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.
143 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
146 JH/02 The smtp transport option "multi_domain" is now expanded.
148 JH/03 The smtp transport now requests PRDR by default, if the server offers
151 JH/04 Certificate name checking on server certificates, when exim is a client,
152 is now done by default. The transport option tls_verify_cert_hostnames
153 can be used to disable this per-host. The build option
154 EXPERIMENTAL_CERTNAMES is withdrawn.
156 JH/05 The value of the tls_verify_certificates smtp transport and main options
157 default to the word "system" to access the system default CA bundle.
158 For GnuTLS, only version 3.0.20 or later.
160 JH/06 Verification of the server certificate for a TLS connection is now tried
161 (but not required) by default. The verification status is now logged by
162 default, for both outbound TLS and client-certificate supplying inbound
165 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
168 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
169 Status Notification (bounce) messages are now MIME format per RFC 3464.
170 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
171 under the control of the dsn_advertise_hosts option, and routers may
172 have a dsn_lasthop option.
174 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
175 default, modifiable by a malware= option. The list separator for
176 the options can now be changed in the usual way. Bug 68.
178 JH/10 The smtp_receive_timeout main option is now expanded before use.
180 JH/11 The incoming_interface log option now also enables logging of the
181 local interface on delivery outgoing connections.
183 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
184 if the interface and destination host and port all match.
186 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
189 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
190 Patch from Andrew Lewis.
192 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
193 now supports optional time-restrictions, weighting, and priority
194 modifiers per server. Patch originally by <rommer@active.by>.
196 JH/16 The spamd_address main option now supports a mixed list of local
197 and remote servers. Remote servers can be IPv6 addresses, and
198 specify a port-range.
200 JH/17 Bug 68: The spamd_address main option now supports an optional
201 timeout value per server.
203 JH/18 Bug 1581: Router and transport options headers_add/remove can
204 now have the list separator specified.
206 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
209 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
212 JH/21 Support for the A6 type of dns record is withdrawn.
214 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
215 rather than the verbs used.
217 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
218 from 255 to 1024 chars.
220 JH/24 Verification callouts now attempt to use TLS by default.
222 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
223 are generic router options now. The defaults didn't change.
225 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
226 Original patch from Alexander Shikoff, worked over by JH.
228 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
229 files if ZCAT_COMMAND is not executable.
231 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
233 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
235 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
236 Normally benign, it bites when the pair was led to by a CNAME;
237 modern usage is to not canoicalize the domain to a CNAME target
238 (and we were inconsistent anyway for A-only vs AAAA+A).
240 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
242 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
243 when evaluating $sender_host_dnssec.
245 JH/31 Check the HELO verification lookup for DNSSEC, adding new
246 $sender_helo_dnssec variable.
248 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
250 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
252 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
254 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
255 documented as working, but never had. Support all but $spam_report.
257 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
258 added for tls authenticator.
263 TL/01 When running the test suite, the README says that variables such as
264 no_msglog_check are global and can be placed anywhere in a specific
265 test's script, however it was observed that placement needed to be near
266 the beginning for it to behave that way. Changed the runtest perl
267 script to read through the entire script once to detect and set these
268 variables, reset to the beginning of the script, and then run through
269 the script parsing/test process like normal.
271 TL/02 The BSD's have an arc4random API. One of the functions to induce
272 adding randomness was arc4random_stir(), but it has been removed in
273 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
274 function when detected.
276 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
277 cause callback expansion.
279 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
280 syntax errors in an expansion can be treated as a string instead of
281 logging or causing an error, due to the internal use of bool_lax
282 instead of bool when processing it.
284 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
285 server certificates when making smtp deliveries.
287 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
289 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
291 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
293 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
294 Merged patch from Sebastian Wiedenroth.
296 JH/05 Fix results-pipe from transport process. Several recipients, combined
297 with certificate use, exposed issues where response data items split
298 over buffer boundaries were not parsed properly. This eventually
299 resulted in duplicates being sent. This issue only became common enough
300 to notice due to the introduction of conection certificate information,
301 the item size being so much larger. Found and fixed by Wolfgang Breyha.
303 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
304 size buffer was used, resulting in syntax errors when an expansion
307 JH/07 Add support for directories of certificates when compiled with a GnuTLS
308 version 3.3.6 or later.
310 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
311 is EXPERIMENTAL_EVENT, the main-configuration and transport options
312 both become "event_action", the variables become $event_name, $event_data
313 and $event_defer_errno. There is a new variable $verify_mode, usable in
314 routers, transports and related events. The tls:cert event is now also
315 raised for inbound connections, if the main configuration event_action
318 TL/06 In test suite, disable OCSP for old versions of openssl which contained
319 early OCSP support, but no stapling (appears to be less than 1.0.0).
321 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
322 server certificate names available under the smtp transport option
323 "tls_verify_cert_hostname" now do not permit multi-component wildcard
326 JH/10 Time-related extraction expansions from certificates now use the main
327 option "timezone" setting for output formatting, and are consistent
328 between OpenSSL and GnuTLS compilations. Bug 1541.
330 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
331 encoded parameter in the incoming message. Bug 1558.
333 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
334 include certificate info, eximon was claiming there were spoolfile
337 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
339 JH/14 Log delivery-related information more consistently, using the sequence
340 "H=<name> [<ip>]" wherever possible.
342 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
343 are problematic for Debian distribution, omit them from the release
346 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
348 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
351 JH/17 Fix a null-indirection in certextract expansions when a nondefault
352 output list separator was used.
357 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
358 checkers that were complaining about end of non-void function with no
361 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
362 This was a regression intruduced in 4.83 by another bugfix.
364 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
366 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
367 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
373 TF/01 Correctly close the server side of TLS when forking for delivery.
375 When a message was received over SMTP with TLS, Exim failed to clear up
376 the incoming connection properly after forking off the child process to
377 deliver the message. In some situations the subsequent outgoing
378 delivery connection happened to have the same fd number as the incoming
379 connection previously had. Exim would try to use TLS and fail, logging
380 a "Bad file descriptor" error.
382 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
383 utilities have not been installed.
385 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
386 temporary space as the ACL may create new global variables.
388 TL/01 LDAP support uses per connection or global context settings, depending
389 upon the detected version of the libraries at build time.
391 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
392 to extract and use the src ip:port in logging and expansions as if it
393 were a direct connection from the outside internet. PPv2 support was
394 updated based on HAProxy spec change in May 2014.
396 JH/02 Add ${listextract {number}{list}{success}{fail}}.
398 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
399 Properly escape header and check for NULL return.
401 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
404 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
406 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
407 characters in header names, implemented as a verify condition.
408 Contributed by Michael Fischer v. Mollard.
410 TL/05 Rename SPF condition results err_perm and err_temp to standardized
411 results permerror and temperror. Previous values are deprecated but
412 still accepted. In a future release, err_perm and err_temp will be
413 completely removed, which will be a backward incompatibility if the
414 ACL tests for either of these two old results. Patch contributed by
415 user bes-internal on the mailing list.
417 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
419 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
420 selectors, in both main and reject logs.
422 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
425 JH/07 Add malware type "sock" for talking to simple daemon.
427 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
429 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
430 routers/transports under cutthrough routing.
432 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
433 numbers. Touch up "bool" conditional to keep the same definition.
435 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
437 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
439 JH/12 Expand items in router/transport headers_add or headers_remove lists
440 individually rather than the list as a whole. Bug 1452.
442 Required for reasonable handling of multiple headers_ options when
443 they may be empty; requires that headers_remove items with embedded
444 colons must have them doubled (or the list-separator changed).
446 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
447 view the policy declared in the DMARC record. Currently, $dmarc_status
448 is a combined value of both the record presence and the result of the
451 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
453 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
454 dnslookup router and the smtp transport (applying to the forward
457 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
458 of ldap servers used for a specific lookup. Patch provided by Heiko
461 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
462 New variable $lookup_dnssec_authenticated for observability.
464 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
465 Patch submitted by Lars Timman.
467 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
469 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
470 Requires trusted mode and valid format message id, aborts otherwise.
471 Patch contributed by Heiko Schlichting.
473 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
474 certextract with support for various fields. Bug 1358.
476 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
477 is requested by default, modifiable by smtp transport option
480 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
481 operate on certificate variables to give certificate fingerprints
482 Also new ${sha256:cert_variable}.
484 JH/23 The PRDR feature is moved from being Experimental into the mainline.
486 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
487 Christian Aistleitner.
489 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
491 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
492 file. Patch from Wolfgang Breyha.
494 JH/25 Expand the coverage of the delivery $host and $host_address to
495 client authenticators run in verify callout. Bug 1476.
497 JH/26 Port service names are now accepted for tls_on_connect_ports, to
498 align with daemon_smtp_ports. Bug 72.
500 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
501 support and error reporting did not work properly.
503 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
504 and is readable. Patch from Andrew Colin Kissa.
506 TL/14 Enhance documentation of ${run expansion and how it parses the
507 commandline after expansion, particularly in the case when an
508 unquoted variable expansion results in an empty value.
510 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
512 PP/02 Fix internal collision of T_APL on systems which support RFC3123
513 by renaming away from it. Addresses GH issue 15, reported by
516 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
518 TL/15 SECURITY: prevent double expansion in math comparison functions
519 (can expand unsanitized data). Not remotely exploitable.
526 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
528 PP/02 Make -n do something, by making it not do something.
529 When combined with -bP, the name of an option is not output.
531 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
534 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
535 $sender_host_name and config options to manage this, and basic check
538 PP/05 DSCP support for outbound connections and control modifier for inbound.
540 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
541 (Only plugin which currently uses this is kerberos4, which nobody should
542 be using, but we should make it available and other future plugins might
543 conceivably use it, even though it would break NAT; stuff *should* be
544 using channel bindings instead).
546 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
547 name; added for Sendmail compatibility; requires admin caller.
548 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
549 just ignore it); requires trusted caller.
550 Also parse but ignore: -Ac -Am -X<logfile>
553 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
555 TL/02 Add +smtp_confirmation as a default logging option.
557 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
558 Patch by Magnus Holmgren from 2007-02-20.
560 TL/04 Bugzilla 1281 - Spec typo.
561 Bugzilla 1283 - Spec typo.
562 Bugzilla 1290 - Spec grammar fixes.
564 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
566 TL/06 Add Experimental DMARC support using libopendmarc libraries.
568 TL/07 Fix an out of order global option causing a segfault. Reported to dev
569 mailing list by by Dmitry Isaikin.
571 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
573 JH/02 Support "G" suffix to numbers in ${if comparisons.
575 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
577 NM/01 Bugzilla 1197 - Spec typo
578 Bugzilla 1196 - Spec examples corrections
580 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
582 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
583 gnutls_enable_pkcs11, but renamed to more accurately indicate its
586 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
587 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
589 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
590 "acl {{name}{arg}...}", and optional args on acl condition
593 JH/05 Permit multiple router/transport headers_add/remove lines.
595 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
597 JH/07 Avoid using a waiting database for a single-message-only transport.
598 Performance patch from Paul Fisher. Bugzilla 1262.
600 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
603 JH/09 Add $headers_added variable, with content from use of ACL modifier
604 add_header (but not yet added to the message). Bugzilla 199.
606 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
607 Pulled from Bugzilla 817 by Wolfgang Breyha.
609 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
611 (nb: this is the same fix as in Exim 4.80.1)
613 JH/11 Add A= logging on delivery lines, and a client_set_id option on
616 JH/12 Add optional authenticated_sender logging to A= and a log_selector
619 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
621 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
622 advertise SMTP AUTH mechanism to us, instead of a generic
623 protocol violation error. Also, make Exim more robust to bad
624 data from the Dovecot auth socket.
626 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
628 When a queue runner is handling a message, Exim first routes the
629 recipient addresses, during which it prunes them based on the retry
630 hints database. After that it attempts to deliver the message to
631 any remaining recipients. It then updates the hints database using
634 So if a recipient address works intermittently, it can get repeatedly
635 deferred at routing time. The retry hints record remains fresh so the
636 address never reaches the final cutoff time.
638 This is a fairly common occurrence when a user is bumping up against
639 their storage quota. Exim had some logic in its local delivery code
640 to deal with this. However it did not apply to per-recipient defers
641 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
643 This change adds a proper retry rule check during routing so that the
644 final cutoff time is checked against the message's age. We only do
645 this check if there is an address retry record and there is not a
646 domain retry record; this implies that previous attempts to handle
647 the address had the retry_use_local_parts option turned on. We use
648 this as an approximation for the destination being like a local
649 delivery, as in LMTP.
651 I suspect this new check makes the old local delivery cutoff check
652 redundant, but I have not verified this so I left the code in place.
654 TF/02 Correct gecos expansion when From: is a prefix of the username.
656 Test 0254 submits a message to Exim with the header
660 When I ran the test suite under the user fanf2, Exim expanded
661 the header to contain my full name, whereas it should have added
662 a Resent-Sender: header. It erroneously treats any prefix of the
663 username as equal to the username.
665 This change corrects that bug.
667 GF/01 DCC debug and logging tidyup
668 Error conditions log to paniclog rather than rejectlog.
669 Debug lines prefixed by "DCC: " to remove any ambiguity.
671 TF/03 Avoid unnecessary rebuilds of lookup-related code.
673 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
674 Bug spotted by Jeremy Harris; was flawed since initial commit.
675 Would have resulted in OCSP responses post-SNI triggering an Exim
676 NULL dereference and crash.
678 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
680 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
681 Bug detection, analysis and fix by Samuel Thibault.
682 Bugzilla 1331, Debian bug #698092.
684 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
686 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
687 Server implementation by Todd Lyons, client by JH.
688 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
689 config variable "prdr_enable" controls whether the server
690 advertises the facility. If the client requests PRDR a new
691 acl_data_smtp_prdr ACL is called once for each recipient, after
692 the body content is received and before the acl_smtp_data ACL.
693 The client is controlled by bolth of: a hosts_try_prdr option
694 on the smtp transport, and the server advertisement.
695 Default client logging of deliveries and rejections involving
696 PRDR are flagged with the string "PRDR".
698 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
699 fclose(). Diagnosis by Todd Lyons.
701 PP/17 Update configure.default to handle IPv6 localhost better.
702 Patch by Alain Williams (plus minor tweaks).
705 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
706 This is now consistent with GnuTLS, and is now documented: the
707 previous undocumented portable approach to treating the option as
708 unset was to force an expansion failure. That still works, and
709 an empty string is now equivalent.
711 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
712 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
713 not performing validation itself.
715 PP/20 Added force_command boolean option to pipe transport.
716 Patch from Nick Koston, of cPanel Inc.
718 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
721 TF/04 Added udpsend ACL modifer and hexquote expansion operator
723 PP/21 Fix eximon continuous updating with timestamped log-files.
724 Broken in a format-string cleanup in 4.80, missed when I repaired the
725 other false fix of the same issue.
726 Report and fix from Heiko Schlichting.
729 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
730 Report from Prashanth Katuri.
732 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
733 It's SecureTransport, so affects any MacOS clients which use the
734 system-integrated TLS libraries, including email clients.
736 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
737 using a MIME ACL for non-SMTP local injection.
738 Report and assistance in diagnosis by Warren Baker.
740 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
742 JH/16 Fix comparisons for 64b. Bugzilla 1385.
744 TL/09 Add expansion variable $authenticated_fail_id to keep track of
745 last id that failed so it may be referenced in subsequent ACL's.
747 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
750 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
751 ldap library initialization, allowing self-signed CA's to be
752 used. Also properly sets require_cert option later in code by
753 using NULL (global ldap config) instead of ldap handle (per
754 session). Bug diagnosis and testing by alxgomz.
756 TL/12 Enhanced documentation in the ratelimit.pl script provided in
757 the src/util/ subdirectory.
759 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
760 renamed to Transport Post Delivery Action by Jeremy Harris, as
763 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
764 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
765 redis_servers = needs to be configured which will be used by the redis
766 lookup. Patch from Warren Baker, of The Packet Hub.
768 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
770 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
771 hostname or reverse DNS when processing a host list. Used suggestions
772 from multiple comments on this bug.
774 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
776 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
777 Missed a few lines, added it to make the runtest require no keyboard
780 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
781 contains upper case chars. Make router use caseful_local_part.
783 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
784 support when GnuTLS has been built with p11-kit.
790 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
792 This, or similar/improved, will also be change PP/11 of 4.82.
798 PP/01 Handle short writes when writing local log-files.
799 In practice, only affects FreeBSD (8 onwards).
800 Bugzilla 1053, with thanks to Dmitry Isaikin.
802 NM/01 Bugzilla 949 - Documentation tweak
804 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
807 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
809 PP/02 Implemented gsasl authenticator.
811 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
813 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
814 `pkg-config foo` for cflags/libs.
816 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
817 with rest of GSASL and with heimdal_gssapi.
819 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
820 `pkg-config foo` for cflags/libs for the TLS implementation.
822 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
823 properties get this fed in as external SSF. A number of robustness
824 and debugging improvements to the cyrus_sasl authenticator.
826 PP/08 cyrus_sasl server now expands the server_realm option.
828 PP/09 Bugzilla 1214 - Log authentication information in reject log.
829 Patch by Jeremy Harris.
831 PP/10 Added dbmjz lookup type.
833 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
835 PP/12 MAIL args handles TAB as well as SP, for better interop with
836 non-compliant senders.
837 Analysis and variant patch by Todd Lyons.
839 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
840 Bug report from Lars Müller <lars@samba.org> (via SUSE),
841 Patch from Dirk Mueller <dmueller@suse.com>
843 PP/13 tls_peerdn now print-escaped for spool files.
844 Observed some $tls_peerdn in wild which contained \n, which resulted
845 in spool file corruption.
847 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
848 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
849 or write after TLS renegotiation, which otherwise led to messages
852 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
853 as a tracking header (ie: a signed header comes before the signature).
854 Patch from Wolfgang Breyha.
856 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
857 comma-sep list; embedded commas doubled.
859 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
861 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
863 Report and patch from Dmitry Banschikov.
865 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
866 Removed SSL_clear() after SSL_new() which led to protocol negotiation
867 failures. We appear to now support TLS1.1+ with Exim.
869 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
870 lets Exim select keys and certificates based upon TLS SNI from client.
871 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
872 before an outbound SMTP session. New log_selector, +tls_sni.
874 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
875 NULL dereference. Report and patch from Alun Jones.
877 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
878 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
879 Not seeing resolver debug output on NetBSD, but suspect this is a
880 resolver implementation change.
882 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
883 Left warnings. Added "eximon gdb" invocation mode.
885 PP/21 Defaulting "accept_8bitmime" to true, not false.
887 PP/22 Added -bw for inetd wait mode support.
889 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
890 locate the relevant includes and libraries. Made this the default.
892 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
893 Bugzilla 1246, report and most of solution from Tomasz Kusy.
895 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
896 This may cause build issues on older platforms.
898 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
899 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
900 gnutls_require_mac & gnutls_require_protocols (no longer supported).
901 Added SNI support via GnuTLS too.
902 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
904 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
906 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
907 Applied second patch from Janne, implementing suggestion to default
908 multiple-strings-in-record handling to match SPF spec.
910 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
912 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
913 read-only, out of scope).
914 Patch from Wolfgang Breyha, report from Stuart Northfield.
916 PP/29 Fix three issues highlighted by clang analyser static analysis.
917 Only crash-plausible issue would require the Cambridge-specific
918 iplookup router and a misconfiguration.
919 Report from Marcin Mirosław.
921 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
923 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
924 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
925 As part of this, removing so much warning spew let me fix some minor
926 real issues in debug logging.
928 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
929 assignment on my part. Fixed.
931 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
932 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
933 Janne Snabb (who went above and beyond: thank you).
935 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
936 string otherwise requires a connection and a bunch more work and it's
937 relatively easy to get wrong. Should also expose TLS library linkage
940 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
941 64-bit ${eval} (JH/03).
943 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
944 GNU libc to support some of the 64-bit stuff, should not lead to
945 conflicts. Defined before os.h is pulled in, so if a given platform
946 needs to override this, it can.
948 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
949 protection layer was required, which is not implemented.
950 Bugzilla 1254, patch from Wolfgang Breyha.
952 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
953 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
954 tls_dhparam take prime identifiers. Also unbreak combination of
955 OpenSSL+DH_params+TLSSNI.
957 PP/39 Disable SSLv2 by default in OpenSSL support.
963 PP/01 Solaris build fix for Oracle's LDAP libraries.
964 Bugzilla 1109, patch from Stephen Usher.
966 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
968 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
971 TF/02 Fix a couple more cases where we did not log the error message
972 when unlink() failed. See also change 4.74-TF/03.
974 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
975 lock up or crash if it happened to be inside a call to libc when it
976 got a SIGUSR1 from exiwhat.
978 The SIGUSR1 handler appends the current process status to the process
979 log which is later printed by exiwhat. It used to use the general
980 purpose logging code to do this, but several functions it calls are
981 not safe for signals.
983 The new output code in the SIGUSR1 handler is specific to the process
984 log, and simple enough that it's easy to inspect for signal safety.
985 Removing some special cases also simplifies the general logging code.
986 Removing the spurious timestamps from the process log simplifies
989 TF/04 Improved ratelimit ACL condition.
991 The /noupdate option has been deprecated in favour of /readonly which
992 has clearer semantics. The /leaky, /strict, and /readonly update modes
993 are mutually exclusive. The update mode is no longer included in the
994 database key; it just determines when the database is updated. (This
995 means that when you upgrde Exim will forget old rate measurements.)
997 Exim now checks that the per_* options are used with an update mode that
998 makes sense for the current ACL. For example, when Exim is processing a
999 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1000 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1001 must specify per_mail/readonly. If you omit the update mode it defaults to
1002 /leaky where that makes sense (as before) or /readonly where required.
1004 The /noupdate option is now undocumented but still supported for
1005 backwards compatibility. It is equivalent to /readonly except that in
1006 ACLs where /readonly is required you may specify /leaky/noupdate or
1007 /strict/noupdate which are treated the same as /readonly.
1009 A useful new feature is the /count= option. This is a generalization
1010 of the per_byte option, so that you can measure the throughput of other
1011 aggregate values. For example, the per_byte option is now equivalent
1012 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1014 The per_rcpt option has been generalized using the /count= mechanism
1015 (though it's more complicated than the per_byte equivalence). When it is
1016 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1017 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1018 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1019 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1020 Note that using per_rcpt with a non-readonly update mode in more than
1021 one ACL will cause the recipients to be double-counted. (The per_mail
1022 and per_byte options don't have this problem.)
1024 The handling of very low rates has changed slightly. If the computed rate
1025 is less than the event's count (usually one) then this event is the first
1026 after a long gap. In this case the rate is set to the same as this event's
1027 count, so that the first message of a spam run is counted properly.
1029 The major new feature is a mechanism for counting the rate of unique
1030 events. The new per_addr option counts the number of different
1031 recipients that someone has sent messages to in the last time period. It
1032 behaves like per_rcpt if all the recipient addresses are different, but
1033 duplicate recipient addresses do not increase the measured rate. Like
1034 the /count= option this is a general mechanism, so the per_addr option
1035 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1036 example, measure the rate that a client uses different sender addresses
1037 with the options per_mail/unique=$sender_address. There are further
1038 details in the main documentation.
1040 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1042 TF/06 Removed a few PCRE remnants.
1044 TF/07 Automatically extract Exim's version number from tags in the git
1045 repository when doing development or release builds.
1047 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1048 Bugzilla 879. Patch from Paul Fisher.
1050 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1051 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1054 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1056 PP/05 Variable $av_failed, true if the AV scanner deferred.
1057 Bugzilla 1078. Patch from John Horne.
1059 PP/06 Stop make process more reliably on build failure.
1060 Bugzilla 1087. Patch from Heiko Schlittermann.
1062 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1063 Bugzilla 1089. Patch from Heiko Schlittermann.
1065 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1066 Bugzilla 1131. Patch from Holger Weiß.
1068 PP/09 Handle IPv6 addresses with SPF.
1069 Bugzilla 860. Patch from Wolfgang Breyha.
1071 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1073 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1076 PP/11 match_* no longer expand right-hand-side by default.
1077 New compile-time build option, EXPAND_LISTMATCH_RHS.
1078 New expansion conditions, "inlist", "inlisti".
1080 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1082 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1084 PP/14 fix log_write() format string regression from TF/03.
1085 Bugzilla 1152. Patch from Dmitry Isaikin.
1091 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1093 PP/02 Harmonised TLS library version reporting; only show if debugging.
1094 Layout now matches that introduced for other libraries in 4.74 PP/03.
1096 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1098 PP/04 New "dns_use_edns0" global option.
1100 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1103 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1104 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1106 TK/01 Updated PolarSSL code to 0.14.2.
1107 Bugzilla 1097. Patch from Andreas Metzler.
1109 PP/07 Catch divide-by-zero in ${eval:...}.
1110 Fixes bugzilla 1102.
1112 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1115 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1116 format-string attack -- SECURITY: remote arbitrary code execution.
1118 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1119 time unintentionally subject to list matching rules, letting the header
1120 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1121 arbitrary string expansion). This allowed for information disclosure.
1123 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1124 INT_MIN/-1 -- value coerced to INT_MAX.
1130 NM/01 Workround for PCRE version dependency in version reporting
1133 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1134 This fixes portability to compilers other than gcc, notably
1135 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1137 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1138 makefiles for portability to HP-UX and POSIX correctness.
1140 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1141 Also via indented variable definition in the Makefile.
1142 (Debugging by Oliver Heesakkers).
1144 PP/02 Restore caching of spamd results with expanded spamd_address.
1145 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1147 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1148 Improves build reliability. Fix from: Frank Elsner
1150 NM/02 Fix wide character breakage in the rfc2047 coding
1151 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1153 NM/03 Allow underscore in dnslist lookups
1154 Fixes bug 1026. Patch from Graeme Fowler
1156 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1157 Code patches from Adam Ciarcinski of NetBSD.
1159 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1162 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1163 is logged, to avoid truncation. Patch from John Horne.
1165 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1166 Patch from Jakob Hirsch.
1168 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1169 SQL string expansion failure details.
1170 Patch from Andrey Oktyabrski.
1172 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1173 Patch from Simon Arlott.
1175 PP/09 New lookups functionality failed to compile on old gcc which rejects
1176 extern declarations in function scope.
1177 Patch from Oliver Fleischmann
1179 PP/10 Use sig_atomic_t for flags set from signal handlers.
1180 Check getgroups() return and improve debugging.
1181 Fixed developed for diagnosis in bug 927 (which turned out to be
1184 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1185 Patch from Mark Zealey.
1187 PP/12 Bugzilla 1056: Improved spamd server selection.
1188 Patch from Mark Zealey.
1190 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1191 Based on patch from Heiko Schlittermann.
1193 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1194 Patch from Uwe Doering, sign-off by Michael Haardt.
1196 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1197 variable declaration deep within a block. Bug and patch from
1200 PP/15 lookups-Makefile IRIX compatibilty coercion.
1202 PP/16 Make DISABLE_DKIM build knob functional.
1204 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1205 Patch by Simon Arlott
1207 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1208 variable argument macros. Our copy now differs from upstream.
1214 TF/01 Failure to get a lock on a hints database can have serious
1215 consequences so log it to the panic log.
1217 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1218 controlled using the smtp_confirmation log selector.
1220 TF/03 Include the error message when we fail to unlink a spool file.
1222 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1223 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1224 for maintaining out-of-tree patches for some time.
1226 PP/01 Bugzilla 139: Documentation and portability issues.
1227 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1228 Handle per-OS dynamic-module compilation flags.
1230 PP/02 Let /dev/null have normal permissions.
1231 The 4.73 fixes were a little too stringent and complained about the
1232 permissions on /dev/null. Exempt it from some checks.
1233 Reported by Andreas M. Kirchwitz.
1235 PP/03 Report version information for many libraries, including
1236 Exim version information for dynamically loaded libraries. Created
1237 version.h, now support a version extension string for distributors
1238 who patch heavily. Dynamic module ABI change.
1240 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1241 privilege escalation vulnerability whereby the Exim run-time user
1242 can cause root to append content of the attacker's choosing to
1245 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1248 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1249 If dropping privileges for untrusted macros, we disabled normal logging
1250 on the basis that it would fail; for the Exim run-time user, this is not
1251 the case, and it resulted in successful deliveries going unlogged.
1252 Fixed. Reported by Andreas Metzler.
1258 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1259 only prepend for the Resent-* case. Fixes regression introduced in
1260 Exim 4.70 by NM/22 for Bugzilla 607.
1262 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1263 increasing numbers of administrators be bitten by this.
1265 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1267 PP/03 Bugzilla 994: added openssl_options main configuration option.
1269 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1271 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1273 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1275 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1276 without a peer certificate, leading to a segfault because of an
1277 assumption that peers always have certificates. Be a little more
1278 paranoid. Problem reported by Martin Tscholak.
1280 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1281 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1282 NB: ClamAV planning to remove STREAM in "middle of 2010".
1283 CL also introduces -bmalware, various -d+acl logging additions and
1284 more caution in buffer sizes.
1286 PP/09 Implemented reverse_ip expansion operator.
1288 PP/10 Bugzilla 937: provide a "debug" ACL control.
1290 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1292 PP/12 Bugzilla 973: Implement --version.
1294 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1296 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1298 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1300 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1301 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1302 ignore trailing whitespace.
1304 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1306 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1309 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1310 Notification from Dr Andrew Aitchison.
1312 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1313 ExtendedDetectionInfo response format.
1314 Notification from John Horne.
1316 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1319 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1320 XSL and documented dependency on system catalogs, with examples of how
1323 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1326 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1327 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1328 configuration file which is writeable by the Exim user or group.
1330 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1331 of configuration files to cover files specified with the -C option if
1332 they are going to be used with root privileges, not just the default
1335 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1336 option (effectively making it always true).
1338 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1339 files to be used while preserving root privileges.
1341 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1342 that rogue child processes cannot use them.
1344 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1345 run-time user, instead of root.
1347 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1348 Exim run-time user without dropping privileges.
1350 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1351 result string, instead of calling string_vformat() twice with the same
1354 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1355 for other users. Others should always drop root privileges if they use
1356 -C on the command line, even for a whitelisted configure file.
1358 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1360 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1366 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1367 $data_path, and $header_path variables; fixed documentation bugs and
1370 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1371 exipick to access non-standard spools, including the "frozen" queue
1374 NM/01 Bugzilla 965: Support mysql stored procedures.
1375 Patch from Alain Williams
1377 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1379 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1380 Patch from Andreas Metzler
1382 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1383 Patch from Kirill Miazine
1385 NM/05 Bugzilla 671: Added umask to procmail example.
1387 JJ/03 installed exipick 20100323.0, fixing doc bug
1389 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1390 directory. Notification and patch from Dan Rosenberg.
1392 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1394 TK/02 Improve log output when DKIM signing operation fails.
1396 MH/01 Treat the transport option dkim_domain as a colon separated
1397 list, not as a single string, and sign the message with each element,
1398 omitting multiple occurences of the same signer.
1400 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1401 Bugzilla 985, 986. Patch by Simon Arlott
1403 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1404 Patch by Simon Arlott
1406 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1407 MBX locking. Notification from Dan Rosenberg.
1413 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1415 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1417 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1419 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1421 NM/04 Bugzilla 915: Flag broken perl installation during build.
1427 TK/01 Added patch by Johannes Berg that expands the main option
1428 "spamd_address" if it starts with a dollar sign.
1430 TK/02 Write list of recipients to X-Envelope-Sender header when building
1431 the mbox-format spool file for content scanning (suggested by Jakob
1434 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1435 (http://www.dcc-servers.net/) support via dccifd. Activated by
1436 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1438 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1439 by Mark Daniel Reidel <mr@df.eu>.
1441 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1442 When building exim an external PCRE library is now needed -
1443 PCRE is a system library on the majority of modern systems.
1444 See entry on PCRE_LIBS in EDITME file.
1446 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1447 conversation. Added nologin parameter to request.
1448 Patch contributed by Kirill Miazine.
1450 TF/01 Do not log submission mode rewrites if they do not change the address.
1452 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1454 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1455 log files in place. Contributed by Roberto Lima.
1457 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1459 TF/03 Bugzilla 615: When checking the local_parts router precondition
1460 after a local_part_suffix or local_part_prefix option, Exim now
1461 does not use the address's named list lookup cache, since this
1462 contains cached lookups for the whole local part.
1464 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1465 Robert Millan. Documentation is in experimental-spec.txt.
1467 TF/04 Bugzilla 668: Fix parallel build (make -j).
1469 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1471 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1472 Patch provided by Jan Srzednicki.
1474 TF/05 Leading white space used to be stripped from $spam_report which
1475 wrecked the formatting. Now it is preserved.
1477 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1478 that they are available at delivery time.
1480 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1482 TF/08 TLS error reporting now respects the incoming_interface and
1483 incoming_port log selectors.
1485 TF/09 Produce a more useful error message if an SMTP transport's hosts
1486 setting expands to an empty string.
1488 NM/06 Bugzilla 744: EXPN did not work under TLS.
1489 Patch provided by Phil Pennock.
1491 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1492 Patch provided by Richard Godbee.
1494 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1495 acl_smtp_notquit, added index entry.
1497 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1498 Patch provided by Eugene Bujak.
1500 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1501 accept(). Patch provided by Maxim Dounin.
1503 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1504 Patch provided by Phil Pennock.
1506 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1508 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1509 Patch provided by Brad "anomie" Jorsch.
1511 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1512 Patch provided by Dean Brooks.
1514 TK/05 Add native DKIM support (does not depend on external libraries).
1516 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1517 Patch provided by Graeme Fowler.
1519 NM/16 Bugzilla 851: Documentation example syntax fix.
1521 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1523 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1526 NM/19 Bugzilla 745: TLS version reporting.
1527 Patch provided by Phil Pennock.
1529 NM/20 Bugzilla 167: bool: condition support.
1530 Patch provided by Phil Pennock.
1532 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1533 clients. Patch provided by Phil Pennock.
1535 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1536 Patch provided by Brad "anomie" Jorsch.
1538 NM/23 Bugzilla 687: Fix misparses in eximstats.
1539 Patch provided by Heiko Schlittermann.
1541 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1542 Patch provided by Heiko Schlittermann.
1544 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1545 plus update to original patch.
1547 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1549 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1550 Patch provided by David Brownlee.
1552 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1554 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1556 NM/30 Bugzilla 888: TLS documentation bugfixes.
1558 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1560 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1561 Unlike the original bugzilla I have changed all shell scripts in src tree.
1563 NM/33 Bugzilla 898: Transport filter timeout fix.
1564 Patch by Todd Rinaldo.
1566 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1567 Patch by Serge Demonchaux.
1569 NM/35 Bugzilla 39: Base64 decode bug fixes.
1570 Patch by Jakob Hirsch.
1572 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1574 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1576 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1578 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1584 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1585 ALT-N's libdkim that I have put here:
1586 http://duncanthrax.net/exim-experimental/
1588 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1589 were called 'true' and it seems that C99 defines that as a reserved
1590 keyword to be used with 'bool' variable types. That means you could
1591 not include C99-style headers which use bools without triggering
1592 build errors in sieve.c.
1594 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1595 as mailq or other aliases. Changed the --help handling significantly
1596 to do whats expected. exim_usage() emits usage/help information.
1598 SC/01 Added the -bylocaldomain option to eximstats.
1600 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1602 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1604 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1610 PH/01 Another patch from the Sieve maintainer.
1612 PH/02 When an IPv6 address is converted to a string for single-key lookup
1613 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1614 dots are used instead of colons so that keys in lsearch files need not
1615 contain colons. This was done some time before quoting was made available
1616 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1617 (notated using the quote facility) so as to distinguish them from IPv4
1618 keys. This meant that lookups for IP addresses in host lists did not work
1619 for iplsearch lookups.
1621 This has been fixed by arranging for IPv6 addresses to be expressed with
1622 colons if the lookup type is iplsearch. This is not incompatible, because
1623 previously such lookups could never work.
1625 The situation is now rather anomolous, since one *can* have colons in
1626 ordinary lsearch keys. However, making the change in all cases is
1627 incompatible and would probably break a number of configurations.
1629 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1632 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1633 conversion specification without a maximum field width, thereby enabling
1634 a rogue spamd server to cause a buffer overflow. While nobody in their
1635 right mind would setup Exim to query an untrusted spamd server, an
1636 attacker that gains access to a server running spamd could potentially
1637 exploit this vulnerability to run arbitrary code as the Exim user.
1639 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1640 $primary_hostname instead of what libspf2 thinks the hosts name is.
1642 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1643 a directory entry by the name of the lookup key. Previously, if a
1644 symlink pointed to a non-existing file or a file in a directory that
1645 Exim lacked permissions to read, a lookup for a key matching that
1646 symlink would fail. Now it is enough that a matching directory entry
1647 exists, symlink or not. (Bugzilla 503.)
1649 PH/03 The body_linecount and body_zerocount variables are now exported in the
1652 PH/04 Added the $dnslist_matched variable.
1654 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1655 This means they are set thereafter only if the connection becomes
1658 PH/06 Added the client_condition to authenticators so that some can be skipped
1659 by clients under certain conditions.
1661 PH/07 The error message for a badly-placed control=no_multiline_responses left
1662 "_responses" off the end of the name.
1664 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1666 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1667 (without spaces) instead of just copying the configuration text.
1669 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1671 PH/11 Added $max_received_linelength.
1673 PH/12 Added +ignore_defer and +include_defer to host lists.
1675 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1676 way in which PCRE > 7.0 is built.
1678 PH/14 Implemented queue_only_load_latch.
1680 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1681 MAIL command. The effect was to mangle the value on 64-bit systems.
1683 PH/16 Another patch from the Sieve maintainer.
1685 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1687 PH/18 If a system quota error occurred while trying to create the file for
1688 a maildir delivery, the message "Mailbox is full" was not appended to the
1689 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1690 only to a quota excession during the actual writing of the file.
1692 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1693 characters?) which causes problems in log lines. The DN values are now
1694 passed through string_printing() before being added to log lines.
1696 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1697 and InterBase are left for another time.)
1699 PH/21 Added message_body_newlines option.
1701 PH/22 Guard against possible overflow in moan_check_errorcopy().
1703 PH/23 POSIX allows open() to be a macro; guard against that.
1705 PH/24 If the recipient of an error message contained an @ in the local part
1706 (suitably quoted, of course), incorrect values were put in $domain and
1707 $local_part during the evaluation of errors_copy.
1713 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1714 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1717 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1718 issue a MAIL command.
1720 PH/02 In an ACL statement such as
1722 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1724 if a client was not listed at all, or was listed with a value other than
1725 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1726 the condition was not true (as it should be), so access was not denied.
1727 The bug was that the ! inversion was incorrectly passed on to the second
1728 item. This has been fixed.
1730 PH/03 Added additional dnslists conditions == and =& which are different from
1731 = and & when the dns lookup returns more than one IP address.
1733 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1734 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1736 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1737 FSYNC, which compiles an option called disable_fsync that allows for
1738 bypassing fsync(). The documentation is heavily laced with warnings.
1740 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1742 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1743 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1744 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1745 including adding "make clean"; (3) Added -fPIC when compiling the test
1746 dynamically loaded module, to get rid of a warning.
1748 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1749 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1750 = 0s. The bug is otherwise harmless.
1752 PH/07 There was a bug in the dovecot authenticator such that the value of
1753 $auth1 could be overwritten, and so not correctly preserved, after a
1754 successful authentication. This usually meant that the value preserved by
1755 the server_setid option was incorrect.
1757 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1759 PH/09 Installed PCRE release 7.0.
1761 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1762 run for batched SMTP input. It is now run at the start of every message
1763 in the batch. While fixing this I discovered that the process information
1764 (output by running exiwhat) was not always getting set for -bs and -bS
1765 input. This is fixed, and it now also says "batched" for BSMTP.
1767 PH/11 Added control=no_pipelining.
1769 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1770 patch, slightly modified), and move the expansion of helo_data till after
1771 the connection is made in the smtp transport (so it can use these
1774 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1776 PH/14 Added log_selector = +pid.
1778 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1780 PH/16 Add ${if forany and ${if forall.
1782 PH/17 Added dsn_from option to vary the From: line in DSNs.
1784 PH/18 Flush SMTP output before performing a callout, unless control =
1785 no_callout_flush is set.
1787 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1788 was true (the default) a successful delivery failed to delete the retry
1789 item, thus causing premature timeout of the address. The bug is now
1792 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1794 PH/21 Long custom messages for fakedefer and fakereject are now split up
1795 into multiline reponses in the same way that messages for "deny" and
1796 other ACL rejections are.
1798 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1799 with slight modification.
1801 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1802 draft, changing the syntax and factoring some duplicate code".
1804 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1805 for deliveries of the second and subsequent messages over the same SMTP
1808 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1809 ${reduce, with only minor "tidies".
1811 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1813 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1814 expansion side effects.
1816 PH/27 When a message times out after an over-quota error from an Exim-imposed
1817 quota, the bounce message says "mailbox is full". This message was not
1818 being given when it was a system quota that was exceeded. It now should
1821 MH/03 Made $recipients available in local_scan(). local_scan() already has
1822 better access to the recipient list through recipients_list[], but
1823 $recipients can be useful in postmaster-provided expansion strings.
1825 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1826 in the case of a MAIL command with additional options following the
1827 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1828 were accidentally chopped off.
1830 PH/29 SMTP synchronization checks are implemented when a command is read -
1831 there is a check that no more input is waiting when there shouldn't be
1832 any. However, for some commands, a delay in an ACL can mean that it is
1833 some time before the response is written. In this time, more input might
1834 arrive, invalidly. So now there are extra checks after an ACL has run for
1835 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1836 pipelining has not been advertised.
1838 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1840 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1841 returned in the SMTP response when smtp_return_error_details was false.
1842 This has been fixed.
1844 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1845 instead of the C I/O that was originally supplied, because problems were
1846 reported on Solaris.
1848 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1849 Exim which did not show up earlier: it was assuming that a call to
1850 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1851 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1852 was a macro that became an assignment, so it seemed to work. This has
1853 changed to a proper function call with a void return, hence the compile
1854 error. Exim's code has been fixed.
1856 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1859 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1861 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1863 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1864 because it turns out to be too big or there is a timeout) there is no
1865 "Completed" line in the log. When some messages of this type were
1866 selected by exigrep, they were listed as "not completed". Others were
1867 picked up by some special patterns. I have improved the selection
1868 criteria to be more general.
1870 PH/38 The host_find_failed option in the manualroute router can now be set
1871 to "ignore", to completely ignore a host whose IP address cannot be
1872 found. If all hosts are ignored, the behaviour is controlled by the new
1873 host_all_ignored option.
1875 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1876 homing or because of multiple MX records with /mx) generated more than
1877 one IP address, and the following item turned out to be the local host,
1878 all the secondary addresses of the first item were incorrectly removed
1879 from the list, along with the local host and any following hosts (which
1880 is what is supposed to happen).
1882 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1883 whoever called Exim into the -H file. In the case of the daemon it was
1884 behaving confusingly. When first started, it used values for whoever
1885 started the daemon, but after a SIGHUP it used the Exim user (because it
1886 calls itself on a restart). I have changed the code so that it now always
1889 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1890 message are rejected with the same error (e.g. no authentication or bad
1891 sender address), and a DATA command is nevertheless sent (as can happen
1892 with PIPELINING or a stupid MUA), the error message that was given to the
1893 RCPT commands is included in the rejection of the DATA command. This is
1894 intended to be helpful for MUAs that show only the final error to their
1897 PH/42 Another patch from the Sieve maintainer.
1899 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1900 Eximstats - Fixed some broken HTML links and added missing column headers
1902 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1903 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1905 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1907 PH/43 Yet another patch from the Sieve maintainer.
1909 PH/44 I found a way to check for a TCP/IP connection going away before sending
1910 the response to the final '.' that terminates a message, but only in the
1911 case where the client has not sent further data following the '.'
1912 (unfortunately, this is allowed). However, in many cases there won't be
1913 any further data because there won't be any more messages to send. A call
1914 to select() can be used: if it shows that the input is "ready", there is
1915 either input waiting, or the socket has been closed. An attempt to read
1916 the next input character can distinguish the two cases. Previously, Exim
1917 would have sent an OK response which the client would never have see.
1918 This could lead to message repetition. This fix should cure that, at
1919 least in a lot of common cases.
1921 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1922 advertised in response to EHLO.
1928 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1929 fixed by 4.65/MH/01 (is this a record?) are fixed:
1931 (i) An empty string was always treated as zero by the numeric comparison
1932 operators. This behaviour has been restored.
1934 (ii) It is documented that the numeric comparison operators always treat
1935 their arguments as decimal numbers. This was broken in that numbers
1936 starting with 0 were being interpreted as octal.
1938 While fixing these problems I realized that there was another issue that
1939 hadn't been noticed. Values of message_size_limit (both the global option
1940 and the transport option) were treated as octal if they started with 0.
1941 The documentation was vague. These values are now always treated as
1942 decimal, and I will make that clear in the documentation.
1948 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1949 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1952 MH/01 Don't check that the operands of numeric comparison operators are
1953 integers when their expansion is in "skipping" mode (fixes bug
1954 introduced by 4.64-PH/07).
1956 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1957 child addresses, Exim now panics and dies. Previously, because the count
1958 is held in a short int, deliveries were likely to be lost. As such a
1959 large number of recipients for a single message is ridiculous
1960 (performance will be very, very poor), I have chosen to impose a limit
1961 rather than extend the field.
1967 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1968 leftover -K file (the existence of which was triggered by #402).
1969 While we were at it, introduced process PID as part of the -K
1970 filename. This should rule out race conditions when creating
1973 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1974 processing considerably. Previous code took too long for large mails,
1975 triggering a timeout which in turn triggers #401.
1977 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1978 in the DK code in transports.c. sendfile() is not really portable,
1979 hence the _LINUX specificness.
1981 TF/01 In the add_headers option to the mail command in an Exim filter,
1982 there was a bug that Exim would claim a syntax error in any
1983 header after the first one which had an odd number of characters
1986 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1987 callout verification, Exim cached a "reject" for the entire domain. This
1988 is correct for most verifications, but it is not correct for a recipient
1989 verification with use_sender or use_postmaster set, because in that case
1990 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1991 case of MAIL FROM:<> rejection from other early rejections (e.g.
1992 rejection of HELO). When verifying a recipient using a non-null MAIL
1993 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1994 Whatever the result of the callout, the value of the domain cache is
1995 left unchanged (for any other kind of callout, getting as far as trying
1996 RCPT means that the domain itself is ok).
1998 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2001 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2002 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2003 session, the daemon ignores ECONNECT errors and logs others; it now
2004 ignores EPIPE as well.
2006 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2007 (quoted-printable decoding).
2009 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2010 later the small subsequent patch to fix an introduced bug.
2012 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2014 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2016 PH/08 An error is now given if message_size_limit is specified negative.
2018 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2019 to be given (somewhat) arbitrary names.
2021 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2024 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2025 miscellaneous code fixes
2027 PH/10 Added the log_reject_target ACL modifier to specify where to log
2030 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2031 hostname. This is wrong, because it relates to the incoming message (and
2032 probably the interface on which it is arriving) and not to the outgoing
2033 callout (which could be using a different interface). This has been
2034 changed to use the value of the helo_data option from the smtp transport
2035 instead - this is what is used when a message is actually being sent. If
2036 there is no remote transport (possible with a router that sets up host
2037 addresses), $smtp_active_hostname is used.
2039 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2040 tweaks were necessary in order to get it to work (see also 21 below):
2041 (a) The code assumed that strncpy() returns a negative number on buffer
2042 overflow, which isn't the case. Replaced with Exim's string_format()
2044 (b) There were several signed/unsigned issues. I just did the minimum
2045 hacking in of casts. There is scope for a larger refactoring.
2046 (c) The code used strcasecmp() which is not a standard C function.
2047 Replaced with Exim's strcmpic() function.
2048 (d) The code set only $1; it now sets $auth1 as well.
2049 (e) A simple test gave the error "authentication client didn't specify
2050 service in request". It would seem that Dovecot has changed its
2051 interface. Fortunately there's a specification; I followed it and
2052 changed what the client sends and it appears to be working now.
2054 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2057 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2058 address A is aliased to B and C, where B exists and C does not. Without
2059 -v the output is "A verified" because verification stops after a
2060 successful redirection if more than one address is generated. However,
2061 with -v the child addresses are also verified. Exim was outputting "A
2062 failed to verify" and then showing the successful verification for C,
2063 with its parentage. It now outputs "B failed to verify", showing B's
2064 parentage before showing the successful verification of C.
2066 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2067 look up a TXT record in a specific list after matching in a combined
2070 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2071 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2072 they consult the DNS. I had assumed they would set it the way they
2073 wanted; and indeed my experiments on Linux seem to show that in some
2074 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2075 To be on the safe side, however, I have now made the interface to
2076 host_find_byname() similar to host_find_bydns(), with an argument
2077 containing the DNS resolver options. The host_find_byname() function now
2078 sets these options at its start, just as host_find_bydns() does. The smtp
2079 transport options dns_qualify_single and dns_search_parents are passed to
2080 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2081 of host_find_byname() use the default settings of RES_DEFNAMES
2082 (qualify_single) but not RES_DNSRCH (search_parents).
2084 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2085 spool_read_header() do less string testing, by means of a preliminary
2086 switch on the second character of optional "-foo" lines. (This is
2087 overdue, caused by the large number of possibilities that now exist.
2088 Originally there were few.) While I was there, I also converted the
2089 str(n)cmp tests so they don't re-test the leading "-" and the first
2090 character, in the hope this might squeeze out yet more improvement.
2092 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2093 flag allowing group syntax was set by the header_syntax check but not
2094 turned off, possible causing trouble later; (2) The flag was not being
2095 set at all for the header_verify test, causing "group"-style headers to
2096 be rejected. I have now set it in this case, and also caused header_
2097 verify to ignore an empty address taken from a group. While doing this, I
2098 came across some other cases where the code for allowing group syntax
2099 while scanning a header line wasn't quite right (mostly, not resetting
2100 the flag correctly in the right place). These bugs could have caused
2101 trouble for malformed header lines. I hope it is now all correct.
2103 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2104 with the "reply" argument non-NULL. The code, however (which originally
2105 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2106 but it didn't always do it. This confused somebody who was copying the
2107 code for some other use. I have removed all the tests.
2109 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2110 feature that was used to support insecure browsers during the U.S. crypto
2111 embargo. It requires special client support, and Exim is probably the
2112 only MTA that supported it -- and would never use it because real RSA is
2113 always available. This code has been removed, because it had the bad
2114 effect of slowing Exim down by computing (never used) parameters for the
2115 RSA_EXPORT functionality.
2117 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2118 authenticator to fail if there's a tab character in the incoming data
2119 (there should never be unless someone is messing about, as it's supposed
2120 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2121 if the connection is using TLS or if the remote IP is the same as the
2122 local IP, and the "valid-client-cert option" if a client certificate has
2125 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2126 authenticators. This can be used for authorization after authentication
2127 succeeds. (In the case of plaintext, it servers for both authentication
2130 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2131 if any retry times were supplied.
2133 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2134 connection, where there is no client IP address to check. In this
2135 situation, the verify now always succeeds.
2137 PH/25 Applied John Jetmore's -Mset patch.
2139 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2141 PH/27 In a string expansion for a processed (not raw) header when multiple
2142 headers of the same name were present, leading whitespace was being
2143 removed from all of them, but trailing whitespace was being removed only
2144 from the last one. Now trailing whitespace is removed from each header
2145 before concatenation. Completely empty headers in a concatenation (as
2146 before) are ignored.
2148 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2149 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2151 PH/29 [Removed. This was a change that I later backed out, and forgot to
2152 correct the ChangeLog entry (that I had efficiently created) before
2153 committing the later change.]
2155 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2156 address errors (4xx response to RCPT) over the same connection as other
2157 messages routed to the same hosts. Such deliveries are always "forced",
2158 so retry times are not inspected. This resulted in far too many retries
2159 for the affected addresses. The effect occurred only when there were more
2160 hosts than the hosts_max_try setting in the smtp transport when it had
2161 the 4xx errors. Those hosts that it had tried were not added to the list
2162 of hosts for which the message was waiting, so if all were tried, there
2163 was no problem. Two fixes have been applied:
2165 (i) If there are any address or message errors in an SMTP delivery, none
2166 of the hosts (tried or untried) are now added to the list of hosts
2167 for which the message is waiting, so the message should not be a
2168 candidate for sending over the same connection that was used for a
2169 successful delivery of some other message. This seems entirely
2170 reasonable: after all the message is NOT "waiting for some host".
2171 This is so "obvious" that I'm not sure why it wasn't done
2172 previously. Hope I haven't missed anything, but it can't do any
2173 harm, as the worst effect is to miss an optimization.
2175 (ii) If, despite (i), such a delivery is accidentally attempted, the
2176 routing retry time is respected, so at least it doesn't keep
2177 hammering the server.
2179 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2180 in ${readsocket because some servers need this prod.
2182 PH/32 Added some extra debug output when updating a wait-xxx database.
2184 PH/33 The hint "could be header name not terminated by colon", which has been
2185 given for certain expansion errors for a long time, was not being given
2186 for the ${if def:h_colon_omitted{... case.
2188 PH/34 The spec says: "With one important exception, whenever a domain list is
2189 being scanned, $domain contains the subject domain." There was at least
2190 one case where this was not true.
2192 PH/35 The error "getsockname() failed: connection reset by peer" was being
2193 written to the panic log as well as the main log, but it isn't really
2194 panic-worthy as it just means the connection died rather early on. I have
2195 removed the panic log writing for the ECONNRESET error when getsockname()
2198 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2199 runs only) independently of the message's sender address. This meant
2200 that, if the 4xx error was in fact related to the sender, a different
2201 message to the same recipient with a different sender could confuse
2202 things. In particualar, this can happen when sending to a greylisting
2203 server, but other circumstances could also provoke similar problems.
2204 I have changed the default so that the retry time for these errors is now
2205 based a combination of the sender and recipient addresses. This change
2206 can be overridden by setting address_retry_include_sender=false in the
2209 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2210 remote server are returned as part of bounce messages. This was not
2211 happening for LMTP over a pipe (the lmtp transport), but now it is the
2212 same for both kinds of LMTP.
2214 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2215 in header lines that were in fact CNAMEs. This is no longer the case.
2217 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2218 and queue runs started by the daemon processed all messages. This has
2219 been fixed so that -R and -S can now usefully be given with -q<time>.
2221 PH/40 Import PCRE release 6.7 (fixes some bugs).
2223 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2225 PH/42 Give an error if -q is specified more than once.
2227 PH/43 Renamed the variables $interface_address and $interface_port as
2228 $received_ip_address and $received_port, to make it clear that these
2229 values apply to message reception, and not to the outgoing interface when
2230 a message is delivered. (The old names remain recognized, of course.)
2232 PH/44 There was no timeout on the connect() call when using a Unix domain
2233 socket in the ${readsocket expansion. There now is.
2235 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2236 be meaningful with "accept".
2238 SC/01 Eximstats V1.43
2239 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2241 SC/02 Eximstats V1.44
2242 Use a glob alias rather than an array ref in the generated
2243 parser. This improves both readability and performance.
2245 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2246 Collect SpamAssassin and rejection statistics.
2247 Don't display local sender or destination tables unless
2248 there is data to show.
2249 Added average volumes into the top table text output.
2251 SC/04 Eximstats V1.46
2252 Collect data on the number of addresses (recipients)
2253 as well as the number of messages.
2255 SC/05 Eximstats V1.47
2256 Added 'Message too big' to the list of mail rejection
2257 reasons (thanks to Marco Gaiarin).
2259 SC/06 Eximstats V1.48
2260 Mainlog lines which have GMT offsets and are too short to
2261 have a flag are now skipped.
2263 SC/07 Eximstats V1.49 (Alain Williams)
2264 Added the -emptyok flag.
2266 SC/08 Eximstats V1.50
2267 Fixes for obtaining the IP address from reject messages.
2269 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2270 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2271 whitesspace changes from 4.64-PH/27
2273 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2276 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2277 are found, allow negative numbers in numeric criteria)
2279 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2281 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2282 to match changes made in 4.64-PH/43
2284 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2286 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2287 "domain" is actually an IP address, and force a failure. This locks out
2288 those revolvers/nameservers that support "A-for-A" lookups, in
2289 contravention of the specifications.
2291 PH/48 When a host name was looked up from an IP address, and the subsequent
2292 forward lookup of the name timed out, the host name was left in
2293 $sender_host_name, contrary to the specification.
2295 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2296 restricted to single-key lookups, Exim was not diagnosing an error if
2297 * or *@ was used with a query-style lookup.
2299 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2301 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2302 long ago, but noone interested enough thought of it. Let's just say that
2303 the "1.1" means that there are some new functions that weren't there at
2304 some point in the past.
2306 PH/51 Error processing for expansion failure of helo_data from an smtp
2307 transport during callout processing was broken.
2309 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2310 tested/used via the -bh/-bhc/-bs options.
2312 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2313 bug, fixed in subsequent PCRE releases).
2315 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2316 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2318 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2324 SC/01 Use a glob alias rather than an array ref in eximstats generated
2325 parser. This improves both readability and performance.
2327 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2328 Don't display local sender or destination tables in eximstats unless
2329 there is data to show.
2330 Added average volumes into the eximstats top table text output.
2332 SC/03 Collect data on the number of addresses (recipients) as well
2333 as the number of messages in eximstats.
2335 TF/01 Correct an error in the documentation for the redirect router. Exim
2336 does (usually) call initgroups() when daemonizing.
2338 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2339 with consistent privilege compared to when running as a daemon.
2341 TF/03 Note in the spec that $authenticated_id is not set for local
2342 submissions from trusted users.
2344 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2345 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2347 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2348 by adding some example configuration directives to the default
2349 configuration file. A little bit of work is required to uncomment the
2350 directives and define how usernames and passwords are checked, but
2351 there is now a framework to start from.
2353 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2354 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2355 without this. I don't know how relevant this is to other LDAP libraries.
2357 PH/02 Add the verb name to the "unknown ACL verb" error.
2359 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2361 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2363 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2364 directory not expanded when it should be if an expanded home directory
2365 was set for the address (which is overridden by the transport).
2367 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2370 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2371 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2372 because it is too late at that time, and has no effect.
2374 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2375 security issue with \' (bugzilla #107). I could not use the
2376 PQescapeStringConn() function, because it needs a PGconn value as one of
2379 PH/08 When testing addresses using -bt, indicate those final addresses that
2380 are duplicates that would not cause an additional delivery. At least one
2381 person was confused, thinking that -bt output corresponded to deliveries.
2382 (Suppressing duplicates isn't a good idea as you lose the information
2383 about possibly different redirections that led to the duplicates.)
2385 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2386 systems where poll() doesn't work, in particular OS X.
2388 PH/10 Added more information to debugging output for retry time not reached.
2390 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2391 operations in malware.c.
2393 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2396 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2397 syslog_duplication set false, log lines that would normally be written
2398 both the the main log and to the reject log were not written to syslog at
2401 PH/14 In the default configuration, change the use of "message" in ACL warn
2402 statements to "add_header".
2404 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2405 not followed by a command (e.g. "seen endif").
2407 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2408 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2411 PH/17 Added extra conditions to the default value of delay_warning_condition
2415 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2416 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2417 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2420 The Auto-Submitted: and various List- headers are standardised, whereas I
2421 don't think Precedence: ever was.
2423 PH/18 Refactored debugging code in route_finduser() to show more information,
2424 in particular, the error code if getpwnam() issues one.
2426 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2427 This is apparently needed in addition to the PH/07 change above to avoid
2428 any possible encoding problems.
2430 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2431 but not after initializing Perl.
2433 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2434 output them only if debugging. By default they are written stderr,
2435 apparently, which is not desirable.
2437 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2440 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2443 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2445 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2446 authenticated or an ident call has been made. Suppress the default
2447 values for $authenticated_id and $authenticated_sender (but permit -oMai
2448 and -oMas) when testing with -bh.
2450 PH/24 Re-jigged the order of the tests in the default configuration so that the
2451 tests for valid domains and recipients precede the DNS black list and CSA
2452 tests, on the grounds that those ones are more expensive.
2454 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2455 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2456 This bug exists in every version of Exim that I still have, right back to
2459 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2460 However, an attempt to turn on case-sensitivity in a regex key by
2461 including (?-i) didn't work because the subject string was already
2462 lowercased, and the effects were non-intuitive. It turns out that a
2463 one-line patch can be used to allow (?-i) to work as expected.
2469 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2470 other effects) broke the use of negated acl sub-conditions.
2472 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2475 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2476 "Deny" causes Exim to reject the incoming connection with a 554 error.
2477 Unfortunately, if there is a major crisis, such as a disk failure,
2478 tcp-wrappers gives "deny", whereas what one would like would be some
2479 kind of temporary error. A kludge has been added to help with this.
2480 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2481 554 error is used if errno is still zero or contains ENOENT (which occurs
2482 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2485 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2487 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2488 errors. Otherwise a message that provokes a temporary error (when other
2489 messages do not) can cause a whole host to time out.
2491 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2492 addresses were routed directly to files or pipes from a redirect router.
2493 File deliveries just didn't batch; pipe deliveries might have suffered
2496 PH/06 A failure to get a lock for a hints database would erroneously always say
2497 "Failed to get write lock", even when it was really a read lock.
2499 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2500 of 0600. This has been changed to use the value of the lockfile_mode
2501 option (which defaults to 0600).
2503 PH/08 Applied small patch from the Sieve maintainer.
2505 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2506 folder from quota calculations, a direct delivery into this folder messed
2507 up the contents of the maildirsize file. This was because the regex was
2508 used only to exclude .Trash (or whatever) when the size of the mailbox
2509 was calculated. There was no check that a delivery was happening into an
2510 excluded directory. This bug has been fixed by ignoring all quota
2511 processing for deliveries into excluded directories.
2513 PH/10 Added the maildirfolder_create_regex option to appendfile.
2519 PH/01 The code for finding all the local interface addresses on a FreeBSD
2520 system running IPv6 was broken. This may well have applied to all BSD
2521 systems, as well as to others that have similar system calls. The broken
2522 code found IPv4 interfaces correctly, but gave incorrect values for the
2523 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2524 that it would not match correctly against @[] and not recognize the IPv6
2527 PH/02 The ipliteral router was not recognizing addresses of the form user@
2528 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2530 PH/03 Added disable_ipv6.
2532 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2533 decoded form, because it is most often used to construct To: headers
2534 lines in autoreplies, and the decoded form may well be syntactically
2535 invalid. However, $reply_address has leading white space removed, and all
2536 newlines turned into spaces so that the autoreply transport does not
2539 PH/05 If group was specified without a user on a router, and no group or user
2540 was specified on a transport, the group from the router was ignored.
2542 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2543 for visible compile-time settings that can be used to change these
2544 numbers, for those that want even more. Backwards compatibility with old
2545 spool files has been maintained. However, going back to a previous Exim
2546 release will lost any variables that are in spool files.
2548 PH/07 Two small changes when running in the test harness: increase delay when
2549 passing a TCP/IP connection to a new process, in case the original
2550 process has to generate a bounce, and remove special handling of
2551 127.0.0.2 (sic), which is no longer necessary.
2553 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2554 be the same on different OS.
2556 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2559 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2560 whether --show-vars was specified or not
2562 JJ/02 exipick: Added support for new ACL variable spool format introduced
2565 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2566 syntactically invalid From: or Reply-to: line, and a filter used this to
2567 generate an autoreply, and therefore failed to obtain an address for the
2568 autoreply, Exim could try to deliver to a non-existent relative file
2569 name, causing unrelated and misleading errors. What now happens is that
2570 it logs this as a hard delivery error, but does not attempt to create a
2573 PH/11 The exinext utility has a -C option for testing purposes, but although
2574 the given file was scanned by exinext itself; it wasn't being passed on
2575 when Exim was called.
2577 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2578 an end-of-file indication when reading a command response.
2580 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2581 compiled. In many other places in Exim, IPv6 addresses are always
2582 recognized, so I have changed this. It also means that IPv4 domain
2583 literals of the form [IPV4:n.n.n.n] are now always recognized.
2585 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2586 used if the router is not running as root, for example, when verifying at
2587 ACL time, or when using -bh. The debugging output from this situation was
2588 non-existent - all you got was a failure to exec. I have made two
2591 (a) Failures to set uid/gid, the current directory, or a process leader
2592 in a subprocess such as that created by queryprogram now generate
2593 suitable debugging ouput when -d is set.
2595 (b) The queryprogram router detects when it is not running as root,
2596 outputs suitable debugging information if -d is set, and then runs
2597 the subprocess without attempting to change uid/gid.
2599 PH/15 Minor change to Makefile for building test_host (undocumented testing
2602 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2603 additional section of a DNS packet that returns MX or SRV records.
2604 Instead, it always explicitly searches for A/AAAA records. This avoids
2605 major problems that occur when a DNS server includes only records of one
2606 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2607 fixed another bug: if SRV records were looked up and the corresponding
2608 address records were *not* found in the additional section, the port
2609 values from the SRV records were lost.
2611 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2612 using the correct key (the original address) when searching the retry
2613 rules in order to find which one to use for generating the retry hint.
2615 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2616 adding the default one. Similarly, if it contains a Reply-To: header, the
2617 errors_reply_to option, if set, is not used.
2619 PH/19 When calculating a retry time, Exim used to measure the "time since
2620 failure" by looking at the "first failed" field in the retry record. Now
2621 it does not use this if it is later than than the arrival time of the
2622 message. Instead it uses the arrival time. This makes for better
2623 behaviour in cases where some deliveries succeed, thus re-setting the
2624 "first failed" field. An example is a quota failure for a huge message
2625 when small messages continue to be delivered. Without this change, the
2626 "time since failure" will always be short, possible causing more frequent
2627 delivery attempts for the huge message than are intended.
2628 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2630 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2631 $1, $2, $3) because the numerical variables can be reset during some
2632 expansion items (e.g. "match"), thereby losing the authentication data.
2634 PH/21 Make -bV show the size of off_t variables so that the test suite can
2635 decide whether to run tests for quotas > 2G.
2637 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2638 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2639 filecount value is greater than 2G or if a quota value is greater than 2G
2640 on a system where the size of off_t is not greater than 4, a panic error
2643 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2644 never match. The debug and -bh output now contains an explicit error
2645 message indicating a malformed IPv4 address or mask.
2647 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2648 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2649 PH/23 above applies.
2651 PH/25 Do not write to syslog when running in the test harness. The only
2652 occasion when this arises is a failure to open the main or panic logs
2653 (for which there is an explicit test).
2655 PH/26 Added the /no_tell option to "control=freeze".
2657 PH/27 If a host name lookup failed very early in a connection, for example, if
2658 the IP address matched host_lookup and the reverse lookup yielded a name
2659 that did not have a forward lookup, an error message of the form "no IP
2660 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2661 could be logged. Now it outputs the IP address instead of "NULL".
2663 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2664 allows the sender and the authenticated sender to be set when
2665 submitting a message from within Exim. Since child_open_exim() is
2666 documented for local_scan(), the new function should be too.
2668 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2669 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2670 results in an empty string is now treated as unset.
2672 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2674 PH/31 Added sender_verify_fail logging option.
2676 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2677 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2678 tidied the source and removed it altogether.
2680 PH/33 When a queue run was abandoned because the load average was too high, a
2681 log line was always written; now it is written only if the queue_run log
2682 selector is set. In addition, the log line for abandonment now contains
2683 information about the queue run such as the pid. This is always present
2684 in "start" and "stop" lines but was omitted from the "abandon" line.
2686 PH/34 Omit spaces between a header name and the colon in the error message that
2687 is given when verify = headers_syntax fails (if there are lots of them,
2688 the message gets confusing).
2690 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2691 names, as there are now some PTR records that contain slashes. This check
2692 is only to protect against broken name servers that fall over on strange
2693 characters, so the fact that it applies to all lookups doesn't matter.
2695 PH/36 Now that the new test suite is complete, we can remove some of the
2696 special code in Exim that was needed for the old test suite. For example,
2697 sorting DNS records because real resolvers return them in an arbitrary
2698 order. The new test suite's fake resolver always returns records in the
2701 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2702 bounces) except when queue_only is set, to avoid logging races between
2703 the different processes.
2705 PH/38 Panic-die if .include specifies a non-absolute path.
2707 PH/39 A tweak to the "H" retry rule from its user.
2709 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2710 a label. They prevented compilation on older perls.
2712 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2713 a warning to be raised on newish perls.
2715 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2716 on queue. Changes to match documented behaviour of showing count of
2717 messages matching specified criteria.
2719 PH/40 Changed the default ident timeout from 30s to 5s.
2721 PH/41 Added support for the use of login_cap features, on those BSD systems
2722 that have them, for controlling the resources used by pipe deliveries.
2724 PH/42 The content-scanning code uses fopen() to create files in which to put
2725 message data. Previously it was not paying any attention to the mode of
2726 the files. Exim runs with umask(0) because the rest of the code creates
2727 files with open(), and sets the required mode explicitly. Thus, these
2728 files were ending up world-writeable. This was not a big issue, because,
2729 being within the spool directory, they were not world-accessible. I have
2730 created a function called modefopen, which takes an additional mode
2731 argument. It sets umask(777), creates the file, chmods it to the required
2732 mode, then resets the umask. All the relevant calls to fopen() in the
2733 content scanning code have been changed to use this function.
2735 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2736 to 24 hours. This avoids potential overflow problems when processing G
2737 and H retry rules. I suspect nobody ever tinkers with this value.
2739 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2741 PH/45 When the plaintext authenticator is running as a client, the server's
2742 challenges are checked to ensure they are valid base64 strings. By
2743 default, the authentication attempt is cancelled if an invalid string is
2744 received. Setting client_ignore_invalid_base64 true ignores these errors.
2745 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2746 they are received. Thus, the responses can be made to depend on the
2747 challenges. If an invalid string is ignored, an empty string is placed in
2750 PH/46 Messages that are created by the autoreply transport now contains a
2751 References: header, in accordance with RFCs 2822 and 3834.
2753 PH/47 Added authenticated_sender_force to the smtp transport.
2755 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2757 PH/49 Installed latest patch from the Sieve maintainer.
2759 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2760 was also set, the appendfile transport was unnecessarily scanning a
2761 directory of message files (e.g. for maildir delivery) to find the count
2762 of files (along with the size), even though it did not need this
2763 information. It now does the scan only if it needs to find either the
2764 size of the count of files.
2766 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2768 PH/52 Two bugs concerned with error handling when the smtp transport is
2771 (i) Exim was not creating retry information for temporary errors given
2772 for individual recipients after the DATA command when the smtp transport
2773 was used in LMTP mode. This meant that they could be retried too
2774 frequently, and not timed out correctly.
2776 (ii) Exim was setting the flag that allows error details to be returned
2777 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2778 recipients that were returned after the DATA command.
2780 PH/53 This is related to PH/52, but is more general: for any failing address,
2781 when detailed error information was permitted to be returned to the
2782 sender, but the error was temporary, then after the final timeout, only
2783 "retry timeout exceeded" was returned. Now it returns the full error as
2784 well as "retry timeout exceeded".
2786 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2787 do this, and (what is worse) MTAs that accept it.
2789 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2790 will now be deprecated.
2792 PH/56 New os.c-cygwin from the Cygwin maintainer.
2794 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2795 formats (previously only available in exim formats via -bpr, -bpru,
2796 and -bpra. Now also available in native and exiqgrep formats)
2798 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2799 with very large, slow to parse queues
2801 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2803 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2805 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2806 responses to authentication challenges, though it was showing the
2807 challenges; (ii) I've removed the CR characters from the debug output for
2810 PH/58 Allow for the insertion of a newline as well as a space when a string
2811 is turned into more than one encoded-word during RFC 2047 encoding. The
2812 Sieve code now uses this.
2814 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2815 data_4xx, lost_connection, tls_required.
2817 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2818 message was being sent as an SMTP response.
2820 PH/61 Add -l and -k options to exicyclog.
2822 PH/62 When verifying, if an address was redirected to one new address, so that
2823 verification continued, and the new address failed or deferred after
2824 having set something in $address_data, the value of $address_data was not
2825 passed back to the ACL. This was different to the case when no
2826 redirection occurred. The value is now passed back in both cases.
2828 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2829 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2830 use login_cap.h, so on its own it isn't the distinguishing feature. The
2831 new name refers directly to the setclassresources() function.
2833 PH/65 Added configuration files for NetBSD3.
2835 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2837 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2838 is preferred over IPv4.
2840 PH/68 The bounce_return_message and bounce_return_body options were not being
2841 honoured for bounces generated during the reception of non-SMTP messages.
2842 In particular, this applied to messages rejected by the ACL. This bug has
2843 been fixed. However, if bounce_return_message is true and bounce_return_
2844 body is false, the headers that are returned for a non-SMTP message
2845 include only those that have been read before the error was detected.
2846 (In the case of an ACL rejection, they have all been read.)
2848 PH/69 The HTML version of the specification is now built in a directory called
2849 spec_html instead of spec.html, because the latter looks like a path with
2850 a MIME-type, and this confuses some software.
2852 PH/70 Catch two compiler warnings in sieve.c.
2854 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2855 function verify_get_ident() calls ip_connect() to connect a socket, but
2856 if the "connect()" function timed out, ip_connect() used to close the
2857 socket. However, verify_get_ident() also closes the socket later, and in
2858 between Exim writes to the log, which may get opened at this point. When
2859 the socket was closed in ip_connect(), the log could get the same file
2860 descriptor number as the socket. This naturally causes chaos. The fix is
2861 not to close the socket in ip_connect(); the socket should be closed by
2862 the function that creates it. There was only one place in the code where
2863 this was missing, in the iplookup router, which I don't think anybody now
2864 uses, but I've fixed it anyway.
2866 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2867 well as to direct DNS lookups. Otherwise the handling of names in host
2868 lists is inconsistent and therefore confusing.
2874 PH/01 Two changes to the default runtime configuration:
2876 (1) Move the checks for relay_from_hosts and authenticated clients from
2877 after to before the (commented out) DNS black list checks.
2879 (2) Add control=submission to the relay_from_hosts and authenticated
2880 clients checks, on the grounds that messages accepted by these
2881 statements are most likely to be submissions.
2883 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2885 (1) Generate an error if the third argument for the ${prvs expansion is
2888 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2891 (3) Reset the variables that are obtained from the first argument of
2892 ${prvscheck and used in the second argument before leaving the code,
2893 because their memory is reclaimed, so using them afterwards may do
2896 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2897 one (it's much easier than Tom thought :-).
2899 (5) Because of (4), we can now allow for the use of $prvscheck_result
2900 inside the third argument.
2902 PH/03 For some reason, the default setting of PATH when running a command from
2903 a pipe transport was just "/usr/bin". I have changed it to
2906 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2907 anything to be listed in the output from -bV.
2909 PH/05 When a filter generated an autoreply, the entire To: header line was
2910 quoted in the delivery log line, like this:
2912 => >A.N.Other <ano@some.domain> <original@ddress> ...
2914 This has been changed so that it extracts the operative address. There
2915 may be more than one such address. If so, they are comma-separated, like
2918 => >ano@some.domain,ona@other.domain <original@ddress> ...
2920 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2921 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2922 not being looked up in the rDNS to get a host name, Exim was showing the
2923 IP address twice in Received: lines, even though the IP addresses were
2924 identical. For example:
2926 Received: from [1.2.3.4] (helo=[1.2.3.4])
2928 However, if the real host name was known, it was omitting the HELO data
2929 if it matched the actual IP address. This has been tidied up so that it
2930 doesn't show the same IP address twice.
2932 PH/07 When both +timestamp and +memory debugging was on, the value given by
2933 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2934 called by the debug printing, thereby overwriting the timestamp buffer.
2935 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2937 PH/08 When the original message was included in an autoreply transport, it
2938 always said "this is a copy of the message, including all the headers",
2939 even if body_only or headers_only was set. It now gives an appropriate
2942 PH/09 Applied a patch from the Sieve maintainer which:
2944 o fixes some comments
2945 o adds the (disabled) notify extension core
2946 o adds some debug output for the result of if/elsif tests
2947 o points to the current vacation draft in the documentation
2948 and documents the missing references header update
2952 o fixes a bug in processing the envelope test (when testing
2953 multiple envelope elements, the last element determinted the
2956 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2957 Electronic Mail") by including:
2959 Auto-submitted: auto-generated
2961 in the messages that it generates (bounce messages and others, such as
2962 warnings). In the case of bounce messages for non-SMTP mesages, there was
2963 also a typo: it was using "Auto_submitted" (underscore instead of
2964 hyphen). Since every message generated by Exim is necessarily in response
2965 to another message, thes have all been changed to:
2967 Auto-Submitted: auto-replied
2969 in accordance with these statements in the RFC:
2971 The auto-replied keyword:
2973 - SHOULD be used on messages sent in direct response to another
2974 message by an automatic process,
2976 - MUST NOT be used on manually-generated messages,
2978 - MAY be used on Delivery Status Notifications (DSNs) and Message
2979 Disposition Notifications (MDNs),
2981 - MUST NOT be used on messages generated by automatic or periodic
2982 processes, except for messages which are automatic responses to
2985 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2986 to the default Received: header definition.
2988 PH/12 Added log selector acl_warn_skipped (default on).
2990 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2991 variables because (a) they are in the wrong storage pool and (b) even if
2992 they were copied, it wouldn't work properly because of the caching.
2994 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2995 checking when decoding. Apparently there are clients that generate
2996 overlong encoded strings. Why am I not surprised?
2998 PH/15 If the first argument of "${if match_address" was not empty, but did not
2999 contain an "@" character, Exim crashed. Now it writes a panic log message
3000 and treats the condition as false.
3002 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3004 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3005 extension "envelope-auth". The code is finished and in agreement with
3006 other implementations, but there is no documentation so far and in fact,
3007 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3008 not changing the active code.
3010 Print executed "if" and "elsif" statements when debugging is used. This
3011 helps a great deal to understand what a filter does.
3013 Document more things not specified clearly in RFC3028. I had all this
3014 sorted out, when out of a sudden new issues came to my mind. Oops."
3016 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3019 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3020 canonical form (as documented). However, after a host name lookup from
3021 the IP address, check_host() was doing a simple string comparison with
3022 addresses acquired from the DNS when checking that the found name did
3023 have the original IP as one of its addresses. Since any found IPv6
3024 addresses are likely to be in abbreviated form, the comparison could
3025 fail. Luckily, there already exists a function for doing the comparison
3026 by converting both addresses to binary, so now that is used instead of
3027 the text comparison.
3029 PH/20 There was another similar case to PH/19, when a complete host name was
3030 given in a host list; looking up its IP address could give an abbreviated
3031 form, whereas the current host's name might or might not be abbreviated.
3032 The same fix has been applied.
3038 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3039 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3042 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3043 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3045 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3047 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3048 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3049 Also, for "Auto-Submitted", treat anything other than "no" as
3050 non-personal, in accordance with RFC 3834. (Previously it treated
3051 anything starting "auto-" as non-personal.)
3053 TF/01 The control=submission/name=... option had a problem with syntax
3054 errors if the name included a slash character. The /name= option
3055 now slurps the rest of the string, so it can include any characters
3056 but it must come last in the list of options (after /sender_retain
3059 PH/05 Some modifications to the interface to the fake nameserver for the new
3067 TK/01 Added the "success_on_redirect" address verification option. See
3068 NewStuff for rationale and an example.
3070 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3072 PH/02 Patch to exigrep to allow it to work on syslog lines.
3074 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3075 fread() to skip over the body file's header line, because in Cygwin the
3076 header line is locked and is inaccessible.
3078 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3079 co-exist for some time) to make it clear that it is the Exim ID that is
3080 referenced, not the Message-ID: header line.
3082 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3083 string_format() function, because snprintf() does not exist on all
3086 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3087 ${stat: expansion item.
3089 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3090 protocol synchronization error", to keep the pedants happy.
3092 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3093 well as for IRIX systems, when gcc is being used. See the host.c source
3096 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3098 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3101 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3102 transport and to the smtp transport in LMTP mode.
3104 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3106 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3107 run a filter in a subprocess. This could lead to confusion in subsequent
3108 lookups in the parent process. There should also be a search_tidyup() at
3109 the end of the subprocess.
3111 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3112 only if the host matched helo_try_verify_hosts, which caused the
3113 verification to occur when the EHLO/HELO command was issued. The ACL just
3114 tested the remembered result. Now, if a previous verification attempt has
3115 not happened, "verify = helo" does it there and then.
3117 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3119 TK/03 Fix log output including CR from clamd.
3121 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3122 exist provoked a memory error which could cause a segfault.
3124 PH/15 Installed PCRE 6.2
3126 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3128 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3129 of the problem. Specifically, suggested +O2 rather than +O1 for the
3132 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3134 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3135 record, the port was not correctly logged when the outgoing_port log
3136 selector was set (it logged the transort's default port).
3138 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3139 fallback_hosts, and "hosts" in the smtp transport.
3141 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3142 host errors such as "Connection refused".
3144 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3145 authentication with radiusclient 0.4.9:
3147 - Error returned from rc_read_config was caught wrongly
3148 - Username/password not passed on to radius server due to wrong length.
3150 The presumption is that some radiusclient API changes for 4.51/PH/17
3151 were not taken care of correctly. The code is still untested by me (my
3152 Linux distribution still has 0.3.2 of radiusclient), but it was
3153 contributed by a Radius user.
3155 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3156 expanding the "port" option of the smtp transport.
3158 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3159 while reading a MIME header. Thanks to Tom Hughes for a patch.
3161 PH/24 Include config.h inside local_scan.h so that configuration settings are
3164 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3165 that in an ACL for RCPT (for example), you can examine exactly what was
3168 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3169 commands, but it was not correctly comparing the address with the actual
3170 client host address. Thus, it would show the EHLO address in Received:
3171 header lines when this was not necessary.
3173 PH/27 Added the % operator to ${eval:}.
3175 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3176 it should be ignoring failures (because with -C, for example, it has lost
3177 privilege). It wasn't ignoring creation failures other than "already
3180 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3183 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3184 because an input line was too long, either on its own, or by virtue of
3185 too many continuations, the temporary file was not being removed, and the
3186 return code was incorrect.
3188 PH/31 Missing "BOOL" in function definition in filtertest.c.
3190 PH/32 Applied Sieve patches from the maintainer.
3192 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3194 PH/33 Added "verify = not_blind".
3196 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3197 Local/Makefile (with some defaults set). These are used in built scripts
3198 such as exicyclog, but they have never been used in the exim_install
3199 script (though there are many overriding facilities there). I have
3200 arranged that the exim_install script now takes note of these two
3203 PH/35 Installed configuration files for Dragonfly.
3205 PH/36 When a locally submitted message by a trusted user did not contain a
3206 From: header, and the sender address was obtained from -f or from an SMTP
3207 MAIL command, and the trusted user did not use -F to supply a sender
3208 name, $originator_name was incorrectly used when constructing a From:
3209 header. Furthermore, $originator_name was used for submission mode
3210 messages from external hosts without From: headers in a similar way,
3211 which is clearly wrong.
3213 PH/37 Added control=suppress_local_fixups.
3215 PH/38 When log_selector = +received_sender was set, and the addition of the
3216 sender made the log line's construction buffer exactly full, or one byte
3217 less than full, an overflow happened when the terminating "\n" was
3220 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3221 when the result of a list match is failure because a DNS lookup failed.
3223 PH/40 RM_COMMAND is now used in the building process.
3225 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3226 the "build-* directories that it finds.
3228 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3229 address in a domain literal was a prefix of an interface address.
3231 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3232 when verifying a sender address, unless rewrite_headers is false.
3234 PH/44 Wrote a long comment about why errors_to addresses are verified as
3235 recipients, not senders.
3237 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3238 the ratelimit ACL was added.
3240 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3242 PH/46 Added extra information about PostgreSQL errors to the error string.
3244 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3245 avoiding the need to install special zones in a real server. This is
3246 backwards compatible; if it can't find the fake resolver, it drops back.
3247 Thus, both old and new test suites can be run.
3249 TF/02 Added util/ratelimit.pl
3251 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3252 clock is set back in time.
3254 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3255 Candler <B.Candler@pobox.com>.
3257 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3258 OK for addresses that are the result of redirections.
3260 PH/48 A number of further additions for the benefit of the new test suite,
3261 including a fake gethostbyname() that interfaces to the fake DNS resolver
3264 TF/06 The fix for widen_domains has also been applied to qualify_single and
3265 search_parents which are the other dnslookup options that can cause
3268 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3271 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3273 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3274 service" if it encountered an MX record with an empty target hostname.
3275 The message is now "an MX or SRV record indicated no SMTP service".
3277 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3278 if the DNS of the sending site is misconfigured. This is quite a
3279 common situation. This change restores the behaviour of treating a
3280 helo verification defer as a failure.
3282 PH/51 If self=fail was set on a router, the bounce message did not include the
3283 actual error message.
3289 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3291 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3292 command itself ended in error, the underlying message about the transport
3293 filter timeout was being overwritten with the pipe command error. Now the
3294 underlying error message should be appended to the second error message.
3296 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3298 PH/02 Reduce the amount of output that "make" produces by default. Full output
3299 can still be requested.
3301 PH/03 The warning log line about a condition test deferring for a "warn" verb
3302 was being output only once per connection, rather than after each
3303 occurrence (because it was using the same function as for successful
3304 "warn" verbs). This seems wrong, so I have changed it.
3306 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3307 it should not have, which might have caused a crash in the right
3308 circumstances, but probably never did.
3310 PH/04 Installed a modified version of Tony Finch's patch to make submission
3311 mode fix the return path as well as the Sender: header line, and to
3312 add a /name= option so that you can make the user's friendly name appear
3315 TF/03 Added the control = fakedefer ACL modifier.
3317 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3318 Mark Lowes for thorough testing.
3320 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3322 TK/03 Merged latest SRS patch from Miles Wilton.
3324 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3325 with the definition in sysexits.h (which is #included earlier).
3326 Fortunately, Exim does not actually use EX_OK. The code used to try to
3327 preserve the sysexits.h value, by assumimg that macro definitions were
3328 scanned for macro replacements. I have been disabused of this notion,
3329 so now the code just undefines EX_OK before #including unistd.h.
3331 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3332 in the smtp transport. When a block could not be written in a single
3333 write() function, the timeout was being re-applied to each part-write.
3334 This seems wrong - if the receiver was accepting one byte at a time it
3335 would take for ever. The timeout is now adjusted when this happens. It
3336 doesn't have to be particularly precise.
3338 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3339 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3341 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3342 without a domain if the check to <postmaster@domain> fails.
3344 SC/01 Eximstats: added -xls and the ability to specify output files
3345 (patch written by Frank Heydlauf).
3347 SC/02 Eximstats: use FileHandles for outputing results.
3349 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3351 SC/04 Eximstats: fixed display of large numbers with -nvr option
3353 SC/05 Eximstats: fixed merging of reports with empty tables.
3355 SC/06 Eximstats: added the -include_original_destination flag
3357 SC/07 Eximstats: removed tabs and trailing whitespace.
3359 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3361 TK/06 MBOX spool code: Add real "From " MBOX separator line
3362 so the .eml file is really in mbox format (even though
3363 most programs do not really care). Patch from Alex Miller.
3365 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3366 The latter is generated from $received_to and is only set if the
3367 message has one envelope recipient. SA can use these headers,
3368 obviously out-of-the-box. Patch from Alex Miller.
3370 PH/08 The ${def test on a variable was returning false if the variable's
3371 value was "0", contrary to what the specification has always said!
3372 The result should be true unless the variable is empty.
3374 PH/09 The syntax error of a character other than { following "${if
3375 def:variable_name" (after optional whitespace) was not being diagnosed.
3376 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3377 accidental colon was present, for example, could give incorrect results.
3379 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3380 result is used (not including appendfile, where other changes are about
3383 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3384 This involved changing a lot of size variables from int to off_t. It
3385 should work with maildirs and everything.
3387 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3388 spamd dying while we are connected to it.
3390 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3393 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3394 function for BDB 4.3.
3396 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3398 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3399 the sources for PCRE have been re-arranged and more files are now
3402 PH/15 The code I had for printing potentially long long variables in PH/11
3403 above was not the best (it lost precision). The length of off_t variables
3404 is now inspected at build time, and an appropriate printing format (%ld
3405 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3406 to be "long long int" or "long int". This is needed for the internal
3407 formatting function string_vformat().
3409 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3410 the configuration file to be ":syslog", then the script "guesses" where
3411 the logs files are, rather than using the compiled in default. In our
3412 case the guess is not the same as the compiled default, so the script
3413 suddenly stopped working when I started to use syslog. The patch checks
3414 to see if log_file_path is "". If so, it attempts to read it from exim
3415 with no configuration file to get the compiled in version, before it
3416 falls back to the previous guessing code."
3418 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3419 implementing BATV in an Exim configuration. See NewStuff for the gory
3422 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3423 Makefile that are specific to HP-UX.
3425 PH/18 If the "use_postmaster" option was set for a recipient callout together
3426 with the "random" option, the postmaster address was used as the MAIL
3427 FROM address for the random test, but not for the subsequent recipient
3428 test. It is now used for both.
3430 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3431 patch removes a few documentation additions to RFC 3028, because the
3432 latest draft now contains them. It adds the new en;ascii-case comparator
3433 and a new error check for 8bit text in MIME parts. Comparator and
3434 require names are now matched exactly. I enabled the subaddress
3435 extension, but it is not well tested yet (read: it works for me)."
3437 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3438 rework some of the code of TK/09 above to avoid the hardwired use of
3439 "%lld" and "long long". Replaced the call to snprintf() with a call to
3442 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3443 records point to non-existent hosts", "retry timeout exceeded", and
3444 "retry time not reached for any host after a long failure period".
3446 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3447 experimental DomainKeys support:
3449 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3450 (2) On an error such as an illegally used "control", the wrong name for
3451 the control was given.
3453 These problems did NOT occur unless DomainKeys support was compiled.
3455 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3457 PH/24 Added ${if match_ip condition.
3459 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3460 databases so that it will be absolutely obvious if a crash occurs in the
3461 DB library. This is a regular occurrence (often caused by mis-matched
3464 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3465 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3466 picked up on a user's system that detects such things. There doesn't seem
3467 to be a gcc warning option for this - only an attribute that has to be
3468 put on the function's prototype. It seems that in Fedora Core 4 they have
3469 set this on a number of new functions. No doubt there will be more in due
3472 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3473 specify a transport. However, if an address that was verified by such a
3474 router was the subject of a callout, Exim crashed because it tried to
3475 read the rcpt_include_affixes from the non-existent transport. Now it
3476 just assumes that the setting of that option is false. This bug was
3477 introduced by 4.51/PH/31.
3479 PH/28 Changed -d+all to exclude +memory, because that information is very
3480 rarely of interest, but it makes the output a lot bigger. People tend to
3481 do -d+all out of habit.
3483 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3484 code in os-type was giving problems when libc.so lives in lib64, like on
3487 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3488 aren't the modern standard, and it seems that some systems' include files
3489 don't always have them. Exim was already checking for some of the newer
3490 ones like T_AAAA, and defining it itself. I've added checks for all the
3491 record types that Exim uses.
3493 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3494 not automatically generating a new one, as it is supposed to. This
3495 prevented TLS from working. If the file did exist, but contained invalid
3496 data, a new version was generated, as expected. It was only the case of a
3497 non-existent file that was broken.
3499 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3500 with a change in libdomainkeys > 0.64.
3502 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3503 from DNS. If the selector record carries the flag, it now has
3504 precedence over the domain-wide flag.
3506 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3508 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3509 the use of a port name that isn't defined in /etc/services) occurred, the
3510 message was deferred as in a normal delivery, and thus remained on the
3511 spool, instead of being failed because of the mua_wrapper setting. This
3512 is now fixed, and I tidied up some of the mua_wrapper messages at the
3515 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3516 the messages in a hash of arrays rather than using individual hashes.
3517 This is a bit cleaner and results in dramatic memory savings, albeit
3518 at a slight CPU cost.
3520 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3521 as requested by Marc Sherman.
3523 SC/10 Eximstats: added histograms for user specified patterns as requested
3526 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3528 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3529 fopen() in the content-scanning modules that did not already have it.
3535 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3536 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3538 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3540 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3542 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3543 to test Sieve filters that use "vacation".
3545 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3546 that changes the way the GnuTLS parameters are stored in the cache file.
3547 The new format can be generated externally. For backward compatibility,
3548 if the data in the cache doesn't make sense, Exim assumes it has read an
3549 old-format file, and it generates new data and writes a new file. This
3550 means that you can't go back to an older release without removing the
3553 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3554 work if there are any delivery delays because "one_time" forces the
3555 parent to be marked "delivered", so its unseen clone is never tried
3556 again. For this reason, Exim now forbids the simultaneous setting of
3559 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3560 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3561 entry if you want to know the details. The fix, however, neglected to
3562 consider the case when local delivery batching is involved. The test for
3563 "previously delivered" was not happening when checking to see if an
3564 address could be batched with a previous (undelivered) one; under
3565 certain circumstances this could lead to multiple deliveries to the same
3568 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3569 in its include files, and this causes problems building Exim.
3571 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3572 header_syntax) but Exim was just ignoring anything given after a slash.
3573 In particular, this caused confusion with an attempt to use "verify =
3574 reverse_host_lookup/defer_ok". An error is now given when options are
3575 supplied for verify items that do not have them. (Maybe reverse_host_
3576 lookup should have a defer_ok option, but that's a different point.)
3578 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3579 defined by RFC 821) to 2048, because there were problems with some AUTH
3580 commands, and RFC 1869 says the size should be increased for extended
3581 SMTP commands that take arguments.
3583 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3586 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3587 "unknown" error; now it says that the functionality isn't in the binary.
3589 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3590 an address' error message when a string expansion fails (syntax or
3591 whatever). Otherwise the password may appear in the log. Following change
3592 PH/42 below, there is no longer a chance of it appearing in a bounce
3595 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3597 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3598 of its addresses was ever tried. (Bugzilla bug #2.)
3600 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3601 the result incorrectly in the debug output. (It correctly added a newline
3602 to what was transported.)
3604 TF/01 Added $received_time.
3606 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3607 commented out examples of how to interface to a virus scanner and to
3608 SpamAssassin. Also added commented examples of av_scanner and
3609 spamd_address settings.
3611 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3612 and controls are allowed in which ACLs. There were a couple of minor
3613 errors. Some of the entries in the conditions table (which is a table of
3614 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3615 negation of where the condition IS allowed.
3617 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3619 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3620 header file does not have a version number, so I've had to invent a new
3621 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3622 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3623 radiusclient), but it was contributed by a Radius user.
3625 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3626 files or directories, for OpenSSL.
3628 PH/19 When an Exim process that is running as root has to create an Exim log
3629 file, it does so in a subprocess that runs as exim:exim so as to get the
3630 ownership right at creation (otherwise, other Exim processes might see
3631 the file with the wrong ownership). There was no test for failure of this
3632 fork() call, which would lead to the process getting stuck as it waited
3633 for a non-existent subprocess. Forks do occasionally fail when resources
3634 run out. I reviewed all the other calls to fork(); they all seem to check
3637 PH/20 When checking for unexpected SMTP input at connect time (before writing
3638 the banner), Exim was not dealing correctly with a non-positive return
3639 from the read() function. If the client had disconnected by this time,
3640 the result was a log entry for a synchronization error with an empty
3641 string after "input=" when read() returned zero. If read() returned -1
3642 (an event I could not check), uninitialized data bytes were printed.
3643 There were reports of junk text (parts of files, etc) appearing after
3646 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3648 PH/22 Added support for macro redefinition, and (re)definition in between
3649 driver and ACL definitions.
3651 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3652 forgetting to use the resulting value; it was using the unexpanded value.
3654 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3655 hadn't been configured. The fix is from Juergen Kreileder, who
3656 understands it better than I do:
3658 "Here's what I see happening with three configured cyrus_sasl
3659 authenticators configured (plain, login, cram-md5):
3661 On startup auth_cyrus_sasl_init() gets called for each of these.
3662 This means three calls to sasl_listmech() without a specified mech_list.
3663 => SASL tests which mechs of all available mechs actually work
3664 => three warnings about OTP not working
3665 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3667 With the patch, sasl_listmech() also gets called three times. But now
3668 SASL's mech_list option is set to the server_mech specified in the the
3669 authenticator. Or in other words, the answer from sasl_listmech()
3670 gets limited to just the mech you're testing for (which is different
3672 => the return list contains just 'plain' or 'login', 'cram-md5' or
3673 nothing depending on the value of ob->server_mech.
3675 I've just tested the patch: Authentication still works fine,
3676 unavailable mechs specified in the exim configuration are still
3677 caught, and the auth.log warnings about OTP are gone."
3679 PH/25 When debugging is enabled, the contents of the command line are added
3680 to the debugging output, even when log_selector=+arguments is not
3683 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3684 answer is "GNU", and only if the return is "GNU/something" is the answer
3687 PH/27 $acl_verify_message is now set immediately after the failure of a
3688 verification in an ACL, and so is available in subsequent modifiers. In
3689 particular, the message can be preserved by coding like this:
3691 warn !verify = sender
3692 set acl_m0 = $acl_verify_message
3694 Previously, $acl_verify_message was set only while expanding "message"
3695 and "log_message" when a very denied access.
3697 PH/28 Modified OS/os.c-Linux with
3699 -#ifndef OS_LOAD_AVERAGE
3700 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3702 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3703 nomenclature these days.)
3705 PH/29 Installed patch from the Sieve maintainer that adds the options
3706 sieve_useraddress and sieve_subaddress to the redirect router.
3708 PH/30 In these circumstances:
3709 . Two addresses routed to the same list of hosts;
3710 . First host does not offer TLS;
3711 . First host accepts first address;
3712 . First host gives temporary error to second address;
3713 . Second host offers TLS and a TLS session is established;
3714 . Second host accepts second address.
3715 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3716 and peerdn, if requested) that were in fact used only for the second
3719 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3720 attention to any local part prefix or suffix that was matched by the
3721 router that accepted the address. It now behaves in the same way as it
3722 does for delivery: the affixes are removed from the local part unless
3723 rcpt_include_affixes is set on the transport.
3725 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3726 timeout during the DATA phase of an incoming message.
3728 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3729 applied a patch sanctioned by the Sieve maintainer.
3731 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3732 the uid or gid is negative. A case of a negative gid caused this to be
3733 noticed. The fix allows for either to be negative.
3735 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3736 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3739 PH/36 The change PH/12 above was broken. Fixed it.
3741 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3742 the grounds that routing the same address twice would always produce the
3743 same answer. This might have been true once, but it is certainly no
3744 longer true now. Routing a child address may depend on the previous
3745 routing that produced that child. Some complicated redirection strategies
3746 went wrong when messages had multiple recipients, and made Exim's
3747 behaviour dependent on the order in which the addresses were given.
3749 I have moved the duplicate checking until after the routing is complete.
3750 Exim scans the addresses that are assigned to local and remote
3751 transports, and removes any duplicates. This means that more work will be
3752 done, as duplicates will always all be routed, but duplicates are
3753 presumably rare, so I don't expect this is of any significance.
3755 For deliveries to pipes, files, and autoreplies, the duplicate checking
3756 still happens during the routing process, since they are not going to be
3759 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3760 It corrects a timeout issue with spamd. This is Ian's comment: "The
3761 background is that sometimes spamd either never reads data from a
3762 connection it has accepted, or it never writes response data. The exiscan
3763 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3764 blindly assumes that writes won't block so it may never time out."
3766 PH/39 Allow G after quota size as well as K and M.
3768 PH/40 The value set for $authenticated_id in an authenticator may not contain
3769 binary zeroes or newlines because the value is written to log lines and
3770 to spool files. There was no check on this. Now the value is run through
3771 the string_printing() function so that such characters are converted to
3772 printable escape sequences.
3774 PH/41 $message_linecount is a new variable that contains the total number of
3775 lines in the message. Compare $body_linecount, which is the count for the
3778 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3779 bounce and delay warning messages, except in certain special cases, which
3782 (a) An SMTP error message from a remote host;
3783 (b) A message specified in a :fail: redirection;
3784 (c) A message specified in a "fail" command in a system filter;
3785 (d) A message specified in a FAIL return from the queryprogram router;
3786 (e) A message specified by the cannot_route_message router option.
3788 In these cases only, Exim does include the error details in bounce and
3789 warning messages. There are also a few cases where bland messages such
3790 as "unrouteable address" or "local delivery error" are given.
3792 PH/43 $value is now also set for the "else" part of a ${run expansion.
3794 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3795 being worked on, but at least Exim now implements the latest version to
3798 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3799 process to complete was treated as a delivery failure, a timeout while
3800 writing the message to the pipe was logged, but erroneously treated as a
3801 successful delivery. Such timeouts include transport filter timeouts. For
3802 consistency with the overall process timeout, these timeouts are now
3803 treated as errors, giving rise to delivery failures by default. However,
3804 there is now a new Boolean option for the pipe transport called
3805 timeout_defer, which, if set TRUE, converts the failures into defers for
3806 both kinds of timeout. A transport filter timeout is now identified in
3809 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3810 systems where "make" and "gmake" are different, calling "gmake" at top
3811 level broke things. I've arranged for the value of $(MAKE) to be passed
3812 from the Makefile to this script so that it can call the same version of
3816 A note about Exim versions 4.44 and 4.50
3817 ----------------------------------------
3819 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3820 changes of various kinds. As a consequence, a big documentation update was
3821 needed. This delayed the release for rather longer than seemed good, especially
3822 in the light of a couple of (minor) security issues. Therefore, the changes
3823 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3824 release. So 4.44 and 4.50 are in effect two different branches that both start
3827 I have left the 4.50 change log unchanged; it contains all the changes since
3828 4.43. The change log for 4.44 is below; many of its items are identical to
3829 those for 4.50. This seems to be the most sensible way to preserve the
3830 historical information.
3836 1. Minor wording change to the doc/README.SIEVE file.
3838 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3839 computation of the current number of files was incorrect.
3841 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3842 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3845 4. Give more explanation in the error message when the command for a transport
3846 filter fails to execute.
3848 5. There are several places where Exim runs a non-Exim command in a
3849 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3850 was being done only for the command run by the queryprogram router. It is
3851 now done for all such subprocesses. The other cases are: ${run, transport
3852 filters, and the commands run by the lmtp and pipe transports.
3854 6. Added CONFIGURE_GROUP build-time option.
3856 7. Some older OS have a limit of 256 on the maximum number of file
3857 descriptors. Exim was using setrlimit() to set 1000 as a large value
3858 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3859 systems. I've change it so that if it can't get 1000, it tries for 256.
3861 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3862 was an oversight, and furthermore, ever since the addition of extra
3863 controls (e.g. 4.43/32), the checks on when to allow different forms of
3864 "control" were broken. There should now be diagnostics for all cases when a
3865 control that does not make sense is encountered.
3867 9. Added the /retain_sender option to "control=submission".
3869 10. $recipients is now available in the predata ACL (oversight).
3871 11. Tidy the search cache before the fork to do a delivery from a message
3872 received from the command line. Otherwise the child will trigger a lookup
3873 failure and thereby defer the delivery if it tries to use (for example) a
3874 cached ldap connection that the parent has called unbind on.
3876 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3877 of $address_data from the recipient verification was clobbered by the
3878 sender verification.
3880 13. The value of address_data from a sender verification is now available in
3881 $sender_address_data in subsequent conditions in the ACL statement.
3883 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3885 15. Added a new option "connect=<time>" to callout options, to set a different
3888 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3889 was its contents. (It was OK if the option was not defined at all.)
3891 17. A "Completed" log line is now written for messages that are removed from
3892 the spool by the -Mrm option.
3894 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3895 information about exactly what failed.
3897 19. Added -dd to debug only the daemon process.
3899 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3900 handles timeouts, both on the server side and network timeouts. Renamed the
3901 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3903 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3904 It is now set to "smtps".
3906 22. $host_address is now set to the target address during the checking of
3907 ignore_target_hosts.
3909 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3910 being passed; this would have caused $sender_host_name to have been used if
3911 matching the list had actually called for a host name (not very likely,
3912 since this list is usually IP addresses). A host name is now passed as
3915 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3916 code that specifies a non-restarting handler (typically sigaction() in
3917 modern systems) in an attempt to fix a rare and obscure crash bug.
3919 25. Narrowed the window for a race in the daemon that could cause it to ignore
3920 SIGCHLD signals. This is not a major problem, because they are used only to
3921 wake it up if nothing else does.
3923 26. A malformed maildirsize file could cause Exim to calculate negative values
3924 for the mailbox size or file count. Odd effects could occur as a result.
3925 The maildirsize information is now recalculated if the size or filecount
3928 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3929 support for a long time. Removed HAVE_SYS_VFS_H.
3931 28. Installed the latest version of exipick from John Jetmore.
3933 29. In an address list, if the pattern was not a regular expression, an empty
3934 subject address (from a bounce message) matched only if the pattern was an
3935 empty string. Non-empty patterns were not even tested. This was the wrong
3936 because it is perfectly reasonable to use an empty address as part of a
3937 database query. An empty address is now tested by patterns that are
3938 lookups. However, all the other forms of pattern expect the subject to
3939 contain a local part and a domain, and therefore, for them, an empty
3940 address still always fails if the pattern is not itself empty.
3942 30. Exim went into a mad DNS loop when attempting to do a callout where the
3943 host was specified on an smtp transport, and looking it up yielded more
3944 than one IP address.
3946 31. Re-factored the code for checking spool and log partition space into a
3947 function that finds that data and another that does the check. The former
3948 is then used to implement four new variables: $spool_space, $log_space,
3949 $spool_inodes, and $log_inodes.
3951 32. The RFC2047 encoding function was originally intended for short strings
3952 such as real names; it was not keeping to the 75-character limit for
3953 encoded words that the RFC imposes. It now respects the limit, and
3954 generates multiple encoded words if necessary. To be on the safe side, I
3955 have increased the buffer size for the ${rfc2047: expansion operator from
3958 33. It is now permitted to omit both strings after an "if" condition; if the
3959 condition is true, the result is "true". As before, when the second string
3960 is omitted, a false condition yields an empty string. This makes it less
3961 cumbersome to write custom ACL and router conditions.
3963 34. Failure to deliver a bounce message always caused it to be frozen, even if
3964 there was an errors_to setting on the router. The errors_to setting is now
3967 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3968 canonical form (fully expanded) before being placed in
3969 $sender_host_address.
3971 36. The table in the code that translates DNS record types into text (T_A to
3972 "A" for instance) was missing entries for NS and CNAME. It is just possible
3973 that this could have caused confusion if both these types were looked up
3974 for the same domain, because the text type is used as part of Exim's
3975 per-process caching. But the chance of anyone hitting this buglet seems
3978 37. The dnsdb lookup has been extended in a number of ways.
3980 (1) There is a new type, "zns", which walks up the domain tree until it
3981 finds some nameserver records. It should be used with care.
3983 (2) There is a new type, "mxh", which is like "mx" except that it returns
3984 just the host names, not the priorities.
3986 (3) It is now possible to give a list of domains (or IP addresses) to be
3987 looked up. The behaviour when one of the lookups defers can be
3988 controlled by a keyword.
3990 (4) It is now possible to specify the separator character for use when
3991 multiple records are returned.
3993 38. The dnslists ACL condition has been extended: it is now possible to supply
3994 a list of IP addresses and/or domains to be looked up in a particular DNS
3997 39. Added log_selector=+queue_time_overall.
3999 40. When running the queue in the test harness, wait just a tad after forking a
4000 delivery process, to get repeatability of debugging output.
4002 41. Include certificate and key file names in error message when GnuTLS fails
4003 to set them up, because the GnuTLS error message doesn't include the name
4004 of the failing file when there is a problem reading it.
4006 42. Allow both -bf and -bF in the same test run.
4008 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4010 44. The "Exiscan patch" is now merged into the mainline Exim source.
4012 45. Sometimes the final signoff response after QUIT could fail to get
4013 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4014 < 0 before doing a fflush(). This bug looks as though it goes back to the
4015 introduction of TLS in release 3.20, but "sometimes" must have been rare
4016 because the tests only now provoked it.
4018 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4019 (this can affect the format of dates).
4021 47. exim_tidydb, when checking for the continued existence of a message for
4022 which it has found a message-specific retry record, was not finding
4023 messages that were in split spool directories. Consequently, it was
4024 deleting retry records that should have stayed in existence.
4026 48. Steve fixed some bugs in eximstats.
4028 49. The SPA authentication driver was not abandoning authentication and moving
4029 on to the next authenticator when an expansion was forced to fail,
4030 contradicting the general specification for all authenticators. Instead it
4031 was generating a temporary error. It now behaves as specified.
4033 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4034 (the order specifies the preference for clients). The order is now AES256,
4035 AES128, 3DES, ARCFOUR128.
4037 51. Small patch to Sieve code - explicitly set From: when generating an
4040 52. Exim crashed if a remote delivery caused a very long error message to be
4041 recorded - for instance if somebody sent an entire SpamAssassin report back
4042 as a large number of 550 error lines. This bug was coincidentally fixed by
4043 increasing the size of one of Exim's internal buffers (big_buffer) that
4044 happened as part of the Exiscan merge. However, to be on the safe side, I
4045 have made the code more robust (and fixed the comments that describe what
4048 53. Now that there can be additional text after "Completed" in log lines (if
4049 the queue_time_overall log selector is set), a one-byte patch to exigrep
4050 was needed to allow it to recognize "Completed" as not the last thing in
4053 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4054 patch that reportedly fixes this has been added. I am not expert enough to
4055 create a test for it. This is what the patch creator wrote:
4057 "I found a little strange behaviour of ldap code when working with
4058 Windows 2003 AD Domain, where users was placed in more than one
4059 Organization Units. When I tried to give exim partial DN, the exit code
4060 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4061 But simultaneously result of request was absolutely normal ldap result,
4062 so I produce this patch..."
4064 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4065 REFERENCE, so I have modified the code to exclude the patch when that macro
4068 55. Some experimental protocols are using DNS PTR records for new purposes. The
4069 keys for these records are domain names, not reversed IP addresses. The
4070 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4071 leaves it alone. Component reversal etc. now happens only for IP addresses.
4074 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4076 57. Double the size of the debug message buffer (to 2048) so that more of very
4077 long debug lines gets shown.
4079 58. The exicyclog utility now does better if the number of log files to keep
4080 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4082 59. Two changes related to the smtp_active_hostname option:
4084 (1) $smtp_active_hostname is now available as a variable.
4085 (2) The default for smtp_banner uses $smtp_active_hostname instead
4086 of $primary_hostname.
4088 60. The host_aton() function is supposed to be passed a string that is known
4089 to be a valid IP address. However, in the case of IPv6 addresses, it was
4090 not checking this. This is a hostage to fortune. Exim now panics and dies
4091 if the condition is not met. A case was found where this could be provoked
4092 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4093 components; fortuitously, this particular loophole had already been fixed
4094 by change 4.50/55 above.
4096 If there are any other similar loopholes, the new check in host_aton()
4097 itself should stop them being exploited. The report I received stated that
4098 data on the command line could provoke the exploit when Exim was running as
4099 exim, but did not say which command line option was involved. All I could
4100 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4101 running as the user.
4104 61. There was a buffer overflow vulnerability in the SPA authentication code
4105 (which came originally from the Samba project). I have added a test to the
4106 spa_base64_to_bits() function which I hope fixes it.
4109 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4110 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4112 63. The daemon start-up calls getloadavg() while still root for those OS that
4113 need the first call to be done as root, but it missed one case: when
4114 deliver_queue_load_max is set with deliver_drop_privilege. This is
4115 necessary for the benefit of the queue runner, because there is no re-exec
4116 when deliver_drop_privilege is set.
4118 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4119 This has been fixed.
4121 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4122 was in use, was not putting the data itself into the right store pool;
4123 consequently, it could be overwritten for a subsequent message in the same
4124 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4127 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4129 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4130 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4131 the function were treating the return as a boolean value, which happened to
4132 work because 0=false and not-0=true, but is not correct code.
4134 68. The host_aton() function was not handling scoped IPv6 addresses (those
4135 with, for example, "%eth0" on the end) correctly.
4137 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4138 negated items (that is, ~something) in unsigned ints. Some compilers
4139 apparently mutter when there is no cast.
4141 70. If an address verification called from an ACL failed, and did not produce a
4142 user-specific message (i.e. there was only a "system" message), nothing was
4143 put in $acl_verify_message. In this situation, it now puts the system
4146 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4147 session; change 4.31/43 added the unwanted input to the log line - except
4148 that it did not do this in the start of session case. It now does.
4150 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4151 This is wrong and can cause the other end to generate a synchronization
4152 error if it is another Exim or anything else that does the synchronization
4153 check. A QUIT command is no longer sent after a timeout.
4155 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4156 during host lookups.
4158 74. The defer_ok option of callout verification was not working if it was used
4159 when verifying addresses in header lines, that is, for this case:
4161 verify = header_sender/callout=defer_ok
4163 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4164 those file descriptors could be used for SMTP connections. If anything
4165 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4166 could be sent to the SMTP client, causing chaos. The daemon now opens
4167 stdin, stdout, and stderr to /dev/null when it puts itself into the
4170 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4171 log by default. The user can override this with suitable Perl magic.
4173 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4174 the log message when discard triggers, was not working for the DATA ACL or
4175 for the non-SMTP ACL.
4177 78. Error message wording change in sieve.c.
4179 79. If smtp_accept_max_per_host was set, the number of connections could be
4180 restricted to fewer than expected, because the daemon was trying to set up
4181 a new connection before checking whether the processes handling previous
4182 connections had finished. The check for completed processes is now done
4183 earlier. On busy systems, this bug wouldn't be noticed because something
4184 else would have woken the daemon, and it would have reaped the completed
4187 80. If a message was submitted locally by a user whose login name contained one
4188 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4189 It caused a spool format error. I have fixed the spool reading code. A
4190 related problem was that the "from" clause in the Received: line became
4191 illegal because of the space(s). It is now covered by ${quote_local_part.
4193 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4196 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4198 83. Patch from Sieve maintainer to fix unterminated string problem in
4199 "vacation" handling.
4201 84. Some minor changes to the Linux configuration files to help with other
4202 OS variants using glibc.
4204 85. One more patch for Sieve to update vacation handling to latest spec.
4207 ----------------------------------------------------
4208 See the note above about the 4.44 and 4.50 releases.
4209 ----------------------------------------------------
4215 1. Change 4.43/35 introduced a bug that caused file counts to be
4216 incorrectly computed when quota_filecount was set in an appendfile
4219 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4220 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4223 3. Give more explanation in the error message when the command for a transport
4224 filter fails to execute.
4226 4. There are several places where Exim runs a non-Exim command in a
4227 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4228 was being done only for the command run by the queryprogram router. It is
4229 now done for all such subprocesses. The other cases are: ${run, transport
4230 filters, and the commands run by the lmtp and pipe transports.
4232 5. Some older OS have a limit of 256 on the maximum number of file
4233 descriptors. Exim was using setrlimit() to set 1000 as a large value
4234 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4235 systems. I've change it so that if it can't get 1000, it tries for 256.
4237 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4238 was an oversight, and furthermore, ever since the addition of extra
4239 controls (e.g. 4.43/32), the checks on when to allow different forms of
4240 "control" were broken. There should now be diagnostics for all cases when a
4241 control that does not make sense is encountered.
4243 7. $recipients is now available in the predata ACL (oversight).
4245 8. Tidy the search cache before the fork to do a delivery from a message
4246 received from the command line. Otherwise the child will trigger a lookup
4247 failure and thereby defer the delivery if it tries to use (for example) a
4248 cached ldap connection that the parent has called unbind on.
4250 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4251 of $address_data from the recipient verification was clobbered by the
4252 sender verification.
4254 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4255 was its contents. (It was OK if the option was not defined at all.)
4257 11. A "Completed" log line is now written for messages that are removed from
4258 the spool by the -Mrm option.
4260 12. $host_address is now set to the target address during the checking of
4261 ignore_target_hosts.
4263 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4264 being passed; this would have caused $sender_host_name to have been used if
4265 matching the list had actually called for a host name (not very likely,
4266 since this list is usually IP addresses). A host name is now passed as
4269 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4270 code that specifies a non-restarting handler (typically sigaction() in
4271 modern systems) in an attempt to fix a rare and obscure crash bug.
4273 15. Narrowed the window for a race in the daemon that could cause it to ignore
4274 SIGCHLD signals. This is not a major problem, because they are used only to
4275 wake it up if nothing else does.
4277 16. A malformed maildirsize file could cause Exim to calculate negative values
4278 for the mailbox size or file count. Odd effects could occur as a result.
4279 The maildirsize information is now recalculated if the size or filecount
4282 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4283 support for a long time. Removed HAVE_SYS_VFS_H.
4285 18. Updated exipick to current release from John Jetmore.
4287 19. Allow an empty sender to be matched against a lookup in an address list.
4288 Previously the only cases considered were a regular expression, or an
4291 20. Exim went into a mad DNS lookup loop when doing a callout where the
4292 host was specified on the transport, if the DNS lookup yielded more than
4295 21. The RFC2047 encoding function was originally intended for short strings
4296 such as real names; it was not keeping to the 75-character limit for
4297 encoded words that the RFC imposes. It now respects the limit, and
4298 generates multiple encoded words if necessary. To be on the safe side, I
4299 have increased the buffer size for the ${rfc2047: expansion operator from
4302 22. Failure to deliver a bounce message always caused it to be frozen, even if
4303 there was an errors_to setting on the router. The errors_to setting is now
4306 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4307 canonical form (fully expanded) before being placed in
4308 $sender_host_address.
4310 24. Updated eximstats to version 1.33
4312 25. Include certificate and key file names in error message when GnuTLS fails
4313 to set them up, because the GnuTLS error message doesn't include the name
4314 of the failing file when there is a problem reading it.
4316 26. Expand error message when OpenSSL has problems setting up cert/key files.
4319 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4320 (this can affect the format of dates).
4322 28. exim_tidydb, when checking for the continued existence of a message for
4323 which it has found a message-specific retry record, was not finding
4324 messages that were in split spool directories. Consequently, it was
4325 deleting retry records that should have stayed in existence.
4327 29. eximstats updated to version 1.35
4328 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4329 1.35 - bugfix such that pie charts by volume are generated correctly
4331 30. The SPA authentication driver was not abandoning authentication and moving
4332 on to the next authenticator when an expansion was forced to fail,
4333 contradicting the general specification for all authenticators. Instead it
4334 was generating a temporary error. It now behaves as specified.
4336 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4337 (the order specifies the preference for clients). The order is now AES256,
4338 AES128, 3DES, ARCFOUR128.
4340 31. Small patch to Sieve code - explicitly set From: when generating an
4343 32. Exim crashed if a remote delivery caused a very long error message to be
4344 recorded - for instance if somebody sent an entire SpamAssassin report back
4345 as a large number of 550 error lines. This bug was coincidentally fixed by
4346 increasing the size of one of Exim's internal buffers (big_buffer) that
4347 happened as part of the Exiscan merge. However, to be on the safe side, I
4348 have made the code more robust (and fixed the comments that describe what
4351 33. Some experimental protocols are using DNS PTR records for new purposes. The
4352 keys for these records are domain names, not reversed IP addresses. The
4353 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4354 leaves it alone. Component reversal etc. now happens only for IP addresses.
4357 34. The host_aton() function is supposed to be passed a string that is known
4358 to be a valid IP address. However, in the case of IPv6 addresses, it was
4359 not checking this. This is a hostage to fortune. Exim now panics and dies
4360 if the condition is not met. A case was found where this could be provoked
4361 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4362 components; fortuitously, this particular loophole had already been fixed
4363 by change 4.50/55 or 4.44/33 above.
4365 If there are any other similar loopholes, the new check in host_aton()
4366 itself should stop them being exploited. The report I received stated that
4367 data on the command line could provoke the exploit when Exim was running as
4368 exim, but did not say which command line option was involved. All I could
4369 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4370 running as the user.
4373 35. There was a buffer overflow vulnerability in the SPA authentication code
4374 (which came originally from the Samba project). I have added a test to the
4375 spa_base64_to_bits() function which I hope fixes it.
4378 36. The daemon start-up calls getloadavg() while still root for those OS that
4379 need the first call to be done as root, but it missed one case: when
4380 deliver_queue_load_max is set with deliver_drop_privilege. This is
4381 necessary for the benefit of the queue runner, because there is no re-exec
4382 when deliver_drop_privilege is set.
4384 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4385 was in use, was not putting the data itself into the right store pool;
4386 consequently, it could be overwritten for a subsequent message in the same
4387 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4390 38. Sometimes the final signoff response after QUIT could fail to get
4391 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4392 < 0 before doing a fflush(). This bug looks as though it goes back to the
4393 introduction of TLS in release 3.20, but "sometimes" must have been rare
4394 because the tests only now provoked it.
4400 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4401 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4402 Now it returns an integer. A number of calls were still expecting a T/F
4403 return. Fortuitously, in all cases, the tests worked in OK situations,
4404 which is the norm. However, things would have gone wrong on any write
4405 failures on the smtp file descriptor. This function is used when sending
4406 messages over SMTP and also when doing verify callouts.
4408 2. When Exim is called to do synchronous delivery of a locally submitted
4409 message (the -odf or -odi options), it no longer closes stderr before doing
4412 3. Implemented the mua_wrapper option.
4414 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4416 5. Implemented the functions header_remove(), header_testname(),
4417 header_add_at_position(), and receive_remove_recipient(), and exported them
4420 6. If an ACL "warn" statement specified the addition of headers, Exim already
4421 inserted X-ACL-Warn: at the start if there was no header name. However, it
4422 was not making this test for the second and subsequent header lines if
4423 there were newlines in the string. This meant that an invalid header could
4424 be inserted if Exim was badly configured.
4426 7. Allow an ACL "warn" statement to add header lines at the start or after all
4427 the Received: headers, as well as at the end.
4429 8. Added the rcpt_4xx retry error code.
4431 9. Added postmaster_mailfrom=xxx to callout verification option.
4433 10. Added mailfrom=xxxx to the callout verification option, for verify=
4436 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4437 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4439 12. Inserted some casts to stop certain compilers warning when using pointer
4440 differences as field lengths or precisions in printf-type calls (mostly
4441 affecting debugging statements).
4443 13. Added optional readline() support for -be (dynamically loaded).
4445 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4446 same clock tick as a message's arrival, so that its received time was the
4447 same as the "first fail" time on the retry record, and that message
4448 remained on the queue past the ultimate address timeout, every queue runner
4449 would try a delivery (because it was past the ultimate address timeout) but
4450 after another failure, the ultimate address timeout, which should have then
4451 bounced the address, did not kick in. This was a "< instead of <=" error;
4452 in most cases the first failure would have been in the next clock tick
4453 after the received time, and all would be well.
4455 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4456 being recognized when the domain list was tested by the match_domain
4457 condition in an expansion string.
4459 16. Added the ${str2b64: operator.
4461 17. Exim was always calling setrlimit() to set a large limit for the number of
4462 processes, without checking whether the existing limit was already
4463 adequate. (It did check for the limit on file descriptors.) Furthermore,
4464 errors from getrlimit() and setrlimit() were being ignored. Now they are
4465 logged to the main and panic logs, but Exim does carry on, to try to do its
4466 job under whatever limits there are.
4468 18. Imported PCRE 5.0.
4470 19. Trivial typo in log message " temporarily refused connection" (the leading
4473 20. If the log selector return_path_on_delivery was set and an address was
4474 redirected to /dev/null, the delivery process crashed because it assumed
4475 that a return path would always be set for a "successful" delivery. In this
4476 case, the whole delivery is bypassed as an optimization, and therefore no
4479 21. Internal re-arrangement: the function for sending a challenge and reading
4480 a response while authentication was assuming a zero-terminated challenge
4481 string. It's now changed to take a pointer and a length, to allow for
4482 binary data in such strings.
4484 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4486 23. Exim was not respecting finduser_retries when seeking the login of the
4487 uid under which it was called; it was always trying 10 times. (The default
4488 setting of finduser_retries is zero.) Also, it was sleeping after the final
4489 failure, which is pointless.
4491 24. Implemented tls_on_connect_ports.
4493 25. Implemented acl_smtp_predata.
4495 26. If the domain in control=submission is set empty, Exim assumes that the
4496 authenticated id is a complete email address when it generates From: or
4497 Sender: header lines.
4499 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4500 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4501 chown and chgrp in /bin and hostname in /usr/bin.
4503 28. Exim was keeping the "process log" file open after each use, just as it
4504 does for the main log. This opens the possibility of it remaining open for
4505 long periods when the USR1 signal hits a daemon. Occasional processlog
4506 errors were reported, that could have been caused by this. Anyway, it seems
4507 much more sensible not to leave this file open at all, so that is what now
4510 29. The long-running daemon process does not normally write to the log once it
4511 has entered its main loop, and it closes the log before doing so. This is
4512 so that log files can straightforwardly be renamed and moved. However,
4513 there are a couple of unusual error situations where the daemon does write
4514 log entries, and I had neglected to close the log afterwards.
4516 30. The text of an SMTP error response that was received during a remote
4517 delivery was being truncated at 512 bytes. This is too short for some of
4518 the long messages that one sometimes sees. I've increased the limit to
4521 31. It is now possible to make retry rules that apply only when a message has a
4522 specific sender, in particular, an empty sender.
4524 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4525 it possible to be selective about when SMTP synchronization is enforced.
4527 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4529 32. Implemented hosts_connection_nolog.
4531 33. Added an ACL for QUIT.
4533 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4536 35. Added mailbox_size and mailbox_filecount to appendfile.
4538 36. Added control = no_multiline_responses to ACLs.
4540 37. There was a bug in the logic of the code that waits for the clock to tick
4541 in the case where the clock went backwards by a substantial amount such
4542 that the microsecond fraction of "now" was more than the microsecond
4543 fraction of "then" (but the whole seconds number was less).
4545 38. Added support for the libradius Radius client library this is found on
4546 FreeBSD (previously only the radiusclient library was supported).
4552 1. When certain lookups returned multiple values in the form name=value, the
4553 quoting of the values was not always being done properly. Specifically:
4554 (a) If the value started with a double quote, but contained no whitespace,
4556 (b) If the value contained whitespace other than a space character (i.e.
4557 tabs or newlines or carriage returns) it was not quoted.
4558 This fix has been applied to the mysql and pgsql lookups by writing a
4559 separate quoting function and calling it from the lookup code. The fix
4560 should probably also be applied to nisplus, ibase and oracle lookups, but
4561 since I cannot test any of those, I have not disturbed their existing code.
4563 2. A hit in the callout cache for a specific address caused a log line with no
4564 reason for rejecting RCPT. Now it says "Previous (cached) callout
4565 verification failure".
4567 3. There was an off-by-one bug in the queryprogram router. An over-long
4568 return line was truncated at 256 instead of 255 characters, thereby
4569 overflowing its buffer with the terminating zero. As well as fixing this, I
4570 have increased the buffer size to 1024 (and made a note to document this).
4572 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4573 when Exim is waiting for an SMTP response from a remote server, Exim
4574 restarts its select() call on the socket, thereby resetting its timeout.
4575 This is not a problem when such interrupts are rare. Somebody set up a cron
4576 job to run exiwhat every 2 minutes, which is less than the normal select()
4577 timeout (5 or 10 minutes). This meant that the select() timeout never
4578 kicked in because it was always reset. I have fixed this by comparing the
4579 time when an interrupt arrives with the time at the start of the first call
4580 to select(). If more time than the timeout has elapsed, the interrupt is
4581 treated as a timeout.
4583 5. Some internal re-factoring in preparation for the addition of Sieve
4584 extensions (by MH). In particular, the "personal" test is moved to a
4585 separate function, and given an option for scanning Cc: and Bcc: (which is
4586 not set for Exim filters).
4588 6. When Exim created an email address using the login of the caller as the
4589 local part (e.g. when creating a From: or Sender: header line), it was not
4590 quoting the local part when it contained special characters such as @.
4592 7. Installed new OpenBSD configuration files.
4594 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4595 try to make them clearer.
4597 9. Callout options, other than the timeout value, were being ignored when
4598 verifying sender addresses in header lines. For example, when using
4600 verify = header_sender/callout=no_cache
4602 the cache was (incorrectly) being used.
4604 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4605 only the Cygwin environment.
4607 11. When return_path_on_delivery was set as a log selector, if different remote
4608 addresses in the same message used different return paths and parallel
4609 remote delivery occurred, the wrong values would sometimes be logged.
4610 (Whenever a remote delivery process finished, the return path value from
4611 the most recently started remote delivery process was logged.)
4613 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4614 lines when AUTH and/or TLS are in use. This is the "received protocol"
4615 field. Exim used to use "asmtp" for authenticated SMTP, without any
4616 indication (in the protocol name) for TLS use. Now it follows the RFC and
4617 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4618 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4619 names appear in log lines as well as in Received: header lines.
4621 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4622 extensions, and comparison tests, and to fix some bugs.
4624 14. Changes to the "personal" filter test:
4626 (1) The test was buggy in that it was just doing the equivalent of
4627 "contains" tests on header lines. For example, if a user's address was
4628 anne@some.where, the "personal" test would incorrectly be true for
4630 To: susanne@some.where
4632 This test is now done by extracting each address from the header in turn,
4633 and checking the entire address. Other tests that are part of "personal"
4634 are now done using regular expressions (for example, to check local parts
4635 of addresses in From: header lines).
4637 (2) The list of non-personal local parts in From: addresses has been
4638 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4639 taken from the Sieve specification recommendations.
4641 (3) If the message contains any header line starting with "List-" it is
4642 treated as non-personal.
4644 (4) The test for "circular" in the Subject: header line has been removed
4645 because it now seems ill-conceived.
4647 15. Minor typos in src/EDITME comments corrected.
4649 16. Installed latest exipick from John Jetmore.
4651 17. If headers_add on a router specified a text string that was too long for
4652 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4653 of string_sprintf() is now avoided.
4655 18. $message_body_size was not set (it was always zero) when running the DATA
4656 ACL and the local_scan() function.
4658 19. For the "mail" command in an Exim filter, no default was being set for
4659 the once_repeat time, causing a random time value to be used if "once" was
4660 specified. (If the value happened to be <= 0, no repeat happened.) The
4661 default is now 0s, meaning "never repeat". The "vacation" command was OK
4662 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4663 (I found it when inspecting the code).
4665 20. There is now an overall timeout for performing a callout verification. It
4666 defaults to 4 times the callout timeout, which applies to individual SMTP
4667 commands during the callout. The overall timeout applies when there is more
4668 than one host that can be tried. The timeout is checked before trying the
4669 next host. This prevents very long delays if there are a large number of
4670 hosts and all are timing out (e.g. when the network connections are timing
4671 out). The value of the overall timeout can be changed by specifying an
4672 additional sub-option for "callout", called "maxwait". For example:
4674 verify = sender/callout=5s,maxwait=20s
4676 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4677 to the end before writing, but this should make it even safer).
4679 22. Exim was forgetting that it had advertised PIPELINING for the second and
4680 subsequent messages on an SMTP connection. It was also not resetting its
4681 memory on STARTTLS and an internal HELO.
4683 23. When Exim logs an SMTP synchronization error within a session, it now
4684 records whether PIPELINING has been advertised or not.
4686 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4687 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4688 rather than long int.
4690 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4692 26. Added the never_mail option to autoreply.
4698 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4699 crash if the getsockname() call failed; this can happen if a connection is
4700 closed very soon after it is established. The problem was simply in the
4701 order in which certain operations were done, causing Exim to try to write
4702 to the SMTP stream before it had set up the file descriptor. The bug has
4703 been fixed by making things happen in the correct order.
4709 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4710 before the connection was closed, thus losing the rejection response.
4712 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4713 some early Solaris releases, but causes trouble in current releases where
4714 socklen_t is defined.
4716 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4719 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4722 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4723 that releasing the top of it at the end releases what was used for sub-
4724 expansions (unless the block got too big). However, discard this block if
4725 the first thing is a variable or header, so that we can use its block when
4726 it is dynamic (useful for very large $message_headers, for example).
4728 6. Lookups now cache *every* query, not just the most recent. A new, separate
4729 store pool is used for this. It can be recovered when all lookup caches are
4730 flushed. Lookups now release memory at the end of their result strings.
4731 This has involved some general refactoring of the lookup sources.
4733 7. Some code has been added to the store_xxx() functions to reduce the amount
4734 of flapping under certain conditions.
4736 8. log_incoming_interface used to affect only the <= reception log lines. Now
4737 it causes the local interface and port to be added to several more SMTP log
4738 lines, for example "SMTP connection from", and rejection lines.
4740 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4742 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4744 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4745 could be overwritten at the end of the current message (or the start of a
4746 new message if it was set in a HELO ACL). The value is now preserved for
4747 the duration of the SMTP connection.
4749 12. If a transport had a headers_rewrite setting, and a matching header line
4750 contained an unqualified address, that address was qualified, even if it
4751 did not match any rewriting rules. The underlying bug was that the values
4752 of the flags that permit the existence of unqualified sender and recipient
4753 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4754 non-local messages, and by -bnq for local messages) were not being
4755 preserved with the message after it was received.
4757 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4758 "next input=" as part of the text comprising the host identity instead of
4759 the correct text. The code was using the same buffer for two different
4760 strings. However, depending on which order the printing function evaluated
4761 its arguments, the bug did not always show up. Under Linux, for example, my
4762 test suite worked just fine.
4764 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4765 doesn't work with some older versions of Perl. It has been changed to "my",
4766 which in any case is probably the better facility to use.
4768 15. A really picky compiler found some instances of statements for creating
4769 error messages that either had too many or two few arguments for the format
4772 16. The size of the buffer for calls to the DNS resolver has been increased
4773 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4774 for addresses that have a lot of PTR records. This alleviates a problem; it
4775 does not fully solve it.
4777 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4778 buffer now truncates the list and logs the incident, which is the same
4779 action as happens when Exim is looking up a host name and its aliases.
4780 Previously in this situation something unpredictable would happen;
4781 sometimes it was "internal error: store_reset failed".
4783 18. If a server dropped the connection unexpectedly when an Exim client was
4784 using GnuTLS and trying to read a response, the client delivery process
4785 crashed while trying to generate an error log message.
4787 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4788 string, for example:
4790 warn message = H1: something\nH2: something
4792 the text was added as a single header line from Exim's point of view
4793 though it ended up OK in the delivered message. However, searching for the
4794 second and subsequent header lines using $h_h2: did not work. This has been
4795 fixed. Similarly, if a system filter added multiple headers in this way,
4796 the routers could not see them.
4798 20. Expanded the error message when iplsearch is called with an invalid key to
4799 suggest using net-iplsearch in a host list.
4801 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4802 ACLs are no longer actually imposed (and a message to that effect is
4805 22. If a "gecos" field in a passwd entry contained escaped characters, in
4806 particular, if it contained a \" sequence, Exim got it wrong when building
4807 a From: or a Sender: header from that name. A second bug also caused
4808 incorrect handling when an unquoted " was present following a character
4809 that needed quoting.
4811 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4812 was not being matched caselessly.
4814 24. Arranged for all hyphens in the exim.8 source to be escaped with
4817 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4818 properly. Recipient callout cache records were still being keyed to include
4819 the sender, even when use_sender was set false. This led to far more
4820 callouts that were necessary. The sender is no longer included in the key
4821 when use_sender is false.
4823 26. Added "control = submission" modifier to ACLs.
4825 27. Added the ${base62d: operator to decode base 62 numbers.
4827 28. dnsdb lookups can now access SRV records.
4829 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4830 the configuration file.
4832 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4833 (-v) mode. This makes the output for a verbose queue run more intelligible.
4835 31. Added a use_postmaster feature to recipient callouts.
4837 32. Added the $body_zerocount variable, containing the number of binary zero
4838 bytes in the message body.
4840 33. The time of last modification of the "new" subdirectory is now used as the
4841 "mailbox time last read" when there is a quota error for a maildir
4844 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4846 35. Added +ignore_unknown as a special item in host lists.
4848 36. Code for decoding IPv6 addresses in host lists is now included, even if
4849 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4850 address was recognized as an IP address, but was then not correctly decoded
4851 into binary, causing unexpected and incorrect effects when compared with
4858 1. Very minor rewording of debugging text in manualroute to say "list of
4859 hosts" instead of "hostlist".
4861 2. If verify=header_syntax was set, and a header line with an unqualified
4862 address (no domain) and a large number of spaces between the end of the
4863 name and the colon was received, the reception process suffered a buffer
4864 overflow, and (when I tested it) crashed. This was caused by some obsolete
4865 code that should have been removed. The fix is to remove it!
4867 3. When running in the test harness, delay a bit after writing a bounce
4868 message to get a bit more predictability in the log output.
4870 4. Added a call to search_tidyup() just before forking a reception process. In
4871 theory, someone could use a lookup in the expansion of smtp_accept_max_
4872 per_host which, without the tidyup, could leave open a database connection.
4874 5. Added the variables $recipient_data and $sender_data which get set from a
4875 lookup success in an ACL "recipients" or "senders" condition, or a router
4876 "senders" option, similar to $domain_data and $local_part_data.
4878 6. Moved the writing of debug_print from before to after the "senders" test
4881 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4882 problems for message scanning, either using a data ACL, or using
4883 local_scan() because the Received: header was not generated till after they
4884 were called (in order to set the time as the time of reception completion).
4885 I have revised the way this works. The header is now generated after the
4886 body is received, but before the ACL or local_scan() are called. After they
4887 are run, the timestamp in the header is updated.
4893 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4894 before starting a queue runner without re-exec. This happened only when
4895 deliver_drop_privilege was set or when the Exim user was set to root. The
4896 effect of the bug was that timeouts during subsequent deliveries caused
4897 crashes instead of being properly handled. The handler is now left at its
4898 default (and expected) setting.
4900 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4901 message, again when deliver_drop_privilege is set or Exim is run as root.
4902 The bug described in (1) was not present in this case, but the tidying up
4903 of the other signals was missing. I have made the two cases consistent.
4905 3. The ignore_target_hosts setting on a manualroute router was being ignored
4906 for hosts that were looked up using the /MX notation.
4908 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4911 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4912 operated on the sender address. After changing the $sender_address to <>
4913 for the sender address verify, Exim was re-instated it as the original
4914 (before rewriting) address, but remembering that it had rewritten it, so it
4915 wasn't rewriting it again. This bug also had the effect of breaking the
4916 sender address verification caching when the sender address was rewritten.
4918 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4919 This has been changed so that if the ip literal address matches
4920 ignore_target_hosts, the router declines.
4922 7. Added expansion conditions match_domain, match_address, and match_local_
4923 part (NOT match_host).
4925 8. The placeholder for the Received: header didn't have a length field set.
4927 9. Added code to Exim itself and to exim_lock to test for a specific race
4928 condition that could lead to file corruption when using MBX delivery. The
4929 issue is with the lockfile that is created in /tmp. If this file is removed
4930 after a process has opened it but before that process has acquired a lock,
4931 there is the potential for a second process to recreate the file and also
4932 acquire a lock. This could lead to two Exim processes writing to the file
4933 at the same time. The added code performs the same test as UW imapd; it
4934 checks after acquiring the lock that its file descriptor still refers to
4935 the same named file.
4937 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4938 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4941 11. Added the smtp_active_hostname option. If used, this will typically be made
4942 to depend on the incoming interface address. Because $interface_address is
4943 not set up until the daemon has forked a reception process, error responses
4944 that can happen earlier (such as "too many connections") no longer contain
4947 12. If an expansion in a condition on a "warn" statement fails because a lookup
4948 defers, the "warn" statement is abandoned, and the next ACL statement is
4949 processed. Previously this caused the whole ACL to be aborted.
4951 13. Added the iplsearch lookup type.
4953 14. Added ident_timeout as a log selector.
4955 15. Added tls_certificate_verified as a log selector.
4957 16. Added a global option tls_require_ciphers (compare the smtp transport
4958 option of the same name). This controls incoming TLS connections.
4960 17. I finally figured out how to make tls_require_ciphers do a similar thing
4961 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4962 before starting the TLS session.
4964 18. Tabs are now shown as \t in -bP output.
4966 19. If the log selector return_path_on_delivery was set, Exim crashed when
4967 bouncing a message because it had too many Received: header lines.
4969 20. If two routers both had headers_remove settings, and the first one included
4970 a superfluous trailing colon, the final name in the first list and the
4971 first name in the second list were incorrectly joined into one item (with a
4972 colon in the middle).
4978 1. Added -C and -D options to the exinext utility, mainly to make it easier
4979 to include in the automated testing, but these could be helpful when
4980 multiple configurations are in use.
4982 2. The exinext utility was not formatting the output nicely when there was
4983 an alternate port involved in the retry record key, nor when there was a
4984 message id as well (for retries that were specific to a specific message
4985 and a specific host). It was also confused by IPv6 addresses, because of
4986 the additional colons they contain. I have fixed the IPv4 problem, and
4987 patched it up to do a reasonable job for IPv6.
4989 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4990 delivery, the log line now contains "pipelined" if PIPELINING was used.
4992 4. An SMTP transport process used to panic and die if the bind() call to set
4993 an explicit outgoing interface failed. This has been changed; it is now
4994 treated in the same way as a connect() failure.
4996 5. A reference to $sender_host_name in the part of a conditional expansion
4997 that was being skipped was still causing a DNS lookup. This no longer
5000 6. The def: expansion condition was not recognizing references to header lines
5001 that used bh_ and bheader_.
5003 7. Added the _cache feature to named lists.
5005 8. The code for checking quota_filecount in the appendfile transport was
5006 allowing one more file than it should have been.
5008 9. For compatibility with Sendmail, the command line option
5016 and sets the incoming protocol and host name (for trusted callers). The
5017 host name and its colon can be omitted when only the protocol is to be set.
5018 Note the Exim already has two private options, -pd and -ps, that refer to
5019 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5020 "s", but I don't think that's a major issue.
5022 10. A number of refactoring changes to the code, none of which should affect
5025 (a) The number of logging options was getting close to filling up the
5026 32-bit word that was used as a bit map. I have split them into two classes:
5027 those that are passed in the argument to log_write(), and those that are
5028 only ever tested independently outside of that function. These are now in
5029 separate 32-bit words, so there is plenty of room for expansion again.
5030 There is no change in the user interface or the logging behaviour.
5032 (b) When building, for example, log lines, the code previously used a
5033 macro that called string_cat() twice, in order to add two strings. This is
5034 not really sufficiently general. Furthermore, there was one instance where
5035 it was actually wrong because one of the argument was used twice, and in
5036 one call a function was used. (As it happened, calling the function twice
5037 did not affect the overall behaviour.) The macro has been replaced by a
5038 function that can join an arbitrary number of extra strings onto a growing
5041 (c) The code for expansion conditions now uses a table and a binary chop
5042 instead of a serial search (which was left over from when there were very
5043 few conditions). Also, it now recognizes conditions like "pam" even when
5044 the relevant support is not compiled in: a suitably worded error message is
5045 given if an attempt is made to use such a condition.
5047 11. Added ${time_interval:xxxxx}.
5049 12. A bug was causing one of the ddress fields not to be passed back correctly
5050 from remote delivery subprocesses. The field in question was not being
5051 subsequently used, so this caused to problems in practice.
5053 13. Added new log selectors queue_time and deliver_time.
5055 14. Might have fixed a bug in maildirsizefile handling that threw up
5056 "unexpected character" debug warnings, and recalculated the data
5057 unnecessarily. In any case, I expanded the warning message to give more
5060 15. Added the message "Restricted characters in address" to the statements in
5061 the default ACL that block characters like @ and % in local parts.
5063 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5064 Three changes have been made:
5066 (a) There was a serious bug; a negative response to MAIL caused the whole
5067 recipient domain to be cached as invalid, thereby blocking all messages
5068 to all local parts at the same domain, from all senders. This bug has
5069 been fixed. The domain is no longer cached after a negative response to
5070 MAIL if the sender used is not empty.
5072 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5075 (c) A new callout option, "use_sender" has been added for people who want
5076 the modified behaviour.
5082 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5085 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5086 indeed breaks things for older releases.
5088 3. Added additional logging to the case where there is a problem reading data
5089 from a filter that is running in a subprocess using a pipe, in order to
5090 try to track down a specific problem.
5092 4. Testing facility fudge: when running in the test harness and attempting
5093 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5094 getting "No route to host". Convert this to a timeout.
5096 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5099 6. Some OS don't have socklen_t but use size_t instead. This affects the
5100 fifth argument of getsockopt() amongst other things. This is now
5101 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5102 can be set for individual OS. I have set it for SunOS5, OSF1, and
5103 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5104 some earlier ones do not.
5106 7. Change 4.30/15 was not doing the test caselessly.
5108 8. The standard form for an IPv6 address literal was being rejected by address
5109 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5110 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5111 this, as well as the form without the "IPv6" on the front (but only when
5112 address literals are enabled, of course).
5114 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5116 10. Exim crashed if a message with an empty sender address specified by -f
5117 encountered a router with an errors_to setting. This could be provoked only
5118 by a command such as
5122 where an empty string was supplied; "<>" did not hit this bug.
5124 11. Installed PCRE release 4.5.
5126 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5127 remained set. It is now erased.
5129 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5130 times from message ids (which are base 36 rather than the normal 62).
5132 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5133 were being counted as actual protocol errors, and logged if the log
5134 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5135 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5136 and DATA following a set of rejected RCPTs do not count as protocol errors.
5137 In other words, Exim assumes they were pipelined, though this may not
5138 actually be the case. Of course, in all cases the client gets an
5139 appropriate error code.
5141 15. If a lookup fails in an ACL condition, a message about the failure may
5142 be available; it is used if testing the ACL cannot continue, because most
5143 such messages specify what the cause of the deferral is. However, some
5144 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5145 that caused an old message to be retained and used if a later statement
5146 caused a defer, replacing the real cause of the deferral.
5148 16. If an IP address had so many PTR records that the DNS lookup buffer
5149 was not large enough to hold them, Exim could crash while trying to process
5150 the truncated data. It now detects and logs this case.
5152 17. Further to 4.21/58, another change has been made: if (and only if) the
5153 first line of a message (the first header line) ends with CRLF, a bare LF
5154 in a subsequent header line has a space inserted after it, so as not to
5155 terminate the header.
5157 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5158 unnecessarily, used atoi() instead of strtol(), and didn't check the
5159 termination when getting file sizes from file names by regex.
5161 19. Completely re-implemented the support for maildirsize files, in the light
5162 of a number of problems with the previous contributed implementation
5163 (4.30/29). In particular:
5165 . If the quota is zero, the maildirsize file is maintained, but no quota is
5168 . If the maildir directory does not exist, it is created before any attempt
5169 to write a maildirsize file.
5171 . The quota value in the file is just a cache; if the quota is changed in
5172 the transport, the new value overrides.
5174 . A regular expression is available for excluding directories from the
5177 20. The autoreply transport checks the characters in options that define the
5178 message's headers; it allows continued headers, but it was checking with
5179 isspace() after an embedded newline instead of explicitly looking for a
5182 21. If all the "regular" hosts to which an address was routed had passed their
5183 expiry times, and had not reached their retry times, the address was
5184 bounced, even if fallback hosts were defined. Now Exim should go on to try
5187 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5188 equivalent code in the SMTP transport. Some hosts send humungous responses
5189 to HELO/EHLO, more than 1024 it seems.
5191 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5192 gives compiler warnings in some environments. I've now done it "properly",
5195 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5196 (because of problems with the built-in one) was declared to return uschar *
5197 instead of char *, causing compiler failure.
5199 25. Fixed a file descriptor leak when processing alias/forward files.
5201 26. Fixed a minor format string issue in dbfn.c.
5203 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5205 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5206 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5209 29. When a sender address is verified, it is cached, to save repeating the test
5210 when there is more than one recipient in a message. However, when the
5211 verification involves a callout, it is possible for different callout
5212 options to be set for different recipients. It is too complicated to keep
5213 track of this in the cache, so now Exim always runs a verification when a
5214 callout is required, relying on the callout cache for the optimization.
5215 The overhead is duplication of the address routing, but this should not be
5218 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5219 to be verified with callout=postmaster, and the main callout worked but the
5220 postmaster check failed, the verification correctly failed. However, if a
5221 subsequent RCPT command asked for sender verification *without* the
5222 postmaster check, incorrect caching caused this verification also to fail,
5225 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5226 it was not caching the DNS options (qualify_single, search_parents) that
5227 were used when the lookup failed. A subsequent lookup with different
5228 options therefore always gave the same answer, though there were cases
5229 where it should not have. (Example: a "domains = !$mx_any" option on a
5230 dnslookup router: the "domains" option is always processed without any
5231 widening, but the router might have qualify_single set.) Now Exim uses the
5232 cached value only when the same options are set.
5234 32. Added John Jetmore's "exipick" utility to the distribution.
5236 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5237 than a timeout (e.g. a certificate is required, and is not provided), an
5238 Exim server now closes the connection immediately. Previously it waited for
5239 the client to close - but if the client is SSL, it seems that they each
5240 wait for each other, leading to a delay before one of them times out.
5242 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5243 maintained 0.8.x compatibility because I don't think many are using it, and
5244 it is clearly obsolete.
5246 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5249 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5250 client certificate was expired. A simple patch fixes this, though I don't
5251 understand the full logic of why the verify callback is called multiple
5254 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5255 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5256 which causes problems with some clients (such as the Certicom SSL Plus
5257 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5258 disables the coutermeasure allowing Eudora to connect."
5260 38. Exim was not checking that a write() to a log file succeeded. This could
5261 lead to Bad Things if a log got too big, in particular if it hit a file
5262 size limit. Exim now panics and dies if it cannot write to a log file, just
5263 as it does if it cannot open a log file.
5265 39. Modified OS/Makefile-Linux so that it now contains
5267 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5269 The two -D definitions ensure that Exim is compiled with large file
5270 support, which makes it possible to handle log files that are bigger than
5273 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5274 instance) a domain was checked against a named list that involved a lookup,
5275 causing $domain_data to be set, then another domain was checked against the
5276 same list, then the first domain was re-checked, the value of $domain_data
5277 after the final check could be wrong. In particular, if the second check
5278 failed, it could be set empty. This bug probably also applied to
5281 41. The strip_trailing_dot option was not being applied to the address given
5282 with the -f command-line option.
5284 42. The code for reading a message's header from the spool was incrementing
5285 $received_count, but never initializing it. This meant that the value was
5286 incorrect (doubled) while delivering a message in the same process in which
5287 it was received. In the most common configuration of Exim, this never
5288 happens - a fresh exec is done - but it can happen when
5289 deliver_drop_privilege is set.
5291 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5292 it now includes up to 150 characters of the unexpected data in the log
5295 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5296 and building data strings. The size of both of these buffers was 10 000
5297 bytes - far larger than anybody would *ever* want, thought I. Needless to
5298 say, somebody hit the limit. I have increased the maximum line length to
5299 20 000 and the maximum data length of concatenated lines to 100 000. I have
5300 also fixed two bugs, because there was no checking on these buffers. Tsk,
5301 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5302 buffer is too small.
5304 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5305 lsearch lookups. Now it does.
5307 46. When parsing a route_list item in a manualroute router, a fixed-length
5308 buffer was used for the list of hosts. I made this 1024 bytes long,
5309 thinking that nobody would ever have a list of hosts that long. Wrong.
5310 Somebody had a whole pile of complicated expansion conditions, and the
5311 string was silently truncated, leading to an expansion error. It turns out
5312 that it is easier to change to an unlimited length (owing to other changes
5313 that have happened since this code was originally written) than to build
5314 structure for giving a limitation error. The length of the item that
5315 expands into the list of hosts is now unlimited.
5317 47. The lsearch lookup could not handle data where the length of text line was
5318 more than 4095 characters. Such lines were truncated, leading to shortened
5319 data being returned. It should now handle lines of any length.
5321 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5322 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5325 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5326 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5327 no longer happens in comment lines. A list of replacements is now placed
5328 at the head of all of the source files, except those whose only change is
5329 to replace PERL_COMMAND in the very first #! line.
5331 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5332 messages on the queue, with a bottom-up merge sort, using code contributed
5333 by Michael Haardt. This should make operations like -bp somewhat faster on
5334 large queues. It won't affect queue runners, except when queue_run_in_order
5337 51. Installed eximstats 1.31 in the distribution.
5339 52. Added support for SRV lookups to the dnslookup router.
5341 53. If an ACL referred to $message_body or $message_body_end, the value was not
5342 reset for any messages that followed in the same SMTP session.
5344 54. The store-handling optimization for building very long strings was not
5345 differentiating between the different store pools. I don't think this
5346 actually made any difference in practice, but I've tidied it.
5348 55. While running the routers to verify a sender address, $sender_address
5349 was still set to the sender address. This is wrong, because when routing to
5350 send a bounce to the sender, it would be empty. Therefore, I have changed
5351 it so that, while verifying a sender address, $sender_address is set to <>.
5352 (There is no change to what happens when verifying a recipient address.)
5354 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5355 target A or AAAA records (if not already returned) without resetting the
5356 qualify_single or search_parents options of the DNS resolver. These are
5357 inappropriate in this case because the targets of MX and SRV records must
5358 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5359 target that, when qualified, matched something in the local domain. These
5360 two options are now turned off when doing these lookups.
5362 57. It seems that at least some releases of Reiserfs (which does not have the
5363 concept of a fixed number of inodes) returns zero and not -1 for the
5364 number of available inodes. This interacted badly with check_spool_inodes,
5365 which assumed that -1 was the "no such thing" setting. What I have done is
5366 to check that the total number of inodes is greater than zero before doing
5367 the test of how many are available.
5369 58. When a "warn" ACL statement has a log_message modifier, the message is
5370 remembered, and not repeated. This is to avoid a lot of repetition when a
5371 message has many recipients that cause the same warning to be written.
5372 Howewer, Exim was preserving the list of already written lines for an
5373 entire SMTP session, which doesn't seem right. The memory is now reset if a
5374 new message is started.
5376 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5377 debugging output unless log_rewrite was also set.
5379 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5380 of (int)(handle) when we know that handle contains (void *)(-1).
5382 61. The Exim daemon panic-logs an error return when it closes the incoming
5383 connection. However "connection reset by peer" seems to be common, and
5384 isn't really an error worthy of noting specially, so that particular error
5387 62. When Exim is trying to find all the local interfaces, it used to panic and
5388 die if the ioctl to get the interface flags failed. However, it seems that
5389 on at least one OS (Solaris 9) it is possible to have an interface that is
5390 included in the list of interfaces, but for which you get a failure error
5391 for this call. This happens when the interface is not "plumbed" into a
5392 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5393 failure of the "get flags" call assumes that the interface is down.
5395 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5396 makes life easier for people who are doing arithmetic on fields extracted
5397 from dates, where you often get leading zeros that should not be
5398 interpreted as octal.
5400 64. Added qualify_domain to the redirect router, to override the global
5403 65. If a pathologically long header line contained very many addresses (the
5404 report of this problem mentioned 10 000) and each of them was rewritten,
5405 Exim could use up a very large amount of memory. (It kept on making new
5406 copies of the header line as it rewrote, and never released the old ones.)
5407 At the expense of a bit more processing, the header rewriting function has
5408 been changed so that it no longer eats memory in this way.
5410 66. The generation of the Received: header has been moved from the time that a
5411 message starts to be received, to the time that it finishes. The timestamp
5412 in the Received: header should now be very close to that of the <= log
5413 line. There are two side-effects of this change:
5415 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5416 logged header lines no longer include the local Received: line, because
5417 it has not yet been created. The same applies to a copy of the message
5418 that is returned to a non-SMTP sender when a message is rejected.
5420 (b) When a filter file is tested using -bf, no additional Received: header
5421 is added to the test message. After some thought, I decided that this
5424 This change does not affect the value of $received_for. It is still set
5425 after address rewriting, but before local_scan() is called.
5427 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5429 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5430 gave an unhelpful panic error message, and a defer error. I have managed to
5431 change this behaviour so that it now rejects any supplied certificate,
5432 which seems right, as the list of acceptable certificates is empty.
5434 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5435 gave an unhelpful defer error. I have not managed to make this reject any
5436 supplied certificates, but the error message it gives is "no certificate
5437 supplied", which is not helpful.
5439 70. exigrep's output now also includes lines that are not associated with any
5440 message, but which match the given pattern. Implemented by a patch from
5441 Martin Sluka, which also tidied up the Perl a bit.
5443 71. Recipient callout verification, like sender verification, was using <> in
5444 the MAIL FROM command. This isn't really the right thing, since the actual
5445 sender may affect whether the remote host accepts the recipient or not. I
5446 have changed it to use the actual sender in the callout; this means that
5447 the cache record is now keyed on a recipient/sender pair, not just the
5448 recipient address. There doesn't seem to be a real danger of callout loops,
5449 since a callout by the remote host to check the sender would use <>.
5450 [SEE ABOVE: changed after hitting problems.]
5452 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5453 temporary errors. However, in the case of such a code being given after
5454 the end of a data transmission (i.e. after ".") Exim was failing to write
5455 a retry record for the message. (Yes, there was some broken host that was
5456 actually sending 8xx at this point.)
5458 73. An unknown lookup type in a host list could cause Exim to panic-die when
5459 the list was checked. (An example that provoked this was putting <; in the
5460 middle of a list instead of at the start.) If this happened during a DATA
5461 ACL check, a -D file could be left lying around. This kind of configuration
5462 error no longer causes Exim to die; instead it causes a defer errror. The
5463 incident is still logged to the main and panic logs.
5465 74. Buglet left over from Exim 3 conversion. The message "too many messages
5466 in one connection" was written to the rejectlog but not the mainlog, except
5467 when address rewriting (yes!) was being logged.
5469 75. Added write_rejectlog option.
5471 76. When a system filter was run not as root (that is, when system_filter_user
5472 was set), the values of the $n variables were not being returned to the
5473 main process; thus, they were not subsequently available in the $sn
5476 77. Added +return_path_on_delivery log selector.
5478 78. A connection timeout was being treated differently from recipients deferred
5479 when testing hosts_max_try with a message that was older than the host's
5480 retry timeout. (The host should not be counted, thus allowing all hosts to
5481 be tried at least once before bouncing.) This may have been the cause of an
5482 occasionally reported bug whereby a message would remain on the queue
5483 longer than the retry timeout, but would be bounced if a delivery was
5484 forced. I say "may" because I never totally pinned down the problem;
5485 setting up timeout/retry tests is difficult. See also the next item.
5487 79. The ultimate address timeout was not being applied to errors that involved
5488 a combination of host plus message (for example, a timeout on a MAIL
5489 command). When an address resolved to a number of possible hosts, and they
5490 were not all tried for each delivery (e.g. because of hosts_max_try), a
5491 message could remain on the queue longer than the retry timeout.
5493 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5496 81. Fixed an obscure SMTP outgoing bug which required at least the following
5497 conditions: (a) there was another message waiting for the same server;
5498 (b) the server returned 5xx to all RCPT commands in the first message so
5499 that the message was not completed; (c) the server dropped the connection
5500 or gave a negative response to the RSET that Exim sends to abort the
5501 transaction. The observed case was a dropped connection after DATA that had
5502 been sent in pipelining mode. That is, the server had advertised PIPELINING
5503 but was not implementing it correctly. The effect of the bug was incorrect
5504 behaviour, such as trying another host, and this could lead to a crash.
5510 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5511 and daemon.c were passed as pointers to ints; they should have been
5512 pointers to socklen_t variables (which are typically unsigned ints).
5514 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5517 3. Fixed a really odd bug that affected only the testing scheme; patching a
5518 certain fixed string in the binary changed the value of another string that
5519 happened to be identical to the end of the original first string.
5521 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5522 name", it returns that address as the IP address. On some operating
5523 systems (e.g. Solaris), it also passes back the IP address string as the
5524 "host name". However, on others (e.g. Linux), it passes back an empty
5525 string. Exim wasn't checking for this, and was changing the host name to an
5526 empty string, assuming it had been canonicized.
5528 5. Although rare, it is permitted to have more than one PTR record for a given
5529 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5530 all the names associated with an address, because they do in Solaris.
5531 However, it seems that they do not in Linux for data that comes from the
5532 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5533 I found this out when I moved to a new Linux workstation and tried to run
5534 the Exim test suite.
5536 To get round this problem I have changed the code so that it now does its
5537 own call to the DNS to look up PTR records when searching for a host name.
5538 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5539 addresses that are only in /etc/hosts are still found.
5541 This behaviour is, however, controlled by an option called host_lookup_
5542 order, which defaults to "bydns:byaddr". If people want to use the other
5543 order, or indeed, just use one or the other means of lookup, they can
5544 specify it in this variable.
5546 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5547 some operating systems, this comes back from gethostbyaddr() as an empty
5548 string, and this is what Exim used to test for. However, it seems that in
5549 other systems, "." is yielded. Exim now tests for this case too.
5551 7. The values of check_spool_space and check_log_space are now held internally
5552 as a number of kilobytes instead of an absolute number of bytes. If a
5553 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5554 kilobyte. This means that much larger values can be stored.
5556 8. Exim monitor: an attempt to get the action menu when not actually pointing
5557 at a message produces an empty menu entitled "No message selected". This
5558 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5559 no entries in it ("Shell widget menu has zero width and/or height"). So I
5560 have added a single, blank menu entry in this case.
5562 9. Added ${quote_local_part.
5564 10. MIME decoding is now applied to the contents of Subject: header lines when
5567 11. Now that a reference to $sender_host_address automatically causes a reverse
5568 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5569 host lookup before query-style lookups in lists that might use this
5570 variable. This has therefore been abolished, and the "net-" prefix is no
5571 longer necessary for query-style lookups.
5573 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5574 have been a typo for LFLAGS, so it has been changed.
5576 13. The install script calls Exim with "-C /dev/null" in order to find the
5577 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5578 to be output. Howeve, since Exim outputs its version number before the
5579 error, it didn't break the script. It just looked ugly. I fixed this by
5580 always allowing "-C /dev/null" if the caller is root.
5582 14. Ignore overlarge ACL variable number when reading spool file - insurance
5583 against a later release with more variables having written the file.
5585 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5586 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5587 this, as well as the form without the "IPv6" on the front.
5589 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5590 OS/Makefile-Darwin file.
5592 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5593 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5595 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5596 variable. However, some OS use ints for inodes. I've added cast to long int
5597 to get rid of the compiler warning.
5599 19. I had forgotten to lock out "/../" in configuration file names when
5600 ALT_CONFIG_PREFIX was set.
5602 20. Routers used for verification do not need to specify transports. However,
5603 if such a router generated a host list, and callout was configured, Exim
5604 crashed, because it could not find a port number from the (non-existent)
5605 transport. It now assumes port 25 in this circumstance.
5607 21. Added the -t option to exigrep.
5609 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5610 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5611 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5612 them is set, LOOKUP_LSEARCH is forced.
5614 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5616 24. Added sender and host information to the "rejected by local_scan()" log
5617 line; previously there was no indication of these.
5619 25. Added .include_if_exists.
5621 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5622 Linux. It turns out that not all file systems support this. Apparently some
5623 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5624 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5625 not supported on the file descriptor, is now ignored when Exim is trying to
5626 sync a directory. This applies only to Linux.
5628 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5630 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5631 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5634 29. Added support for maildirsize files from supplied patch (modified a bit).
5636 30. The use of :fail: followed by an empty string could lead Exim to respond to
5637 sender verification failures with (e.g.):
5639 550 Verification failed for <xxx>
5640 550 Sender verify failed
5642 where the first response line was missing the '-' that indicates it is not
5643 the final line of the response.
5645 31. The loop for finding the name of the user that called Exim had a hardwired
5646 limit of 10; it now uses the value of finduser_retries, which is used for
5647 all other user lookups.
5649 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5652 33. Exim was neglecting to zero errno before one call of strtol() when
5653 expanding a string and expecting an integer value. On some systems this
5654 resulted in spurious "integer overflow" errors. Also, it was casting the
5655 result into an int without checking.
5657 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5658 did not work. The code looks as if it has *never* worked, though it appears
5659 to have been documented since at least releast 1.62. I have made it work.
5661 35. The "timeout_DNS" error in retry rules, also documented since at least
5662 1.62, also never worked. As it isn't clear exactly what this means, and
5663 clearly it isn't a major issue, I have abolished the feature by treating it
5664 as "timeout", and writing a warning to the main and panic logs.
5666 36. The display of retry rules for -brt wasn't always showing the error code
5669 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5670 timeout_connect_A, timeout_connect_MX.
5672 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5673 to the empty sender.
5675 39. The daemon was not analysing the content of -oX till after it had closed
5676 stderr and disconnected from the controlling terminal. This meant that any
5677 syntax errors were only noted on the panic log, and the return code from
5678 the command was 0. By re-arranging the code a little, I've made the
5679 decoding happen first, so such errors now appear on stderr, and the return
5680 code is 1. However, the actual setting up of the sockets still happens in
5681 the disconnected process, so errors there are still only recorded on the
5684 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5685 connections (as happens on some IP stacks) was logged at start up time as
5686 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5687 it from "IPv6 and IPv4", which means that two separate sockets are being
5690 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5691 says whether AF_INET or AF_INET6 was passed as an argument.
5693 42. Exiwhat output was messed up when time zones were included in log
5696 43. Exiwhat now gives more information about the daemon's listening ports,
5697 and whether -tls-on-connect was used.
5699 44. The "port" option of the smtp transport is now expanded.
5701 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5702 silently ignored. Now an error message is written to the main and panic
5705 46. There's a new ACL modifier called "logwrite" which writes to a log file
5706 as soon as it is encountered.
5708 47. Added $local_user_uid and $local_user_gid at routing time.
5710 48. Exim crashed when trying to verify a sender address that was being
5713 49. Exim was recognizing only a space character after ".include". It now also
5714 recognizes a tab character.
5716 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5717 extracting the relevant information from the specification. The man page no
5718 longer contains scrambled data for the -d option, and I've added a section
5719 at the front about calling Exim under different names.
5721 51. Added "extra_headers" argument to the "mail" command in filter files.
5723 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5726 53. Installed eximstats 1.29.
5728 54. Added transport_filter_timeout as a generic transport option.
5730 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5731 Cc: header lines. This was required by RFC 822, but it not required by RFC
5734 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5735 incoming messages that did not have them. Now it does so only if the
5736 message originates locally, that is, if there is no associated remote host
5737 address. When Resent- header lines are present, this applies to the Resent-
5738 lines rather than the non-Resent- lines.
5740 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5741 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5743 58. Messages for configuration errors now include the name of the main
5744 configuration file - useful now that there may be more than one file in a
5745 list (.included file names were always shown).
5747 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5748 for those rare installations that do not start the daemon as root or run it
5749 setuid root. I've cut out the call to initgroups() if the daemon is not
5752 60. The Exim user and group can now be bound into the binary as text strings
5753 that are looked up at the start of Exim's processing.
5755 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5757 62. Added $mailstore_basename variable.
5759 63. Installed patch to sieve.c from Michael Haardt.
5761 64. When Exim failed to open the panic log after failing to open the main log,
5762 the original message it was trying to log was written to stderr and debug
5763 output, but if they were not available (the usual case in production), it
5764 was lost. Now it is written to syslog before the two lines that record the
5765 failures to open the logs.
5767 65. Users' Exim filters run in subprocesses under the user's uid. It is
5768 possible for a "deliver" command or an alias in a "personal" command to
5769 provoke an address rewrite. If logging of address rewriting is configured,
5770 this fails because the process is not running as root or exim. There may be
5771 a better way of dealing with this, but for the moment (because 4.30 needs
5772 to be released), I have disabled address rewrite logging when running a
5773 filter in a non-root, non-exim process.
5779 1. The buildconfig auxiliary program wasn't quoting the value set for
5780 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5781 not defined. This bug was masked in 4.22 by the effect that was fixed in
5784 2. Some messages that were rejected after a message id was allocated were
5785 shown as "incomplete" by exigrep. It no longer does this for messages that
5786 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5788 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5789 have allow_domain_literals set, the ID did not get logged in the <= line.
5790 Domain literals are now always recognized in Message-ID: header lines.
5792 4. The first argument for a ${extract expansion item is the key name or field
5793 number. Leading and trailing spaces in this item were not being ignored,
5794 causing some misleading effects.
5796 5. When deliver_drop_privilege was set, single queue runner processes started
5797 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5798 same command in the process it spins off) were not dropping privilege.
5800 6. When the daemon running as "exim" started a queue runner, it always
5801 re-executed Exim in the spun-off process. This is a waste of effort when
5802 deliver_drop_privilege is set. The new process now just calls the
5803 queue-runner function directly.
5809 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5812 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5813 ignored. Though the use of -oP was forcing the writing of a pid file, it
5814 was always written to the default place.
5816 3. If the message "no IP address found for host xxxx" is generated during
5817 incoming verification, it is now followed by identification of the incoming
5818 connection (so you can more easily find what provoked it).
5820 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5822 5. Added some features to "harden" Exim a bit more against certain attacks:
5824 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5825 be put in Local/Makefile. This is like the never_users runtime option,
5826 but it cannot be overridden. The default setting is "root".
5828 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5829 prefix string with which any file named in a -C command line option
5832 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5833 is retained for -C and -D only if the caller of Exim is root. Without
5834 it, the exim user may also use -C and -D and retain privilege.
5836 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5837 command line option is disabled.
5839 6. Macro names set by the -D option must start with an upper case letter, just
5840 like macro names defined in the configuration file.
5842 7. Added "dereference=" facility to LDAP.
5844 8. Two instances of the typo "uknown" in the source files are fixed.
5846 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5847 the Configure-Makefile script screwed up while processing it.
5849 10. Incorporated PCRE 4.4.
5851 11. The SMTP synchronization check was not operating right at the start of an
5852 SMTP session. For example, it could not catch a HELO sent before the client
5853 waited for the greeting. There is now a check for outstanding input at the
5854 point when the greeting is written. Because of the duplex, asynchronous
5855 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5856 way, but not yet received, when the check is performed.
5858 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5859 on TCP/IP sockets, because this apparently causes some broken clients to
5862 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5863 unchanged) from the Cygwin maintainer.
5865 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5866 was supported instead of testing for mbx. Effectively a typo.
5868 15. The spa authenticator server code was not checking that the input it
5869 received was valid base64.
5871 16. The debug output line for the "set" modifier in ACLs was not showing the
5872 name of the variable that was being set.
5874 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5876 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5877 lookup on its own. Something else was needed to trigger the lookup. For
5878 example, a match in host_lookup or the need for a host name in a host list.
5879 Now, if $sender_host_name is referenced and the host name has not yet been
5880 looked up, a lookup is performed. If the lookup fails, the variable remains
5881 empty, and $host_lookup_failed is set to "1".
5883 19. Added "eqi" as a case-independent comparison operator.
5885 20. The saslauthd authentication condition could segfault if neither service
5886 nor realm was specified.
5888 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5889 error message that was logged was misleading, and incoming SMTP
5890 connections were dropped. The message is now more accurate, and temporary
5891 errors are given to SMTP connections.
5893 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5894 (except RSET) with a 421 error, until QUIT is received. However, it was
5895 failing to send a response to QUIT.
5897 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5898 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5899 ACL is now run after the helo_try_verify_hosts code.
5901 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5902 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5903 case-independent, so other case variants are also recognized). Apparently
5904 some systems use these upper case variants.
5906 25. If more than two messages were waiting for the same host, and a transport
5907 filter was specified for the transport, Exim sent two messages over the
5908 same TCP/IP connection, and then failed with "socket operation on non-
5909 socket" when it tried to send the third.
5911 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5913 27. The extern definition of crypt16() in expand.c was not being excluded when
5914 the OS had its own crypt16() function.
5916 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5917 as a preferred synonym for return_size_limit, both as an option and as an
5920 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5922 30. Changed the default configuration ACL to relax the local part checking rule
5923 for addresses that are not in any local domains. For these addresses,
5924 slashes and pipe symbols are allowed within local parts, but the sequence
5925 /../ is explicitly forbidden.
5927 31. SPA server authentication was not clearing the challenge buffer before
5930 32. log_message in a "warn" ACL statement was writing to the reject log as
5931 well as to the main log, which contradicts the documentation and doesn't
5932 seem right (because no rejection is happening). So I have stopped it.
5934 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5935 However, I am unable to do any testing of this.
5937 34. Fixed an infelicity in the appendfile transport. When checking directories
5938 for a mailbox, to see if any needed to be created, it was accidentally
5939 using path names with one or more superfluous leading slashes; tracing
5940 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5942 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5943 contents are added to the log line that is written for every discarded
5944 recipient. (Previously a log_message setting was ignored.)
5946 36. The ${quote: operator now quotes the string if it is empty.
5948 37. The install script runs exim in order to find its version number. If for
5949 some reason other than non-existence or emptiness, which it checks, it
5950 could not run './exim', it was installing it with an empty version number,
5951 i.e. as "exim-". This error state is now caught, and the installation is
5954 38. An argument was missing from the function that creates an error message
5955 when Exim fails to connect to the socket for saslauthd authentication.
5956 This could cause Exim to crash, or give a corrupted message.
5958 39. Added isip, isip4, and isip6 to ${if conditions.
5960 40. The ACL variables $acl_xx are now saved with the message, and can be
5961 accessed later in routers, transports, and filters.
5963 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5964 strings in the file are not string-expanded.
5966 42. If a MAIL command specified a SIZE value that was too large to fit into an
5967 int variable, the check against message_size_limit failed. Such values are
5968 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5969 day this will have to be increased, but I don't think I want to be around
5970 when emails are that large.
5977 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5978 iconv() is not standard in FreeBSD.
5980 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5981 IPv6 enabled. The observed symptom was a segmentation fault on return from
5982 the function os_common_find_running_interfaces() in src/os.c.
5984 3. In the check_special_case() function in daemon.c I had used "errno" as an
5985 argument name, which causes warnings on some systems. This was basically a
5986 typo, since it was named "eno" in the comments!
5988 4. The code that waits for the clock to tick (at a resolution of some fraction
5989 of a second) so as to ensure message-id uniqueness was always waiting for
5990 at least one whole tick, when it could have waited for less. [This is
5991 almost certainly not relevant at current processor speeds, where it is
5992 unlikely to ever wait at all. But we try to future-proof.]
5994 5. The function that sleeps for a time interval that includes fractions of a
5995 second contained a race. It did not block SIGALRM between setting the
5996 timer, and suspending (a couple of lines later). If the interval was short
5997 and the sigsuspend() was delayed until after it had expired, the suspension
5998 never ended. On busy systems this could lead to processes getting stuck for
6001 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6002 process, before it forks any delivery processes. The open lookup caching
6003 mechanism meant that the open file or database connection was passed into
6004 the delivery process. The problem was that delivery processes always tidy
6005 up cached lookup data. This could cause a problem for the next delivery
6006 process started by the queue runner, because the external queue runner
6007 process does not know about the closure. So the next delivery process
6008 still has data in the lookup cache. In the case of a file lookup, there was
6009 no problem because closing a file descriptor in a subprocess doesn't affect
6010 the parent. However, if the lookup was caching a connection to a database,
6011 the connection was closed, and the second delivery process was likely to
6012 see errors such as "PGSQL: query failed: server closed the connection
6013 unexpectedly". The problem has been fixed by closing all cached lookups
6014 in a queue runner before running a delivery process.
6016 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6017 seem to have the "const" qualifier which it has on other OS. I've
6020 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6021 *of the same type* (client or server) with the same public name that an
6022 error should be diagnosed.
6024 9. When Exim looked up a host name for an IP address, but failed to find the
6025 original IP address when looking up the host name (a safety check), it
6026 output the message "<ip address> does not match any IP for NULL", which was
6027 confusing, to say the least. The bug was that the host name should have
6028 appeared instead of "NULL".
6030 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6031 user that is built into the binary, and the -C or -D options is used, root
6032 privilege is dropped before the configuration file is read. In addition,
6033 logging is switched to stderr instead of the normal log files. If the
6034 configuration then re-defines the Exim user, the unprivileged environment
6035 is probably not what is expected, so Exim logs a panic warning message (but
6038 However, if deliver_drop_privilege is set, the unprivileged state may well
6039 be exactly what is intended, so the warning has been cut out in that case,
6040 and Exim is allowed to try to write to its normal log files.
6046 1. smtp_return_error_details was not giving details for temporary sender
6047 or receiver verification errors.
6049 2. Diagnose a configuration error if two authenticators have the same public
6052 3. Exim used not to create the message log file for a message until the first
6053 delivery attempt. This could be confusing when incoming messages were held
6054 for policy or load reasons. The message log file is now created at the time
6055 the message is received, and an initial "Received" line is written to it.
6057 4. The automatically generated man page for command line options had a minor
6058 bug that caused no ill effects; however, a more serious problem was that
6059 the procedure for building the man page automatically didn't always
6060 operate. Consequently, release 4.20 contains an out-of-date version. This
6061 shouldn't happen again.
6063 5. When building Exim with embedded Perl support, the script that builds the
6064 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6065 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6067 6. The freeze_tell option was not being used for messages that were frozen on
6068 arrival, either by an ACL or by local_scan().
6070 7. Added the smtp_incomplete_transaction log selector.
6072 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6073 was accepting AUTH without a new EHLO.
6075 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6076 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6079 10. Logging of TCP/IP connections (when configured) now happens in the main
6080 daemon process instead of the child process, so that the TCP/IP connection
6081 count is more accurate (but it can never be perfect).
6083 11. The use of "drop" in a nested ACL was not being handled correctly in the
6084 outer ACL. Now, if condition failure induced by the nested "drop" causes
6085 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6086 or "require"), the connection is dropped.
6088 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6089 that yield "discard" can now be used with an "accept" or a "discard" verb,
6090 but an error is generated for any others (because I can't see a useful way
6091 to define what should happen).
6093 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6094 are now processed in the same way as lines in the Exim configuration file.
6095 In particular, continuation lines are supported.
6097 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6099 15. Added -ti meaning -t -i.
6101 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6102 domains, and warn by logging if others are found.
6104 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6105 returned from GIFCONF when Exim is trying to find the list of interfaces on
6106 a host. The code in os.c has been modified to copy each ifreq to an aligned
6107 structure in all cases.
6109 Also, in some cases, the returned ifreq's were being copied to a 'struct
6110 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6111 means the last couple of bytes of an IPv6 address could be chopped if the
6112 ifreq contained only a normal sockaddr (14 bytes storage).
6114 18. Named domain lists were not supported in the hosts_treat_as_local option.
6115 An entry such as +xxxx was not recognized, and was treated as a literal
6118 19. Ensure that header lines added by a DATA ACL are included in the reject log
6119 if the ACL subsequently rejects the message.
6121 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6122 MD5 (which is deprecated).
6124 21. When testing a filter file using -bf, Exim was writing a message when it
6125 took the sender from a "From " line in the message, but it was not doing so
6126 when it took $return_path from a Return-Path: header line. It now does.
6128 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6129 with a valid header line field name (a series of printing characters
6130 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6132 23. Changed "disc" in the source to "disk" to conform to the documentation and
6133 the book and for uniformity.
6135 24. Ignore Sendmail's -Ooption=value command line item.
6137 25. When execve() failed while trying to run a command in a pipe transport,
6138 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6139 could be confused with a return value of 69 from the command itself. This
6140 has been changed to 127, the value the shell returns if it is asked to run
6141 a non-existent command. The wording for the related log line suggests a
6142 non-existent command as the problem.
6144 26. If received_header_text expands to an empty string, do not add a Received:
6145 header line to the message. (Well, it adds a token one on the spool, but
6146 marks it "old" so that it doesn't get used or transmitted.)
6148 27. Installed eximstats 1.28 (addition of -nt option).
6150 28. There was no check for failure on the call to getsockname() in the daemon
6151 code. This can fail if there is a shortage of resources on the system, with
6152 ENOMEM, for example. A temporary error is now given on failure.
6154 29. Contrary to the C standard, it seems that in some environments, the
6155 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6156 program. Exim now does this explicitly; it affects the formatting of
6157 timestamps using strftime().
6159 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6160 complaints. I've now initialized all the variables, to avoid this.
6162 32. Header lines added by a system filter were not being "seen" during
6163 transport-time rewrites.
6165 33. The info_callback() function passed to OpenSSL is set up with type void
6166 (*)(SSL *, int, int), as described somewhere. However, when calling the
6167 function (actually a macro) that sets it up, the type void(*)() is
6168 expected. I've put in a cast to prevent warnings from picky compilers.
6170 34. If a DNS black list lookup found a CNAME record, but there were no A
6171 records associated with the domain it pointed at, Exim crashed.
6173 35. If a DNS black list lookup returned more than one A record, Exim ignored
6174 all but the first. It now scans all returned addresses if a particular IP
6175 value is being sought. In this situation, the contents of the
6176 $dnslist_value variable are a list of all the addresses, separated by a
6179 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6180 to accept a host name and all its aliases if the forward lookup for any of
6181 them yielded the IP address of the incoming connection. Now it accepts only
6182 those names whose forward lookup yields the correct IP address. Any other
6183 names are discarded. This closes a loophole whereby a rogue DNS
6184 administrator could create reverse DNS records to break through a
6185 wildcarded host restriction in an ACL.
6187 37. If a user filter or a system filter that ran in a subprocess used any of
6188 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6189 the wrong values were passed to the pipe command ($thisaddress had the
6190 value of $0, $0 had the value of $1, etc). This bug was introduced by
6191 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6193 38. Improved the line breaking for long SMTP error messages from ACLs.
6194 Previously, if there was no break point between 40 and 75 characters, Exim
6195 left the rest of the message alone. Two changes have been made: (a) I've
6196 reduced the minimum length to 35 characters; (b) if it can't find a break
6197 point between 35 and 75 characters, it looks ahead and uses the first one
6198 that it finds. This may give the occasional overlong line, but at least the
6199 remaining text gets split now.
6201 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6202 file descriptors might be low, and that setting 1000 would always raise it.
6203 It turns out that in some environments, the limit is already over 1000 and
6204 that lowering it causes trouble. So now Exim takes care not to decrease it.
6206 40. When delivering a message, the value of $return_path is set to $sender_
6207 address at the start of routing (routers may change the value). By an
6208 oversight, this default was not being set up when an address was tested by
6209 -bt or -bv, which affected the outcome if any router or filter referred to
6212 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6213 log only when "message" or "log_message" are set. However, if one of the
6214 conditions was an address verification, or a call to a nested ACL, the
6215 messages generated by the underlying test were being passed through. This
6216 no longer happens. The underlying message is available in $acl_verify_
6217 message for both "message" and "log_message" expansions, so it can be
6218 passed through if needed.
6220 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6221 new expansion $bh_ to give the encoded byte string without charset
6222 translation. Translation happens only if iconv() is available; HAVE_ICONV
6223 indicates this at build time. HEADERS_CHARSET gives the charset to
6224 translate to; headers_charset can change it in the configuration, and
6225 "headers charset" can change it in an individual filter file.
6227 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6228 that creates RFC 2047 encoded "words" labels them as that charset instead
6229 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6230 expansion operator; (ii) when Exim creates a From: line for a local
6231 message; (iii) when a header line is rewritten to include a "phrase" part.
6233 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6234 buggy, causing it to skip the first lines of messages whose message ID
6235 ended in 'D'. This would not have bitten before Exim release 4.14, because
6236 message IDs were unlikely to end in 'D' before then. The effect was to have
6237 incorrect size information for certain domains.
6239 45. #include "config.h" was missing at the start of the crypt16.c module. This
6240 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6243 46. If there was a timeout during a "random" callout check, Exim treated it as
6244 a failure of the random address, and carried on sending RSET and the real
6245 address. If the delay was just some slowness somewhere, the response to the
6246 original RCPT would be taken as a response to RSET and so on, causing
6247 mayhem of various kinds.
6249 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6250 when I implemented it. It didn't allow for the fact that some option values
6251 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6252 the right test for positive values.
6254 48. Domain names in DNS records are case-independent. Exim always looks them up
6255 in lower case. Some resolvers return domain names in exactly the case they
6256 appear in the zone file, that is, they may contain uppercase letters. Not
6257 all resolvers do this - some return always lower case. Exim was treating a
6258 change of case by a resolver as a change of domain, similar to a widening
6259 of a domain abbreviation. This triggered its re-routing code and so it was
6260 trying to route what was effectively the same domain again. This normally
6261 caused routing to fail (because the router wouldn't handle the domain
6262 twice). Now Exim checks for this case specially, and just changes the
6263 casing of the domain that it ultimately uses when it transmits the message
6266 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6269 50. If a filter generated a file delivery with a non-absolute name (possible if
6270 no home directory exists for the router), the forbid_file option was not
6273 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6274 the existing equality matching.
6276 52. Exim was using ints instead of ino_t variables in some places where it was
6277 dealing with inode numbers.
6279 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6280 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6281 called TMPDIR, and if it finds it is different, it changes its value.
6283 54. The smtp_printf() function is now made available to local_scan() so
6284 additional output lines can be written before returning. There is also an
6285 smtp_fflush() function to enable the detection of a dropped connection.
6286 The variables smtp_input and smtp_batched_input are exported to
6289 55. Changed the default runtime configuration: the message "Unknown user"
6290 has been removed from the ACL, and instead placed on the localuser router,
6291 using the cannot_route_message feature. This means that any verification
6292 failures that generate their own messages won't get overridden. Similarly,
6293 the "Unrouteable address" message that was in the ACL for unverifiable
6294 relay addresses has also been removed.
6296 56. Added hosts_avoid_esmtp to the smtp transport.
6298 57. The exicyclog script was not checking for the esoteric option
6299 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6300 will work only if exicyclog is run under the appropriate euid.
6302 58. Following a discussion on the list, the rules by which Exim recognises line
6303 endings on incoming messages have been changed. The -dropcr and drop_cr
6304 options are now no-ops, retained only for backwards compatibility. The
6305 following line terminators are recognized: LF CRLF CR. However, special
6306 processing applies to CR:
6308 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6309 nor a local message in the state where . is a terminator.
6311 (ii) If a bare CR is encountered in a header line, an extra space is added
6312 after the line terminator so as not to end the header. The reasoning
6313 behind this is that bare CRs in header lines are most likely either
6314 to be mistakes, or people trying to play silly games.
6316 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6317 was being incorrectly given as 18 bytes larger than it should have been.
6318 This is a VOB (very old bug).
6320 60. This may never have affected anything current, but just in case it has:
6321 When the local host is found other than at the start of a list of hosts,
6322 the local host, those with the same MX, and any that follow, are discarded.
6323 When the list in question was part of a longer list of hosts, the following
6324 hosts (not currently being processed) were also being discarded. This no
6325 longer happens. I'm not sure if this situation could ever has previously
6328 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6331 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6332 header. This is something that is recommended in a new Internet Draft, and
6333 is something that is documented as being done by Sendmail. There are two
6334 possible values. For messages generated by the autoreply transport, Exim
6337 Auto-Submitted: auto-replied
6339 whereas for all other generated messages (e.g. bounces) it adds
6341 Auto-Submitted: auto-generated
6343 63. The "personal" condition in filters now includes a test for the
6344 Auto-Submitted: header. If it contains the string "auto-" the message it
6345 not considered personal.
6347 64. Added rcpt_include_affixes as a generic transport option.
6349 65. Added queue_only_override (default true).
6351 66. Added the syslog_duplication option.
6353 67. If what should have been the first header line of a message consisted of
6354 a space followed by a colon, Exim was mis-interpreting it as a header line.
6355 It isn't of course - it is syntactically invalid and should therefore be
6356 treated as the start of the message body. The misbehaviour could have
6357 caused a number of strange effects, including loss of data in subsequent
6358 header lines, and spool format errors.
6360 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6361 client host had authenticated. This control can now be exercised by an ACL
6362 for more flexibility.
6364 69. By default, callouts do not happen when testing with -bh. There is now a
6365 variant, -bhc, which does actually run the callout code, including
6366 consulting and updating the callout cache.
6368 70. Added support for saslauthd authentication, courtesy of Alexander
6371 71. If statvfs() failed on the spool or log directories while checking their
6372 size for availability, Exim confusingly gave the error "space shortage".
6373 Furthermore, in debugging mode it crashed with a floating point exception.
6374 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6375 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6376 serious problem, Exim now writes to the main and panic logs when this
6377 happens, with details of the failure. It then refuses to accept the
6378 incoming message, giving the message "spool directory problem" or "log
6379 directory problem" with a 421 code for SMTP messages.
6381 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6382 0, 1, and 2 exist, because some OS complain for execs without them (see
6383 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6384 descriptors. However, the code omitted to check that the open succeeded,
6385 causing mysterious errors if for some reason the permissions on /dev/null
6386 got screwed. Now Exim writes a message to the main and panic logs, and
6387 bombs out if it can't open /dev/null.
6389 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6390 interact so that it is all more flexible. It is supposed to remain
6391 backwards compatible. Also added extra_local_interfaces.
6393 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6394 to bomb out with an assertion failure - to the client this appears as a
6395 connection drop. This problem occurs in the part of the code that was taken
6396 from the Samba project. Fortunately, the assertion is in a very simple
6397 function, so I have fixed this by reproducing the function inline in the
6398 one place where it is called, and arranging for authentication to fail
6399 instead of killing the process with assert().
6401 75. The SPA client code was not working when the server requested OEM rather
6402 than Unicode encoding.
6404 76. Added code to make require_files with a specific uid setting more usable in
6405 the case where statting the file as root fails - usually a non-root-mounted
6406 NFS file system. When this happens and the failure is EACCES, Exim now
6407 forks a subprocess and does the per-uid checking as the relevant uid.
6409 77. Added process_log_path.
6411 78. If log_file_path was not explicitly set, a setting of check_log_space or
6412 check_log_inodes was ignored.
6414 79. If a space check for the spool or log partitions fails, the incident is now
6415 logged. Of course, in the latter case the data may get lost...
6417 80. Added the %p formatting code to string_format() so that it can be used to
6418 print addresses in debug_print(). Adjusted all the address printing in the
6419 debugging in store.c to use %p rather than %d.
6421 81. There was a concern that a line of code in smtp_in.c could overflow a
6422 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6423 initially expressed, the concern was not well-founded, because trailing
6424 spaces are removed early. However, if the trailing spaces were followed by
6425 a NULL, they did not get removed, so the overflow was possible. Two fixes
6428 (a) I re-wrote the offending code in a cleaner fashion.
6429 (b) If an incoming SMTP command contains a NULL character, it is rejected
6432 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6433 runs initgroups(), so that if the Exim user is in any additional groups,
6434 they will be used during message reception.
6440 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob