sign-off by Michael Haardt.
1231 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1232 variable declaration deep within a block. Bug and patch from
1235 PP/15 lookups-Makefile IRIX compatibilty coercion.
1237 PP/16 Make DISABLE_DKIM build knob functional.
1239 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1240 Patch by Simon Arlott
1242 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1243 variable argument macros. Our copy now differs from upstream.
1249 TF/01 Failure to get a lock on a hints database can have serious
1250 consequences so log it to the panic log.
1252 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1253 controlled using the smtp_confirmation log selector.
1255 TF/03 Include the error message when we fail to unlink a spool file.
1257 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1258 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1259 for maintaining out-of-tree patches for some time.
1261 PP/01 Bugzilla 139: Documentation and portability issues.
1262 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1263 Handle per-OS dynamic-module compilation flags.
1265 PP/02 Let /dev/null have normal permissions.
1266 The 4.73 fixes were a little too stringent and complained about the
1267 permissions on /dev/null. Exempt it from some checks.
1268 Reported by Andreas M. Kirchwitz.
1270 PP/03 Report version information for many libraries, including
1271 Exim version information for dynamically loaded libraries. Created
1272 version.h, now support a version extension string for distributors
1273 who patch heavily. Dynamic module ABI change.
1275 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1276 privilege escalation vulnerability whereby the Exim run-time user
1277 can cause root to append content of the attacker's choosing to
1280 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1283 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1284 If dropping privileges for untrusted macros, we disabled normal logging
1285 on the basis that it would fail; for the Exim run-time user, this is not
1286 the case, and it resulted in successful deliveries going unlogged.
1287 Fixed. Reported by Andreas Metzler.
1293 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1294 only prepend for the Resent-* case. Fixes regression introduced in
1295 Exim 4.70 by NM/22 for Bugzilla 607.
1297 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1298 increasing numbers of administrators be bitten by this.
1300 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1302 PP/03 Bugzilla 994: added openssl_options main configuration option.
1304 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1306 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1308 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1310 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1311 without a peer certificate, leading to a segfault because of an
1312 assumption that peers always have certificates. Be a little more
1313 paranoid. Problem reported by Martin Tscholak.
1315 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1316 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1317 NB: ClamAV planning to remove STREAM in "middle of 2010".
1318 CL also introduces -bmalware, various -d+acl logging additions and
1319 more caution in buffer sizes.
1321 PP/09 Implemented reverse_ip expansion operator.
1323 PP/10 Bugzilla 937: provide a "debug" ACL control.
1325 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1327 PP/12 Bugzilla 973: Implement --version.
1329 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1331 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1333 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1335 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1336 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1337 ignore trailing whitespace.
1339 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1341 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1344 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1345 Notification from Dr Andrew Aitchison.
1347 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1348 ExtendedDetectionInfo response format.
1349 Notification from John Horne.
1351 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1354 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1355 XSL and documented dependency on system catalogs, with examples of how
1358 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1361 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1362 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1363 configuration file which is writeable by the Exim user or group.
1365 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1366 of configuration files to cover files specified with the -C option if
1367 they are going to be used with root privileges, not just the default
1370 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1371 option (effectively making it always true).
1373 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1374 files to be used while preserving root privileges.
1376 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1377 that rogue child processes cannot use them.
1379 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1380 run-time user, instead of root.
1382 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1383 Exim run-time user without dropping privileges.
1385 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1386 result string, instead of calling string_vformat() twice with the same
1389 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1390 for other users. Others should always drop root privileges if they use
1391 -C on the command line, even for a whitelisted configure file.
1393 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1395 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1401 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1402 $data_path, and $header_path variables; fixed documentation bugs and
1405 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1406 exipick to access non-standard spools, including the "frozen" queue
1409 NM/01 Bugzilla 965: Support mysql stored procedures.
1410 Patch from Alain Williams
1412 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1414 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1415 Patch from Andreas Metzler
1417 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1418 Patch from Kirill Miazine
1420 NM/05 Bugzilla 671: Added umask to procmail example.
1422 JJ/03 installed exipick 20100323.0, fixing doc bug
1424 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1425 directory. Notification and patch from Dan Rosenberg.
1427 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1429 TK/02 Improve log output when DKIM signing operation fails.
1431 MH/01 Treat the transport option dkim_domain as a colon separated
1432 list, not as a single string, and sign the message with each element,
1433 omitting multiple occurences of the same signer.
1435 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1436 Bugzilla 985, 986. Patch by Simon Arlott
1438 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1439 Patch by Simon Arlott
1441 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1442 MBX locking. Notification from Dan Rosenberg.
1448 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1450 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1452 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1454 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1456 NM/04 Bugzilla 915: Flag broken perl installation during build.
1462 TK/01 Added patch by Johannes Berg that expands the main option
1463 "spamd_address" if it starts with a dollar sign.
1465 TK/02 Write list of recipients to X-Envelope-Sender header when building
1466 the mbox-format spool file for content scanning (suggested by Jakob
1469 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1470 (http://www.dcc-servers.net/) support via dccifd. Activated by
1471 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1473 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1474 by Mark Daniel Reidel <mr@df.eu>.
1476 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1477 When building exim an external PCRE library is now needed -
1478 PCRE is a system library on the majority of modern systems.
1479 See entry on PCRE_LIBS in EDITME file.
1481 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1482 conversation. Added nologin parameter to request.
1483 Patch contributed by Kirill Miazine.
1485 TF/01 Do not log submission mode rewrites if they do not change the address.
1487 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1489 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1490 log files in place. Contributed by Roberto Lima.
1492 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1494 TF/03 Bugzilla 615: When checking the local_parts router precondition
1495 after a local_part_suffix or local_part_prefix option, Exim now
1496 does not use the address's named list lookup cache, since this
1497 contains cached lookups for the whole local part.
1499 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1500 Robert Millan. Documentation is in experimental-spec.txt.
1502 TF/04 Bugzilla 668: Fix parallel build (make -j).
1504 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1506 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1507 Patch provided by Jan Srzednicki.
1509 TF/05 Leading white space used to be stripped from $spam_report which
1510 wrecked the formatting. Now it is preserved.
1512 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1513 that they are available at delivery time.
1515 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1517 TF/08 TLS error reporting now respects the incoming_interface and
1518 incoming_port log selectors.
1520 TF/09 Produce a more useful error message if an SMTP transport's hosts
1521 setting expands to an empty string.
1523 NM/06 Bugzilla 744: EXPN did not work under TLS.
1524 Patch provided by Phil Pennock.
1526 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1527 Patch provided by Richard Godbee.
1529 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1530 acl_smtp_notquit, added index entry.
1532 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1533 Patch provided by Eugene Bujak.
1535 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1536 accept(). Patch provided by Maxim Dounin.
1538 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1539 Patch provided by Phil Pennock.
1541 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1543 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1544 Patch provided by Brad "anomie" Jorsch.
1546 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1547 Patch provided by Dean Brooks.
1549 TK/05 Add native DKIM support (does not depend on external libraries).
1551 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1552 Patch provided by Graeme Fowler.
1554 NM/16 Bugzilla 851: Documentation example syntax fix.
1556 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1558 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1561 NM/19 Bugzilla 745: TLS version reporting.
1562 Patch provided by Phil Pennock.
1564 NM/20 Bugzilla 167: bool: condition support.
1565 Patch provided by Phil Pennock.
1567 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1568 clients. Patch provided by Phil Pennock.
1570 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1571 Patch provided by Brad "anomie" Jorsch.
1573 NM/23 Bugzilla 687: Fix misparses in eximstats.
1574 Patch provided by Heiko Schlittermann.
1576 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1577 Patch provided by Heiko Schlittermann.
1579 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1580 plus update to original patch.
1582 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1584 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1585 Patch provided by David Brownlee.
1587 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1589 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1591 NM/30 Bugzilla 888: TLS documentation bugfixes.
1593 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1595 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1596 Unlike the original bugzilla I have changed all shell scripts in src tree.
1598 NM/33 Bugzilla 898: Transport filter timeout fix.
1599 Patch by Todd Rinaldo.
1601 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1602 Patch by Serge Demonchaux.
1604 NM/35 Bugzilla 39: Base64 decode bug fixes.
1605 Patch by Jakob Hirsch.
1607 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1609 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1611 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1613 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1619 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1620 ALT-N's libdkim that I have put here:
1621 http://duncanthrax.net/exim-experimental/
1623 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1624 were called 'true' and it seems that C99 defines that as a reserved
1625 keyword to be used with 'bool' variable types. That means you could
1626 not include C99-style headers which use bools without triggering
1627 build errors in sieve.c.
1629 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1630 as mailq or other aliases. Changed the --help handling significantly
1631 to do whats expected. exim_usage() emits usage/help information.
1633 SC/01 Added the -bylocaldomain option to eximstats.
1635 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1637 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1639 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1645 PH/01 Another patch from the Sieve maintainer.
1647 PH/02 When an IPv6 address is converted to a string for single-key lookup
1648 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1649 dots are used instead of colons so that keys in lsearch files need not
1650 contain colons. This was done some time before quoting was made available
1651 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1652 (notated using the quote facility) so as to distinguish them from IPv4
1653 keys. This meant that lookups for IP addresses in host lists did not work
1654 for iplsearch lookups.
1656 This has been fixed by arranging for IPv6 addresses to be expressed with
1657 colons if the lookup type is iplsearch. This is not incompatible, because
1658 previously such lookups could never work.
1660 The situation is now rather anomolous, since one *can* have colons in
1661 ordinary lsearch keys. However, making the change in all cases is
1662 incompatible and would probably break a number of configurations.
1664 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1667 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1668 conversion specification without a maximum field width, thereby enabling
1669 a rogue spamd server to cause a buffer overflow. While nobody in their
1670 right mind would setup Exim to query an untrusted spamd server, an
1671 attacker that gains access to a server running spamd could potentially
1672 exploit this vulnerability to run arbitrary code as the Exim user.
1674 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1675 $primary_hostname instead of what libspf2 thinks the hosts name is.
1677 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1678 a directory entry by the name of the lookup key. Previously, if a
1679 symlink pointed to a non-existing file or a file in a directory that
1680 Exim lacked permissions to read, a lookup for a key matching that
1681 symlink would fail. Now it is enough that a matching directory entry
1682 exists, symlink or not. (Bugzilla 503.)
1684 PH/03 The body_linecount and body_zerocount variables are now exported in the
1687 PH/04 Added the $dnslist_matched variable.
1689 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1690 This means they are set thereafter only if the connection becomes
1693 PH/06 Added the client_condition to authenticators so that some can be skipped
1694 by clients under certain conditions.
1696 PH/07 The error message for a badly-placed control=no_multiline_responses left
1697 "_responses" off the end of the name.
1699 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1701 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1702 (without spaces) instead of just copying the configuration text.
1704 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1706 PH/11 Added $max_received_linelength.
1708 PH/12 Added +ignore_defer and +include_defer to host lists.
1710 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1711 way in which PCRE > 7.0 is built.
1713 PH/14 Implemented queue_only_load_latch.
1715 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1716 MAIL command. The effect was to mangle the value on 64-bit systems.
1718 PH/16 Another patch from the Sieve maintainer.
1720 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1722 PH/18 If a system quota error occurred while trying to create the file for
1723 a maildir delivery, the message "Mailbox is full" was not appended to the
1724 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1725 only to a quota excession during the actual writing of the file.
1727 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1728 characters?) which causes problems in log lines. The DN values are now
1729 passed through string_printing() before being added to log lines.
1731 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1732 and InterBase are left for another time.)
1734 PH/21 Added message_body_newlines option.
1736 PH/22 Guard against possible overflow in moan_check_errorcopy().
1738 PH/23 POSIX allows open() to be a macro; guard against that.
1740 PH/24 If the recipient of an error message contained an @ in the local part
1741 (suitably quoted, of course), incorrect values were put in $domain and
1742 $local_part during the evaluation of errors_copy.
1748 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1749 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1752 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1753 issue a MAIL command.
1755 PH/02 In an ACL statement such as
1757 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1759 if a client was not listed at all, or was listed with a value other than
1760 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1761 the condition was not true (as it should be), so access was not denied.
1762 The bug was that the ! inversion was incorrectly passed on to the second
1763 item. This has been fixed.
1765 PH/03 Added additional dnslists conditions == and =& which are different from
1766 = and & when the dns lookup returns more than one IP address.
1768 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1769 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1771 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1772 FSYNC, which compiles an option called disable_fsync that allows for
1773 bypassing fsync(). The documentation is heavily laced with warnings.
1775 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1777 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1778 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1779 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1780 including adding "make clean"; (3) Added -fPIC when compiling the test
1781 dynamically loaded module, to get rid of a warning.
1783 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1784 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1785 = 0s. The bug is otherwise harmless.
1787 PH/07 There was a bug in the dovecot authenticator such that the value of
1788 $auth1 could be overwritten, and so not correctly preserved, after a
1789 successful authentication. This usually meant that the value preserved by
1790 the server_setid option was incorrect.
1792 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1794 PH/09 Installed PCRE release 7.0.
1796 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1797 run for batched SMTP input. It is now run at the start of every message
1798 in the batch. While fixing this I discovered that the process information
1799 (output by running exiwhat) was not always getting set for -bs and -bS
1800 input. This is fixed, and it now also says "batched" for BSMTP.
1802 PH/11 Added control=no_pipelining.
1804 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1805 patch, slightly modified), and move the expansion of helo_data till after
1806 the connection is made in the smtp transport (so it can use these
1809 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1811 PH/14 Added log_selector = +pid.
1813 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1815 PH/16 Add ${if forany and ${if forall.
1817 PH/17 Added dsn_from option to vary the From: line in DSNs.
1819 PH/18 Flush SMTP output before performing a callout, unless control =
1820 no_callout_flush is set.
1822 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1823 was true (the default) a successful delivery failed to delete the retry
1824 item, thus causing premature timeout of the address. The bug is now
1827 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1829 PH/21 Long custom messages for fakedefer and fakereject are now split up
1830 into multiline reponses in the same way that messages for "deny" and
1831 other ACL rejections are.
1833 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1834 with slight modification.
1836 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1837 draft, changing the syntax and factoring some duplicate code".
1839 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1840 for deliveries of the second and subsequent messages over the same SMTP
1843 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1844 ${reduce, with only minor "tidies".
1846 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1848 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1849 expansion side effects.
1851 PH/27 When a message times out after an over-quota error from an Exim-imposed
1852 quota, the bounce message says "mailbox is full". This message was not
1853 being given when it was a system quota that was exceeded. It now should
1856 MH/03 Made $recipients available in local_scan(). local_scan() already has
1857 better access to the recipient list through recipients_list[], but
1858 $recipients can be useful in postmaster-provided expansion strings.
1860 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1861 in the case of a MAIL command with additional options following the
1862 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1863 were accidentally chopped off.
1865 PH/29 SMTP synchronization checks are implemented when a command is read -
1866 there is a check that no more input is waiting when there shouldn't be
1867 any. However, for some commands, a delay in an ACL can mean that it is
1868 some time before the response is written. In this time, more input might
1869 arrive, invalidly. So now there are extra checks after an ACL has run for
1870 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1871 pipelining has not been advertised.
1873 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1875 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1876 returned in the SMTP response when smtp_return_error_details was false.
1877 This has been fixed.
1879 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1880 instead of the C I/O that was originally supplied, because problems were
1881 reported on Solaris.
1883 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1884 Exim which did not show up earlier: it was assuming that a call to
1885 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1886 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1887 was a macro that became an assignment, so it seemed to work. This has
1888 changed to a proper function call with a void return, hence the compile
1889 error. Exim's code has been fixed.
1891 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1894 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1896 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1898 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1899 because it turns out to be too big or there is a timeout) there is no
1900 "Completed" line in the log. When some messages of this type were
1901 selected by exigrep, they were listed as "not completed". Others were
1902 picked up by some special patterns. I have improved the selection
1903 criteria to be more general.
1905 PH/38 The host_find_failed option in the manualroute router can now be set
1906 to "ignore", to completely ignore a host whose IP address cannot be
1907 found. If all hosts are ignored, the behaviour is controlled by the new
1908 host_all_ignored option.
1910 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1911 homing or because of multiple MX records with /mx) generated more than
1912 one IP address, and the following item turned out to be the local host,
1913 all the secondary addresses of the first item were incorrectly removed
1914 from the list, along with the local host and any following hosts (which
1915 is what is supposed to happen).
1917 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1918 whoever called Exim into the -H file. In the case of the daemon it was
1919 behaving confusingly. When first started, it used values for whoever
1920 started the daemon, but after a SIGHUP it used the Exim user (because it
1921 calls itself on a restart). I have changed the code so that it now always
1924 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1925 message are rejected with the same error (e.g. no authentication or bad
1926 sender address), and a DATA command is nevertheless sent (as can happen
1927 with PIPELINING or a stupid MUA), the error message that was given to the
1928 RCPT commands is included in the rejection of the DATA command. This is
1929 intended to be helpful for MUAs that show only the final error to their
1932 PH/42 Another patch from the Sieve maintainer.
1934 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1935 Eximstats - Fixed some broken HTML links and added missing column headers
1937 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1938 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1940 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1942 PH/43 Yet another patch from the Sieve maintainer.
1944 PH/44 I found a way to check for a TCP/IP connection going away before sending
1945 the response to the final '.' that terminates a message, but only in the
1946 case where the client has not sent further data following the '.'
1947 (unfortunately, this is allowed). However, in many cases there won't be
1948 any further data because there won't be any more messages to send. A call
1949 to select() can be used: if it shows that the input is "ready", there is
1950 either input waiting, or the socket has been closed. An attempt to read
1951 the next input character can distinguish the two cases. Previously, Exim
1952 would have sent an OK response which the client would never have see.
1953 This could lead to message repetition. This fix should cure that, at
1954 least in a lot of common cases.
1956 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1957 advertised in response to EHLO.
1963 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1964 fixed by 4.65/MH/01 (is this a record?) are fixed:
1966 (i) An empty string was always treated as zero by the numeric comparison
1967 operators. This behaviour has been restored.
1969 (ii) It is documented that the numeric comparison operators always treat
1970 their arguments as decimal numbers. This was broken in that numbers
1971 starting with 0 were being interpreted as octal.
1973 While fixing these problems I realized that there was another issue that
1974 hadn't been noticed. Values of message_size_limit (both the global option
1975 and the transport option) were treated as octal if they started with 0.
1976 The documentation was vague. These values are now always treated as
1977 decimal, and I will make that clear in the documentation.
1983 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1984 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1987 MH/01 Don't check that the operands of numeric comparison operators are
1988 integers when their expansion is in "skipping" mode (fixes bug
1989 introduced by 4.64-PH/07).
1991 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1992 child addresses, Exim now panics and dies. Previously, because the count
1993 is held in a short int, deliveries were likely to be lost. As such a
1994 large number of recipients for a single message is ridiculous
1995 (performance will be very, very poor), I have chosen to impose a limit
1996 rather than extend the field.
2002 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2003 leftover -K file (the existence of which was triggered by #402).
2004 While we were at it, introduced process PID as part of the -K
2005 filename. This should rule out race conditions when creating
2008 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2009 processing considerably. Previous code took too long for large mails,
2010 triggering a timeout which in turn triggers #401.
2012 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2013 in the DK code in transports.c. sendfile() is not really portable,
2014 hence the _LINUX specificness.
2016 TF/01 In the add_headers option to the mail command in an Exim filter,
2017 there was a bug that Exim would claim a syntax error in any
2018 header after the first one which had an odd number of characters
2021 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2022 callout verification, Exim cached a "reject" for the entire domain. This
2023 is correct for most verifications, but it is not correct for a recipient
2024 verification with use_sender or use_postmaster set, because in that case
2025 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2026 case of MAIL FROM:<> rejection from other early rejections (e.g.
2027 rejection of HELO). When verifying a recipient using a non-null MAIL
2028 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2029 Whatever the result of the callout, the value of the domain cache is
2030 left unchanged (for any other kind of callout, getting as far as trying
2031 RCPT means that the domain itself is ok).
2033 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2036 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2037 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2038 session, the daemon ignores ECONNECT errors and logs others; it now
2039 ignores EPIPE as well.
2041 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2042 (quoted-printable decoding).
2044 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2045 later the small subsequent patch to fix an introduced bug.
2047 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2049 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2051 PH/08 An error is now given if message_size_limit is specified negative.
2053 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2054 to be given (somewhat) arbitrary names.
2056 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2059 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2060 miscellaneous code fixes
2062 PH/10 Added the log_reject_target ACL modifier to specify where to log
2065 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2066 hostname. This is wrong, because it relates to the incoming message (and
2067 probably the interface on which it is arriving) and not to the outgoing
2068 callout (which could be using a different interface). This has been
2069 changed to use the value of the helo_data option from the smtp transport
2070 instead - this is what is used when a message is actually being sent. If
2071 there is no remote transport (possible with a router that sets up host
2072 addresses), $smtp_active_hostname is used.
2074 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2075 tweaks were necessary in order to get it to work (see also 21 below):
2076 (a) The code assumed that strncpy() returns a negative number on buffer
2077 overflow, which isn't the case. Replaced with Exim's string_format()
2079 (b) There were several signed/unsigned issues. I just did the minimum
2080 hacking in of casts. There is scope for a larger refactoring.
2081 (c) The code used strcasecmp() which is not a standard C function.
2082 Replaced with Exim's strcmpic() function.
2083 (d) The code set only $1; it now sets $auth1 as well.
2084 (e) A simple test gave the error "authentication client didn't specify
2085 service in request". It would seem that Dovecot has changed its
2086 interface. Fortunately there's a specification; I followed it and
2087 changed what the client sends and it appears to be working now.
2089 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2092 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2093 address A is aliased to B and C, where B exists and C does not. Without
2094 -v the output is "A verified" because verification stops after a
2095 successful redirection if more than one address is generated. However,
2096 with -v the child addresses are also verified. Exim was outputting "A
2097 failed to verify" and then showing the successful verification for C,
2098 with its parentage. It now outputs "B failed to verify", showing B's
2099 parentage before showing the successful verification of C.
2101 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2102 look up a TXT record in a specific list after matching in a combined
2105 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2106 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2107 they consult the DNS. I had assumed they would set it the way they
2108 wanted; and indeed my experiments on Linux seem to show that in some
2109 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2110 To be on the safe side, however, I have now made the interface to
2111 host_find_byname() similar to host_find_bydns(), with an argument
2112 containing the DNS resolver options. The host_find_byname() function now
2113 sets these options at its start, just as host_find_bydns() does. The smtp
2114 transport options dns_qualify_single and dns_search_parents are passed to
2115 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2116 of host_find_byname() use the default settings of RES_DEFNAMES
2117 (qualify_single) but not RES_DNSRCH (search_parents).
2119 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2120 spool_read_header() do less string testing, by means of a preliminary
2121 switch on the second character of optional "-foo" lines. (This is
2122 overdue, caused by the large number of possibilities that now exist.
2123 Originally there were few.) While I was there, I also converted the
2124 str(n)cmp tests so they don't re-test the leading "-" and the first
2125 character, in the hope this might squeeze out yet more improvement.
2127 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2128 flag allowing group syntax was set by the header_syntax check but not
2129 turned off, possible causing trouble later; (2) The flag was not being
2130 set at all for the header_verify test, causing "group"-style headers to
2131 be rejected. I have now set it in this case, and also caused header_
2132 verify to ignore an empty address taken from a group. While doing this, I
2133 came across some other cases where the code for allowing group syntax
2134 while scanning a header line wasn't quite right (mostly, not resetting
2135 the flag correctly in the right place). These bugs could have caused
2136 trouble for malformed header lines. I hope it is now all correct.
2138 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2139 with the "reply" argument non-NULL. The code, however (which originally
2140 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2141 but it didn't always do it. This confused somebody who was copying the
2142 code for some other use. I have removed all the tests.
2144 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2145 feature that was used to support insecure browsers during the U.S. crypto
2146 embargo. It requires special client support, and Exim is probably the
2147 only MTA that supported it -- and would never use it because real RSA is
2148 always available. This code has been removed, because it had the bad
2149 effect of slowing Exim down by computing (never used) parameters for the
2150 RSA_EXPORT functionality.
2152 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2153 authenticator to fail if there's a tab character in the incoming data
2154 (there should never be unless someone is messing about, as it's supposed
2155 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2156 if the connection is using TLS or if the remote IP is the same as the
2157 local IP, and the "valid-client-cert option" if a client certificate has
2160 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2161 authenticators. This can be used for authorization after authentication
2162 succeeds. (In the case of plaintext, it servers for both authentication
2165 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2166 if any retry times were supplied.
2168 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2169 connection, where there is no client IP address to check. In this
2170 situation, the verify now always succeeds.
2172 PH/25 Applied John Jetmore's -Mset patch.
2174 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2176 PH/27 In a string expansion for a processed (not raw) header when multiple
2177 headers of the same name were present, leading whitespace was being
2178 removed from all of them, but trailing whitespace was being removed only
2179 from the last one. Now trailing whitespace is removed from each header
2180 before concatenation. Completely empty headers in a concatenation (as
2181 before) are ignored.
2183 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2184 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2186 PH/29 [Removed. This was a change that I later backed out, and forgot to
2187 correct the ChangeLog entry (that I had efficiently created) before
2188 committing the later change.]
2190 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2191 address errors (4xx response to RCPT) over the same connection as other
2192 messages routed to the same hosts. Such deliveries are always "forced",
2193 so retry times are not inspected. This resulted in far too many retries
2194 for the affected addresses. The effect occurred only when there were more
2195 hosts than the hosts_max_try setting in the smtp transport when it had
2196 the 4xx errors. Those hosts that it had tried were not added to the list
2197 of hosts for which the message was waiting, so if all were tried, there
2198 was no problem. Two fixes have been applied:
2200 (i) If there are any address or message errors in an SMTP delivery, none
2201 of the hosts (tried or untried) are now added to the list of hosts
2202 for which the message is waiting, so the message should not be a
2203 candidate for sending over the same connection that was used for a
2204 successful delivery of some other message. This seems entirely
2205 reasonable: after all the message is NOT "waiting for some host".
2206 This is so "obvious" that I'm not sure why it wasn't done
2207 previously. Hope I haven't missed anything, but it can't do any
2208 harm, as the worst effect is to miss an optimization.
2210 (ii) If, despite (i), such a delivery is accidentally attempted, the
2211 routing retry time is respected, so at least it doesn't keep
2212 hammering the server.
2214 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2215 in ${readsocket because some servers need this prod.
2217 PH/32 Added some extra debug output when updating a wait-xxx database.
2219 PH/33 The hint "could be header name not terminated by colon", which has been
2220 given for certain expansion errors for a long time, was not being given
2221 for the ${if def:h_colon_omitted{... case.
2223 PH/34 The spec says: "With one important exception, whenever a domain list is
2224 being scanned, $domain contains the subject domain." There was at least
2225 one case where this was not true.
2227 PH/35 The error "getsockname() failed: connection reset by peer" was being
2228 written to the panic log as well as the main log, but it isn't really
2229 panic-worthy as it just means the connection died rather early on. I have
2230 removed the panic log writing for the ECONNRESET error when getsockname()
2233 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2234 runs only) independently of the message's sender address. This meant
2235 that, if the 4xx error was in fact related to the sender, a different
2236 message to the same recipient with a different sender could confuse
2237 things. In particualar, this can happen when sending to a greylisting
2238 server, but other circumstances could also provoke similar problems.
2239 I have changed the default so that the retry time for these errors is now
2240 based a combination of the sender and recipient addresses. This change
2241 can be overridden by setting address_retry_include_sender=false in the
2244 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2245 remote server are returned as part of bounce messages. This was not
2246 happening for LMTP over a pipe (the lmtp transport), but now it is the
2247 same for both kinds of LMTP.
2249 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2250 in header lines that were in fact CNAMEs. This is no longer the case.
2252 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2253 and queue runs started by the daemon processed all messages. This has
2254 been fixed so that -R and -S can now usefully be given with -q<time>.
2256 PH/40 Import PCRE release 6.7 (fixes some bugs).
2258 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2260 PH/42 Give an error if -q is specified more than once.
2262 PH/43 Renamed the variables $interface_address and $interface_port as
2263 $received_ip_address and $received_port, to make it clear that these
2264 values apply to message reception, and not to the outgoing interface when
2265 a message is delivered. (The old names remain recognized, of course.)
2267 PH/44 There was no timeout on the connect() call when using a Unix domain
2268 socket in the ${readsocket expansion. There now is.
2270 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2271 be meaningful with "accept".
2273 SC/01 Eximstats V1.43
2274 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2276 SC/02 Eximstats V1.44
2277 Use a glob alias rather than an array ref in the generated
2278 parser. This improves both readability and performance.
2280 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2281 Collect SpamAssassin and rejection statistics.
2282 Don't display local sender or destination tables unless
2283 there is data to show.
2284 Added average volumes into the top table text output.
2286 SC/04 Eximstats V1.46
2287 Collect data on the number of addresses (recipients)
2288 as well as the number of messages.
2290 SC/05 Eximstats V1.47
2291 Added 'Message too big' to the list of mail rejection
2292 reasons (thanks to Marco Gaiarin).
2294 SC/06 Eximstats V1.48
2295 Mainlog lines which have GMT offsets and are too short to
2296 have a flag are now skipped.
2298 SC/07 Eximstats V1.49 (Alain Williams)
2299 Added the -emptyok flag.
2301 SC/08 Eximstats V1.50
2302 Fixes for obtaining the IP address from reject messages.
2304 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2305 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2306 whitesspace changes from 4.64-PH/27
2308 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2311 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2312 are found, allow negative numbers in numeric criteria)
2314 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2316 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2317 to match changes made in 4.64-PH/43
2319 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2321 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2322 "domain" is actually an IP address, and force a failure. This locks out
2323 those revolvers/nameservers that support "A-for-A" lookups, in
2324 contravention of the specifications.
2326 PH/48 When a host name was looked up from an IP address, and the subsequent
2327 forward lookup of the name timed out, the host name was left in
2328 $sender_host_name, contrary to the specification.
2330 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2331 restricted to single-key lookups, Exim was not diagnosing an error if
2332 * or *@ was used with a query-style lookup.
2334 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2336 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2337 long ago, but noone interested enough thought of it. Let's just say that
2338 the "1.1" means that there are some new functions that weren't there at
2339 some point in the past.
2341 PH/51 Error processing for expansion failure of helo_data from an smtp
2342 transport during callout processing was broken.
2344 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2345 tested/used via the -bh/-bhc/-bs options.
2347 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2348 bug, fixed in subsequent PCRE releases).
2350 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2351 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2353 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2359 SC/01 Use a glob alias rather than an array ref in eximstats generated
2360 parser. This improves both readability and performance.
2362 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2363 Don't display local sender or destination tables in eximstats unless
2364 there is data to show.
2365 Added average volumes into the eximstats top table text output.
2367 SC/03 Collect data on the number of addresses (recipients) as well
2368 as the number of messages in eximstats.
2370 TF/01 Correct an error in the documentation for the redirect router. Exim
2371 does (usually) call initgroups() when daemonizing.
2373 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2374 with consistent privilege compared to when running as a daemon.
2376 TF/03 Note in the spec that $authenticated_id is not set for local
2377 submissions from trusted users.
2379 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2380 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2382 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2383 by adding some example configuration directives to the default
2384 configuration file. A little bit of work is required to uncomment the
2385 directives and define how usernames and passwords are checked, but
2386 there is now a framework to start from.
2388 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2389 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2390 without this. I don't know how relevant this is to other LDAP libraries.
2392 PH/02 Add the verb name to the "unknown ACL verb" error.
2394 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2396 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2398 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2399 directory not expanded when it should be if an expanded home directory
2400 was set for the address (which is overridden by the transport).
2402 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2405 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2406 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2407 because it is too late at that time, and has no effect.
2409 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2410 security issue with \' (bugzilla #107). I could not use the
2411 PQescapeStringConn() function, because it needs a PGconn value as one of
2414 PH/08 When testing addresses using -bt, indicate those final addresses that
2415 are duplicates that would not cause an additional delivery. At least one
2416 person was confused, thinking that -bt output corresponded to deliveries.
2417 (Suppressing duplicates isn't a good idea as you lose the information
2418 about possibly different redirections that led to the duplicates.)
2420 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2421 systems where poll() doesn't work, in particular OS X.
2423 PH/10 Added more information to debugging output for retry time not reached.
2425 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2426 operations in malware.c.
2428 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2431 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2432 syslog_duplication set false, log lines that would normally be written
2433 both the the main log and to the reject log were not written to syslog at
2436 PH/14 In the default configuration, change the use of "message" in ACL warn
2437 statements to "add_header".
2439 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2440 not followed by a command (e.g. "seen endif").
2442 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2443 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2446 PH/17 Added extra conditions to the default value of delay_warning_condition
2450 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2451 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2452 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2455 The Auto-Submitted: and various List- headers are standardised, whereas I
2456 don't think Precedence: ever was.
2458 PH/18 Refactored debugging code in route_finduser() to show more information,
2459 in particular, the error code if getpwnam() issues one.
2461 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2462 This is apparently needed in addition to the PH/07 change above to avoid
2463 any possible encoding problems.
2465 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2466 but not after initializing Perl.
2468 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2469 output them only if debugging. By default they are written stderr,
2470 apparently, which is not desirable.
2472 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2475 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2478 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2480 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2481 authenticated or an ident call has been made. Suppress the default
2482 values for $authenticated_id and $authenticated_sender (but permit -oMai
2483 and -oMas) when testing with -bh.
2485 PH/24 Re-jigged the order of the tests in the default configuration so that the
2486 tests for valid domains and recipients precede the DNS black list and CSA
2487 tests, on the grounds that those ones are more expensive.
2489 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2490 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2491 This bug exists in every version of Exim that I still have, right back to
2494 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2495 However, an attempt to turn on case-sensitivity in a regex key by
2496 including (?-i) didn't work because the subject string was already
2497 lowercased, and the effects were non-intuitive. It turns out that a
2498 one-line patch can be used to allow (?-i) to work as expected.
2504 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2505 other effects) broke the use of negated acl sub-conditions.
2507 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2510 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2511 "Deny" causes Exim to reject the incoming connection with a 554 error.
2512 Unfortunately, if there is a major crisis, such as a disk failure,
2513 tcp-wrappers gives "deny", whereas what one would like would be some
2514 kind of temporary error. A kludge has been added to help with this.
2515 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2516 554 error is used if errno is still zero or contains ENOENT (which occurs
2517 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2520 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2522 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2523 errors. Otherwise a message that provokes a temporary error (when other
2524 messages do not) can cause a whole host to time out.
2526 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2527 addresses were routed directly to files or pipes from a redirect router.
2528 File deliveries just didn't batch; pipe deliveries might have suffered
2531 PH/06 A failure to get a lock for a hints database would erroneously always say
2532 "Failed to get write lock", even when it was really a read lock.
2534 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2535 of 0600. This has been changed to use the value of the lockfile_mode
2536 option (which defaults to 0600).
2538 PH/08 Applied small patch from the Sieve maintainer.
2540 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2541 folder from quota calculations, a direct delivery into this folder messed
2542 up the contents of the maildirsize file. This was because the regex was
2543 used only to exclude .Trash (or whatever) when the size of the mailbox
2544 was calculated. There was no check that a delivery was happening into an
2545 excluded directory. This bug has been fixed by ignoring all quota
2546 processing for deliveries into excluded directories.
2548 PH/10 Added the maildirfolder_create_regex option to appendfile.
2554 PH/01 The code for finding all the local interface addresses on a FreeBSD
2555 system running IPv6 was broken. This may well have applied to all BSD
2556 systems, as well as to others that have similar system calls. The broken
2557 code found IPv4 interfaces correctly, but gave incorrect values for the
2558 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2559 that it would not match correctly against @[] and not recognize the IPv6
2562 PH/02 The ipliteral router was not recognizing addresses of the form user@
2563 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2565 PH/03 Added disable_ipv6.
2567 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2568 decoded form, because it is most often used to construct To: headers
2569 lines in autoreplies, and the decoded form may well be syntactically
2570 invalid. However, $reply_address has leading white space removed, and all
2571 newlines turned into spaces so that the autoreply transport does not
2574 PH/05 If group was specified without a user on a router, and no group or user
2575 was specified on a transport, the group from the router was ignored.
2577 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2578 for visible compile-time settings that can be used to change these
2579 numbers, for those that want even more. Backwards compatibility with old
2580 spool files has been maintained. However, going back to a previous Exim
2581 release will lost any variables that are in spool files.
2583 PH/07 Two small changes when running in the test harness: increase delay when
2584 passing a TCP/IP connection to a new process, in case the original
2585 process has to generate a bounce, and remove special handling of
2586 127.0.0.2 (sic), which is no longer necessary.
2588 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2589 be the same on different OS.
2591 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2594 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2595 whether --show-vars was specified or not
2597 JJ/02 exipick: Added support for new ACL variable spool format introduced
2600 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2601 syntactically invalid From: or Reply-to: line, and a filter used this to
2602 generate an autoreply, and therefore failed to obtain an address for the
2603 autoreply, Exim could try to deliver to a non-existent relative file
2604 name, causing unrelated and misleading errors. What now happens is that
2605 it logs this as a hard delivery error, but does not attempt to create a
2608 PH/11 The exinext utility has a -C option for testing purposes, but although
2609 the given file was scanned by exinext itself; it wasn't being passed on
2610 when Exim was called.
2612 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2613 an end-of-file indication when reading a command response.
2615 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2616 compiled. In many other places in Exim, IPv6 addresses are always
2617 recognized, so I have changed this. It also means that IPv4 domain
2618 literals of the form [IPV4:n.n.n.n] are now always recognized.
2620 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2621 used if the router is not running as root, for example, when verifying at
2622 ACL time, or when using -bh. The debugging output from this situation was
2623 non-existent - all you got was a failure to exec. I have made two
2626 (a) Failures to set uid/gid, the current directory, or a process leader
2627 in a subprocess such as that created by queryprogram now generate
2628 suitable debugging ouput when -d is set.
2630 (b) The queryprogram router detects when it is not running as root,
2631 outputs suitable debugging information if -d is set, and then runs
2632 the subprocess without attempting to change uid/gid.
2634 PH/15 Minor change to Makefile for building test_host (undocumented testing
2637 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2638 additional section of a DNS packet that returns MX or SRV records.
2639 Instead, it always explicitly searches for A/AAAA records. This avoids
2640 major problems that occur when a DNS server includes only records of one
2641 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2642 fixed another bug: if SRV records were looked up and the corresponding
2643 address records were *not* found in the additional section, the port
2644 values from the SRV records were lost.
2646 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2647 using the correct key (the original address) when searching the retry
2648 rules in order to find which one to use for generating the retry hint.
2650 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2651 adding the default one. Similarly, if it contains a Reply-To: header, the
2652 errors_reply_to option, if set, is not used.
2654 PH/19 When calculating a retry time, Exim used to measure the "time since
2655 failure" by looking at the "first failed" field in the retry record. Now
2656 it does not use this if it is later than than the arrival time of the
2657 message. Instead it uses the arrival time. This makes for better
2658 behaviour in cases where some deliveries succeed, thus re-setting the
2659 "first failed" field. An example is a quota failure for a huge message
2660 when small messages continue to be delivered. Without this change, the
2661 "time since failure" will always be short, possible causing more frequent
2662 delivery attempts for the huge message than are intended.
2663 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2665 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2666 $1, $2, $3) because the numerical variables can be reset during some
2667 expansion items (e.g. "match"), thereby losing the authentication data.
2669 PH/21 Make -bV show the size of off_t variables so that the test suite can
2670 decide whether to run tests for quotas > 2G.
2672 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2673 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2674 filecount value is greater than 2G or if a quota value is greater than 2G
2675 on a system where the size of off_t is not greater than 4, a panic error
2678 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2679 never match. The debug and -bh output now contains an explicit error
2680 message indicating a malformed IPv4 address or mask.
2682 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2683 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2684 PH/23 above applies.
2686 PH/25 Do not write to syslog when running in the test harness. The only
2687 occasion when this arises is a failure to open the main or panic logs
2688 (for which there is an explicit test).
2690 PH/26 Added the /no_tell option to "control=freeze".
2692 PH/27 If a host name lookup failed very early in a connection, for example, if
2693 the IP address matched host_lookup and the reverse lookup yielded a name
2694 that did not have a forward lookup, an error message of the form "no IP
2695 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2696 could be logged. Now it outputs the IP address instead of "NULL".
2698 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2699 allows the sender and the authenticated sender to be set when
2700 submitting a message from within Exim. Since child_open_exim() is
2701 documented for local_scan(), the new function should be too.
2703 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2704 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2705 results in an empty string is now treated as unset.
2707 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2709 PH/31 Added sender_verify_fail logging option.
2711 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2712 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2713 tidied the source and removed it altogether.
2715 PH/33 When a queue run was abandoned because the load average was too high, a
2716 log line was always written; now it is written only if the queue_run log
2717 selector is set. In addition, the log line for abandonment now contains
2718 information about the queue run such as the pid. This is always present
2719 in "start" and "stop" lines but was omitted from the "abandon" line.
2721 PH/34 Omit spaces between a header name and the colon in the error message that
2722 is given when verify = headers_syntax fails (if there are lots of them,
2723 the message gets confusing).
2725 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2726 names, as there are now some PTR records that contain slashes. This check
2727 is only to protect against broken name servers that fall over on strange
2728 characters, so the fact that it applies to all lookups doesn't matter.
2730 PH/36 Now that the new test suite is complete, we can remove some of the
2731 special code in Exim that was needed for the old test suite. For example,
2732 sorting DNS records because real resolvers return them in an arbitrary
2733 order. The new test suite's fake resolver always returns records in the
2736 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2737 bounces) except when queue_only is set, to avoid logging races between
2738 the different processes.
2740 PH/38 Panic-die if .include specifies a non-absolute path.
2742 PH/39 A tweak to the "H" retry rule from its user.
2744 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2745 a label. They prevented compilation on older perls.
2747 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2748 a warning to be raised on newish perls.
2750 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2751 on queue. Changes to match documented behaviour of showing count of
2752 messages matching specified criteria.
2754 PH/40 Changed the default ident timeout from 30s to 5s.
2756 PH/41 Added support for the use of login_cap features, on those BSD systems
2757 that have them, for controlling the resources used by pipe deliveries.
2759 PH/42 The content-scanning code uses fopen() to create files in which to put
2760 message data. Previously it was not paying any attention to the mode of
2761 the files. Exim runs with umask(0) because the rest of the code creates
2762 files with open(), and sets the required mode explicitly. Thus, these
2763 files were ending up world-writeable. This was not a big issue, because,
2764 being within the spool directory, they were not world-accessible. I have
2765 created a function called modefopen, which takes an additional mode
2766 argument. It sets umask(777), creates the file, chmods it to the required
2767 mode, then resets the umask. All the relevant calls to fopen() in the
2768 content scanning code have been changed to use this function.
2770 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2771 to 24 hours. This avoids potential overflow problems when processing G
2772 and H retry rules. I suspect nobody ever tinkers with this value.
2774 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2776 PH/45 When the plaintext authenticator is running as a client, the server's
2777 challenges are checked to ensure they are valid base64 strings. By
2778 default, the authentication attempt is cancelled if an invalid string is
2779 received. Setting client_ignore_invalid_base64 true ignores these errors.
2780 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2781 they are received. Thus, the responses can be made to depend on the
2782 challenges. If an invalid string is ignored, an empty string is placed in
2785 PH/46 Messages that are created by the autoreply transport now contains a
2786 References: header, in accordance with RFCs 2822 and 3834.
2788 PH/47 Added authenticated_sender_force to the smtp transport.
2790 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2792 PH/49 Installed latest patch from the Sieve maintainer.
2794 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2795 was also set, the appendfile transport was unnecessarily scanning a
2796 directory of message files (e.g. for maildir delivery) to find the count
2797 of files (along with the size), even though it did not need this
2798 information. It now does the scan only if it needs to find either the
2799 size of the count of files.
2801 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2803 PH/52 Two bugs concerned with error handling when the smtp transport is
2806 (i) Exim was not creating retry information for temporary errors given
2807 for individual recipients after the DATA command when the smtp transport
2808 was used in LMTP mode. This meant that they could be retried too
2809 frequently, and not timed out correctly.
2811 (ii) Exim was setting the flag that allows error details to be returned
2812 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2813 recipients that were returned after the DATA command.
2815 PH/53 This is related to PH/52, but is more general: for any failing address,
2816 when detailed error information was permitted to be returned to the
2817 sender, but the error was temporary, then after the final timeout, only
2818 "retry timeout exceeded" was returned. Now it returns the full error as
2819 well as "retry timeout exceeded".
2821 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2822 do this, and (what is worse) MTAs that accept it.
2824 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2825 will now be deprecated.
2827 PH/56 New os.c-cygwin from the Cygwin maintainer.
2829 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2830 formats (previously only available in exim formats via -bpr, -bpru,
2831 and -bpra. Now also available in native and exiqgrep formats)
2833 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2834 with very large, slow to parse queues
2836 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2838 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2840 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2841 responses to authentication challenges, though it was showing the
2842 challenges; (ii) I've removed the CR characters from the debug output for
2845 PH/58 Allow for the insertion of a newline as well as a space when a string
2846 is turned into more than one encoded-word during RFC 2047 encoding. The
2847 Sieve code now uses this.
2849 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2850 data_4xx, lost_connection, tls_required.
2852 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2853 message was being sent as an SMTP response.
2855 PH/61 Add -l and -k options to exicyclog.
2857 PH/62 When verifying, if an address was redirected to one new address, so that
2858 verification continued, and the new address failed or deferred after
2859 having set something in $address_data, the value of $address_data was not
2860 passed back to the ACL. This was different to the case when no
2861 redirection occurred. The value is now passed back in both cases.
2863 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2864 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2865 use login_cap.h, so on its own it isn't the distinguishing feature. The
2866 new name refers directly to the setclassresources() function.
2868 PH/65 Added configuration files for NetBSD3.
2870 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2872 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2873 is preferred over IPv4.
2875 PH/68 The bounce_return_message and bounce_return_body options were not being
2876 honoured for bounces generated during the reception of non-SMTP messages.
2877 In particular, this applied to messages rejected by the ACL. This bug has
2878 been fixed. However, if bounce_return_message is true and bounce_return_
2879 body is false, the headers that are returned for a non-SMTP message
2880 include only those that have been read before the error was detected.
2881 (In the case of an ACL rejection, they have all been read.)
2883 PH/69 The HTML version of the specification is now built in a directory called
2884 spec_html instead of spec.html, because the latter looks like a path with
2885 a MIME-type, and this confuses some software.
2887 PH/70 Catch two compiler warnings in sieve.c.
2889 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2890 function verify_get_ident() calls ip_connect() to connect a socket, but
2891 if the "connect()" function timed out, ip_connect() used to close the
2892 socket. However, verify_get_ident() also closes the socket later, and in
2893 between Exim writes to the log, which may get opened at this point. When
2894 the socket was closed in ip_connect(), the log could get the same file
2895 descriptor number as the socket. This naturally causes chaos. The fix is
2896 not to close the socket in ip_connect(); the socket should be closed by
2897 the function that creates it. There was only one place in the code where
2898 this was missing, in the iplookup router, which I don't think anybody now
2899 uses, but I've fixed it anyway.
2901 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2902 well as to direct DNS lookups. Otherwise the handling of names in host
2903 lists is inconsistent and therefore confusing.
2909 PH/01 Two changes to the default runtime configuration:
2911 (1) Move the checks for relay_from_hosts and authenticated clients from
2912 after to before the (commented out) DNS black list checks.
2914 (2) Add control=submission to the relay_from_hosts and authenticated
2915 clients checks, on the grounds that messages accepted by these
2916 statements are most likely to be submissions.
2918 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2920 (1) Generate an error if the third argument for the ${prvs expansion is
2923 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2926 (3) Reset the variables that are obtained from the first argument of
2927 ${prvscheck and used in the second argument before leaving the code,
2928 because their memory is reclaimed, so using them afterwards may do
2931 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2932 one (it's much easier than Tom thought :-).
2934 (5) Because of (4), we can now allow for the use of $prvscheck_result
2935 inside the third argument.
2937 PH/03 For some reason, the default setting of PATH when running a command from
2938 a pipe transport was just "/usr/bin". I have changed it to
2941 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2942 anything to be listed in the output from -bV.
2944 PH/05 When a filter generated an autoreply, the entire To: header line was
2945 quoted in the delivery log line, like this:
2947 => >A.N.Other <ano@some.domain> <original@ddress> ...
2949 This has been changed so that it extracts the operative address. There
2950 may be more than one such address. If so, they are comma-separated, like
2953 => >ano@some.domain,ona@other.domain <original@ddress> ...
2955 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2956 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2957 not being looked up in the rDNS to get a host name, Exim was showing the
2958 IP address twice in Received: lines, even though the IP addresses were
2959 identical. For example:
2961 Received: from [1.2.3.4] (helo=[1.2.3.4])
2963 However, if the real host name was known, it was omitting the HELO data
2964 if it matched the actual IP address. This has been tidied up so that it
2965 doesn't show the same IP address twice.
2967 PH/07 When both +timestamp and +memory debugging was on, the value given by
2968 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2969 called by the debug printing, thereby overwriting the timestamp buffer.
2970 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2972 PH/08 When the original message was included in an autoreply transport, it
2973 always said "this is a copy of the message, including all the headers",
2974 even if body_only or headers_only was set. It now gives an appropriate
2977 PH/09 Applied a patch from the Sieve maintainer which:
2979 o fixes some comments
2980 o adds the (disabled) notify extension core
2981 o adds some debug output for the result of if/elsif tests
2982 o points to the current vacation draft in the documentation
2983 and documents the missing references header update
2987 o fixes a bug in processing the envelope test (when testing
2988 multiple envelope elements, the last element determinted the
2991 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2992 Electronic Mail") by including:
2994 Auto-submitted: auto-generated
2996 in the messages that it generates (bounce messages and others, such as
2997 warnings). In the case of bounce messages for non-SMTP mesages, there was
2998 also a typo: it was using "Auto_submitted" (underscore instead of
2999 hyphen). Since every message generated by Exim is necessarily in response
3000 to another message, thes have all been changed to:
3002 Auto-Submitted: auto-replied
3004 in accordance with these statements in the RFC:
3006 The auto-replied keyword:
3008 - SHOULD be used on messages sent in direct response to another
3009 message by an automatic process,
3011 - MUST NOT be used on manually-generated messages,
3013 - MAY be used on Delivery Status Notifications (DSNs) and Message
3014 Disposition Notifications (MDNs),
3016 - MUST NOT be used on messages generated by automatic or periodic
3017 processes, except for messages which are automatic responses to
3020 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3021 to the default Received: header definition.
3023 PH/12 Added log selector acl_warn_skipped (default on).
3025 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3026 variables because (a) they are in the wrong storage pool and (b) even if
3027 they were copied, it wouldn't work properly because of the caching.
3029 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3030 checking when decoding. Apparently there are clients that generate
3031 overlong encoded strings. Why am I not surprised?
3033 PH/15 If the first argument of "${if match_address" was not empty, but did not
3034 contain an "@" character, Exim crashed. Now it writes a panic log message
3035 and treats the condition as false.
3037 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3039 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3040 extension "envelope-auth". The code is finished and in agreement with
3041 other implementations, but there is no documentation so far and in fact,
3042 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3043 not changing the active code.
3045 Print executed "if" and "elsif" statements when debugging is used. This
3046 helps a great deal to understand what a filter does.
3048 Document more things not specified clearly in RFC3028. I had all this
3049 sorted out, when out of a sudden new issues came to my mind. Oops."
3051 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3054 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3055 canonical form (as documented). However, after a host name lookup from
3056 the IP address, check_host() was doing a simple string comparison with
3057 addresses acquired from the DNS when checking that the found name did
3058 have the original IP as one of its addresses. Since any found IPv6
3059 addresses are likely to be in abbreviated form, the comparison could
3060 fail. Luckily, there already exists a function for doing the comparison
3061 by converting both addresses to binary, so now that is used instead of
3062 the text comparison.
3064 PH/20 There was another similar case to PH/19, when a complete host name was
3065 given in a host list; looking up its IP address could give an abbreviated
3066 form, whereas the current host's name might or might not be abbreviated.
3067 The same fix has been applied.
3073 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3074 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3077 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3078 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3080 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3082 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3083 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3084 Also, for "Auto-Submitted", treat anything other than "no" as
3085 non-personal, in accordance with RFC 3834. (Previously it treated
3086 anything starting "auto-" as non-personal.)
3088 TF/01 The control=submission/name=... option had a problem with syntax
3089 errors if the name included a slash character. The /name= option
3090 now slurps the rest of the string, so it can include any characters
3091 but it must come last in the list of options (after /sender_retain
3094 PH/05 Some modifications to the interface to the fake nameserver for the new
3102 TK/01 Added the "success_on_redirect" address verification option. See
3103 NewStuff for rationale and an example.
3105 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3107 PH/02 Patch to exigrep to allow it to work on syslog lines.
3109 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3110 fread() to skip over the body file's header line, because in Cygwin the
3111 header line is locked and is inaccessible.
3113 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3114 co-exist for some time) to make it clear that it is the Exim ID that is
3115 referenced, not the Message-ID: header line.
3117 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3118 string_format() function, because snprintf() does not exist on all
3121 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3122 ${stat: expansion item.
3124 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3125 protocol synchronization error", to keep the pedants happy.
3127 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3128 well as for IRIX systems, when gcc is being used. See the host.c source
3131 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3133 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3136 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3137 transport and to the smtp transport in LMTP mode.
3139 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3141 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3142 run a filter in a subprocess. This could lead to confusion in subsequent
3143 lookups in the parent process. There should also be a search_tidyup() at
3144 the end of the subprocess.
3146 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3147 only if the host matched helo_try_verify_hosts, which caused the
3148 verification to occur when the EHLO/HELO command was issued. The ACL just
3149 tested the remembered result. Now, if a previous verification attempt has
3150 not happened, "verify = helo" does it there and then.
3152 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3154 TK/03 Fix log output including CR from clamd.
3156 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3157 exist provoked a memory error which could cause a segfault.
3159 PH/15 Installed PCRE 6.2
3161 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3163 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3164 of the problem. Specifically, suggested +O2 rather than +O1 for the
3167 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3169 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3170 record, the port was not correctly logged when the outgoing_port log
3171 selector was set (it logged the transort's default port).
3173 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3174 fallback_hosts, and "hosts" in the smtp transport.
3176 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3177 host errors such as "Connection refused".
3179 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3180 authentication with radiusclient 0.4.9:
3182 - Error returned from rc_read_config was caught wrongly
3183 - Username/password not passed on to radius server due to wrong length.
3185 The presumption is that some radiusclient API changes for 4.51/PH/17
3186 were not taken care of correctly. The code is still untested by me (my
3187 Linux distribution still has 0.3.2 of radiusclient), but it was
3188 contributed by a Radius user.
3190 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3191 expanding the "port" option of the smtp transport.
3193 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3194 while reading a MIME header. Thanks to Tom Hughes for a patch.
3196 PH/24 Include config.h inside local_scan.h so that configuration settings are
3199 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3200 that in an ACL for RCPT (for example), you can examine exactly what was
3203 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3204 commands, but it was not correctly comparing the address with the actual
3205 client host address. Thus, it would show the EHLO address in Received:
3206 header lines when this was not necessary.
3208 PH/27 Added the % operator to ${eval:}.
3210 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3211 it should be ignoring failures (because with -C, for example, it has lost
3212 privilege). It wasn't ignoring creation failures other than "already
3215 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3218 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3219 because an input line was too long, either on its own, or by virtue of
3220 too many continuations, the temporary file was not being removed, and the
3221 return code was incorrect.
3223 PH/31 Missing "BOOL" in function definition in filtertest.c.
3225 PH/32 Applied Sieve patches from the maintainer.
3227 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3229 PH/33 Added "verify = not_blind".
3231 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3232 Local/Makefile (with some defaults set). These are used in built scripts
3233 such as exicyclog, but they have never been used in the exim_install
3234 script (though there are many overriding facilities there). I have
3235 arranged that the exim_install script now takes note of these two
3238 PH/35 Installed configuration files for Dragonfly.
3240 PH/36 When a locally submitted message by a trusted user did not contain a
3241 From: header, and the sender address was obtained from -f or from an SMTP
3242 MAIL command, and the trusted user did not use -F to supply a sender
3243 name, $originator_name was incorrectly used when constructing a From:
3244 header. Furthermore, $originator_name was used for submission mode
3245 messages from external hosts without From: headers in a similar way,
3246 which is clearly wrong.
3248 PH/37 Added control=suppress_local_fixups.
3250 PH/38 When log_selector = +received_sender was set, and the addition of the
3251 sender made the log line's construction buffer exactly full, or one byte
3252 less than full, an overflow happened when the terminating "\n" was
3255 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3256 when the result of a list match is failure because a DNS lookup failed.
3258 PH/40 RM_COMMAND is now used in the building process.
3260 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3261 the "build-* directories that it finds.
3263 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3264 address in a domain literal was a prefix of an interface address.
3266 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3267 when verifying a sender address, unless rewrite_headers is false.
3269 PH/44 Wrote a long comment about why errors_to addresses are verified as
3270 recipients, not senders.
3272 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3273 the ratelimit ACL was added.
3275 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3277 PH/46 Added extra information about PostgreSQL errors to the error string.
3279 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3280 avoiding the need to install special zones in a real server. This is
3281 backwards compatible; if it can't find the fake resolver, it drops back.
3282 Thus, both old and new test suites can be run.
3284 TF/02 Added util/ratelimit.pl
3286 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3287 clock is set back in time.
3289 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3290 Candler <B.Candler@pobox.com>.
3292 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3293 OK for addresses that are the result of redirections.
3295 PH/48 A number of further additions for the benefit of the new test suite,
3296 including a fake gethostbyname() that interfaces to the fake DNS resolver
3299 TF/06 The fix for widen_domains has also been applied to qualify_single and
3300 search_parents which are the other dnslookup options that can cause
3303 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3306 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3308 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3309 service" if it encountered an MX record with an empty target hostname.
3310 The message is now "an MX or SRV record indicated no SMTP service".
3312 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3313 if the DNS of the sending site is misconfigured. This is quite a
3314 common situation. This change restores the behaviour of treating a
3315 helo verification defer as a failure.
3317 PH/51 If self=fail was set on a router, the bounce message did not include the
3318 actual error message.
3324 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3326 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3327 command itself ended in error, the underlying message about the transport
3328 filter timeout was being overwritten with the pipe command error. Now the
3329 underlying error message should be appended to the second error message.
3331 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3333 PH/02 Reduce the amount of output that "make" produces by default. Full output
3334 can still be requested.
3336 PH/03 The warning log line about a condition test deferring for a "warn" verb
3337 was being output only once per connection, rather than after each
3338 occurrence (because it was using the same function as for successful
3339 "warn" verbs). This seems wrong, so I have changed it.
3341 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3342 it should not have, which might have caused a crash in the right
3343 circumstances, but probably never did.
3345 PH/04 Installed a modified version of Tony Finch's patch to make submission
3346 mode fix the return path as well as the Sender: header line, and to
3347 add a /name= option so that you can make the user's friendly name appear
3350 TF/03 Added the control = fakedefer ACL modifier.
3352 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3353 Mark Lowes for thorough testing.
3355 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3357 TK/03 Merged latest SRS patch from Miles Wilton.
3359 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3360 with the definition in sysexits.h (which is #included earlier).
3361 Fortunately, Exim does not actually use EX_OK. The code used to try to
3362 preserve the sysexits.h value, by assumimg that macro definitions were
3363 scanned for macro replacements. I have been disabused of this notion,
3364 so now the code just undefines EX_OK before #including unistd.h.
3366 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3367 in the smtp transport. When a block could not be written in a single
3368 write() function, the timeout was being re-applied to each part-write.
3369 This seems wrong - if the receiver was accepting one byte at a time it
3370 would take for ever. The timeout is now adjusted when this happens. It
3371 doesn't have to be particularly precise.
3373 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3374 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3376 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3377 without a domain if the check to <postmaster@domain> fails.
3379 SC/01 Eximstats: added -xls and the ability to specify output files
3380 (patch written by Frank Heydlauf).
3382 SC/02 Eximstats: use FileHandles for outputing results.
3384 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3386 SC/04 Eximstats: fixed display of large numbers with -nvr option
3388 SC/05 Eximstats: fixed merging of reports with empty tables.
3390 SC/06 Eximstats: added the -include_original_destination flag
3392 SC/07 Eximstats: removed tabs and trailing whitespace.
3394 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3396 TK/06 MBOX spool code: Add real "From " MBOX separator line
3397 so the .eml file is really in mbox format (even though
3398 most programs do not really care). Patch from Alex Miller.
3400 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3401 The latter is generated from $received_to and is only set if the
3402 message has one envelope recipient. SA can use these headers,
3403 obviously out-of-the-box. Patch from Alex Miller.
3405 PH/08 The ${def test on a variable was returning false if the variable's
3406 value was "0", contrary to what the specification has always said!
3407 The result should be true unless the variable is empty.
3409 PH/09 The syntax error of a character other than { following "${if
3410 def:variable_name" (after optional whitespace) was not being diagnosed.
3411 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3412 accidental colon was present, for example, could give incorrect results.
3414 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3415 result is used (not including appendfile, where other changes are about
3418 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3419 This involved changing a lot of size variables from int to off_t. It
3420 should work with maildirs and everything.
3422 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3423 spamd dying while we are connected to it.
3425 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3428 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3429 function for BDB 4.3.
3431 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3433 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3434 the sources for PCRE have been re-arranged and more files are now
3437 PH/15 The code I had for printing potentially long long variables in PH/11
3438 above was not the best (it lost precision). The length of off_t variables
3439 is now inspected at build time, and an appropriate printing format (%ld
3440 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3441 to be "long long int" or "long int". This is needed for the internal
3442 formatting function string_vformat().
3444 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3445 the configuration file to be ":syslog", then the script "guesses" where
3446 the logs files are, rather than using the compiled in default. In our
3447 case the guess is not the same as the compiled default, so the script
3448 suddenly stopped working when I started to use syslog. The patch checks
3449 to see if log_file_path is "". If so, it attempts to read it from exim
3450 with no configuration file to get the compiled in version, before it
3451 falls back to the previous guessing code."
3453 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3454 implementing BATV in an Exim configuration. See NewStuff for the gory
3457 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3458 Makefile that are specific to HP-UX.
3460 PH/18 If the "use_postmaster" option was set for a recipient callout together
3461 with the "random" option, the postmaster address was used as the MAIL
3462 FROM address for the random test, but not for the subsequent recipient
3463 test. It is now used for both.
3465 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3466 patch removes a few documentation additions to RFC 3028, because the
3467 latest draft now contains them. It adds the new en;ascii-case comparator
3468 and a new error check for 8bit text in MIME parts. Comparator and
3469 require names are now matched exactly. I enabled the subaddress
3470 extension, but it is not well tested yet (read: it works for me)."
3472 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3473 rework some of the code of TK/09 above to avoid the hardwired use of
3474 "%lld" and "long long". Replaced the call to snprintf() with a call to
3477 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3478 records point to non-existent hosts", "retry timeout exceeded", and
3479 "retry time not reached for any host after a long failure period".
3481 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3482 experimental DomainKeys support:
3484 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3485 (2) On an error such as an illegally used "control", the wrong name for
3486 the control was given.
3488 These problems did NOT occur unless DomainKeys support was compiled.
3490 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3492 PH/24 Added ${if match_ip condition.
3494 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3495 databases so that it will be absolutely obvious if a crash occurs in the
3496 DB library. This is a regular occurrence (often caused by mis-matched
3499 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3500 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3501 picked up on a user's system that detects such things. There doesn't seem
3502 to be a gcc warning option for this - only an attribute that has to be
3503 put on the function's prototype. It seems that in Fedora Core 4 they have
3504 set this on a number of new functions. No doubt there will be more in due
3507 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3508 specify a transport. However, if an address that was verified by such a
3509 router was the subject of a callout, Exim crashed because it tried to
3510 read the rcpt_include_affixes from the non-existent transport. Now it
3511 just assumes that the setting of that option is false. This bug was
3512 introduced by 4.51/PH/31.
3514 PH/28 Changed -d+all to exclude +memory, because that information is very
3515 rarely of interest, but it makes the output a lot bigger. People tend to
3516 do -d+all out of habit.
3518 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3519 code in os-type was giving problems when libc.so lives in lib64, like on
3522 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3523 aren't the modern standard, and it seems that some systems' include files
3524 don't always have them. Exim was already checking for some of the newer
3525 ones like T_AAAA, and defining it itself. I've added checks for all the
3526 record types that Exim uses.
3528 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3529 not automatically generating a new one, as it is supposed to. This
3530 prevented TLS from working. If the file did exist, but contained invalid
3531 data, a new version was generated, as expected. It was only the case of a
3532 non-existent file that was broken.
3534 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3535 with a change in libdomainkeys > 0.64.
3537 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3538 from DNS. If the selector record carries the flag, it now has
3539 precedence over the domain-wide flag.
3541 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3543 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3544 the use of a port name that isn't defined in /etc/services) occurred, the
3545 message was deferred as in a normal delivery, and thus remained on the
3546 spool, instead of being failed because of the mua_wrapper setting. This
3547 is now fixed, and I tidied up some of the mua_wrapper messages at the
3550 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3551 the messages in a hash of arrays rather than using individual hashes.
3552 This is a bit cleaner and results in dramatic memory savings, albeit
3553 at a slight CPU cost.
3555 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3556 as requested by Marc Sherman.
3558 SC/10 Eximstats: added histograms for user specified patterns as requested
3561 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3563 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3564 fopen() in the content-scanning modules that did not already have it.
3570 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3571 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3573 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3575 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3577 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3578 to test Sieve filters that use "vacation".
3580 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3581 that changes the way the GnuTLS parameters are stored in the cache file.
3582 The new format can be generated externally. For backward compatibility,
3583 if the data in the cache doesn't make sense, Exim assumes it has read an
3584 old-format file, and it generates new data and writes a new file. This
3585 means that you can't go back to an older release without removing the
3588 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3589 work if there are any delivery delays because "one_time" forces the
3590 parent to be marked "delivered", so its unseen clone is never tried
3591 again. For this reason, Exim now forbids the simultaneous setting of
3594 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3595 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3596 entry if you want to know the details. The fix, however, neglected to
3597 consider the case when local delivery batching is involved. The test for
3598 "previously delivered" was not happening when checking to see if an
3599 address could be batched with a previous (undelivered) one; under
3600 certain circumstances this could lead to multiple deliveries to the same
3603 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3604 in its include files, and this causes problems building Exim.
3606 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3607 header_syntax) but Exim was just ignoring anything given after a slash.
3608 In particular, this caused confusion with an attempt to use "verify =
3609 reverse_host_lookup/defer_ok". An error is now given when options are
3610 supplied for verify items that do not have them. (Maybe reverse_host_
3611 lookup should have a defer_ok option, but that's a different point.)
3613 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3614 defined by RFC 821) to 2048, because there were problems with some AUTH
3615 commands, and RFC 1869 says the size should be increased for extended
3616 SMTP commands that take arguments.
3618 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3621 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3622 "unknown" error; now it says that the functionality isn't in the binary.
3624 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3625 an address' error message when a string expansion fails (syntax or
3626 whatever). Otherwise the password may appear in the log. Following change
3627 PH/42 below, there is no longer a chance of it appearing in a bounce
3630 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3632 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3633 of its addresses was ever tried. (Bugzilla bug #2.)
3635 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3636 the result incorrectly in the debug output. (It correctly added a newline
3637 to what was transported.)
3639 TF/01 Added $received_time.
3641 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3642 commented out examples of how to interface to a virus scanner and to
3643 SpamAssassin. Also added commented examples of av_scanner and
3644 spamd_address settings.
3646 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3647 and controls are allowed in which ACLs. There were a couple of minor
3648 errors. Some of the entries in the conditions table (which is a table of
3649 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3650 negation of where the condition IS allowed.
3652 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3654 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3655 header file does not have a version number, so I've had to invent a new
3656 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3657 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3658 radiusclient), but it was contributed by a Radius user.
3660 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3661 files or directories, for OpenSSL.
3663 PH/19 When an Exim process that is running as root has to create an Exim log
3664 file, it does so in a subprocess that runs as exim:exim so as to get the
3665 ownership right at creation (otherwise, other Exim processes might see
3666 the file with the wrong ownership). There was no test for failure of this
3667 fork() call, which would lead to the process getting stuck as it waited
3668 for a non-existent subprocess. Forks do occasionally fail when resources
3669 run out. I reviewed all the other calls to fork(); they all seem to check
3672 PH/20 When checking for unexpected SMTP input at connect time (before writing
3673 the banner), Exim was not dealing correctly with a non-positive return
3674 from the read() function. If the client had disconnected by this time,
3675 the result was a log entry for a synchronization error with an empty
3676 string after "input=" when read() returned zero. If read() returned -1
3677 (an event I could not check), uninitialized data bytes were printed.
3678 There were reports of junk text (parts of files, etc) appearing after
3681 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3683 PH/22 Added support for macro redefinition, and (re)definition in between
3684 driver and ACL definitions.
3686 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3687 forgetting to use the resulting value; it was using the unexpanded value.
3689 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3690 hadn't been configured. The fix is from Juergen Kreileder, who
3691 understands it better than I do:
3693 "Here's what I see happening with three configured cyrus_sasl
3694 authenticators configured (plain, login, cram-md5):
3696 On startup auth_cyrus_sasl_init() gets called for each of these.
3697 This means three calls to sasl_listmech() without a specified mech_list.
3698 => SASL tests which mechs of all available mechs actually work
3699 => three warnings about OTP not working
3700 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3702 With the patch, sasl_listmech() also gets called three times. But now
3703 SASL's mech_list option is set to the server_mech specified in the the
3704 authenticator. Or in other words, the answer from sasl_listmech()
3705 gets limited to just the mech you're testing for (which is different
3707 => the return list contains just 'plain' or 'login', 'cram-md5' or
3708 nothing depending on the value of ob->server_mech.
3710 I've just tested the patch: Authentication still works fine,
3711 unavailable mechs specified in the exim configuration are still
3712 caught, and the auth.log warnings about OTP are gone."
3714 PH/25 When debugging is enabled, the contents of the command line are added
3715 to the debugging output, even when log_selector=+arguments is not
3718 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3719 answer is "GNU", and only if the return is "GNU/something" is the answer
3722 PH/27 $acl_verify_message is now set immediately after the failure of a
3723 verification in an ACL, and so is available in subsequent modifiers. In
3724 particular, the message can be preserved by coding like this:
3726 warn !verify = sender
3727 set acl_m0 = $acl_verify_message
3729 Previously, $acl_verify_message was set only while expanding "message"
3730 and "log_message" when a very denied access.
3732 PH/28 Modified OS/os.c-Linux with
3734 -#ifndef OS_LOAD_AVERAGE
3735 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3737 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3738 nomenclature these days.)
3740 PH/29 Installed patch from the Sieve maintainer that adds the options
3741 sieve_useraddress and sieve_subaddress to the redirect router.
3743 PH/30 In these circumstances:
3744 . Two addresses routed to the same list of hosts;
3745 . First host does not offer TLS;
3746 . First host accepts first address;
3747 . First host gives temporary error to second address;
3748 . Second host offers TLS and a TLS session is established;
3749 . Second host accepts second address.
3750 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3751 and peerdn, if requested) that were in fact used only for the second
3754 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3755 attention to any local part prefix or suffix that was matched by the
3756 router that accepted the address. It now behaves in the same way as it
3757 does for delivery: the affixes are removed from the local part unless
3758 rcpt_include_affixes is set on the transport.
3760 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3761 timeout during the DATA phase of an incoming message.
3763 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3764 applied a patch sanctioned by the Sieve maintainer.
3766 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3767 the uid or gid is negative. A case of a negative gid caused this to be
3768 noticed. The fix allows for either to be negative.
3770 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3771 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3774 PH/36 The change PH/12 above was broken. Fixed it.
3776 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3777 the grounds that routing the same address twice would always produce the
3778 same answer. This might have been true once, but it is certainly no
3779 longer true now. Routing a child address may depend on the previous
3780 routing that produced that child. Some complicated redirection strategies
3781 went wrong when messages had multiple recipients, and made Exim's
3782 behaviour dependent on the order in which the addresses were given.
3784 I have moved the duplicate checking until after the routing is complete.
3785 Exim scans the addresses that are assigned to local and remote
3786 transports, and removes any duplicates. This means that more work will be
3787 done, as duplicates will always all be routed, but duplicates are
3788 presumably rare, so I don't expect this is of any significance.
3790 For deliveries to pipes, files, and autoreplies, the duplicate checking
3791 still happens during the routing process, since they are not going to be
3794 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3795 It corrects a timeout issue with spamd. This is Ian's comment: "The
3796 background is that sometimes spamd either never reads data from a
3797 connection it has accepted, or it never writes response data. The exiscan
3798 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3799 blindly assumes that writes won't block so it may never time out."
3801 PH/39 Allow G after quota size as well as K and M.
3803 PH/40 The value set for $authenticated_id in an authenticator may not contain
3804 binary zeroes or newlines because the value is written to log lines and
3805 to spool files. There was no check on this. Now the value is run through
3806 the string_printing() function so that such characters are converted to
3807 printable escape sequences.
3809 PH/41 $message_linecount is a new variable that contains the total number of
3810 lines in the message. Compare $body_linecount, which is the count for the
3813 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3814 bounce and delay warning messages, except in certain special cases, which
3817 (a) An SMTP error message from a remote host;
3818 (b) A message specified in a :fail: redirection;
3819 (c) A message specified in a "fail" command in a system filter;
3820 (d) A message specified in a FAIL return from the queryprogram router;
3821 (e) A message specified by the cannot_route_message router option.
3823 In these cases only, Exim does include the error details in bounce and
3824 warning messages. There are also a few cases where bland messages such
3825 as "unrouteable address" or "local delivery error" are given.
3827 PH/43 $value is now also set for the "else" part of a ${run expansion.
3829 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3830 being worked on, but at least Exim now implements the latest version to
3833 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3834 process to complete was treated as a delivery failure, a timeout while
3835 writing the message to the pipe was logged, but erroneously treated as a
3836 successful delivery. Such timeouts include transport filter timeouts. For
3837 consistency with the overall process timeout, these timeouts are now
3838 treated as errors, giving rise to delivery failures by default. However,
3839 there is now a new Boolean option for the pipe transport called
3840 timeout_defer, which, if set TRUE, converts the failures into defers for
3841 both kinds of timeout. A transport filter timeout is now identified in
3844 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3845 systems where "make" and "gmake" are different, calling "gmake" at top
3846 level broke things. I've arranged for the value of $(MAKE) to be passed
3847 from the Makefile to this script so that it can call the same version of
3851 A note about Exim versions 4.44 and 4.50
3852 ----------------------------------------
3854 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3855 changes of various kinds. As a consequence, a big documentation update was
3856 needed. This delayed the release for rather longer than seemed good, especially
3857 in the light of a couple of (minor) security issues. Therefore, the changes
3858 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3859 release. So 4.44 and 4.50 are in effect two different branches that both start
3862 I have left the 4.50 change log unchanged; it contains all the changes since
3863 4.43. The change log for 4.44 is below; many of its items are identical to
3864 those for 4.50. This seems to be the most sensible way to preserve the
3865 historical information.
3871 1. Minor wording change to the doc/README.SIEVE file.
3873 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3874 computation of the current number of files was incorrect.
3876 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3877 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3880 4. Give more explanation in the error message when the command for a transport
3881 filter fails to execute.
3883 5. There are several places where Exim runs a non-Exim command in a
3884 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3885 was being done only for the command run by the queryprogram router. It is
3886 now done for all such subprocesses. The other cases are: ${run, transport
3887 filters, and the commands run by the lmtp and pipe transports.
3889 6. Added CONFIGURE_GROUP build-time option.
3891 7. Some older OS have a limit of 256 on the maximum number of file
3892 descriptors. Exim was using setrlimit() to set 1000 as a large value
3893 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3894 systems. I've change it so that if it can't get 1000, it tries for 256.
3896 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3897 was an oversight, and furthermore, ever since the addition of extra
3898 controls (e.g. 4.43/32), the checks on when to allow different forms of
3899 "control" were broken. There should now be diagnostics for all cases when a
3900 control that does not make sense is encountered.
3902 9. Added the /retain_sender option to "control=submission".
3904 10. $recipients is now available in the predata ACL (oversight).
3906 11. Tidy the search cache before the fork to do a delivery from a message
3907 received from the command line. Otherwise the child will trigger a lookup
3908 failure and thereby defer the delivery if it tries to use (for example) a
3909 cached ldap connection that the parent has called unbind on.
3911 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3912 of $address_data from the recipient verification was clobbered by the
3913 sender verification.
3915 13. The value of address_data from a sender verification is now available in
3916 $sender_address_data in subsequent conditions in the ACL statement.
3918 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3920 15. Added a new option "connect=<time>" to callout options, to set a different
3923 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3924 was its contents. (It was OK if the option was not defined at all.)
3926 17. A "Completed" log line is now written for messages that are removed from
3927 the spool by the -Mrm option.
3929 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3930 information about exactly what failed.
3932 19. Added -dd to debug only the daemon process.
3934 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3935 handles timeouts, both on the server side and network timeouts. Renamed the
3936 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3938 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3939 It is now set to "smtps".
3941 22. $host_address is now set to the target address during the checking of
3942 ignore_target_hosts.
3944 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3945 being passed; this would have caused $sender_host_name to have been used if
3946 matching the list had actually called for a host name (not very likely,
3947 since this list is usually IP addresses). A host name is now passed as
3950 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3951 code that specifies a non-restarting handler (typically sigaction() in
3952 modern systems) in an attempt to fix a rare and obscure crash bug.
3954 25. Narrowed the window for a race in the daemon that could cause it to ignore
3955 SIGCHLD signals. This is not a major problem, because they are used only to
3956 wake it up if nothing else does.
3958 26. A malformed maildirsize file could cause Exim to calculate negative values
3959 for the mailbox size or file count. Odd effects could occur as a result.
3960 The maildirsize information is now recalculated if the size or filecount
3963 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3964 support for a long time. Removed HAVE_SYS_VFS_H.
3966 28. Installed the latest version of exipick from John Jetmore.
3968 29. In an address list, if the pattern was not a regular expression, an empty
3969 subject address (from a bounce message) matched only if the pattern was an
3970 empty string. Non-empty patterns were not even tested. This was the wrong
3971 because it is perfectly reasonable to use an empty address as part of a
3972 database query. An empty address is now tested by patterns that are
3973 lookups. However, all the other forms of pattern expect the subject to
3974 contain a local part and a domain, and therefore, for them, an empty
3975 address still always fails if the pattern is not itself empty.
3977 30. Exim went into a mad DNS loop when attempting to do a callout where the
3978 host was specified on an smtp transport, and looking it up yielded more
3979 than one IP address.
3981 31. Re-factored the code for checking spool and log partition space into a
3982 function that finds that data and another that does the check. The former
3983 is then used to implement four new variables: $spool_space, $log_space,
3984 $spool_inodes, and $log_inodes.
3986 32. The RFC2047 encoding function was originally intended for short strings
3987 such as real names; it was not keeping to the 75-character limit for
3988 encoded words that the RFC imposes. It now respects the limit, and
3989 generates multiple encoded words if necessary. To be on the safe side, I
3990 have increased the buffer size for the ${rfc2047: expansion operator from
3993 33. It is now permitted to omit both strings after an "if" condition; if the
3994 condition is true, the result is "true". As before, when the second string
3995 is omitted, a false condition yields an empty string. This makes it less
3996 cumbersome to write custom ACL and router conditions.
3998 34. Failure to deliver a bounce message always caused it to be frozen, even if
3999 there was an errors_to setting on the router. The errors_to setting is now
4002 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4003 canonical form (fully expanded) before being placed in
4004 $sender_host_address.
4006 36. The table in the code that translates DNS record types into text (T_A to
4007 "A" for instance) was missing entries for NS and CNAME. It is just possible
4008 that this could have caused confusion if both these types were looked up
4009 for the same domain, because the text type is used as part of Exim's
4010 per-process caching. But the chance of anyone hitting this buglet seems
4013 37. The dnsdb lookup has been extended in a number of ways.
4015 (1) There is a new type, "zns", which walks up the domain tree until it
4016 finds some nameserver records. It should be used with care.
4018 (2) There is a new type, "mxh", which is like "mx" except that it returns
4019 just the host names, not the priorities.
4021 (3) It is now possible to give a list of domains (or IP addresses) to be
4022 looked up. The behaviour when one of the lookups defers can be
4023 controlled by a keyword.
4025 (4) It is now possible to specify the separator character for use when
4026 multiple records are returned.
4028 38. The dnslists ACL condition has been extended: it is now possible to supply
4029 a list of IP addresses and/or domains to be looked up in a particular DNS
4032 39. Added log_selector=+queue_time_overall.
4034 40. When running the queue in the test harness, wait just a tad after forking a
4035 delivery process, to get repeatability of debugging output.
4037 41. Include certificate and key file names in error message when GnuTLS fails
4038 to set them up, because the GnuTLS error message doesn't include the name
4039 of the failing file when there is a problem reading it.
4041 42. Allow both -bf and -bF in the same test run.
4043 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4045 44. The "Exiscan patch" is now merged into the mainline Exim source.
4047 45. Sometimes the final signoff response after QUIT could fail to get
4048 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4049 < 0 before doing a fflush(). This bug looks as though it goes back to the
4050 introduction of TLS in release 3.20, but "sometimes" must have been rare
4051 because the tests only now provoked it.
4053 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4054 (this can affect the format of dates).
4056 47. exim_tidydb, when checking for the continued existence of a message for
4057 which it has found a message-specific retry record, was not finding
4058 messages that were in split spool directories. Consequently, it was
4059 deleting retry records that should have stayed in existence.
4061 48. Steve fixed some bugs in eximstats.
4063 49. The SPA authentication driver was not abandoning authentication and moving
4064 on to the next authenticator when an expansion was forced to fail,
4065 contradicting the general specification for all authenticators. Instead it
4066 was generating a temporary error. It now behaves as specified.
4068 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4069 (the order specifies the preference for clients). The order is now AES256,
4070 AES128, 3DES, ARCFOUR128.
4072 51. Small patch to Sieve code - explicitly set From: when generating an
4075 52. Exim crashed if a remote delivery caused a very long error message to be
4076 recorded - for instance if somebody sent an entire SpamAssassin report back
4077 as a large number of 550 error lines. This bug was coincidentally fixed by
4078 increasing the size of one of Exim's internal buffers (big_buffer) that
4079 happened as part of the Exiscan merge. However, to be on the safe side, I
4080 have made the code more robust (and fixed the comments that describe what
4083 53. Now that there can be additional text after "Completed" in log lines (if
4084 the queue_time_overall log selector is set), a one-byte patch to exigrep
4085 was needed to allow it to recognize "Completed" as not the last thing in
4088 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4089 patch that reportedly fixes this has been added. I am not expert enough to
4090 create a test for it. This is what the patch creator wrote:
4092 "I found a little strange behaviour of ldap code when working with
4093 Windows 2003 AD Domain, where users was placed in more than one
4094 Organization Units. When I tried to give exim partial DN, the exit code
4095 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4096 But simultaneously result of request was absolutely normal ldap result,
4097 so I produce this patch..."
4099 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4100 REFERENCE, so I have modified the code to exclude the patch when that macro
4103 55. Some experimental protocols are using DNS PTR records for new purposes. The
4104 keys for these records are domain names, not reversed IP addresses. The
4105 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4106 leaves it alone. Component reversal etc. now happens only for IP addresses.
4109 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4111 57. Double the size of the debug message buffer (to 2048) so that more of very
4112 long debug lines gets shown.
4114 58. The exicyclog utility now does better if the number of log files to keep
4115 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4117 59. Two changes related to the smtp_active_hostname option:
4119 (1) $smtp_active_hostname is now available as a variable.
4120 (2) The default for smtp_banner uses $smtp_active_hostname instead
4121 of $primary_hostname.
4123 60. The host_aton() function is supposed to be passed a string that is known
4124 to be a valid IP address. However, in the case of IPv6 addresses, it was
4125 not checking this. This is a hostage to fortune. Exim now panics and dies
4126 if the condition is not met. A case was found where this could be provoked
4127 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4128 components; fortuitously, this particular loophole had already been fixed
4129 by change 4.50/55 above.
4131 If there are any other similar loopholes, the new check in host_aton()
4132 itself should stop them being exploited. The report I received stated that
4133 data on the command line could provoke the exploit when Exim was running as
4134 exim, but did not say which command line option was involved. All I could
4135 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4136 running as the user.
4139 61. There was a buffer overflow vulnerability in the SPA authentication code
4140 (which came originally from the Samba project). I have added a test to the
4141 spa_base64_to_bits() function which I hope fixes it.
4144 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4145 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4147 63. The daemon start-up calls getloadavg() while still root for those OS that
4148 need the first call to be done as root, but it missed one case: when
4149 deliver_queue_load_max is set with deliver_drop_privilege. This is
4150 necessary for the benefit of the queue runner, because there is no re-exec
4151 when deliver_drop_privilege is set.
4153 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4154 This has been fixed.
4156 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4157 was in use, was not putting the data itself into the right store pool;
4158 consequently, it could be overwritten for a subsequent message in the same
4159 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4162 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4164 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4165 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4166 the function were treating the return as a boolean value, which happened to
4167 work because 0=false and not-0=true, but is not correct code.
4169 68. The host_aton() function was not handling scoped IPv6 addresses (those
4170 with, for example, "%eth0" on the end) correctly.
4172 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4173 negated items (that is, ~something) in unsigned ints. Some compilers
4174 apparently mutter when there is no cast.
4176 70. If an address verification called from an ACL failed, and did not produce a
4177 user-specific message (i.e. there was only a "system" message), nothing was
4178 put in $acl_verify_message. In this situation, it now puts the system
4181 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4182 session; change 4.31/43 added the unwanted input to the log line - except
4183 that it did not do this in the start of session case. It now does.
4185 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4186 This is wrong and can cause the other end to generate a synchronization
4187 error if it is another Exim or anything else that does the synchronization
4188 check. A QUIT command is no longer sent after a timeout.
4190 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4191 during host lookups.
4193 74. The defer_ok option of callout verification was not working if it was used
4194 when verifying addresses in header lines, that is, for this case:
4196 verify = header_sender/callout=defer_ok
4198 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4199 those file descriptors could be used for SMTP connections. If anything
4200 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4201 could be sent to the SMTP client, causing chaos. The daemon now opens
4202 stdin, stdout, and stderr to /dev/null when it puts itself into the
4205 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4206 log by default. The user can override this with suitable Perl magic.
4208 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4209 the log message when discard triggers, was not working for the DATA ACL or
4210 for the non-SMTP ACL.
4212 78. Error message wording change in sieve.c.
4214 79. If smtp_accept_max_per_host was set, the number of connections could be
4215 restricted to fewer than expected, because the daemon was trying to set up
4216 a new connection before checking whether the processes handling previous
4217 connections had finished. The check for completed processes is now done
4218 earlier. On busy systems, this bug wouldn't be noticed because something
4219 else would have woken the daemon, and it would have reaped the completed
4222 80. If a message was submitted locally by a user whose login name contained one
4223 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4224 It caused a spool format error. I have fixed the spool reading code. A
4225 related problem was that the "from" clause in the Received: line became
4226 illegal because of the space(s). It is now covered by ${quote_local_part.
4228 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4231 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4233 83. Patch from Sieve maintainer to fix unterminated string problem in
4234 "vacation" handling.
4236 84. Some minor changes to the Linux configuration files to help with other
4237 OS variants using glibc.
4239 85. One more patch for Sieve to update vacation handling to latest spec.
4242 ----------------------------------------------------
4243 See the note above about the 4.44 and 4.50 releases.
4244 ----------------------------------------------------
4250 1. Change 4.43/35 introduced a bug that caused file counts to be
4251 incorrectly computed when quota_filecount was set in an appendfile
4254 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4255 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4258 3. Give more explanation in the error message when the command for a transport
4259 filter fails to execute.
4261 4. There are several places where Exim runs a non-Exim command in a
4262 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4263 was being done only for the command run by the queryprogram router. It is
4264 now done for all such subprocesses. The other cases are: ${run, transport
4265 filters, and the commands run by the lmtp and pipe transports.
4267 5. Some older OS have a limit of 256 on the maximum number of file
4268 descriptors. Exim was using setrlimit() to set 1000 as a large value
4269 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4270 systems. I've change it so that if it can't get 1000, it tries for 256.
4272 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4273 was an oversight, and furthermore, ever since the addition of extra
4274 controls (e.g. 4.43/32), the checks on when to allow different forms of
4275 "control" were broken. There should now be diagnostics for all cases when a
4276 control that does not make sense is encountered.
4278 7. $recipients is now available in the predata ACL (oversight).
4280 8. Tidy the search cache before the fork to do a delivery from a message
4281 received from the command line. Otherwise the child will trigger a lookup
4282 failure and thereby defer the delivery if it tries to use (for example) a
4283 cached ldap connection that the parent has called unbind on.
4285 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4286 of $address_data from the recipient verification was clobbered by the
4287 sender verification.
4289 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4290 was its contents. (It was OK if the option was not defined at all.)
4292 11. A "Completed" log line is now written for messages that are removed from
4293 the spool by the -Mrm option.
4295 12. $host_address is now set to the target address during the checking of
4296 ignore_target_hosts.
4298 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4299 being passed; this would have caused $sender_host_name to have been used if
4300 matching the list had actually called for a host name (not very likely,
4301 since this list is usually IP addresses). A host name is now passed as
4304 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4305 code that specifies a non-restarting handler (typically sigaction() in
4306 modern systems) in an attempt to fix a rare and obscure crash bug.
4308 15. Narrowed the window for a race in the daemon that could cause it to ignore
4309 SIGCHLD signals. This is not a major problem, because they are used only to
4310 wake it up if nothing else does.
4312 16. A malformed maildirsize file could cause Exim to calculate negative values
4313 for the mailbox size or file count. Odd effects could occur as a result.
4314 The maildirsize information is now recalculated if the size or filecount
4317 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4318 support for a long time. Removed HAVE_SYS_VFS_H.
4320 18. Updated exipick to current release from John Jetmore.
4322 19. Allow an empty sender to be matched against a lookup in an address list.
4323 Previously the only cases considered were a regular expression, or an
4326 20. Exim went into a mad DNS lookup loop when doing a callout where the
4327 host was specified on the transport, if the DNS lookup yielded more than
4330 21. The RFC2047 encoding function was originally intended for short strings
4331 such as real names; it was not keeping to the 75-character limit for
4332 encoded words that the RFC imposes. It now respects the limit, and
4333 generates multiple encoded words if necessary. To be on the safe side, I
4334 have increased the buffer size for the ${rfc2047: expansion operator from
4337 22. Failure to deliver a bounce message always caused it to be frozen, even if
4338 there was an errors_to setting on the router. The errors_to setting is now
4341 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4342 canonical form (fully expanded) before being placed in
4343 $sender_host_address.
4345 24. Updated eximstats to version 1.33
4347 25. Include certificate and key file names in error message when GnuTLS fails
4348 to set them up, because the GnuTLS error message doesn't include the name
4349 of the failing file when there is a problem reading it.
4351 26. Expand error message when OpenSSL has problems setting up cert/key files.
4354 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4355 (this can affect the format of dates).
4357 28. exim_tidydb, when checking for the continued existence of a message for
4358 which it has found a message-specific retry record, was not finding
4359 messages that were in split spool directories. Consequently, it was
4360 deleting retry records that should have stayed in existence.
4362 29. eximstats updated to version 1.35
4363 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4364 1.35 - bugfix such that pie charts by volume are generated correctly
4366 30. The SPA authentication driver was not abandoning authentication and moving
4367 on to the next authenticator when an expansion was forced to fail,
4368 contradicting the general specification for all authenticators. Instead it
4369 was generating a temporary error. It now behaves as specified.
4371 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4372 (the order specifies the preference for clients). The order is now AES256,
4373 AES128, 3DES, ARCFOUR128.
4375 31. Small patch to Sieve code - explicitly set From: when generating an
4378 32. Exim crashed if a remote delivery caused a very long error message to be
4379 recorded - for instance if somebody sent an entire SpamAssassin report back
4380 as a large number of 550 error lines. This bug was coincidentally fixed by
4381 increasing the size of one of Exim's internal buffers (big_buffer) that
4382 happened as part of the Exiscan merge. However, to be on the safe side, I
4383 have made the code more robust (and fixed the comments that describe what
4386 33. Some experimental protocols are using DNS PTR records for new purposes. The
4387 keys for these records are domain names, not reversed IP addresses. The
4388 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4389 leaves it alone. Component reversal etc. now happens only for IP addresses.
4392 34. The host_aton() function is supposed to be passed a string that is known
4393 to be a valid IP address. However, in the case of IPv6 addresses, it was
4394 not checking this. This is a hostage to fortune. Exim now panics and dies
4395 if the condition is not met. A case was found where this could be provoked
4396 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4397 components; fortuitously, this particular loophole had already been fixed
4398 by change 4.50/55 or 4.44/33 above.
4400 If there are any other similar loopholes, the new check in host_aton()
4401 itself should stop them being exploited. The report I received stated that
4402 data on the command line could provoke the exploit when Exim was running as
4403 exim, but did not say which command line option was involved. All I could
4404 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4405 running as the user.
4408 35. There was a buffer overflow vulnerability in the SPA authentication code
4409 (which came originally from the Samba project). I have added a test to the
4410 spa_base64_to_bits() function which I hope fixes it.
4413 36. The daemon start-up calls getloadavg() while still root for those OS that
4414 need the first call to be done as root, but it missed one case: when
4415 deliver_queue_load_max is set with deliver_drop_privilege. This is
4416 necessary for the benefit of the queue runner, because there is no re-exec
4417 when deliver_drop_privilege is set.
4419 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4420 was in use, was not putting the data itself into the right store pool;
4421 consequently, it could be overwritten for a subsequent message in the same
4422 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4425 38. Sometimes the final signoff response after QUIT could fail to get
4426 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4427 < 0 before doing a fflush(). This bug looks as though it goes back to the
4428 introduction of TLS in release 3.20, but "sometimes" must have been rare
4429 because the tests only now provoked it.
4435 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4436 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4437 Now it returns an integer. A number of calls were still expecting a T/F
4438 return. Fortuitously, in all cases, the tests worked in OK situations,
4439 which is the norm. However, things would have gone wrong on any write
4440 failures on the smtp file descriptor. This function is used when sending
4441 messages over SMTP and also when doing verify callouts.
4443 2. When Exim is called to do synchronous delivery of a locally submitted
4444 message (the -odf or -odi options), it no longer closes stderr before doing
4447 3. Implemented the mua_wrapper option.
4449 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4451 5. Implemented the functions header_remove(), header_testname(),
4452 header_add_at_position(), and receive_remove_recipient(), and exported them
4455 6. If an ACL "warn" statement specified the addition of headers, Exim already
4456 inserted X-ACL-Warn: at the start if there was no header name. However, it
4457 was not making this test for the second and subsequent header lines if
4458 there were newlines in the string. This meant that an invalid header could
4459 be inserted if Exim was badly configured.
4461 7. Allow an ACL "warn" statement to add header lines at the start or after all
4462 the Received: headers, as well as at the end.
4464 8. Added the rcpt_4xx retry error code.
4466 9. Added postmaster_mailfrom=xxx to callout verification option.
4468 10. Added mailfrom=xxxx to the callout verification option, for verify=
4471 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4472 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4474 12. Inserted some casts to stop certain compilers warning when using pointer
4475 differences as field lengths or precisions in printf-type calls (mostly
4476 affecting debugging statements).
4478 13. Added optional readline() support for -be (dynamically loaded).
4480 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4481 same clock tick as a message's arrival, so that its received time was the
4482 same as the "first fail" time on the retry record, and that message
4483 remained on the queue past the ultimate address timeout, every queue runner
4484 would try a delivery (because it was past the ultimate address timeout) but
4485 after another failure, the ultimate address timeout, which should have then
4486 bounced the address, did not kick in. This was a "< instead of <=" error;
4487 in most cases the first failure would have been in the next clock tick
4488 after the received time, and all would be well.
4490 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4491 being recognized when the domain list was tested by the match_domain
4492 condition in an expansion string.
4494 16. Added the ${str2b64: operator.
4496 17. Exim was always calling setrlimit() to set a large limit for the number of
4497 processes, without checking whether the existing limit was already
4498 adequate. (It did check for the limit on file descriptors.) Furthermore,
4499 errors from getrlimit() and setrlimit() were being ignored. Now they are
4500 logged to the main and panic logs, but Exim does carry on, to try to do its
4501 job under whatever limits there are.
4503 18. Imported PCRE 5.0.
4505 19. Trivial typo in log message " temporarily refused connection" (the leading
4508 20. If the log selector return_path_on_delivery was set and an address was
4509 redirected to /dev/null, the delivery process crashed because it assumed
4510 that a return path would always be set for a "successful" delivery. In this
4511 case, the whole delivery is bypassed as an optimization, and therefore no
4514 21. Internal re-arrangement: the function for sending a challenge and reading
4515 a response while authentication was assuming a zero-terminated challenge
4516 string. It's now changed to take a pointer and a length, to allow for
4517 binary data in such strings.
4519 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4521 23. Exim was not respecting finduser_retries when seeking the login of the
4522 uid under which it was called; it was always trying 10 times. (The default
4523 setting of finduser_retries is zero.) Also, it was sleeping after the final
4524 failure, which is pointless.
4526 24. Implemented tls_on_connect_ports.
4528 25. Implemented acl_smtp_predata.
4530 26. If the domain in control=submission is set empty, Exim assumes that the
4531 authenticated id is a complete email address when it generates From: or
4532 Sender: header lines.
4534 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4535 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4536 chown and chgrp in /bin and hostname in /usr/bin.
4538 28. Exim was keeping the "process log" file open after each use, just as it
4539 does for the main log. This opens the possibility of it remaining open for
4540 long periods when the USR1 signal hits a daemon. Occasional processlog
4541 errors were reported, that could have been caused by this. Anyway, it seems
4542 much more sensible not to leave this file open at all, so that is what now
4545 29. The long-running daemon process does not normally write to the log once it
4546 has entered its main loop, and it closes the log before doing so. This is
4547 so that log files can straightforwardly be renamed and moved. However,
4548 there are a couple of unusual error situations where the daemon does write
4549 log entries, and I had neglected to close the log afterwards.
4551 30. The text of an SMTP error response that was received during a remote
4552 delivery was being truncated at 512 bytes. This is too short for some of
4553 the long messages that one sometimes sees. I've increased the limit to
4556 31. It is now possible to make retry rules that apply only when a message has a
4557 specific sender, in particular, an empty sender.
4559 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4560 it possible to be selective about when SMTP synchronization is enforced.
4562 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4564 32. Implemented hosts_connection_nolog.
4566 33. Added an ACL for QUIT.
4568 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4571 35. Added mailbox_size and mailbox_filecount to appendfile.
4573 36. Added control = no_multiline_responses to ACLs.
4575 37. There was a bug in the logic of the code that waits for the clock to tick
4576 in the case where the clock went backwards by a substantial amount such
4577 that the microsecond fraction of "now" was more than the microsecond
4578 fraction of "then" (but the whole seconds number was less).
4580 38. Added support for the libradius Radius client library this is found on
4581 FreeBSD (previously only the radiusclient library was supported).
4587 1. When certain lookups returned multiple values in the form name=value, the
4588 quoting of the values was not always being done properly. Specifically:
4589 (a) If the value started with a double quote, but contained no whitespace,
4591 (b) If the value contained whitespace other than a space character (i.e.
4592 tabs or newlines or carriage returns) it was not quoted.
4593 This fix has been applied to the mysql and pgsql lookups by writing a
4594 separate quoting function and calling it from the lookup code. The fix
4595 should probably also be applied to nisplus, ibase and oracle lookups, but
4596 since I cannot test any of those, I have not disturbed their existing code.
4598 2. A hit in the callout cache for a specific address caused a log line with no
4599 reason for rejecting RCPT. Now it says "Previous (cached) callout
4600 verification failure".
4602 3. There was an off-by-one bug in the queryprogram router. An over-long
4603 return line was truncated at 256 instead of 255 characters, thereby
4604 overflowing its buffer with the terminating zero. As well as fixing this, I
4605 have increased the buffer size to 1024 (and made a note to document this).
4607 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4608 when Exim is waiting for an SMTP response from a remote server, Exim
4609 restarts its select() call on the socket, thereby resetting its timeout.
4610 This is not a problem when such interrupts are rare. Somebody set up a cron
4611 job to run exiwhat every 2 minutes, which is less than the normal select()
4612 timeout (5 or 10 minutes). This meant that the select() timeout never
4613 kicked in because it was always reset. I have fixed this by comparing the
4614 time when an interrupt arrives with the time at the start of the first call
4615 to select(). If more time than the timeout has elapsed, the interrupt is
4616 treated as a timeout.
4618 5. Some internal re-factoring in preparation for the addition of Sieve
4619 extensions (by MH). In particular, the "personal" test is moved to a
4620 separate function, and given an option for scanning Cc: and Bcc: (which is
4621 not set for Exim filters).
4623 6. When Exim created an email address using the login of the caller as the
4624 local part (e.g. when creating a From: or Sender: header line), it was not
4625 quoting the local part when it contained special characters such as @.
4627 7. Installed new OpenBSD configuration files.
4629 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4630 try to make them clearer.
4632 9. Callout options, other than the timeout value, were being ignored when
4633 verifying sender addresses in header lines. For example, when using
4635 verify = header_sender/callout=no_cache
4637 the cache was (incorrectly) being used.
4639 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4640 only the Cygwin environment.
4642 11. When return_path_on_delivery was set as a log selector, if different remote
4643 addresses in the same message used different return paths and parallel
4644 remote delivery occurred, the wrong values would sometimes be logged.
4645 (Whenever a remote delivery process finished, the return path value from
4646 the most recently started remote delivery process was logged.)
4648 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4649 lines when AUTH and/or TLS are in use. This is the "received protocol"
4650 field. Exim used to use "asmtp" for authenticated SMTP, without any
4651 indication (in the protocol name) for TLS use. Now it follows the RFC and
4652 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4653 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4654 names appear in log lines as well as in Received: header lines.
4656 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4657 extensions, and comparison tests, and to fix some bugs.
4659 14. Changes to the "personal" filter test:
4661 (1) The test was buggy in that it was just doing the equivalent of
4662 "contains" tests on header lines. For example, if a user's address was
4663 anne@some.where, the "personal" test would incorrectly be true for
4665 To: susanne@some.where
4667 This test is now done by extracting each address from the header in turn,
4668 and checking the entire address. Other tests that are part of "personal"
4669 are now done using regular expressions (for example, to check local parts
4670 of addresses in From: header lines).
4672 (2) The list of non-personal local parts in From: addresses has been
4673 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4674 taken from the Sieve specification recommendations.
4676 (3) If the message contains any header line starting with "List-" it is
4677 treated as non-personal.
4679 (4) The test for "circular" in the Subject: header line has been removed
4680 because it now seems ill-conceived.
4682 15. Minor typos in src/EDITME comments corrected.
4684 16. Installed latest exipick from John Jetmore.
4686 17. If headers_add on a router specified a text string that was too long for
4687 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4688 of string_sprintf() is now avoided.
4690 18. $message_body_size was not set (it was always zero) when running the DATA
4691 ACL and the local_scan() function.
4693 19. For the "mail" command in an Exim filter, no default was being set for
4694 the once_repeat time, causing a random time value to be used if "once" was
4695 specified. (If the value happened to be <= 0, no repeat happened.) The
4696 default is now 0s, meaning "never repeat". The "vacation" command was OK
4697 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4698 (I found it when inspecting the code).
4700 20. There is now an overall timeout for performing a callout verification. It
4701 defaults to 4 times the callout timeout, which applies to individual SMTP
4702 commands during the callout. The overall timeout applies when there is more
4703 than one host that can be tried. The timeout is checked before trying the
4704 next host. This prevents very long delays if there are a large number of
4705 hosts and all are timing out (e.g. when the network connections are timing
4706 out). The value of the overall timeout can be changed by specifying an
4707 additional sub-option for "callout", called "maxwait". For example:
4709 verify = sender/callout=5s,maxwait=20s
4711 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4712 to the end before writing, but this should make it even safer).
4714 22. Exim was forgetting that it had advertised PIPELINING for the second and
4715 subsequent messages on an SMTP connection. It was also not resetting its
4716 memory on STARTTLS and an internal HELO.
4718 23. When Exim logs an SMTP synchronization error within a session, it now
4719 records whether PIPELINING has been advertised or not.
4721 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4722 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4723 rather than long int.
4725 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4727 26. Added the never_mail option to autoreply.
4733 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4734 crash if the getsockname() call failed; this can happen if a connection is
4735 closed very soon after it is established. The problem was simply in the
4736 order in which certain operations were done, causing Exim to try to write
4737 to the SMTP stream before it had set up the file descriptor. The bug has
4738 been fixed by making things happen in the correct order.
4744 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4745 before the connection was closed, thus losing the rejection response.
4747 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4748 some early Solaris releases, but causes trouble in current releases where
4749 socklen_t is defined.
4751 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4754 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4757 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4758 that releasing the top of it at the end releases what was used for sub-
4759 expansions (unless the block got too big). However, discard this block if
4760 the first thing is a variable or header, so that we can use its block when
4761 it is dynamic (useful for very large $message_headers, for example).
4763 6. Lookups now cache *every* query, not just the most recent. A new, separate
4764 store pool is used for this. It can be recovered when all lookup caches are
4765 flushed. Lookups now release memory at the end of their result strings.
4766 This has involved some general refactoring of the lookup sources.
4768 7. Some code has been added to the store_xxx() functions to reduce the amount
4769 of flapping under certain conditions.
4771 8. log_incoming_interface used to affect only the <= reception log lines. Now
4772 it causes the local interface and port to be added to several more SMTP log
4773 lines, for example "SMTP connection from", and rejection lines.
4775 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4777 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4779 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4780 could be overwritten at the end of the current message (or the start of a
4781 new message if it was set in a HELO ACL). The value is now preserved for
4782 the duration of the SMTP connection.
4784 12. If a transport had a headers_rewrite setting, and a matching header line
4785 contained an unqualified address, that address was qualified, even if it
4786 did not match any rewriting rules. The underlying bug was that the values
4787 of the flags that permit the existence of unqualified sender and recipient
4788 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4789 non-local messages, and by -bnq for local messages) were not being
4790 preserved with the message after it was received.
4792 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4793 "next input=" as part of the text comprising the host identity instead of
4794 the correct text. The code was using the same buffer for two different
4795 strings. However, depending on which order the printing function evaluated
4796 its arguments, the bug did not always show up. Under Linux, for example, my
4797 test suite worked just fine.
4799 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4800 doesn't work with some older versions of Perl. It has been changed to "my",
4801 which in any case is probably the better facility to use.
4803 15. A really picky compiler found some instances of statements for creating
4804 error messages that either had too many or two few arguments for the format
4807 16. The size of the buffer for calls to the DNS resolver has been increased
4808 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4809 for addresses that have a lot of PTR records. This alleviates a problem; it
4810 does not fully solve it.
4812 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4813 buffer now truncates the list and logs the incident, which is the same
4814 action as happens when Exim is looking up a host name and its aliases.
4815 Previously in this situation something unpredictable would happen;
4816 sometimes it was "internal error: store_reset failed".
4818 18. If a server dropped the connection unexpectedly when an Exim client was
4819 using GnuTLS and trying to read a response, the client delivery process
4820 crashed while trying to generate an error log message.
4822 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4823 string, for example:
4825 warn message = H1: something\nH2: something
4827 the text was added as a single header line from Exim's point of view
4828 though it ended up OK in the delivered message. However, searching for the
4829 second and subsequent header lines using $h_h2: did not work. This has been
4830 fixed. Similarly, if a system filter added multiple headers in this way,
4831 the routers could not see them.
4833 20. Expanded the error message when iplsearch is called with an invalid key to
4834 suggest using net-iplsearch in a host list.
4836 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4837 ACLs are no longer actually imposed (and a message to that effect is
4840 22. If a "gecos" field in a passwd entry contained escaped characters, in
4841 particular, if it contained a \" sequence, Exim got it wrong when building
4842 a From: or a Sender: header from that name. A second bug also caused
4843 incorrect handling when an unquoted " was present following a character
4844 that needed quoting.
4846 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4847 was not being matched caselessly.
4849 24. Arranged for all hyphens in the exim.8 source to be escaped with
4852 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4853 properly. Recipient callout cache records were still being keyed to include
4854 the sender, even when use_sender was set false. This led to far more
4855 callouts that were necessary. The sender is no longer included in the key
4856 when use_sender is false.
4858 26. Added "control = submission" modifier to ACLs.
4860 27. Added the ${base62d: operator to decode base 62 numbers.
4862 28. dnsdb lookups can now access SRV records.
4864 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4865 the configuration file.
4867 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4868 (-v) mode. This makes the output for a verbose queue run more intelligible.
4870 31. Added a use_postmaster feature to recipient callouts.
4872 32. Added the $body_zerocount variable, containing the number of binary zero
4873 bytes in the message body.
4875 33. The time of last modification of the "new" subdirectory is now used as the
4876 "mailbox time last read" when there is a quota error for a maildir
4879 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4881 35. Added +ignore_unknown as a special item in host lists.
4883 36. Code for decoding IPv6 addresses in host lists is now included, even if
4884 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4885 address was recognized as an IP address, but was then not correctly decoded
4886 into binary, causing unexpected and incorrect effects when compared with
4893 1. Very minor rewording of debugging text in manualroute to say "list of
4894 hosts" instead of "hostlist".
4896 2. If verify=header_syntax was set, and a header line with an unqualified
4897 address (no domain) and a large number of spaces between the end of the
4898 name and the colon was received, the reception process suffered a buffer
4899 overflow, and (when I tested it) crashed. This was caused by some obsolete
4900 code that should have been removed. The fix is to remove it!
4902 3. When running in the test harness, delay a bit after writing a bounce
4903 message to get a bit more predictability in the log output.
4905 4. Added a call to search_tidyup() just before forking a reception process. In
4906 theory, someone could use a lookup in the expansion of smtp_accept_max_
4907 per_host which, without the tidyup, could leave open a database connection.
4909 5. Added the variables $recipient_data and $sender_data which get set from a
4910 lookup success in an ACL "recipients" or "senders" condition, or a router
4911 "senders" option, similar to $domain_data and $local_part_data.
4913 6. Moved the writing of debug_print from before to after the "senders" test
4916 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4917 problems for message scanning, either using a data ACL, or using
4918 local_scan() because the Received: header was not generated till after they
4919 were called (in order to set the time as the time of reception completion).
4920 I have revised the way this works. The header is now generated after the
4921 body is received, but before the ACL or local_scan() are called. After they
4922 are run, the timestamp in the header is updated.
4928 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4929 before starting a queue runner without re-exec. This happened only when
4930 deliver_drop_privilege was set or when the Exim user was set to root. The
4931 effect of the bug was that timeouts during subsequent deliveries caused
4932 crashes instead of being properly handled. The handler is now left at its
4933 default (and expected) setting.
4935 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4936 message, again when deliver_drop_privilege is set or Exim is run as root.
4937 The bug described in (1) was not present in this case, but the tidying up
4938 of the other signals was missing. I have made the two cases consistent.
4940 3. The ignore_target_hosts setting on a manualroute router was being ignored
4941 for hosts that were looked up using the /MX notation.
4943 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4946 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4947 operated on the sender address. After changing the $sender_address to <>
4948 for the sender address verify, Exim was re-instated it as the original
4949 (before rewriting) address, but remembering that it had rewritten it, so it
4950 wasn't rewriting it again. This bug also had the effect of breaking the
4951 sender address verification caching when the sender address was rewritten.
4953 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4954 This has been changed so that if the ip literal address matches
4955 ignore_target_hosts, the router declines.
4957 7. Added expansion conditions match_domain, match_address, and match_local_
4958 part (NOT match_host).
4960 8. The placeholder for the Received: header didn't have a length field set.
4962 9. Added code to Exim itself and to exim_lock to test for a specific race
4963 condition that could lead to file corruption when using MBX delivery. The
4964 issue is with the lockfile that is created in /tmp. If this file is removed
4965 after a process has opened it but before that process has acquired a lock,
4966 there is the potential for a second process to recreate the file and also
4967 acquire a lock. This could lead to two Exim processes writing to the file
4968 at the same time. The added code performs the same test as UW imapd; it
4969 checks after acquiring the lock that its file descriptor still refers to
4970 the same named file.
4972 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4973 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4976 11. Added the smtp_active_hostname option. If used, this will typically be made
4977 to depend on the incoming interface address. Because $interface_address is
4978 not set up until the daemon has forked a reception process, error responses
4979 that can happen earlier (such as "too many connections") no longer contain
4982 12. If an expansion in a condition on a "warn" statement fails because a lookup
4983 defers, the "warn" statement is abandoned, and the next ACL statement is
4984 processed. Previously this caused the whole ACL to be aborted.
4986 13. Added the iplsearch lookup type.
4988 14. Added ident_timeout as a log selector.
4990 15. Added tls_certificate_verified as a log selector.
4992 16. Added a global option tls_require_ciphers (compare the smtp transport
4993 option of the same name). This controls incoming TLS connections.
4995 17. I finally figured out how to make tls_require_ciphers do a similar thing
4996 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4997 before starting the TLS session.
4999 18. Tabs are now shown as \t in -bP output.
5001 19. If the log selector return_path_on_delivery was set, Exim crashed when
5002 bouncing a message because it had too many Received: header lines.
5004 20. If two routers both had headers_remove settings, and the first one included
5005 a superfluous trailing colon, the final name in the first list and the
5006 first name in the second list were incorrectly joined into one item (with a
5007 colon in the middle).
5013 1. Added -C and -D options to the exinext utility, mainly to make it easier
5014 to include in the automated testing, but these could be helpful when
5015 multiple configurations are in use.
5017 2. The exinext utility was not formatting the output nicely when there was
5018 an alternate port involved in the retry record key, nor when there was a
5019 message id as well (for retries that were specific to a specific message
5020 and a specific host). It was also confused by IPv6 addresses, because of
5021 the additional colons they contain. I have fixed the IPv4 problem, and
5022 patched it up to do a reasonable job for IPv6.
5024 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5025 delivery, the log line now contains "pipelined" if PIPELINING was used.
5027 4. An SMTP transport process used to panic and die if the bind() call to set
5028 an explicit outgoing interface failed. This has been changed; it is now
5029 treated in the same way as a connect() failure.
5031 5. A reference to $sender_host_name in the part of a conditional expansion
5032 that was being skipped was still causing a DNS lookup. This no longer
5035 6. The def: expansion condition was not recognizing references to header lines
5036 that used bh_ and bheader_.
5038 7. Added the _cache feature to named lists.
5040 8. The code for checking quota_filecount in the appendfile transport was
5041 allowing one more file than it should have been.
5043 9. For compatibility with Sendmail, the command line option
5051 and sets the incoming protocol and host name (for trusted callers). The
5052 host name and its colon can be omitted when only the protocol is to be set.
5053 Note the Exim already has two private options, -pd and -ps, that refer to
5054 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5055 "s", but I don't think that's a major issue.
5057 10. A number of refactoring changes to the code, none of which should affect
5060 (a) The number of logging options was getting close to filling up the
5061 32-bit word that was used as a bit map. I have split them into two classes:
5062 those that are passed in the argument to log_write(), and those that are
5063 only ever tested independently outside of that function. These are now in
5064 separate 32-bit words, so there is plenty of room for expansion again.
5065 There is no change in the user interface or the logging behaviour.
5067 (b) When building, for example, log lines, the code previously used a
5068 macro that called string_cat() twice, in order to add two strings. This is
5069 not really sufficiently general. Furthermore, there was one instance where
5070 it was actually wrong because one of the argument was used twice, and in
5071 one call a function was used. (As it happened, calling the function twice
5072 did not affect the overall behaviour.) The macro has been replaced by a
5073 function that can join an arbitrary number of extra strings onto a growing
5076 (c) The code for expansion conditions now uses a table and a binary chop
5077 instead of a serial search (which was left over from when there were very
5078 few conditions). Also, it now recognizes conditions like "pam" even when
5079 the relevant support is not compiled in: a suitably worded error message is
5080 given if an attempt is made to use such a condition.
5082 11. Added ${time_interval:xxxxx}.
5084 12. A bug was causing one of the ddress fields not to be passed back correctly
5085 from remote delivery subprocesses. The field in question was not being
5086 subsequently used, so this caused to problems in practice.
5088 13. Added new log selectors queue_time and deliver_time.
5090 14. Might have fixed a bug in maildirsizefile handling that threw up
5091 "unexpected character" debug warnings, and recalculated the data
5092 unnecessarily. In any case, I expanded the warning message to give more
5095 15. Added the message "Restricted characters in address" to the statements in
5096 the default ACL that block characters like @ and % in local parts.
5098 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5099 Three changes have been made:
5101 (a) There was a serious bug; a negative response to MAIL caused the whole
5102 recipient domain to be cached as invalid, thereby blocking all messages
5103 to all local parts at the same domain, from all senders. This bug has
5104 been fixed. The domain is no longer cached after a negative response to
5105 MAIL if the sender used is not empty.
5107 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5110 (c) A new callout option, "use_sender" has been added for people who want
5111 the modified behaviour.
5117 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5120 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5121 indeed breaks things for older releases.
5123 3. Added additional logging to the case where there is a problem reading data
5124 from a filter that is running in a subprocess using a pipe, in order to
5125 try to track down a specific problem.
5127 4. Testing facility fudge: when running in the test harness and attempting
5128 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5129 getting "No route to host". Convert this to a timeout.
5131 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5134 6. Some OS don't have socklen_t but use size_t instead. This affects the
5135 fifth argument of getsockopt() amongst other things. This is now
5136 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5137 can be set for individual OS. I have set it for SunOS5, OSF1, and
5138 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5139 some earlier ones do not.
5141 7. Change 4.30/15 was not doing the test caselessly.
5143 8. The standard form for an IPv6 address literal was being rejected by address
5144 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5145 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5146 this, as well as the form without the "IPv6" on the front (but only when
5147 address literals are enabled, of course).
5149 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5151 10. Exim crashed if a message with an empty sender address specified by -f
5152 encountered a router with an errors_to setting. This could be provoked only
5153 by a command such as
5157 where an empty string was supplied; "<>" did not hit this bug.
5159 11. Installed PCRE release 4.5.
5161 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5162 remained set. It is now erased.
5164 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5165 times from message ids (which are base 36 rather than the normal 62).
5167 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5168 were being counted as actual protocol errors, and logged if the log
5169 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5170 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5171 and DATA following a set of rejected RCPTs do not count as protocol errors.
5172 In other words, Exim assumes they were pipelined, though this may not
5173 actually be the case. Of course, in all cases the client gets an
5174 appropriate error code.
5176 15. If a lookup fails in an ACL condition, a message about the failure may
5177 be available; it is used if testing the ACL cannot continue, because most
5178 such messages specify what the cause of the deferral is. However, some
5179 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5180 that caused an old message to be retained and used if a later statement
5181 caused a defer, replacing the real cause of the deferral.
5183 16. If an IP address had so many PTR records that the DNS lookup buffer
5184 was not large enough to hold them, Exim could crash while trying to process
5185 the truncated data. It now detects and logs this case.
5187 17. Further to 4.21/58, another change has been made: if (and only if) the
5188 first line of a message (the first header line) ends with CRLF, a bare LF
5189 in a subsequent header line has a space inserted after it, so as not to
5190 terminate the header.
5192 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5193 unnecessarily, used atoi() instead of strtol(), and didn't check the
5194 termination when getting file sizes from file names by regex.
5196 19. Completely re-implemented the support for maildirsize files, in the light
5197 of a number of problems with the previous contributed implementation
5198 (4.30/29). In particular:
5200 . If the quota is zero, the maildirsize file is maintained, but no quota is
5203 . If the maildir directory does not exist, it is created before any attempt
5204 to write a maildirsize file.
5206 . The quota value in the file is just a cache; if the quota is changed in
5207 the transport, the new value overrides.
5209 . A regular expression is available for excluding directories from the
5212 20. The autoreply transport checks the characters in options that define the
5213 message's headers; it allows continued headers, but it was checking with
5214 isspace() after an embedded newline instead of explicitly looking for a
5217 21. If all the "regular" hosts to which an address was routed had passed their
5218 expiry times, and had not reached their retry times, the address was
5219 bounced, even if fallback hosts were defined. Now Exim should go on to try
5222 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5223 equivalent code in the SMTP transport. Some hosts send humungous responses
5224 to HELO/EHLO, more than 1024 it seems.
5226 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5227 gives compiler warnings in some environments. I've now done it "properly",
5230 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5231 (because of problems with the built-in one) was declared to return uschar *
5232 instead of char *, causing compiler failure.
5234 25. Fixed a file descriptor leak when processing alias/forward files.
5236 26. Fixed a minor format string issue in dbfn.c.
5238 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5240 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5241 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5244 29. When a sender address is verified, it is cached, to save repeating the test
5245 when there is more than one recipient in a message. However, when the
5246 verification involves a callout, it is possible for different callout
5247 options to be set for different recipients. It is too complicated to keep
5248 track of this in the cache, so now Exim always runs a verification when a
5249 callout is required, relying on the callout cache for the optimization.
5250 The overhead is duplication of the address routing, but this should not be
5253 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5254 to be verified with callout=postmaster, and the main callout worked but the
5255 postmaster check failed, the verification correctly failed. However, if a
5256 subsequent RCPT command asked for sender verification *without* the
5257 postmaster check, incorrect caching caused this verification also to fail,
5260 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5261 it was not caching the DNS options (qualify_single, search_parents) that
5262 were used when the lookup failed. A subsequent lookup with different
5263 options therefore always gave the same answer, though there were cases
5264 where it should not have. (Example: a "domains = !$mx_any" option on a
5265 dnslookup router: the "domains" option is always processed without any
5266 widening, but the router might have qualify_single set.) Now Exim uses the
5267 cached value only when the same options are set.
5269 32. Added John Jetmore's "exipick" utility to the distribution.
5271 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5272 than a timeout (e.g. a certificate is required, and is not provided), an
5273 Exim server now closes the connection immediately. Previously it waited for
5274 the client to close - but if the client is SSL, it seems that they each
5275 wait for each other, leading to a delay before one of them times out.
5277 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5278 maintained 0.8.x compatibility because I don't think many are using it, and
5279 it is clearly obsolete.
5281 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5284 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5285 client certificate was expired. A simple patch fixes this, though I don't
5286 understand the full logic of why the verify callback is called multiple
5289 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5290 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5291 which causes problems with some clients (such as the Certicom SSL Plus
5292 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5293 disables the coutermeasure allowing Eudora to connect."
5295 38. Exim was not checking that a write() to a log file succeeded. This could
5296 lead to Bad Things if a log got too big, in particular if it hit a file
5297 size limit. Exim now panics and dies if it cannot write to a log file, just
5298 as it does if it cannot open a log file.
5300 39. Modified OS/Makefile-Linux so that it now contains
5302 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5304 The two -D definitions ensure that Exim is compiled with large file
5305 support, which makes it possible to handle log files that are bigger than
5308 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5309 instance) a domain was checked against a named list that involved a lookup,
5310 causing $domain_data to be set, then another domain was checked against the
5311 same list, then the first domain was re-checked, the value of $domain_data
5312 after the final check could be wrong. In particular, if the second check
5313 failed, it could be set empty. This bug probably also applied to
5316 41. The strip_trailing_dot option was not being applied to the address given
5317 with the -f command-line option.
5319 42. The code for reading a message's header from the spool was incrementing
5320 $received_count, but never initializing it. This meant that the value was
5321 incorrect (doubled) while delivering a message in the same process in which
5322 it was received. In the most common configuration of Exim, this never
5323 happens - a fresh exec is done - but it can happen when
5324 deliver_drop_privilege is set.
5326 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5327 it now includes up to 150 characters of the unexpected data in the log
5330 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5331 and building data strings. The size of both of these buffers was 10 000
5332 bytes - far larger than anybody would *ever* want, thought I. Needless to
5333 say, somebody hit the limit. I have increased the maximum line length to
5334 20 000 and the maximum data length of concatenated lines to 100 000. I have
5335 also fixed two bugs, because there was no checking on these buffers. Tsk,
5336 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5337 buffer is too small.
5339 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5340 lsearch lookups. Now it does.
5342 46. When parsing a route_list item in a manualroute router, a fixed-length
5343 buffer was used for the list of hosts. I made this 1024 bytes long,
5344 thinking that nobody would ever have a list of hosts that long. Wrong.
5345 Somebody had a whole pile of complicated expansion conditions, and the
5346 string was silently truncated, leading to an expansion error. It turns out
5347 that it is easier to change to an unlimited length (owing to other changes
5348 that have happened since this code was originally written) than to build
5349 structure for giving a limitation error. The length of the item that
5350 expands into the list of hosts is now unlimited.
5352 47. The lsearch lookup could not handle data where the length of text line was
5353 more than 4095 characters. Such lines were truncated, leading to shortened
5354 data being returned. It should now handle lines of any length.
5356 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5357 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5360 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5361 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5362 no longer happens in comment lines. A list of replacements is now placed
5363 at the head of all of the source files, except those whose only change is
5364 to replace PERL_COMMAND in the very first #! line.
5366 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5367 messages on the queue, with a bottom-up merge sort, using code contributed
5368 by Michael Haardt. This should make operations like -bp somewhat faster on
5369 large queues. It won't affect queue runners, except when queue_run_in_order
5372 51. Installed eximstats 1.31 in the distribution.
5374 52. Added support for SRV lookups to the dnslookup router.
5376 53. If an ACL referred to $message_body or $message_body_end, the value was not
5377 reset for any messages that followed in the same SMTP session.
5379 54. The store-handling optimization for building very long strings was not
5380 differentiating between the different store pools. I don't think this
5381 actually made any difference in practice, but I've tidied it.
5383 55. While running the routers to verify a sender address, $sender_address
5384 was still set to the sender address. This is wrong, because when routing to
5385 send a bounce to the sender, it would be empty. Therefore, I have changed
5386 it so that, while verifying a sender address, $sender_address is set to <>.
5387 (There is no change to what happens when verifying a recipient address.)
5389 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5390 target A or AAAA records (if not already returned) without resetting the
5391 qualify_single or search_parents options of the DNS resolver. These are
5392 inappropriate in this case because the targets of MX and SRV records must
5393 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5394 target that, when qualified, matched something in the local domain. These
5395 two options are now turned off when doing these lookups.
5397 57. It seems that at least some releases of Reiserfs (which does not have the
5398 concept of a fixed number of inodes) returns zero and not -1 for the
5399 number of available inodes. This interacted badly with check_spool_inodes,
5400 which assumed that -1 was the "no such thing" setting. What I have done is
5401 to check that the total number of inodes is greater than zero before doing
5402 the test of how many are available.
5404 58. When a "warn" ACL statement has a log_message modifier, the message is
5405 remembered, and not repeated. This is to avoid a lot of repetition when a
5406 message has many recipients that cause the same warning to be written.
5407 Howewer, Exim was preserving the list of already written lines for an
5408 entire SMTP session, which doesn't seem right. The memory is now reset if a
5409 new message is started.
5411 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5412 debugging output unless log_rewrite was also set.
5414 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5415 of (int)(handle) when we know that handle contains (void *)(-1).
5417 61. The Exim daemon panic-logs an error return when it closes the incoming
5418 connection. However "connection reset by peer" seems to be common, and
5419 isn't really an error worthy of noting specially, so that particular error
5422 62. When Exim is trying to find all the local interfaces, it used to panic and
5423 die if the ioctl to get the interface flags failed. However, it seems that
5424 on at least one OS (Solaris 9) it is possible to have an interface that is
5425 included in the list of interfaces, but for which you get a failure error
5426 for this call. This happens when the interface is not "plumbed" into a
5427 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5428 failure of the "get flags" call assumes that the interface is down.
5430 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5431 makes life easier for people who are doing arithmetic on fields extracted
5432 from dates, where you often get leading zeros that should not be
5433 interpreted as octal.
5435 64. Added qualify_domain to the redirect router, to override the global
5438 65. If a pathologically long header line contained very many addresses (the
5439 report of this problem mentioned 10 000) and each of them was rewritten,
5440 Exim could use up a very large amount of memory. (It kept on making new
5441 copies of the header line as it rewrote, and never released the old ones.)
5442 At the expense of a bit more processing, the header rewriting function has
5443 been changed so that it no longer eats memory in this way.
5445 66. The generation of the Received: header has been moved from the time that a
5446 message starts to be received, to the time that it finishes. The timestamp
5447 in the Received: header should now be very close to that of the <= log
5448 line. There are two side-effects of this change:
5450 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5451 logged header lines no longer include the local Received: line, because
5452 it has not yet been created. The same applies to a copy of the message
5453 that is returned to a non-SMTP sender when a message is rejected.
5455 (b) When a filter file is tested using -bf, no additional Received: header
5456 is added to the test message. After some thought, I decided that this
5459 This change does not affect the value of $received_for. It is still set
5460 after address rewriting, but before local_scan() is called.
5462 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5464 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5465 gave an unhelpful panic error message, and a defer error. I have managed to
5466 change this behaviour so that it now rejects any supplied certificate,
5467 which seems right, as the list of acceptable certificates is empty.
5469 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5470 gave an unhelpful defer error. I have not managed to make this reject any
5471 supplied certificates, but the error message it gives is "no certificate
5472 supplied", which is not helpful.
5474 70. exigrep's output now also includes lines that are not associated with any
5475 message, but which match the given pattern. Implemented by a patch from
5476 Martin Sluka, which also tidied up the Perl a bit.
5478 71. Recipient callout verification, like sender verification, was using <> in
5479 the MAIL FROM command. This isn't really the right thing, since the actual
5480 sender may affect whether the remote host accepts the recipient or not. I
5481 have changed it to use the actual sender in the callout; this means that
5482 the cache record is now keyed on a recipient/sender pair, not just the
5483 recipient address. There doesn't seem to be a real danger of callout loops,
5484 since a callout by the remote host to check the sender would use <>.
5485 [SEE ABOVE: changed after hitting problems.]
5487 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5488 temporary errors. However, in the case of such a code being given after
5489 the end of a data transmission (i.e. after ".") Exim was failing to write
5490 a retry record for the message. (Yes, there was some broken host that was
5491 actually sending 8xx at this point.)
5493 73. An unknown lookup type in a host list could cause Exim to panic-die when
5494 the list was checked. (An example that provoked this was putting <; in the
5495 middle of a list instead of at the start.) If this happened during a DATA
5496 ACL check, a -D file could be left lying around. This kind of configuration
5497 error no longer causes Exim to die; instead it causes a defer errror. The
5498 incident is still logged to the main and panic logs.
5500 74. Buglet left over from Exim 3 conversion. The message "too many messages
5501 in one connection" was written to the rejectlog but not the mainlog, except
5502 when address rewriting (yes!) was being logged.
5504 75. Added write_rejectlog option.
5506 76. When a system filter was run not as root (that is, when system_filter_user
5507 was set), the values of the $n variables were not being returned to the
5508 main process; thus, they were not subsequently available in the $sn
5511 77. Added +return_path_on_delivery log selector.
5513 78. A connection timeout was being treated differently from recipients deferred
5514 when testing hosts_max_try with a message that was older than the host's
5515 retry timeout. (The host should not be counted, thus allowing all hosts to
5516 be tried at least once before bouncing.) This may have been the cause of an
5517 occasionally reported bug whereby a message would remain on the queue
5518 longer than the retry timeout, but would be bounced if a delivery was
5519 forced. I say "may" because I never totally pinned down the problem;
5520 setting up timeout/retry tests is difficult. See also the next item.
5522 79. The ultimate address timeout was not being applied to errors that involved
5523 a combination of host plus message (for example, a timeout on a MAIL
5524 command). When an address resolved to a number of possible hosts, and they
5525 were not all tried for each delivery (e.g. because of hosts_max_try), a
5526 message could remain on the queue longer than the retry timeout.
5528 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5531 81. Fixed an obscure SMTP outgoing bug which required at least the following
5532 conditions: (a) there was another message waiting for the same server;
5533 (b) the server returned 5xx to all RCPT commands in the first message so
5534 that the message was not completed; (c) the server dropped the connection
5535 or gave a negative response to the RSET that Exim sends to abort the
5536 transaction. The observed case was a dropped connection after DATA that had
5537 been sent in pipelining mode. That is, the server had advertised PIPELINING
5538 but was not implementing it correctly. The effect of the bug was incorrect
5539 behaviour, such as trying another host, and this could lead to a crash.
5545 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5546 and daemon.c were passed as pointers to ints; they should have been
5547 pointers to socklen_t variables (which are typically unsigned ints).
5549 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5552 3. Fixed a really odd bug that affected only the testing scheme; patching a
5553 certain fixed string in the binary changed the value of another string that
5554 happened to be identical to the end of the original first string.
5556 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5557 name", it returns that address as the IP address. On some operating
5558 systems (e.g. Solaris), it also passes back the IP address string as the
5559 "host name". However, on others (e.g. Linux), it passes back an empty
5560 string. Exim wasn't checking for this, and was changing the host name to an
5561 empty string, assuming it had been canonicized.
5563 5. Although rare, it is permitted to have more than one PTR record for a given
5564 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5565 all the names associated with an address, because they do in Solaris.
5566 However, it seems that they do not in Linux for data that comes from the
5567 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5568 I found this out when I moved to a new Linux workstation and tried to run
5569 the Exim test suite.
5571 To get round this problem I have changed the code so that it now does its
5572 own call to the DNS to look up PTR records when searching for a host name.
5573 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5574 addresses that are only in /etc/hosts are still found.
5576 This behaviour is, however, controlled by an option called host_lookup_
5577 order, which defaults to "bydns:byaddr". If people want to use the other
5578 order, or indeed, just use one or the other means of lookup, they can
5579 specify it in this variable.
5581 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5582 some operating systems, this comes back from gethostbyaddr() as an empty
5583 string, and this is what Exim used to test for. However, it seems that in
5584 other systems, "." is yielded. Exim now tests for this case too.
5586 7. The values of check_spool_space and check_log_space are now held internally
5587 as a number of kilobytes instead of an absolute number of bytes. If a
5588 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5589 kilobyte. This means that much larger values can be stored.
5591 8. Exim monitor: an attempt to get the action menu when not actually pointing
5592 at a message produces an empty menu entitled "No message selected". This
5593 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5594 no entries in it ("Shell widget menu has zero width and/or height"). So I
5595 have added a single, blank menu entry in this case.
5597 9. Added ${quote_local_part.
5599 10. MIME decoding is now applied to the contents of Subject: header lines when
5602 11. Now that a reference to $sender_host_address automatically causes a reverse
5603 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5604 host lookup before query-style lookups in lists that might use this
5605 variable. This has therefore been abolished, and the "net-" prefix is no
5606 longer necessary for query-style lookups.
5608 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5609 have been a typo for LFLAGS, so it has been changed.
5611 13. The install script calls Exim with "-C /dev/null" in order to find the
5612 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5613 to be output. Howeve, since Exim outputs its version number before the
5614 error, it didn't break the script. It just looked ugly. I fixed this by
5615 always allowing "-C /dev/null" if the caller is root.
5617 14. Ignore overlarge ACL variable number when reading spool file - insurance
5618 against a later release with more variables having written the file.
5620 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5621 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5622 this, as well as the form without the "IPv6" on the front.
5624 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5625 OS/Makefile-Darwin file.
5627 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5628 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5630 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5631 variable. However, some OS use ints for inodes. I've added cast to long int
5632 to get rid of the compiler warning.
5634 19. I had forgotten to lock out "/../" in configuration file names when
5635 ALT_CONFIG_PREFIX was set.
5637 20. Routers used for verification do not need to specify transports. However,
5638 if such a router generated a host list, and callout was configured, Exim
5639 crashed, because it could not find a port number from the (non-existent)
5640 transport. It now assumes port 25 in this circumstance.
5642 21. Added the -t option to exigrep.
5644 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5645 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5646 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5647 them is set, LOOKUP_LSEARCH is forced.
5649 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5651 24. Added sender and host information to the "rejected by local_scan()" log
5652 line; previously there was no indication of these.
5654 25. Added .include_if_exists.
5656 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5657 Linux. It turns out that not all file systems support this. Apparently some
5658 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5659 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5660 not supported on the file descriptor, is now ignored when Exim is trying to
5661 sync a directory. This applies only to Linux.
5663 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5665 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5666 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5669 29. Added support for maildirsize files from supplied patch (modified a bit).
5671 30. The use of :fail: followed by an empty string could lead Exim to respond to
5672 sender verification failures with (e.g.):
5674 550 Verification failed for <xxx>
5675 550 Sender verify failed
5677 where the first response line was missing the '-' that indicates it is not
5678 the final line of the response.
5680 31. The loop for finding the name of the user that called Exim had a hardwired
5681 limit of 10; it now uses the value of finduser_retries, which is used for
5682 all other user lookups.
5684 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5687 33. Exim was neglecting to zero errno before one call of strtol() when
5688 expanding a string and expecting an integer value. On some systems this
5689 resulted in spurious "integer overflow" errors. Also, it was casting the
5690 result into an int without checking.
5692 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5693 did not work. The code looks as if it has *never* worked, though it appears
5694 to have been documented since at least releast 1.62. I have made it work.
5696 35. The "timeout_DNS" error in retry rules, also documented since at least
5697 1.62, also never worked. As it isn't clear exactly what this means, and
5698 clearly it isn't a major issue, I have abolished the feature by treating it
5699 as "timeout", and writing a warning to the main and panic logs.
5701 36. The display of retry rules for -brt wasn't always showing the error code
5704 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5705 timeout_connect_A, timeout_connect_MX.
5707 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5708 to the empty sender.
5710 39. The daemon was not analysing the content of -oX till after it had closed
5711 stderr and disconnected from the controlling terminal. This meant that any
5712 syntax errors were only noted on the panic log, and the return code from
5713 the command was 0. By re-arranging the code a little, I've made the
5714 decoding happen first, so such errors now appear on stderr, and the return
5715 code is 1. However, the actual setting up of the sockets still happens in
5716 the disconnected process, so errors there are still only recorded on the
5719 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5720 connections (as happens on some IP stacks) was logged at start up time as
5721 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5722 it from "IPv6 and IPv4", which means that two separate sockets are being
5725 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5726 says whether AF_INET or AF_INET6 was passed as an argument.
5728 42. Exiwhat output was messed up when time zones were included in log
5731 43. Exiwhat now gives more information about the daemon's listening ports,
5732 and whether -tls-on-connect was used.
5734 44. The "port" option of the smtp transport is now expanded.
5736 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5737 silently ignored. Now an error message is written to the main and panic
5740 46. There's a new ACL modifier called "logwrite" which writes to a log file
5741 as soon as it is encountered.
5743 47. Added $local_user_uid and $local_user_gid at routing time.
5745 48. Exim crashed when trying to verify a sender address that was being
5748 49. Exim was recognizing only a space character after ".include". It now also
5749 recognizes a tab character.
5751 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5752 extracting the relevant information from the specification. The man page no
5753 longer contains scrambled data for the -d option, and I've added a section
5754 at the front about calling Exim under different names.
5756 51. Added "extra_headers" argument to the "mail" command in filter files.
5758 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5761 53. Installed eximstats 1.29.
5763 54. Added transport_filter_timeout as a generic transport option.
5765 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5766 Cc: header lines. This was required by RFC 822, but it not required by RFC
5769 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5770 incoming messages that did not have them. Now it does so only if the
5771 message originates locally, that is, if there is no associated remote host
5772 address. When Resent- header lines are present, this applies to the Resent-
5773 lines rather than the non-Resent- lines.
5775 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5776 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5778 58. Messages for configuration errors now include the name of the main
5779 configuration file - useful now that there may be more than one file in a
5780 list (.included file names were always shown).
5782 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5783 for those rare installations that do not start the daemon as root or run it
5784 setuid root. I've cut out the call to initgroups() if the daemon is not
5787 60. The Exim user and group can now be bound into the binary as text strings
5788 that are looked up at the start of Exim's processing.
5790 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5792 62. Added $mailstore_basename variable.
5794 63. Installed patch to sieve.c from Michael Haardt.
5796 64. When Exim failed to open the panic log after failing to open the main log,
5797 the original message it was trying to log was written to stderr and debug
5798 output, but if they were not available (the usual case in production), it
5799 was lost. Now it is written to syslog before the two lines that record the
5800 failures to open the logs.
5802 65. Users' Exim filters run in subprocesses under the user's uid. It is
5803 possible for a "deliver" command or an alias in a "personal" command to
5804 provoke an address rewrite. If logging of address rewriting is configured,
5805 this fails because the process is not running as root or exim. There may be
5806 a better way of dealing with this, but for the moment (because 4.30 needs
5807 to be released), I have disabled address rewrite logging when running a
5808 filter in a non-root, non-exim process.
5814 1. The buildconfig auxiliary program wasn't quoting the value set for
5815 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5816 not defined. This bug was masked in 4.22 by the effect that was fixed in
5819 2. Some messages that were rejected after a message id was allocated were
5820 shown as "incomplete" by exigrep. It no longer does this for messages that
5821 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5823 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5824 have allow_domain_literals set, the ID did not get logged in the <= line.
5825 Domain literals are now always recognized in Message-ID: header lines.
5827 4. The first argument for a ${extract expansion item is the key name or field
5828 number. Leading and trailing spaces in this item were not being ignored,
5829 causing some misleading effects.
5831 5. When deliver_drop_privilege was set, single queue runner processes started
5832 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5833 same command in the process it spins off) were not dropping privilege.
5835 6. When the daemon running as "exim" started a queue runner, it always
5836 re-executed Exim in the spun-off process. This is a waste of effort when
5837 deliver_drop_privilege is set. The new process now just calls the
5838 queue-runner function directly.
5844 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5847 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5848 ignored. Though the use of -oP was forcing the writing of a pid file, it
5849 was always written to the default place.
5851 3. If the message "no IP address found for host xxxx" is generated during
5852 incoming verification, it is now followed by identification of the incoming
5853 connection (so you can more easily find what provoked it).
5855 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5857 5. Added some features to "harden" Exim a bit more against certain attacks:
5859 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5860 be put in Local/Makefile. This is like the never_users runtime option,
5861 but it cannot be overridden. The default setting is "root".
5863 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5864 prefix string with which any file named in a -C command line option
5867 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5868 is retained for -C and -D only if the caller of Exim is root. Without
5869 it, the exim user may also use -C and -D and retain privilege.
5871 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5872 command line option is disabled.
5874 6. Macro names set by the -D option must start with an upper case letter, just
5875 like macro names defined in the configuration file.
5877 7. Added "dereference=" facility to LDAP.
5879 8. Two instances of the typo "uknown" in the source files are fixed.
5881 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5882 the Configure-Makefile script screwed up while processing it.
5884 10. Incorporated PCRE 4.4.
5886 11. The SMTP synchronization check was not operating right at the start of an
5887 SMTP session. For example, it could not catch a HELO sent before the client
5888 waited for the greeting. There is now a check for outstanding input at the
5889 point when the greeting is written. Because of the duplex, asynchronous
5890 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5891 way, but not yet received, when the check is performed.
5893 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5894 on TCP/IP sockets, because this apparently causes some broken clients to
5897 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5898 unchanged) from the Cygwin maintainer.
5900 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5901 was supported instead of testing for mbx. Effectively a typo.
5903 15. The spa authenticator server code was not checking that the input it
5904 received was valid base64.
5906 16. The debug output line for the "set" modifier in ACLs was not showing the
5907 name of the variable that was being set.
5909 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5911 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5912 lookup on its own. Something else was needed to trigger the lookup. For
5913 example, a match in host_lookup or the need for a host name in a host list.
5914 Now, if $sender_host_name is referenced and the host name has not yet been
5915 looked up, a lookup is performed. If the lookup fails, the variable remains
5916 empty, and $host_lookup_failed is set to "1".
5918 19. Added "eqi" as a case-independent comparison operator.
5920 20. The saslauthd authentication condition could segfault if neither service
5921 nor realm was specified.
5923 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5924 error message that was logged was misleading, and incoming SMTP
5925 connections were dropped. The message is now more accurate, and temporary
5926 errors are given to SMTP connections.
5928 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5929 (except RSET) with a 421 error, until QUIT is received. However, it was
5930 failing to send a response to QUIT.
5932 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5933 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5934 ACL is now run after the helo_try_verify_hosts code.
5936 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5937 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5938 case-independent, so other case variants are also recognized). Apparently
5939 some systems use these upper case variants.
5941 25. If more than two messages were waiting for the same host, and a transport
5942 filter was specified for the transport, Exim sent two messages over the
5943 same TCP/IP connection, and then failed with "socket operation on non-
5944 socket" when it tried to send the third.
5946 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5948 27. The extern definition of crypt16() in expand.c was not being excluded when
5949 the OS had its own crypt16() function.
5951 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5952 as a preferred synonym for return_size_limit, both as an option and as an
5955 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5957 30. Changed the default configuration ACL to relax the local part checking rule
5958 for addresses that are not in any local domains. For these addresses,
5959 slashes and pipe symbols are allowed within local parts, but the sequence
5960 /../ is explicitly forbidden.
5962 31. SPA server authentication was not clearing the challenge buffer before
5965 32. log_message in a "warn" ACL statement was writing to the reject log as
5966 well as to the main log, which contradicts the documentation and doesn't
5967 seem right (because no rejection is happening). So I have stopped it.
5969 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5970 However, I am unable to do any testing of this.
5972 34. Fixed an infelicity in the appendfile transport. When checking directories
5973 for a mailbox, to see if any needed to be created, it was accidentally
5974 using path names with one or more superfluous leading slashes; tracing
5975 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5977 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5978 contents are added to the log line that is written for every discarded
5979 recipient. (Previously a log_message setting was ignored.)
5981 36. The ${quote: operator now quotes the string if it is empty.
5983 37. The install script runs exim in order to find its version number. If for
5984 some reason other than non-existence or emptiness, which it checks, it
5985 could not run './exim', it was installing it with an empty version number,
5986 i.e. as "exim-". This error state is now caught, and the installation is
5989 38. An argument was missing from the function that creates an error message
5990 when Exim fails to connect to the socket for saslauthd authentication.
5991 This could cause Exim to crash, or give a corrupted message.
5993 39. Added isip, isip4, and isip6 to ${if conditions.
5995 40. The ACL variables $acl_xx are now saved with the message, and can be
5996 accessed later in routers, transports, and filters.
5998 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5999 strings in the file are not string-expanded.
6001 42. If a MAIL command specified a SIZE value that was too large to fit into an
6002 int variable, the check against message_size_limit failed. Such values are
6003 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6004 day this will have to be increased, but I don't think I want to be around
6005 when emails are that large.
6012 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6013 iconv() is not standard in FreeBSD.
6015 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6016 IPv6 enabled. The observed symptom was a segmentation fault on return from
6017 the function os_common_find_running_interfaces() in src/os.c.
6019 3. In the check_special_case() function in daemon.c I had used "errno" as an
6020 argument name, which causes warnings on some systems. This was basically a
6021 typo, since it was named "eno" in the comments!
6023 4. The code that waits for the clock to tick (at a resolution of some fraction
6024 of a second) so as to ensure message-id uniqueness was always waiting for
6025 at least one whole tick, when it could have waited for less. [This is
6026 almost certainly not relevant at current processor speeds, where it is
6027 unlikely to ever wait at all. But we try to future-proof.]
6029 5. The function that sleeps for a time interval that includes fractions of a
6030 second contained a race. It did not block SIGALRM between setting the
6031 timer, and suspending (a couple of lines later). If the interval was short
6032 and the sigsuspend() was delayed until after it had expired, the suspension
6033 never ended. On busy systems this could lead to processes getting stuck for
6036 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6037 process, before it forks any delivery processes. The open lookup caching
6038 mechanism meant that the open file or database connection was passed into
6039 the delivery process. The problem was that delivery processes always tidy
6040 up cached lookup data. This could cause a problem for the next delivery
6041 process started by the queue runner, because the external queue runner
6042 process does not know about the closure. So the next delivery process
6043 still has data in the lookup cache. In the case of a file lookup, there was
6044 no problem because closing a file descriptor in a subprocess doesn't affect
6045 the parent. However, if the lookup was caching a connection to a database,
6046 the connection was closed, and the second delivery process was likely to
6047 see errors such as "PGSQL: query failed: server closed the connection
6048 unexpectedly". The problem has been fixed by closing all cached lookups
6049 in a queue runner before running a delivery process.
6051 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6052 seem to have the "const" qualifier which it has on other OS. I've
6055 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6056 *of the same type* (client or server) with the same public name that an
6057 error should be diagnosed.
6059 9. When Exim looked up a host name for an IP address, but failed to find the
6060 original IP address when looking up the host name (a safety check), it
6061 output the message "<ip address> does not match any IP for NULL", which was
6062 confusing, to say the least. The bug was that the host name should have
6063 appeared instead of "NULL".
6065 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6066 user that is built into the binary, and the -C or -D options is used, root
6067 privilege is dropped before the configuration file is read. In addition,
6068 logging is switched to stderr instead of the normal log files. If the
6069 configuration then re-defines the Exim user, the unprivileged environment
6070 is probably not what is expected, so Exim logs a panic warning message (but
6073 However, if deliver_drop_privilege is set, the unprivileged state may well
6074 be exactly what is intended, so the warning has been cut out in that case,
6075 and Exim is allowed to try to write to its normal log files.
6081 1. smtp_return_error_details was not giving details for temporary sender
6082 or receiver verification errors.
6084 2. Diagnose a configuration error if two authenticators have the same public
6087 3. Exim used not to create the message log file for a message until the first
6088 delivery attempt. This could be confusing when incoming messages were held
6089 for policy or load reasons. The message log file is now created at the time
6090 the message is received, and an initial "Received" line is written to it.
6092 4. The automatically generated man page for command line options had a minor
6093 bug that caused no ill effects; however, a more serious problem was that
6094 the procedure for building the man page automatically didn't always
6095 operate. Consequently, release 4.20 contains an out-of-date version. This
6096 shouldn't happen again.
6098 5. When building Exim with embedded Perl support, the script that builds the
6099 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6100 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6102 6. The freeze_tell option was not being used for messages that were frozen on
6103 arrival, either by an ACL or by local_scan().
6105 7. Added the smtp_incomplete_transaction log selector.
6107 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6108 was accepting AUTH without a new EHLO.
6110 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6111 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6114 10. Logging of TCP/IP connections (when configured) now happens in the main
6115 daemon process instead of the child process, so that the TCP/IP connection
6116 count is more accurate (but it can never be perfect).
6118 11. The use of "drop" in a nested ACL was not being handled correctly in the
6119 outer ACL. Now, if condition failure induced by the nested "drop" causes
6120 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6121 or "require"), the connection is dropped.
6123 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6124 that yield "discard" can now be used with an "accept" or a "discard" verb,
6125 but an error is generated for any others (because I can't see a useful way
6126 to define what should happen).
6128 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6129 are now processed in the same way as lines in the Exim configuration file.
6130 In particular, continuation lines are supported.
6132 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6134 15. Added -ti meaning -t -i.
6136 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6137 domains, and warn by logging if others are found.
6139 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6140 returned from GIFCONF when Exim is trying to find the list of interfaces on
6141 a host. The code in os.c has been modified to copy each ifreq to an aligned
6142 structure in all cases.
6144 Also, in some cases, the returned ifreq's were being copied to a 'struct
6145 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6146 means the last couple of bytes of an IPv6 address could be chopped if the
6147 ifreq contained only a normal sockaddr (14 bytes storage).
6149 18. Named domain lists were not supported in the hosts_treat_as_local option.
6150 An entry such as +xxxx was not recognized, and was treated as a literal
6153 19. Ensure that header lines added by a DATA ACL are included in the reject log
6154 if the ACL subsequently rejects the message.
6156 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6157 MD5 (which is deprecated).
6159 21. When testing a filter file using -bf, Exim was writing a message when it
6160 took the sender from a "From " line in the message, but it was not doing so
6161 when it took $return_path from a Return-Path: header line. It now does.
6163 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6164 with a valid header line field name (a series of printing characters
6165 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6167 23. Changed "disc" in the source to "disk" to conform to the documentation and
6168 the book and for uniformity.
6170 24. Ignore Sendmail's -Ooption=value command line item.
6172 25. When execve() failed while trying to run a command in a pipe transport,
6173 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6174 could be confused with a return value of 69 from the command itself. This
6175 has been changed to 127, the value the shell returns if it is asked to run
6176 a non-existent command. The wording for the related log line suggests a
6177 non-existent command as the problem.
6179 26. If received_header_text expands to an empty string, do not add a Received:
6180 header line to the message. (Well, it adds a token one on the spool, but
6181 marks it "old" so that it doesn't get used or transmitted.)
6183 27. Installed eximstats 1.28 (addition of -nt option).
6185 28. There was no check for failure on the call to getsockname() in the daemon
6186 code. This can fail if there is a shortage of resources on the system, with
6187 ENOMEM, for example. A temporary error is now given on failure.
6189 29. Contrary to the C standard, it seems that in some environments, the
6190 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6191 program. Exim now does this explicitly; it affects the formatting of
6192 timestamps using strftime().
6194 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6195 complaints. I've now initialized all the variables, to avoid this.
6197 32. Header lines added by a system filter were not being "seen" during
6198 transport-time rewrites.
6200 33. The info_callback() function passed to OpenSSL is set up with type void
6201 (*)(SSL *, int, int), as described somewhere. However, when calling the
6202 function (actually a macro) that sets it up, the type void(*)() is
6203 expected. I've put in a cast to prevent warnings from picky compilers.
6205 34. If a DNS black list lookup found a CNAME record, but there were no A
6206 records associated with the domain it pointed at, Exim crashed.
6208 35. If a DNS black list lookup returned more than one A record, Exim ignored
6209 all but the first. It now scans all returned addresses if a particular IP
6210 value is being sought. In this situation, the contents of the
6211 $dnslist_value variable are a list of all the addresses, separated by a
6214 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6215 to accept a host name and all its aliases if the forward lookup for any of
6216 them yielded the IP address of the incoming connection. Now it accepts only
6217 those names whose forward lookup yields the correct IP address. Any other
6218 names are discarded. This closes a loophole whereby a rogue DNS
6219 administrator could create reverse DNS records to break through a
6220 wildcarded host restriction in an ACL.
6222 37. If a user filter or a system filter that ran in a subprocess used any of
6223 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6224 the wrong values were passed to the pipe command ($thisaddress had the
6225 value of $0, $0 had the value of $1, etc). This bug was introduced by
6226 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6228 38. Improved the line breaking for long SMTP error messages from ACLs.
6229 Previously, if there was no break point between 40 and 75 characters, Exim
6230 left the rest of the message alone. Two changes have been made: (a) I've
6231 reduced the minimum length to 35 characters; (b) if it can't find a break
6232 point between 35 and 75 characters, it looks ahead and uses the first one
6233 that it finds. This may give the occasional overlong line, but at least the
6234 remaining text gets split now.
6236 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6237 file descriptors might be low, and that setting 1000 would always raise it.
6238 It turns out that in some environments, the limit is already over 1000 and
6239 that lowering it causes trouble. So now Exim takes care not to decrease it.
6241 40. When delivering a message, the value of $return_path is set to $sender_
6242 address at the start of routing (routers may change the value). By an
6243 oversight, this default was not being set up when an address was tested by
6244 -bt or -bv, which affected the outcome if any router or filter referred to
6247 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6248 log only when "message" or "log_message" are set. However, if one of the
6249 conditions was an address verification, or a call to a nested ACL, the
6250 messages generated by the underlying test were being passed through. This
6251 no longer happens. The underlying message is available in $acl_verify_
6252 message for both "message" and "log_message" expansions, so it can be
6253 passed through if needed.
6255 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6256 new expansion $bh_ to give the encoded byte string without charset
6257 translation. Translation happens only if iconv() is available; HAVE_ICONV
6258 indicates this at build time. HEADERS_CHARSET gives the charset to
6259 translate to; headers_charset can change it in the configuration, and
6260 "headers charset" can change it in an individual filter file.
6262 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6263 that creates RFC 2047 encoded "words" labels them as that charset instead
6264 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6265 expansion operator; (ii) when Exim creates a From: line for a local
6266 message; (iii) when a header line is rewritten to include a "phrase" part.
6268 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6269 buggy, causing it to skip the first lines of messages whose message ID
6270 ended in 'D'. This would not have bitten before Exim release 4.14, because
6271 message IDs were unlikely to end in 'D' before then. The effect was to have
6272 incorrect size information for certain domains.
6274 45. #include "config.h" was missing at the start of the crypt16.c module. This
6275 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6278 46. If there was a timeout during a "random" callout check, Exim treated it as
6279 a failure of the random address, and carried on sending RSET and the real
6280 address. If the delay was just some slowness somewhere, the response to the
6281 original RCPT would be taken as a response to RSET and so on, causing
6282 mayhem of various kinds.
6284 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6285 when I implemented it. It didn't allow for the fact that some option values
6286 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6287 the right test for positive values.
6289 48. Domain names in DNS records are case-independent. Exim always looks them up
6290 in lower case. Some resolvers return domain names in exactly the case they
6291 appear in the zone file, that is, they may contain uppercase letters. Not
6292 all resolvers do this - some return always lower case. Exim was treating a
6293 change of case by a resolver as a change of domain, similar to a widening
6294 of a domain abbreviation. This triggered its re-routing code and so it was
6295 trying to route what was effectively the same domain again. This normally
6296 caused routing to fail (because the router wouldn't handle the domain
6297 twice). Now Exim checks for this case specially, and just changes the
6298 casing of the domain that it ultimately uses when it transmits the message
6301 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6304 50. If a filter generated a file delivery with a non-absolute name (possible if
6305 no home directory exists for the router), the forbid_file option was not
6308 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6309 the existing equality matching.
6311 52. Exim was using ints instead of ino_t variables in some places where it was
6312 dealing with inode numbers.
6314 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6315 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6316 called TMPDIR, and if it finds it is different, it changes its value.
6318 54. The smtp_printf() function is now made available to local_scan() so
6319 additional output lines can be written before returning. There is also an
6320 smtp_fflush() function to enable the detection of a dropped connection.
6321 The variables smtp_input and smtp_batched_input are exported to
6324 55. Changed the default runtime configuration: the message "Unknown user"
6325 has been removed from the ACL, and instead placed on the localuser router,
6326 using the cannot_route_message feature. This means that any verification
6327 failures that generate their own messages won't get overridden. Similarly,
6328 the "Unrouteable address" message that was in the ACL for unverifiable
6329 relay addresses has also been removed.
6331 56. Added hosts_avoid_esmtp to the smtp transport.
6333 57. The exicyclog script was not checking for the esoteric option
6334 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6335 will work only if exicyclog is run under the appropriate euid.
6337 58. Following a discussion on the list, the rules by which Exim recognises line
6338 endings on incoming messages have been changed. The -dropcr and drop_cr
6339 options are now no-ops, retained only for backwards compatibility. The
6340 following line terminators are recognized: LF CRLF CR. However, special
6341 processing applies to CR:
6343 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6344 nor a local message in the state where . is a terminator.
6346 (ii) If a bare CR is encountered in a header line, an extra space is added
6347 after the line terminator so as not to end the header. The reasoning
6348 behind this is that bare CRs in header lines are most likely either
6349 to be mistakes, or people trying to play silly games.
6351 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6352 was being incorrectly given as 18 bytes larger than it should have been.
6353 This is a VOB (very old bug).
6355 60. This may never have affected anything current, but just in case it has:
6356 When the local host is found other than at the start of a list of hosts,
6357 the local host, those with the same MX, and any that follow, are discarded.
6358 When the list in question was part of a longer list of hosts, the following
6359 hosts (not currently being processed) were also being discarded. This no
6360 longer happens. I'm not sure if this situation could ever has previously
6363 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6366 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6367 header. This is something that is recommended in a new Internet Draft, and
6368 is something that is documented as being done by Sendmail. There are two
6369 possible values. For messages generated by the autoreply transport, Exim
6372 Auto-Submitted: auto-replied
6374 whereas for all other generated messages (e.g. bounces) it adds
6376 Auto-Submitted: auto-generated
6378 63. The "personal" condition in filters now includes a test for the
6379 Auto-Submitted: header. If it contains the string "auto-" the message it
6380 not considered personal.
6382 64. Added rcpt_include_affixes as a generic transport option.
6384 65. Added queue_only_override (default true).
6386 66. Added the syslog_duplication option.
6388 67. If what should have been the first header line of a message consisted of
6389 a space followed by a colon, Exim was mis-interpreting it as a header line.
6390 It isn't of course - it is syntactically invalid and should therefore be
6391 treated as the start of the message body. The misbehaviour could have
6392 caused a number of strange effects, including loss of data in subsequent
6393 header lines, and spool format errors.
6395 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6396 client host had authenticated. This control can now be exercised by an ACL
6397 for more flexibility.
6399 69. By default, callouts do not happen when testing with -bh. There is now a
6400 variant, -bhc, which does actually run the callout code, including
6401 consulting and updating the callout cache.
6403 70. Added support for saslauthd authentication, courtesy of Alexander
6406 71. If statvfs() failed on the spool or log directories while checking their
6407 size for availability, Exim confusingly gave the error "space shortage".
6408 Furthermore, in debugging mode it crashed with a floating point exception.
6409 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6410 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6411 serious problem, Exim now writes to the main and panic logs when this
6412 happens, with details of the failure. It then refuses to accept the
6413 incoming message, giving the message "spool directory problem" or "log
6414 directory problem" with a 421 code for SMTP messages.
6416 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6417 0, 1, and 2 exist, because some OS complain for execs without them (see
6418 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6419 descriptors. However, the code omitted to check that the open succeeded,
6420 causing mysterious errors if for some reason the permissions on /dev/null
6421 got screwed. Now Exim writes a message to the main and panic logs, and
6422 bombs out if it can't open /dev/null.
6424 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6425 interact so that it is all more flexible. It is supposed to remain
6426 backwards compatible. Also added extra_local_interfaces.
6428 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6429 to bomb out with an assertion failure - to the client this appears as a
6430 connection drop. This problem occurs in the part of the code that was taken
6431 from the Samba project. Fortunately, the assertion is in a very simple
6432 function, so I have fixed this by reproducing the function inline in the
6433 one place where it is called, and arranging for authentication to fail
6434 instead of killing the process with assert().
6436 75. The SPA client code was not working when the server requested OEM rather
6437 than Unicode encoding.
6439 76. Added code to make require_files with a specific uid setting more usable in
6440 the case where statting the file as root fails - usually a non-root-mounted
6441 NFS file system. When this happens and the failure is EACCES, Exim now
6442 forks a subprocess and does the per-uid checking as the relevant uid.
6444 77. Added process_log_path.
6446 78. If log_file_path was not explicitly set, a setting of check_log_space or
6447 check_log_inodes was ignored.
6449 79. If a space check for the spool or log partitions fails, the incident is now
6450 logged. Of course, in the latter case the data may get lost...
6452 80. Added the %p formatting code to string_format() so that it can be used to
6453 print addresses in debug_print(). Adjusted all the address printing in the
6454 debugging in store.c to use %p rather than %d.
6456 81. There was a concern that a line of code in smtp_in.c could overflow a
6457 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6458 initially expressed, the concern was not well-founded, because trailing
6459 spaces are removed early. However, if the trailing spaces were followed by
6460 a NULL, they did not get removed, so the overflow was possible. Two fixes
6463 (a) I re-wrote the offending code in a cleaner fashion.
6464 (b) If an incoming SMTP command contains a NULL character, it is rejected
6467 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6468 runs initgroups(), so that if the Exim user is in any additional groups,
6469 they will be used during message reception.
6475 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob