1 Change log file for Exim from version 4.21
2 -------------------------------------------
7 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
8 and 3.4.4 - once the server is enabled to respond to an OCSP request
9 it does even when not requested, resulting in a stapling non-aware
10 client dropping the TLS connection.
12 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
13 support variable-length bit vectors. No functional change.
15 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
16 The I= interface field on outgoing lines is now after the H= remote
17 host field, same as incoming lines. There is a separate
18 outgoing_interface log selector which allows you to disable the
21 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
22 If not running log_selector +smtp_connection the mainlog would be held
23 open indefinitely after a "too many connections" event, including to a
24 deleted file after a log rotate. Leave the per net connection logging
25 leaving it open for efficiency as that will be quickly detected by the
26 check on the next write.
28 HS/01 Bug 1671: Fix post transport crash.
29 Processing the wait-<transport> messages could crash the delivery
30 process if the message IDs didn't exist for some reason. When
31 using 'split_spool_directory=yes' the construction of the spool
32 file name failed already, exposing the same netto behaviour.
34 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
35 mime_regex ACL conditions.
37 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
38 to DSN fail messages (bounces): remote IP, remote greeting, remote response
39 to HELO, local diagnostic string.
41 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
42 log line to debug. Even when configured with a tls authenticator many
43 client connections are expected to not authenticate in this way, so
44 an authenticate fail is not an error.
46 HS/02 Add the Exim version string to the process info. This way exiwhat
47 gives some more detail about the running daemon.
49 JH/06 Bug 1395: time-limit cacheing of DNS lookups, to the TTL value. This may
50 matter for fast-change records such as DNSBLs.
52 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
53 retry record, even if constant. There may be multiple transports with
54 different interface settings and the retry behaviour needs to be kept
57 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
59 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
61 JH/10 Bug 840: fix log_defer_output option of pipe transport
63 JH/11 Bug 830: use same host for all RCPTS of a message, even under
64 hosts_randomize. This matters a lot when combined with mua_wrapper.
66 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
67 ${quote_pgsql:<string>} operator.
69 JH/13 Bug 1708: avoid misaligned access in cached lookup.
71 JH/14 Change header file name for freeradius-client. Relevant if compiling
72 with Radius support; from the Gentoo tree and checked under Fedora.
74 JH/15 Bug 1712: Introduce $prdr_requested flag variable
76 JH/16 Bug 1714: Permit an empty string as expansion result for transport
77 option transport_filter, meaning no filtering.
79 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
81 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
82 defaults to "*" (all hosts). The variable is now available when not built
83 with TLS, default unset, mainly to enable keeping the testuite sane.
84 If a server certificate is not supplied (via tls_certificate) an error is
85 logged, and clients will find TLS connections fail on startup. Presumably
86 they will retry in-clear.
87 Packagers of Exim are strongly encouraged to create a server certificate
90 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
91 with the $config_file variable.
93 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
94 in transport context, after the attempt, and per-recipient. The latter type
95 is per host attempted. The event data is the error message, and the errno
96 information encodes the lookup type (A vs. MX) used for the (first) host,
97 and the trailing two digits of the smtp 4xx reponse.
99 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
100 to write to mainlog (or rejectlog, paniclog) in the window between file
101 creation and permissions/ownership being changed. Particularly affects
102 installations where exicyclog is run as root, rather than exim user;
103 result is that the running daemon panics and dies.
105 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
107 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
108 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
109 "pri" and "weight". Note that the previous implicit priority given by the
110 list order is no longer honoured.
112 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalisation
115 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
116 by defining SUPPORT_SOCKS.
118 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
119 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
120 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
121 variables are renamed to proxy_{local,external}_{address,port}.
123 JH/27 Move Internationalisation support from Experimental to mainline, enabled
124 for a build by defining SUPPORT_I18N
126 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
127 of the query string, and make ${quote_redis:} do that quoting.
129 JH/29 Move Events support from Experimental to mainline, enabled by default
130 and removable for a build by defining DISABLE_EVENT.
132 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
134 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
135 cached by the daemon.
137 JH/32 Move Redis support from Experimental to mainline, enabled for a build
138 by defining LOOKUP_REDIS. The libhiredis library is required.
140 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
141 keys are given for lookup.
146 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
149 JH/02 The smtp transport option "multi_domain" is now expanded.
151 JH/03 The smtp transport now requests PRDR by default, if the server offers
154 JH/04 Certificate name checking on server certificates, when exim is a client,
155 is now done by default. The transport option tls_verify_cert_hostnames
156 can be used to disable this per-host. The build option
157 EXPERIMENTAL_CERTNAMES is withdrawn.
159 JH/05 The value of the tls_verify_certificates smtp transport and main options
160 default to the word "system" to access the system default CA bundle.
161 For GnuTLS, only version 3.0.20 or later.
163 JH/06 Verification of the server certificate for a TLS connection is now tried
164 (but not required) by default. The verification status is now logged by
165 default, for both outbound TLS and client-certificate supplying inbound
168 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
171 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
172 Status Notification (bounce) messages are now MIME format per RFC 3464.
173 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
174 under the control of the dsn_advertise_hosts option, and routers may
175 have a dsn_lasthop option.
177 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
178 default, modifiable by a malware= option. The list separator for
179 the options can now be changed in the usual way. Bug 68.
181 JH/10 The smtp_receive_timeout main option is now expanded before use.
183 JH/11 The incoming_interface log option now also enables logging of the
184 local interface on delivery outgoing connections.
186 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
187 if the interface and destination host and port all match.
189 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
192 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
193 Patch from Andrew Lewis.
195 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
196 now supports optional time-restrictions, weighting, and priority
197 modifiers per server. Patch originally by <rommer@active.by>.
199 JH/16 The spamd_address main option now supports a mixed list of local
200 and remote servers. Remote servers can be IPv6 addresses, and
201 specify a port-range.
203 JH/17 Bug 68: The spamd_address main option now supports an optional
204 timeout value per server.
206 JH/18 Bug 1581: Router and transport options headers_add/remove can
207 now have the list separator specified.
209 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
212 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
215 JH/21 Support for the A6 type of dns record is withdrawn.
217 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
218 rather than the verbs used.
220 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
221 from 255 to 1024 chars.
223 JH/24 Verification callouts now attempt to use TLS by default.
225 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
226 are generic router options now. The defaults didn't change.
228 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
229 Original patch from Alexander Shikoff, worked over by JH.
231 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
232 files if ZCAT_COMMAND is not executable.
234 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
236 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
238 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
239 Normally benign, it bites when the pair was led to by a CNAME;
240 modern usage is to not canoicalize the domain to a CNAME target
241 (and we were inconsistent anyway for A-only vs AAAA+A).
243 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
245 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
246 when evaluating $sender_host_dnssec.
248 JH/31 Check the HELO verification lookup for DNSSEC, adding new
249 $sender_helo_dnssec variable.
251 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
253 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
255 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
257 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
258 documented as working, but never had. Support all but $spam_report.
260 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
261 added for tls authenticator.
266 TL/01 When running the test suite, the README says that variables such as
267 no_msglog_check are global and can be placed anywhere in a specific
268 test's script, however it was observed that placement needed to be near
269 the beginning for it to behave that way. Changed the runtest perl
270 script to read through the entire script once to detect and set these
271 variables, reset to the beginning of the script, and then run through
272 the script parsing/test process like normal.
274 TL/02 The BSD's have an arc4random API. One of the functions to induce
275 adding randomness was arc4random_stir(), but it has been removed in
276 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
277 function when detected.
279 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
280 cause callback expansion.
282 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
283 syntax errors in an expansion can be treated as a string instead of
284 logging or causing an error, due to the internal use of bool_lax
285 instead of bool when processing it.
287 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
288 server certificates when making smtp deliveries.
290 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
292 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
294 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
296 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
297 Merged patch from Sebastian Wiedenroth.
299 JH/05 Fix results-pipe from transport process. Several recipients, combined
300 with certificate use, exposed issues where response data items split
301 over buffer boundaries were not parsed properly. This eventually
302 resulted in duplicates being sent. This issue only became common enough
303 to notice due to the introduction of conection certificate information,
304 the item size being so much larger. Found and fixed by Wolfgang Breyha.
306 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
307 size buffer was used, resulting in syntax errors when an expansion
310 JH/07 Add support for directories of certificates when compiled with a GnuTLS
311 version 3.3.6 or later.
313 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
314 is EXPERIMENTAL_EVENT, the main-configuration and transport options
315 both become "event_action", the variables become $event_name, $event_data
316 and $event_defer_errno. There is a new variable $verify_mode, usable in
317 routers, transports and related events. The tls:cert event is now also
318 raised for inbound connections, if the main configuration event_action
321 TL/06 In test suite, disable OCSP for old versions of openssl which contained
322 early OCSP support, but no stapling (appears to be less than 1.0.0).
324 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
325 server certificate names available under the smtp transport option
326 "tls_verify_cert_hostname" now do not permit multi-component wildcard
329 JH/10 Time-related extraction expansions from certificates now use the main
330 option "timezone" setting for output formatting, and are consistent
331 between OpenSSL and GnuTLS compilations. Bug 1541.
333 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
334 encoded parameter in the incoming message. Bug 1558.
336 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
337 include certificate info, eximon was claiming there were spoolfile
340 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
342 JH/14 Log delivery-related information more consistently, using the sequence
343 "H=<name> [<ip>]" wherever possible.
345 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
346 are problematic for Debian distribution, omit them from the release
349 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
351 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
354 JH/17 Fix a null-indirection in certextract expansions when a nondefault
355 output list separator was used.
360 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
361 checkers that were complaining about end of non-void function with no
364 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
365 This was a regression intruduced in 4.83 by another bugfix.
367 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
369 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
370 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
376 TF/01 Correctly close the server side of TLS when forking for delivery.
378 When a message was received over SMTP with TLS, Exim failed to clear up
379 the incoming connection properly after forking off the child process to
380 deliver the message. In some situations the subsequent outgoing
381 delivery connection happened to have the same fd number as the incoming
382 connection previously had. Exim would try to use TLS and fail, logging
383 a "Bad file descriptor" error.
385 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
386 utilities have not been installed.
388 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
389 temporary space as the ACL may create new global variables.
391 TL/01 LDAP support uses per connection or global context settings, depending
392 upon the detected version of the libraries at build time.
394 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
395 to extract and use the src ip:port in logging and expansions as if it
396 were a direct connection from the outside internet. PPv2 support was
397 updated based on HAProxy spec change in May 2014.
399 JH/02 Add ${listextract {number}{list}{success}{fail}}.
401 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
402 Properly escape header and check for NULL return.
404 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
407 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
409 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
410 characters in header names, implemented as a verify condition.
411 Contributed by Michael Fischer v. Mollard.
413 TL/05 Rename SPF condition results err_perm and err_temp to standardized
414 results permerror and temperror. Previous values are deprecated but
415 still accepted. In a future release, err_perm and err_temp will be
416 completely removed, which will be a backward incompatibility if the
417 ACL tests for either of these two old results. Patch contributed by
418 user bes-internal on the mailing list.
420 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
422 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
423 selectors, in both main and reject logs.
425 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
428 JH/07 Add malware type "sock" for talking to simple daemon.
430 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
432 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
433 routers/transports under cutthrough routing.
435 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
436 numbers. Touch up "bool" conditional to keep the same definition.
438 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
440 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
442 JH/12 Expand items in router/transport headers_add or headers_remove lists
443 individually rather than the list as a whole. Bug 1452.
445 Required for reasonable handling of multiple headers_ options when
446 they may be empty; requires that headers_remove items with embedded
447 colons must have them doubled (or the list-separator changed).
449 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
450 view the policy declared in the DMARC record. Currently, $dmarc_status
451 is a combined value of both the record presence and the result of the
454 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
456 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
457 dnslookup router and the smtp transport (applying to the forward
460 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
461 of ldap servers used for a specific lookup. Patch provided by Heiko
464 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
465 New variable $lookup_dnssec_authenticated for observability.
467 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
468 Patch submitted by Lars Timman.
470 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
472 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
473 Requires trusted mode and valid format message id, aborts otherwise.
474 Patch contributed by Heiko Schlichting.
476 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
477 certextract with support for various fields. Bug 1358.
479 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
480 is requested by default, modifiable by smtp transport option
483 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
484 operate on certificate variables to give certificate fingerprints
485 Also new ${sha256:cert_variable}.
487 JH/23 The PRDR feature is moved from being Experimental into the mainline.
489 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
490 Christian Aistleitner.
492 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
494 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
495 file. Patch from Wolfgang Breyha.
497 JH/25 Expand the coverage of the delivery $host and $host_address to
498 client authenticators run in verify callout. Bug 1476.
500 JH/26 Port service names are now accepted for tls_on_connect_ports, to
501 align with daemon_smtp_ports. Bug 72.
503 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
504 support and error reporting did not work properly.
506 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
507 and is readable. Patch from Andrew Colin Kissa.
509 TL/14 Enhance documentation of ${run expansion and how it parses the
510 commandline after expansion, particularly in the case when an
511 unquoted variable expansion results in an empty value.
513 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
515 PP/02 Fix internal collision of T_APL on systems which support RFC3123
516 by renaming away from it. Addresses GH issue 15, reported by
519 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
521 TL/15 SECURITY: prevent double expansion in math comparison functions
522 (can expand unsanitized data). Not remotely exploitable.
529 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
531 PP/02 Make -n do something, by making it not do something.
532 When combined with -bP, the name of an option is not output.
534 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
537 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
538 $sender_host_name and config options to manage this, and basic check
541 PP/05 DSCP support for outbound connections and control modifier for inbound.
543 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
544 (Only plugin which currently uses this is kerberos4, which nobody should
545 be using, but we should make it available and other future plugins might
546 conceivably use it, even though it would break NAT; stuff *should* be
547 using channel bindings instead).
549 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
550 name; added for Sendmail compatibility; requires admin caller.
551 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
552 just ignore it); requires trusted caller.
553 Also parse but ignore: -Ac -Am -X<logfile>
556 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
558 TL/02 Add +smtp_confirmation as a default logging option.
560 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
561 Patch by Magnus Holmgren from 2007-02-20.
563 TL/04 Bugzilla 1281 - Spec typo.
564 Bugzilla 1283 - Spec typo.
565 Bugzilla 1290 - Spec grammar fixes.
567 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
569 TL/06 Add Experimental DMARC support using libopendmarc libraries.
571 TL/07 Fix an out of order global option causing a segfault. Reported to dev
572 mailing list by by Dmitry Isaikin.
574 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
576 JH/02 Support "G" suffix to numbers in ${if comparisons.
578 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
580 NM/01 Bugzilla 1197 - Spec typo
581 Bugzilla 1196 - Spec examples corrections
583 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
585 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
586 gnutls_enable_pkcs11, but renamed to more accurately indicate its
589 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
590 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
592 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
593 "acl {{name}{arg}...}", and optional args on acl condition
596 JH/05 Permit multiple router/transport headers_add/remove lines.
598 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
600 JH/07 Avoid using a waiting database for a single-message-only transport.
601 Performance patch from Paul Fisher. Bugzilla 1262.
603 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
606 JH/09 Add $headers_added variable, with content from use of ACL modifier
607 add_header (but not yet added to the message). Bugzilla 199.
609 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
610 Pulled from Bugzilla 817 by Wolfgang Breyha.
612 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
614 (nb: this is the same fix as in Exim 4.80.1)
616 JH/11 Add A= logging on delivery lines, and a client_set_id option on
619 JH/12 Add optional authenticated_sender logging to A= and a log_selector
622 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
624 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
625 advertise SMTP AUTH mechanism to us, instead of a generic
626 protocol violation error. Also, make Exim more robust to bad
627 data from the Dovecot auth socket.
629 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
631 When a queue runner is handling a message, Exim first routes the
632 recipient addresses, during which it prunes them based on the retry
633 hints database. After that it attempts to deliver the message to
634 any remaining recipients. It then updates the hints database using
637 So if a recipient address works intermittently, it can get repeatedly
638 deferred at routing time. The retry hints record remains fresh so the
639 address never reaches the final cutoff time.
641 This is a fairly common occurrence when a user is bumping up against
642 their storage quota. Exim had some logic in its local delivery code
643 to deal with this. However it did not apply to per-recipient defers
644 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
646 This change adds a proper retry rule check during routing so that the
647 final cutoff time is checked against the message's age. We only do
648 this check if there is an address retry record and there is not a
649 domain retry record; this implies that previous attempts to handle
650 the address had the retry_use_local_parts option turned on. We use
651 this as an approximation for the destination being like a local
652 delivery, as in LMTP.
654 I suspect this new check makes the old local delivery cutoff check
655 redundant, but I have not verified this so I left the code in place.
657 TF/02 Correct gecos expansion when From: is a prefix of the username.
659 Test 0254 submits a message to Exim with the header
663 When I ran the test suite under the user fanf2, Exim expanded
664 the header to contain my full name, whereas it should have added
665 a Resent-Sender: header. It erroneously treats any prefix of the
666 username as equal to the username.
668 This change corrects that bug.
670 GF/01 DCC debug and logging tidyup
671 Error conditions log to paniclog rather than rejectlog.
672 Debug lines prefixed by "DCC: " to remove any ambiguity.
674 TF/03 Avoid unnecessary rebuilds of lookup-related code.
676 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
677 Bug spotted by Jeremy Harris; was flawed since initial commit.
678 Would have resulted in OCSP responses post-SNI triggering an Exim
679 NULL dereference and crash.
681 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
683 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
684 Bug detection, analysis and fix by Samuel Thibault.
685 Bugzilla 1331, Debian bug #698092.
687 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
689 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
690 Server implementation by Todd Lyons, client by JH.
691 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
692 config variable "prdr_enable" controls whether the server
693 advertises the facility. If the client requests PRDR a new
694 acl_data_smtp_prdr ACL is called once for each recipient, after
695 the body content is received and before the acl_smtp_data ACL.
696 The client is controlled by bolth of: a hosts_try_prdr option
697 on the smtp transport, and the server advertisement.
698 Default client logging of deliveries and rejections involving
699 PRDR are flagged with the string "PRDR".
701 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
702 fclose(). Diagnosis by Todd Lyons.
704 PP/17 Update configure.default to handle IPv6 localhost better.
705 Patch by Alain Williams (plus minor tweaks).
708 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
709 This is now consistent with GnuTLS, and is now documented: the
710 previous undocumented portable approach to treating the option as
711 unset was to force an expansion failure. That still works, and
712 an empty string is now equivalent.
714 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
715 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
716 not performing validation itself.
718 PP/20 Added force_command boolean option to pipe transport.
719 Patch from Nick Koston, of cPanel Inc.
721 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
724 TF/04 Added udpsend ACL modifer and hexquote expansion operator
726 PP/21 Fix eximon continuous updating with timestamped log-files.
727 Broken in a format-string cleanup in 4.80, missed when I repaired the
728 other false fix of the same issue.
729 Report and fix from Heiko Schlichting.
732 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
733 Report from Prashanth Katuri.
735 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
736 It's SecureTransport, so affects any MacOS clients which use the
737 system-integrated TLS libraries, including email clients.
739 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
740 using a MIME ACL for non-SMTP local injection.
741 Report and assistance in diagnosis by Warren Baker.
743 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
745 JH/16 Fix comparisons for 64b. Bugzilla 1385.
747 TL/09 Add expansion variable $authenticated_fail_id to keep track of
748 last id that failed so it may be referenced in subsequent ACL's.
750 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
753 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
754 ldap library initialization, allowing self-signed CA's to be
755 used. Also properly sets require_cert option later in code by
756 using NULL (global ldap config) instead of ldap handle (per
757 session). Bug diagnosis and testing by alxgomz.
759 TL/12 Enhanced documentation in the ratelimit.pl script provided in
760 the src/util/ subdirectory.
762 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
763 renamed to Transport Post Delivery Action by Jeremy Harris, as
766 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
767 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
768 redis_servers = needs to be configured which will be used by the redis
769 lookup. Patch from Warren Baker, of The Packet Hub.
771 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
773 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
774 hostname or reverse DNS when processing a host list. Used suggestions
775 from multiple comments on this bug.
777 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
779 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
780 Missed a few lines, added it to make the runtest require no keyboard
783 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
784 contains upper case chars. Make router use caseful_local_part.
786 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
787 support when GnuTLS has been built with p11-kit.
793 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
795 This, or similar/improved, will also be change PP/11 of 4.82.
801 PP/01 Handle short writes when writing local log-files.
802 In practice, only affects FreeBSD (8 onwards).
803 Bugzilla 1053, with thanks to Dmitry Isaikin.
805 NM/01 Bugzilla 949 - Documentation tweak
807 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
810 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
812 PP/02 Implemented gsasl authenticator.
814 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
816 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
817 `pkg-config foo` for cflags/libs.
819 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
820 with rest of GSASL and with heimdal_gssapi.
822 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
823 `pkg-config foo` for cflags/libs for the TLS implementation.
825 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
826 properties get this fed in as external SSF. A number of robustness
827 and debugging improvements to the cyrus_sasl authenticator.
829 PP/08 cyrus_sasl server now expands the server_realm option.
831 PP/09 Bugzilla 1214 - Log authentication information in reject log.
832 Patch by Jeremy Harris.
834 PP/10 Added dbmjz lookup type.
836 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
838 PP/12 MAIL args handles TAB as well as SP, for better interop with
839 non-compliant senders.
840 Analysis and variant patch by Todd Lyons.
842 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
843 Bug report from Lars Müller <lars@samba.org> (via SUSE),
844 Patch from Dirk Mueller <dmueller@suse.com>
846 PP/13 tls_peerdn now print-escaped for spool files.
847 Observed some $tls_peerdn in wild which contained \n, which resulted
848 in spool file corruption.
850 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
851 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
852 or write after TLS renegotiation, which otherwise led to messages
855 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
856 as a tracking header (ie: a signed header comes before the signature).
857 Patch from Wolfgang Breyha.
859 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
860 comma-sep list; embedded commas doubled.
862 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
864 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
866 Report and patch from Dmitry Banschikov.
868 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
869 Removed SSL_clear() after SSL_new() which led to protocol negotiation
870 failures. We appear to now support TLS1.1+ with Exim.
872 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
873 lets Exim select keys and certificates based upon TLS SNI from client.
874 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
875 before an outbound SMTP session. New log_selector, +tls_sni.
877 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
878 NULL dereference. Report and patch from Alun Jones.
880 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
881 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
882 Not seeing resolver debug output on NetBSD, but suspect this is a
883 resolver implementation change.
885 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
886 Left warnings. Added "eximon gdb" invocation mode.
888 PP/21 Defaulting "accept_8bitmime" to true, not false.
890 PP/22 Added -bw for inetd wait mode support.
892 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
893 locate the relevant includes and libraries. Made this the default.
895 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
896 Bugzilla 1246, report and most of solution from Tomasz Kusy.
898 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
899 This may cause build issues on older platforms.
901 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
902 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
903 gnutls_require_mac & gnutls_require_protocols (no longer supported).
904 Added SNI support via GnuTLS too.
905 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
907 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
909 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
910 Applied second patch from Janne, implementing suggestion to default
911 multiple-strings-in-record handling to match SPF spec.
913 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
915 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
916 read-only, out of scope).
917 Patch from Wolfgang Breyha, report from Stuart Northfield.
919 PP/29 Fix three issues highlighted by clang analyser static analysis.
920 Only crash-plausible issue would require the Cambridge-specific
921 iplookup router and a misconfiguration.
922 Report from Marcin Mirosław.
924 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
926 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
927 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
928 As part of this, removing so much warning spew let me fix some minor
929 real issues in debug logging.
931 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
932 assignment on my part. Fixed.
934 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
935 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
936 Janne Snabb (who went above and beyond: thank you).
938 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
939 string otherwise requires a connection and a bunch more work and it's
940 relatively easy to get wrong. Should also expose TLS library linkage
943 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
944 64-bit ${eval} (JH/03).
946 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
947 GNU libc to support some of the 64-bit stuff, should not lead to
948 conflicts. Defined before os.h is pulled in, so if a given platform
949 needs to override this, it can.
951 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
952 protection layer was required, which is not implemented.
953 Bugzilla 1254, patch from Wolfgang Breyha.
955 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
956 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
957 tls_dhparam take prime identifiers. Also unbreak combination of
958 OpenSSL+DH_params+TLSSNI.
960 PP/39 Disable SSLv2 by default in OpenSSL support.
966 PP/01 Solaris build fix for Oracle's LDAP libraries.
967 Bugzilla 1109, patch from Stephen Usher.
969 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
971 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
974 TF/02 Fix a couple more cases where we did not log the error message
975 when unlink() failed. See also change 4.74-TF/03.
977 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
978 lock up or crash if it happened to be inside a call to libc when it
979 got a SIGUSR1 from exiwhat.
981 The SIGUSR1 handler appends the current process status to the process
982 log which is later printed by exiwhat. It used to use the general
983 purpose logging code to do this, but several functions it calls are
984 not safe for signals.
986 The new output code in the SIGUSR1 handler is specific to the process
987 log, and simple enough that it's easy to inspect for signal safety.
988 Removing some special cases also simplifies the general logging code.
989 Removing the spurious timestamps from the process log simplifies
992 TF/04 Improved ratelimit ACL condition.
994 The /noupdate option has been deprecated in favour of /readonly which
995 has clearer semantics. The /leaky, /strict, and /readonly update modes
996 are mutually exclusive. The update mode is no longer included in the
997 database key; it just determines when the database is updated. (This
998 means that when you upgrde Exim will forget old rate measurements.)
1000 Exim now checks that the per_* options are used with an update mode that
1001 makes sense for the current ACL. For example, when Exim is processing a
1002 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1003 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1004 must specify per_mail/readonly. If you omit the update mode it defaults to
1005 /leaky where that makes sense (as before) or /readonly where required.
1007 The /noupdate option is now undocumented but still supported for
1008 backwards compatibility. It is equivalent to /readonly except that in
1009 ACLs where /readonly is required you may specify /leaky/noupdate or
1010 /strict/noupdate which are treated the same as /readonly.
1012 A useful new feature is the /count= option. This is a generalization
1013 of the per_byte option, so that you can measure the throughput of other
1014 aggregate values. For example, the per_byte option is now equivalent
1015 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1017 The per_rcpt option has been generalized using the /count= mechanism
1018 (though it's more complicated than the per_byte equivalence). When it is
1019 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1020 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1021 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1022 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1023 Note that using per_rcpt with a non-readonly update mode in more than
1024 one ACL will cause the recipients to be double-counted. (The per_mail
1025 and per_byte options don't have this problem.)
1027 The handling of very low rates has changed slightly. If the computed rate
1028 is less than the event's count (usually one) then this event is the first
1029 after a long gap. In this case the rate is set to the same as this event's
1030 count, so that the first message of a spam run is counted properly.
1032 The major new feature is a mechanism for counting the rate of unique
1033 events. The new per_addr option counts the number of different
1034 recipients that someone has sent messages to in the last time period. It
1035 behaves like per_rcpt if all the recipient addresses are different, but
1036 duplicate recipient addresses do not increase the measured rate. Like
1037 the /count= option this is a general mechanism, so the per_addr option
1038 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1039 example, measure the rate that a client uses different sender addresses
1040 with the options per_mail/unique=$sender_address. There are further
1041 details in the main documentation.
1043 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1045 TF/06 Removed a few PCRE remnants.
1047 TF/07 Automatically extract Exim's version number from tags in the git
1048 repository when doing development or release builds.
1050 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1051 Bugzilla 879. Patch from Paul Fisher.
1053 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1054 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1057 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1059 PP/05 Variable $av_failed, true if the AV scanner deferred.
1060 Bugzilla 1078. Patch from John Horne.
1062 PP/06 Stop make process more reliably on build failure.
1063 Bugzilla 1087. Patch from Heiko Schlittermann.
1065 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1066 Bugzilla 1089. Patch from Heiko Schlittermann.
1068 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1069 Bugzilla 1131. Patch from Holger Weiß.
1071 PP/09 Handle IPv6 addresses with SPF.
1072 Bugzilla 860. Patch from Wolfgang Breyha.
1074 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1076 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1079 PP/11 match_* no longer expand right-hand-side by default.
1080 New compile-time build option, EXPAND_LISTMATCH_RHS.
1081 New expansion conditions, "inlist", "inlisti".
1083 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1085 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1087 PP/14 fix log_write() format string regression from TF/03.
1088 Bugzilla 1152. Patch from Dmitry Isaikin.
1094 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1096 PP/02 Harmonised TLS library version reporting; only show if debugging.
1097 Layout now matches that introduced for other libraries in 4.74 PP/03.
1099 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1101 PP/04 New "dns_use_edns0" global option.
1103 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1106 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1107 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1109 TK/01 Updated PolarSSL code to 0.14.2.
1110 Bugzilla 1097. Patch from Andreas Metzler.
1112 PP/07 Catch divide-by-zero in ${eval:...}.
1113 Fixes bugzilla 1102.
1115 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1118 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1119 format-string attack -- SECURITY: remote arbitrary code execution.
1121 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1122 time unintentionally subject to list matching rules, letting the header
1123 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1124 arbitrary string expansion). This allowed for information disclosure.
1126 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1127 INT_MIN/-1 -- value coerced to INT_MAX.
1133 NM/01 Workround for PCRE version dependency in version reporting
1136 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1137 This fixes portability to compilers other than gcc, notably
1138 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1140 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1141 makefiles for portability to HP-UX and POSIX correctness.
1143 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1144 Also via indented variable definition in the Makefile.
1145 (Debugging by Oliver Heesakkers).
1147 PP/02 Restore caching of spamd results with expanded spamd_address.
1148 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1150 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1151 Improves build reliability. Fix from: Frank Elsner
1153 NM/02 Fix wide character breakage in the rfc2047 coding
1154 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1156 NM/03 Allow underscore in dnslist lookups
1157 Fixes bug 1026. Patch from Graeme Fowler
1159 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1160 Code patches from Adam Ciarcinski of NetBSD.
1162 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1165 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1166 is logged, to avoid truncation. Patch from John Horne.
1168 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1169 Patch from Jakob Hirsch.
1171 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1172 SQL string expansion failure details.
1173 Patch from Andrey Oktyabrski.
1175 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1176 Patch from Simon Arlott.
1178 PP/09 New lookups functionality failed to compile on old gcc which rejects
1179 extern declarations in function scope.
1180 Patch from Oliver Fleischmann
1182 PP/10 Use sig_atomic_t for flags set from signal handlers.
1183 Check getgroups() return and improve debugging.
1184 Fixed developed for diagnosis in bug 927 (which turned out to be
1187 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1188 Patch from Mark Zealey.
1190 PP/12 Bugzilla 1056: Improved spamd server selection.
1191 Patch from Mark Zealey.
1193 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1194 Based on patch from Heiko Schlittermann.
1196 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1197 Patch from Uwe Doering, sign-off by Michael Haardt.
1199 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1200 variable declaration deep within a block. Bug and patch from
1203 PP/15 lookups-Makefile IRIX compatibilty coercion.
1205 PP/16 Make DISABLE_DKIM build knob functional.
1207 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1208 Patch by Simon Arlott
1210 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1211 variable argument macros. Our copy now differs from upstream.
1217 TF/01 Failure to get a lock on a hints database can have serious
1218 consequences so log it to the panic log.
1220 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1221 controlled using the smtp_confirmation log selector.
1223 TF/03 Include the error message when we fail to unlink a spool file.
1225 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1226 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1227 for maintaining out-of-tree patches for some time.
1229 PP/01 Bugzilla 139: Documentation and portability issues.
1230 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1231 Handle per-OS dynamic-module compilation flags.
1233 PP/02 Let /dev/null have normal permissions.
1234 The 4.73 fixes were a little too stringent and complained about the
1235 permissions on /dev/null. Exempt it from some checks.
1236 Reported by Andreas M. Kirchwitz.
1238 PP/03 Report version information for many libraries, including
1239 Exim version information for dynamically loaded libraries. Created
1240 version.h, now support a version extension string for distributors
1241 who patch heavily. Dynamic module ABI change.
1243 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1244 privilege escalation vulnerability whereby the Exim run-time user
1245 can cause root to append content of the attacker's choosing to
1248 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1251 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1252 If dropping privileges for untrusted macros, we disabled normal logging
1253 on the basis that it would fail; for the Exim run-time user, this is not
1254 the case, and it resulted in successful deliveries going unlogged.
1255 Fixed. Reported by Andreas Metzler.
1261 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1262 only prepend for the Resent-* case. Fixes regression introduced in
1263 Exim 4.70 by NM/22 for Bugzilla 607.
1265 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1266 increasing numbers of administrators be bitten by this.
1268 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1270 PP/03 Bugzilla 994: added openssl_options main configuration option.
1272 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1274 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1276 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1278 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1279 without a peer certificate, leading to a segfault because of an
1280 assumption that peers always have certificates. Be a little more
1281 paranoid. Problem reported by Martin Tscholak.
1283 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1284 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1285 NB: ClamAV planning to remove STREAM in "middle of 2010".
1286 CL also introduces -bmalware, various -d+acl logging additions and
1287 more caution in buffer sizes.
1289 PP/09 Implemented reverse_ip expansion operator.
1291 PP/10 Bugzilla 937: provide a "debug" ACL control.
1293 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1295 PP/12 Bugzilla 973: Implement --version.
1297 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1299 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1301 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1303 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1304 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1305 ignore trailing whitespace.
1307 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1309 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1312 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1313 Notification from Dr Andrew Aitchison.
1315 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1316 ExtendedDetectionInfo response format.
1317 Notification from John Horne.
1319 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1322 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1323 XSL and documented dependency on system catalogs, with examples of how
1326 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1329 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1330 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1331 configuration file which is writeable by the Exim user or group.
1333 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1334 of configuration files to cover files specified with the -C option if
1335 they are going to be used with root privileges, not just the default
1338 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1339 option (effectively making it always true).
1341 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1342 files to be used while preserving root privileges.
1344 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1345 that rogue child processes cannot use them.
1347 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1348 run-time user, instead of root.
1350 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1351 Exim run-time user without dropping privileges.
1353 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1354 result string, instead of calling string_vformat() twice with the same
1357 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1358 for other users. Others should always drop root privileges if they use
1359 -C on the command line, even for a whitelisted configure file.
1361 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1363 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1369 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1370 $data_path, and $header_path variables; fixed documentation bugs and
1373 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1374 exipick to access non-standard spools, including the "frozen" queue
1377 NM/01 Bugzilla 965: Support mysql stored procedures.
1378 Patch from Alain Williams
1380 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1382 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1383 Patch from Andreas Metzler
1385 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1386 Patch from Kirill Miazine
1388 NM/05 Bugzilla 671: Added umask to procmail example.
1390 JJ/03 installed exipick 20100323.0, fixing doc bug
1392 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1393 directory. Notification and patch from Dan Rosenberg.
1395 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1397 TK/02 Improve log output when DKIM signing operation fails.
1399 MH/01 Treat the transport option dkim_domain as a colon separated
1400 list, not as a single string, and sign the message with each element,
1401 omitting multiple occurences of the same signer.
1403 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1404 Bugzilla 985, 986. Patch by Simon Arlott
1406 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1407 Patch by Simon Arlott
1409 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1410 MBX locking. Notification from Dan Rosenberg.
1416 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1418 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1420 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1422 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1424 NM/04 Bugzilla 915: Flag broken perl installation during build.
1430 TK/01 Added patch by Johannes Berg that expands the main option
1431 "spamd_address" if it starts with a dollar sign.
1433 TK/02 Write list of recipients to X-Envelope-Sender header when building
1434 the mbox-format spool file for content scanning (suggested by Jakob
1437 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1438 (http://www.dcc-servers.net/) support via dccifd. Activated by
1439 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1441 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1442 by Mark Daniel Reidel <mr@df.eu>.
1444 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1445 When building exim an external PCRE library is now needed -
1446 PCRE is a system library on the majority of modern systems.
1447 See entry on PCRE_LIBS in EDITME file.
1449 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1450 conversation. Added nologin parameter to request.
1451 Patch contributed by Kirill Miazine.
1453 TF/01 Do not log submission mode rewrites if they do not change the address.
1455 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1457 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1458 log files in place. Contributed by Roberto Lima.
1460 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1462 TF/03 Bugzilla 615: When checking the local_parts router precondition
1463 after a local_part_suffix or local_part_prefix option, Exim now
1464 does not use the address's named list lookup cache, since this
1465 contains cached lookups for the whole local part.
1467 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1468 Robert Millan. Documentation is in experimental-spec.txt.
1470 TF/04 Bugzilla 668: Fix parallel build (make -j).
1472 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1474 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1475 Patch provided by Jan Srzednicki.
1477 TF/05 Leading white space used to be stripped from $spam_report which
1478 wrecked the formatting. Now it is preserved.
1480 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1481 that they are available at delivery time.
1483 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1485 TF/08 TLS error reporting now respects the incoming_interface and
1486 incoming_port log selectors.
1488 TF/09 Produce a more useful error message if an SMTP transport's hosts
1489 setting expands to an empty string.
1491 NM/06 Bugzilla 744: EXPN did not work under TLS.
1492 Patch provided by Phil Pennock.
1494 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1495 Patch provided by Richard Godbee.
1497 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1498 acl_smtp_notquit, added index entry.
1500 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1501 Patch provided by Eugene Bujak.
1503 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1504 accept(). Patch provided by Maxim Dounin.
1506 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1507 Patch provided by Phil Pennock.
1509 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1511 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1512 Patch provided by Brad "anomie" Jorsch.
1514 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1515 Patch provided by Dean Brooks.
1517 TK/05 Add native DKIM support (does not depend on external libraries).
1519 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1520 Patch provided by Graeme Fowler.
1522 NM/16 Bugzilla 851: Documentation example syntax fix.
1524 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1526 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1529 NM/19 Bugzilla 745: TLS version reporting.
1530 Patch provided by Phil Pennock.
1532 NM/20 Bugzilla 167: bool: condition support.
1533 Patch provided by Phil Pennock.
1535 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1536 clients. Patch provided by Phil Pennock.
1538 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1539 Patch provided by Brad "anomie" Jorsch.
1541 NM/23 Bugzilla 687: Fix misparses in eximstats.
1542 Patch provided by Heiko Schlittermann.
1544 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1545 Patch provided by Heiko Schlittermann.
1547 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1548 plus update to original patch.
1550 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1552 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1553 Patch provided by David Brownlee.
1555 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1557 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1559 NM/30 Bugzilla 888: TLS documentation bugfixes.
1561 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1563 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1564 Unlike the original bugzilla I have changed all shell scripts in src tree.
1566 NM/33 Bugzilla 898: Transport filter timeout fix.
1567 Patch by Todd Rinaldo.
1569 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1570 Patch by Serge Demonchaux.
1572 NM/35 Bugzilla 39: Base64 decode bug fixes.
1573 Patch by Jakob Hirsch.
1575 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1577 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1579 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1581 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1587 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1588 ALT-N's libdkim that I have put here:
1589 http://duncanthrax.net/exim-experimental/
1591 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1592 were called 'true' and it seems that C99 defines that as a reserved
1593 keyword to be used with 'bool' variable types. That means you could
1594 not include C99-style headers which use bools without triggering
1595 build errors in sieve.c.
1597 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1598 as mailq or other aliases. Changed the --help handling significantly
1599 to do whats expected. exim_usage() emits usage/help information.
1601 SC/01 Added the -bylocaldomain option to eximstats.
1603 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1605 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1607 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1613 PH/01 Another patch from the Sieve maintainer.
1615 PH/02 When an IPv6 address is converted to a string for single-key lookup
1616 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1617 dots are used instead of colons so that keys in lsearch files need not
1618 contain colons. This was done some time before quoting was made available
1619 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1620 (notated using the quote facility) so as to distinguish them from IPv4
1621 keys. This meant that lookups for IP addresses in host lists did not work
1622 for iplsearch lookups.
1624 This has been fixed by arranging for IPv6 addresses to be expressed with
1625 colons if the lookup type is iplsearch. This is not incompatible, because
1626 previously such lookups could never work.
1628 The situation is now rather anomolous, since one *can* have colons in
1629 ordinary lsearch keys. However, making the change in all cases is
1630 incompatible and would probably break a number of configurations.
1632 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1635 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1636 conversion specification without a maximum field width, thereby enabling
1637 a rogue spamd server to cause a buffer overflow. While nobody in their
1638 right mind would setup Exim to query an untrusted spamd server, an
1639 attacker that gains access to a server running spamd could potentially
1640 exploit this vulnerability to run arbitrary code as the Exim user.
1642 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1643 $primary_hostname instead of what libspf2 thinks the hosts name is.
1645 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1646 a directory entry by the name of the lookup key. Previously, if a
1647 symlink pointed to a non-existing file or a file in a directory that
1648 Exim lacked permissions to read, a lookup for a key matching that
1649 symlink would fail. Now it is enough that a matching directory entry
1650 exists, symlink or not. (Bugzilla 503.)
1652 PH/03 The body_linecount and body_zerocount variables are now exported in the
1655 PH/04 Added the $dnslist_matched variable.
1657 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1658 This means they are set thereafter only if the connection becomes
1661 PH/06 Added the client_condition to authenticators so that some can be skipped
1662 by clients under certain conditions.
1664 PH/07 The error message for a badly-placed control=no_multiline_responses left
1665 "_responses" off the end of the name.
1667 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1669 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1670 (without spaces) instead of just copying the configuration text.
1672 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1674 PH/11 Added $max_received_linelength.
1676 PH/12 Added +ignore_defer and +include_defer to host lists.
1678 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1679 way in which PCRE > 7.0 is built.
1681 PH/14 Implemented queue_only_load_latch.
1683 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1684 MAIL command. The effect was to mangle the value on 64-bit systems.
1686 PH/16 Another patch from the Sieve maintainer.
1688 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1690 PH/18 If a system quota error occurred while trying to create the file for
1691 a maildir delivery, the message "Mailbox is full" was not appended to the
1692 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1693 only to a quota excession during the actual writing of the file.
1695 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1696 characters?) which causes problems in log lines. The DN values are now
1697 passed through string_printing() before being added to log lines.
1699 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1700 and InterBase are left for another time.)
1702 PH/21 Added message_body_newlines option.
1704 PH/22 Guard against possible overflow in moan_check_errorcopy().
1706 PH/23 POSIX allows open() to be a macro; guard against that.
1708 PH/24 If the recipient of an error message contained an @ in the local part
1709 (suitably quoted, of course), incorrect values were put in $domain and
1710 $local_part during the evaluation of errors_copy.
1716 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1717 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1720 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1721 issue a MAIL command.
1723 PH/02 In an ACL statement such as
1725 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1727 if a client was not listed at all, or was listed with a value other than
1728 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1729 the condition was not true (as it should be), so access was not denied.
1730 The bug was that the ! inversion was incorrectly passed on to the second
1731 item. This has been fixed.
1733 PH/03 Added additional dnslists conditions == and =& which are different from
1734 = and & when the dns lookup returns more than one IP address.
1736 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1737 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1739 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1740 FSYNC, which compiles an option called disable_fsync that allows for
1741 bypassing fsync(). The documentation is heavily laced with warnings.
1743 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1745 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1746 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1747 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1748 including adding "make clean"; (3) Added -fPIC when compiling the test
1749 dynamically loaded module, to get rid of a warning.
1751 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1752 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1753 = 0s. The bug is otherwise harmless.
1755 PH/07 There was a bug in the dovecot authenticator such that the value of
1756 $auth1 could be overwritten, and so not correctly preserved, after a
1757 successful authentication. This usually meant that the value preserved by
1758 the server_setid option was incorrect.
1760 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1762 PH/09 Installed PCRE release 7.0.
1764 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1765 run for batched SMTP input. It is now run at the start of every message
1766 in the batch. While fixing this I discovered that the process information
1767 (output by running exiwhat) was not always getting set for -bs and -bS
1768 input. This is fixed, and it now also says "batched" for BSMTP.
1770 PH/11 Added control=no_pipelining.
1772 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1773 patch, slightly modified), and move the expansion of helo_data till after
1774 the connection is made in the smtp transport (so it can use these
1777 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1779 PH/14 Added log_selector = +pid.
1781 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1783 PH/16 Add ${if forany and ${if forall.
1785 PH/17 Added dsn_from option to vary the From: line in DSNs.
1787 PH/18 Flush SMTP output before performing a callout, unless control =
1788 no_callout_flush is set.
1790 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1791 was true (the default) a successful delivery failed to delete the retry
1792 item, thus causing premature timeout of the address. The bug is now
1795 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1797 PH/21 Long custom messages for fakedefer and fakereject are now split up
1798 into multiline reponses in the same way that messages for "deny" and
1799 other ACL rejections are.
1801 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1802 with slight modification.
1804 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1805 draft, changing the syntax and factoring some duplicate code".
1807 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1808 for deliveries of the second and subsequent messages over the same SMTP
1811 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1812 ${reduce, with only minor "tidies".
1814 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1816 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1817 expansion side effects.
1819 PH/27 When a message times out after an over-quota error from an Exim-imposed
1820 quota, the bounce message says "mailbox is full". This message was not
1821 being given when it was a system quota that was exceeded. It now should
1824 MH/03 Made $recipients available in local_scan(). local_scan() already has
1825 better access to the recipient list through recipients_list[], but
1826 $recipients can be useful in postmaster-provided expansion strings.
1828 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1829 in the case of a MAIL command with additional options following the
1830 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1831 were accidentally chopped off.
1833 PH/29 SMTP synchronization checks are implemented when a command is read -
1834 there is a check that no more input is waiting when there shouldn't be
1835 any. However, for some commands, a delay in an ACL can mean that it is
1836 some time before the response is written. In this time, more input might
1837 arrive, invalidly. So now there are extra checks after an ACL has run for
1838 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1839 pipelining has not been advertised.
1841 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1843 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1844 returned in the SMTP response when smtp_return_error_details was false.
1845 This has been fixed.
1847 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1848 instead of the C I/O that was originally supplied, because problems were
1849 reported on Solaris.
1851 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1852 Exim which did not show up earlier: it was assuming that a call to
1853 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1854 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1855 was a macro that became an assignment, so it seemed to work. This has
1856 changed to a proper function call with a void return, hence the compile
1857 error. Exim's code has been fixed.
1859 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1862 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1864 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1866 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1867 because it turns out to be too big or there is a timeout) there is no
1868 "Completed" line in the log. When some messages of this type were
1869 selected by exigrep, they were listed as "not completed". Others were
1870 picked up by some special patterns. I have improved the selection
1871 criteria to be more general.
1873 PH/38 The host_find_failed option in the manualroute router can now be set
1874 to "ignore", to completely ignore a host whose IP address cannot be
1875 found. If all hosts are ignored, the behaviour is controlled by the new
1876 host_all_ignored option.
1878 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1879 homing or because of multiple MX records with /mx) generated more than
1880 one IP address, and the following item turned out to be the local host,
1881 all the secondary addresses of the first item were incorrectly removed
1882 from the list, along with the local host and any following hosts (which
1883 is what is supposed to happen).
1885 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1886 whoever called Exim into the -H file. In the case of the daemon it was
1887 behaving confusingly. When first started, it used values for whoever
1888 started the daemon, but after a SIGHUP it used the Exim user (because it
1889 calls itself on a restart). I have changed the code so that it now always
1892 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1893 message are rejected with the same error (e.g. no authentication or bad
1894 sender address), and a DATA command is nevertheless sent (as can happen
1895 with PIPELINING or a stupid MUA), the error message that was given to the
1896 RCPT commands is included in the rejection of the DATA command. This is
1897 intended to be helpful for MUAs that show only the final error to their
1900 PH/42 Another patch from the Sieve maintainer.
1902 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1903 Eximstats - Fixed some broken HTML links and added missing column headers
1905 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1906 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1908 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1910 PH/43 Yet another patch from the Sieve maintainer.
1912 PH/44 I found a way to check for a TCP/IP connection going away before sending
1913 the response to the final '.' that terminates a message, but only in the
1914 case where the client has not sent further data following the '.'
1915 (unfortunately, this is allowed). However, in many cases there won't be
1916 any further data because there won't be any more messages to send. A call
1917 to select() can be used: if it shows that the input is "ready", there is
1918 either input waiting, or the socket has been closed. An attempt to read
1919 the next input character can distinguish the two cases. Previously, Exim
1920 would have sent an OK response which the client would never have see.
1921 This could lead to message repetition. This fix should cure that, at
1922 least in a lot of common cases.
1924 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1925 advertised in response to EHLO.
1931 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1932 fixed by 4.65/MH/01 (is this a record?) are fixed:
1934 (i) An empty string was always treated as zero by the numeric comparison
1935 operators. This behaviour has been restored.
1937 (ii) It is documented that the numeric comparison operators always treat
1938 their arguments as decimal numbers. This was broken in that numbers
1939 starting with 0 were being interpreted as octal.
1941 While fixing these problems I realized that there was another issue that
1942 hadn't been noticed. Values of message_size_limit (both the global option
1943 and the transport option) were treated as octal if they started with 0.
1944 The documentation was vague. These values are now always treated as
1945 decimal, and I will make that clear in the documentation.
1951 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1952 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1955 MH/01 Don't check that the operands of numeric comparison operators are
1956 integers when their expansion is in "skipping" mode (fixes bug
1957 introduced by 4.64-PH/07).
1959 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1960 child addresses, Exim now panics and dies. Previously, because the count
1961 is held in a short int, deliveries were likely to be lost. As such a
1962 large number of recipients for a single message is ridiculous
1963 (performance will be very, very poor), I have chosen to impose a limit
1964 rather than extend the field.
1970 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1971 leftover -K file (the existence of which was triggered by #402).
1972 While we were at it, introduced process PID as part of the -K
1973 filename. This should rule out race conditions when creating
1976 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1977 processing considerably. Previous code took too long for large mails,
1978 triggering a timeout which in turn triggers #401.
1980 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1981 in the DK code in transports.c. sendfile() is not really portable,
1982 hence the _LINUX specificness.
1984 TF/01 In the add_headers option to the mail command in an Exim filter,
1985 there was a bug that Exim would claim a syntax error in any
1986 header after the first one which had an odd number of characters
1989 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1990 callout verification, Exim cached a "reject" for the entire domain. This
1991 is correct for most verifications, but it is not correct for a recipient
1992 verification with use_sender or use_postmaster set, because in that case
1993 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1994 case of MAIL FROM:<> rejection from other early rejections (e.g.
1995 rejection of HELO). When verifying a recipient using a non-null MAIL
1996 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1997 Whatever the result of the callout, the value of the domain cache is
1998 left unchanged (for any other kind of callout, getting as far as trying
1999 RCPT means that the domain itself is ok).
2001 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2004 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2005 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2006 session, the daemon ignores ECONNECT errors and logs others; it now
2007 ignores EPIPE as well.
2009 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2010 (quoted-printable decoding).
2012 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2013 later the small subsequent patch to fix an introduced bug.
2015 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2017 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2019 PH/08 An error is now given if message_size_limit is specified negative.
2021 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2022 to be given (somewhat) arbitrary names.
2024 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2027 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2028 miscellaneous code fixes
2030 PH/10 Added the log_reject_target ACL modifier to specify where to log
2033 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2034 hostname. This is wrong, because it relates to the incoming message (and
2035 probably the interface on which it is arriving) and not to the outgoing
2036 callout (which could be using a different interface). This has been
2037 changed to use the value of the helo_data option from the smtp transport
2038 instead - this is what is used when a message is actually being sent. If
2039 there is no remote transport (possible with a router that sets up host
2040 addresses), $smtp_active_hostname is used.
2042 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2043 tweaks were necessary in order to get it to work (see also 21 below):
2044 (a) The code assumed that strncpy() returns a negative number on buffer
2045 overflow, which isn't the case. Replaced with Exim's string_format()
2047 (b) There were several signed/unsigned issues. I just did the minimum
2048 hacking in of casts. There is scope for a larger refactoring.
2049 (c) The code used strcasecmp() which is not a standard C function.
2050 Replaced with Exim's strcmpic() function.
2051 (d) The code set only $1; it now sets $auth1 as well.
2052 (e) A simple test gave the error "authentication client didn't specify
2053 service in request". It would seem that Dovecot has changed its
2054 interface. Fortunately there's a specification; I followed it and
2055 changed what the client sends and it appears to be working now.
2057 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2060 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2061 address A is aliased to B and C, where B exists and C does not. Without
2062 -v the output is "A verified" because verification stops after a
2063 successful redirection if more than one address is generated. However,
2064 with -v the child addresses are also verified. Exim was outputting "A
2065 failed to verify" and then showing the successful verification for C,
2066 with its parentage. It now outputs "B failed to verify", showing B's
2067 parentage before showing the successful verification of C.
2069 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2070 look up a TXT record in a specific list after matching in a combined
2073 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2074 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2075 they consult the DNS. I had assumed they would set it the way they
2076 wanted; and indeed my experiments on Linux seem to show that in some
2077 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2078 To be on the safe side, however, I have now made the interface to
2079 host_find_byname() similar to host_find_bydns(), with an argument
2080 containing the DNS resolver options. The host_find_byname() function now
2081 sets these options at its start, just as host_find_bydns() does. The smtp
2082 transport options dns_qualify_single and dns_search_parents are passed to
2083 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2084 of host_find_byname() use the default settings of RES_DEFNAMES
2085 (qualify_single) but not RES_DNSRCH (search_parents).
2087 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2088 spool_read_header() do less string testing, by means of a preliminary
2089 switch on the second character of optional "-foo" lines. (This is
2090 overdue, caused by the large number of possibilities that now exist.
2091 Originally there were few.) While I was there, I also converted the
2092 str(n)cmp tests so they don't re-test the leading "-" and the first
2093 character, in the hope this might squeeze out yet more improvement.
2095 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2096 flag allowing group syntax was set by the header_syntax check but not
2097 turned off, possible causing trouble later; (2) The flag was not being
2098 set at all for the header_verify test, causing "group"-style headers to
2099 be rejected. I have now set it in this case, and also caused header_
2100 verify to ignore an empty address taken from a group. While doing this, I
2101 came across some other cases where the code for allowing group syntax
2102 while scanning a header line wasn't quite right (mostly, not resetting
2103 the flag correctly in the right place). These bugs could have caused
2104 trouble for malformed header lines. I hope it is now all correct.
2106 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2107 with the "reply" argument non-NULL. The code, however (which originally
2108 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2109 but it didn't always do it. This confused somebody who was copying the
2110 code for some other use. I have removed all the tests.
2112 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2113 feature that was used to support insecure browsers during the U.S. crypto
2114 embargo. It requires special client support, and Exim is probably the
2115 only MTA that supported it -- and would never use it because real RSA is
2116 always available. This code has been removed, because it had the bad
2117 effect of slowing Exim down by computing (never used) parameters for the
2118 RSA_EXPORT functionality.
2120 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2121 authenticator to fail if there's a tab character in the incoming data
2122 (there should never be unless someone is messing about, as it's supposed
2123 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2124 if the connection is using TLS or if the remote IP is the same as the
2125 local IP, and the "valid-client-cert option" if a client certificate has
2128 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2129 authenticators. This can be used for authorization after authentication
2130 succeeds. (In the case of plaintext, it servers for both authentication
2133 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2134 if any retry times were supplied.
2136 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2137 connection, where there is no client IP address to check. In this
2138 situation, the verify now always succeeds.
2140 PH/25 Applied John Jetmore's -Mset patch.
2142 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2144 PH/27 In a string expansion for a processed (not raw) header when multiple
2145 headers of the same name were present, leading whitespace was being
2146 removed from all of them, but trailing whitespace was being removed only
2147 from the last one. Now trailing whitespace is removed from each header
2148 before concatenation. Completely empty headers in a concatenation (as
2149 before) are ignored.
2151 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2152 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2154 PH/29 [Removed. This was a change that I later backed out, and forgot to
2155 correct the ChangeLog entry (that I had efficiently created) before
2156 committing the later change.]
2158 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2159 address errors (4xx response to RCPT) over the same connection as other
2160 messages routed to the same hosts. Such deliveries are always "forced",
2161 so retry times are not inspected. This resulted in far too many retries
2162 for the affected addresses. The effect occurred only when there were more
2163 hosts than the hosts_max_try setting in the smtp transport when it had
2164 the 4xx errors. Those hosts that it had tried were not added to the list
2165 of hosts for which the message was waiting, so if all were tried, there
2166 was no problem. Two fixes have been applied:
2168 (i) If there are any address or message errors in an SMTP delivery, none
2169 of the hosts (tried or untried) are now added to the list of hosts
2170 for which the message is waiting, so the message should not be a
2171 candidate for sending over the same connection that was used for a
2172 successful delivery of some other message. This seems entirely
2173 reasonable: after all the message is NOT "waiting for some host".
2174 This is so "obvious" that I'm not sure why it wasn't done
2175 previously. Hope I haven't missed anything, but it can't do any
2176 harm, as the worst effect is to miss an optimization.
2178 (ii) If, despite (i), such a delivery is accidentally attempted, the
2179 routing retry time is respected, so at least it doesn't keep
2180 hammering the server.
2182 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2183 in ${readsocket because some servers need this prod.
2185 PH/32 Added some extra debug output when updating a wait-xxx database.
2187 PH/33 The hint "could be header name not terminated by colon", which has been
2188 given for certain expansion errors for a long time, was not being given
2189 for the ${if def:h_colon_omitted{... case.
2191 PH/34 The spec says: "With one important exception, whenever a domain list is
2192 being scanned, $domain contains the subject domain." There was at least
2193 one case where this was not true.
2195 PH/35 The error "getsockname() failed: connection reset by peer" was being
2196 written to the panic log as well as the main log, but it isn't really
2197 panic-worthy as it just means the connection died rather early on. I have
2198 removed the panic log writing for the ECONNRESET error when getsockname()
2201 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2202 runs only) independently of the message's sender address. This meant
2203 that, if the 4xx error was in fact related to the sender, a different
2204 message to the same recipient with a different sender could confuse
2205 things. In particualar, this can happen when sending to a greylisting
2206 server, but other circumstances could also provoke similar problems.
2207 I have changed the default so that the retry time for these errors is now
2208 based a combination of the sender and recipient addresses. This change
2209 can be overridden by setting address_retry_include_sender=false in the
2212 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2213 remote server are returned as part of bounce messages. This was not
2214 happening for LMTP over a pipe (the lmtp transport), but now it is the
2215 same for both kinds of LMTP.
2217 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2218 in header lines that were in fact CNAMEs. This is no longer the case.
2220 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2221 and queue runs started by the daemon processed all messages. This has
2222 been fixed so that -R and -S can now usefully be given with -q<time>.
2224 PH/40 Import PCRE release 6.7 (fixes some bugs).
2226 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2228 PH/42 Give an error if -q is specified more than once.
2230 PH/43 Renamed the variables $interface_address and $interface_port as
2231 $received_ip_address and $received_port, to make it clear that these
2232 values apply to message reception, and not to the outgoing interface when
2233 a message is delivered. (The old names remain recognized, of course.)
2235 PH/44 There was no timeout on the connect() call when using a Unix domain
2236 socket in the ${readsocket expansion. There now is.
2238 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2239 be meaningful with "accept".
2241 SC/01 Eximstats V1.43
2242 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2244 SC/02 Eximstats V1.44
2245 Use a glob alias rather than an array ref in the generated
2246 parser. This improves both readability and performance.
2248 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2249 Collect SpamAssassin and rejection statistics.
2250 Don't display local sender or destination tables unless
2251 there is data to show.
2252 Added average volumes into the top table text output.
2254 SC/04 Eximstats V1.46
2255 Collect data on the number of addresses (recipients)
2256 as well as the number of messages.
2258 SC/05 Eximstats V1.47
2259 Added 'Message too big' to the list of mail rejection
2260 reasons (thanks to Marco Gaiarin).
2262 SC/06 Eximstats V1.48
2263 Mainlog lines which have GMT offsets and are too short to
2264 have a flag are now skipped.
2266 SC/07 Eximstats V1.49 (Alain Williams)
2267 Added the -emptyok flag.
2269 SC/08 Eximstats V1.50
2270 Fixes for obtaining the IP address from reject messages.
2272 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2273 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2274 whitesspace changes from 4.64-PH/27
2276 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2279 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2280 are found, allow negative numbers in numeric criteria)
2282 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2284 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2285 to match changes made in 4.64-PH/43
2287 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2289 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2290 "domain" is actually an IP address, and force a failure. This locks out
2291 those revolvers/nameservers that support "A-for-A" lookups, in
2292 contravention of the specifications.
2294 PH/48 When a host name was looked up from an IP address, and the subsequent
2295 forward lookup of the name timed out, the host name was left in
2296 $sender_host_name, contrary to the specification.
2298 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2299 restricted to single-key lookups, Exim was not diagnosing an error if
2300 * or *@ was used with a query-style lookup.
2302 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2304 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2305 long ago, but noone interested enough thought of it. Let's just say that
2306 the "1.1" means that there are some new functions that weren't there at
2307 some point in the past.
2309 PH/51 Error processing for expansion failure of helo_data from an smtp
2310 transport during callout processing was broken.
2312 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2313 tested/used via the -bh/-bhc/-bs options.
2315 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2316 bug, fixed in subsequent PCRE releases).
2318 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2319 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2321 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2327 SC/01 Use a glob alias rather than an array ref in eximstats generated
2328 parser. This improves both readability and performance.
2330 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2331 Don't display local sender or destination tables in eximstats unless
2332 there is data to show.
2333 Added average volumes into the eximstats top table text output.
2335 SC/03 Collect data on the number of addresses (recipients) as well
2336 as the number of messages in eximstats.
2338 TF/01 Correct an error in the documentation for the redirect router. Exim
2339 does (usually) call initgroups() when daemonizing.
2341 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2342 with consistent privilege compared to when running as a daemon.
2344 TF/03 Note in the spec that $authenticated_id is not set for local
2345 submissions from trusted users.
2347 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2348 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2350 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2351 by adding some example configuration directives to the default
2352 configuration file. A little bit of work is required to uncomment the
2353 directives and define how usernames and passwords are checked, but
2354 there is now a framework to start from.
2356 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2357 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2358 without this. I don't know how relevant this is to other LDAP libraries.
2360 PH/02 Add the verb name to the "unknown ACL verb" error.
2362 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2364 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2366 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2367 directory not expanded when it should be if an expanded home directory
2368 was set for the address (which is overridden by the transport).
2370 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2373 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2374 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2375 because it is too late at that time, and has no effect.
2377 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2378 security issue with \' (bugzilla #107). I could not use the
2379 PQescapeStringConn() function, because it needs a PGconn value as one of
2382 PH/08 When testing addresses using -bt, indicate those final addresses that
2383 are duplicates that would not cause an additional delivery. At least one
2384 person was confused, thinking that -bt output corresponded to deliveries.
2385 (Suppressing duplicates isn't a good idea as you lose the information
2386 about possibly different redirections that led to the duplicates.)
2388 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2389 systems where poll() doesn't work, in particular OS X.
2391 PH/10 Added more information to debugging output for retry time not reached.
2393 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2394 operations in malware.c.
2396 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2399 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2400 syslog_duplication set false, log lines that would normally be written
2401 both the the main log and to the reject log were not written to syslog at
2404 PH/14 In the default configuration, change the use of "message" in ACL warn
2405 statements to "add_header".
2407 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2408 not followed by a command (e.g. "seen endif").
2410 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2411 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2414 PH/17 Added extra conditions to the default value of delay_warning_condition
2418 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2419 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2420 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2423 The Auto-Submitted: and various List- headers are standardised, whereas I
2424 don't think Precedence: ever was.
2426 PH/18 Refactored debugging code in route_finduser() to show more information,
2427 in particular, the error code if getpwnam() issues one.
2429 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2430 This is apparently needed in addition to the PH/07 change above to avoid
2431 any possible encoding problems.
2433 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2434 but not after initializing Perl.
2436 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2437 output them only if debugging. By default they are written stderr,
2438 apparently, which is not desirable.
2440 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2443 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2446 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2448 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2449 authenticated or an ident call has been made. Suppress the default
2450 values for $authenticated_id and $authenticated_sender (but permit -oMai
2451 and -oMas) when testing with -bh.
2453 PH/24 Re-jigged the order of the tests in the default configuration so that the
2454 tests for valid domains and recipients precede the DNS black list and CSA
2455 tests, on the grounds that those ones are more expensive.
2457 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2458 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2459 This bug exists in every version of Exim that I still have, right back to
2462 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2463 However, an attempt to turn on case-sensitivity in a regex key by
2464 including (?-i) didn't work because the subject string was already
2465 lowercased, and the effects were non-intuitive. It turns out that a
2466 one-line patch can be used to allow (?-i) to work as expected.
2472 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2473 other effects) broke the use of negated acl sub-conditions.
2475 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2478 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2479 "Deny" causes Exim to reject the incoming connection with a 554 error.
2480 Unfortunately, if there is a major crisis, such as a disk failure,
2481 tcp-wrappers gives "deny", whereas what one would like would be some
2482 kind of temporary error. A kludge has been added to help with this.
2483 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2484 554 error is used if errno is still zero or contains ENOENT (which occurs
2485 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2488 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2490 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2491 errors. Otherwise a message that provokes a temporary error (when other
2492 messages do not) can cause a whole host to time out.
2494 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2495 addresses were routed directly to files or pipes from a redirect router.
2496 File deliveries just didn't batch; pipe deliveries might have suffered
2499 PH/06 A failure to get a lock for a hints database would erroneously always say
2500 "Failed to get write lock", even when it was really a read lock.
2502 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2503 of 0600. This has been changed to use the value of the lockfile_mode
2504 option (which defaults to 0600).
2506 PH/08 Applied small patch from the Sieve maintainer.
2508 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2509 folder from quota calculations, a direct delivery into this folder messed
2510 up the contents of the maildirsize file. This was because the regex was
2511 used only to exclude .Trash (or whatever) when the size of the mailbox
2512 was calculated. There was no check that a delivery was happening into an
2513 excluded directory. This bug has been fixed by ignoring all quota
2514 processing for deliveries into excluded directories.
2516 PH/10 Added the maildirfolder_create_regex option to appendfile.
2522 PH/01 The code for finding all the local interface addresses on a FreeBSD
2523 system running IPv6 was broken. This may well have applied to all BSD
2524 systems, as well as to others that have similar system calls. The broken
2525 code found IPv4 interfaces correctly, but gave incorrect values for the
2526 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2527 that it would not match correctly against @[] and not recognize the IPv6
2530 PH/02 The ipliteral router was not recognizing addresses of the form user@
2531 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2533 PH/03 Added disable_ipv6.
2535 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2536 decoded form, because it is most often used to construct To: headers
2537 lines in autoreplies, and the decoded form may well be syntactically
2538 invalid. However, $reply_address has leading white space removed, and all
2539 newlines turned into spaces so that the autoreply transport does not
2542 PH/05 If group was specified without a user on a router, and no group or user
2543 was specified on a transport, the group from the router was ignored.
2545 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2546 for visible compile-time settings that can be used to change these
2547 numbers, for those that want even more. Backwards compatibility with old
2548 spool files has been maintained. However, going back to a previous Exim
2549 release will lost any variables that are in spool files.
2551 PH/07 Two small changes when running in the test harness: increase delay when
2552 passing a TCP/IP connection to a new process, in case the original
2553 process has to generate a bounce, and remove special handling of
2554 127.0.0.2 (sic), which is no longer necessary.
2556 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2557 be the same on different OS.
2559 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2562 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2563 whether --show-vars was specified or not
2565 JJ/02 exipick: Added support for new ACL variable spool format introduced
2568 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2569 syntactically invalid From: or Reply-to: line, and a filter used this to
2570 generate an autoreply, and therefore failed to obtain an address for the
2571 autoreply, Exim could try to deliver to a non-existent relative file
2572 name, causing unrelated and misleading errors. What now happens is that
2573 it logs this as a hard delivery error, but does not attempt to create a
2576 PH/11 The exinext utility has a -C option for testing purposes, but although
2577 the given file was scanned by exinext itself; it wasn't being passed on
2578 when Exim was called.
2580 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2581 an end-of-file indication when reading a command response.
2583 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2584 compiled. In many other places in Exim, IPv6 addresses are always
2585 recognized, so I have changed this. It also means that IPv4 domain
2586 literals of the form [IPV4:n.n.n.n] are now always recognized.
2588 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2589 used if the router is not running as root, for example, when verifying at
2590 ACL time, or when using -bh. The debugging output from this situation was
2591 non-existent - all you got was a failure to exec. I have made two
2594 (a) Failures to set uid/gid, the current directory, or a process leader
2595 in a subprocess such as that created by queryprogram now generate
2596 suitable debugging ouput when -d is set.
2598 (b) The queryprogram router detects when it is not running as root,
2599 outputs suitable debugging information if -d is set, and then runs
2600 the subprocess without attempting to change uid/gid.
2602 PH/15 Minor change to Makefile for building test_host (undocumented testing
2605 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2606 additional section of a DNS packet that returns MX or SRV records.
2607 Instead, it always explicitly searches for A/AAAA records. This avoids
2608 major problems that occur when a DNS server includes only records of one
2609 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2610 fixed another bug: if SRV records were looked up and the corresponding
2611 address records were *not* found in the additional section, the port
2612 values from the SRV records were lost.
2614 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2615 using the correct key (the original address) when searching the retry
2616 rules in order to find which one to use for generating the retry hint.
2618 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2619 adding the default one. Similarly, if it contains a Reply-To: header, the
2620 errors_reply_to option, if set, is not used.
2622 PH/19 When calculating a retry time, Exim used to measure the "time since
2623 failure" by looking at the "first failed" field in the retry record. Now
2624 it does not use this if it is later than than the arrival time of the
2625 message. Instead it uses the arrival time. This makes for better
2626 behaviour in cases where some deliveries succeed, thus re-setting the
2627 "first failed" field. An example is a quota failure for a huge message
2628 when small messages continue to be delivered. Without this change, the
2629 "time since failure" will always be short, possible causing more frequent
2630 delivery attempts for the huge message than are intended.
2631 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2633 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2634 $1, $2, $3) because the numerical variables can be reset during some
2635 expansion items (e.g. "match"), thereby losing the authentication data.
2637 PH/21 Make -bV show the size of off_t variables so that the test suite can
2638 decide whether to run tests for quotas > 2G.
2640 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2641 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2642 filecount value is greater than 2G or if a quota value is greater than 2G
2643 on a system where the size of off_t is not greater than 4, a panic error
2646 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2647 never match. The debug and -bh output now contains an explicit error
2648 message indicating a malformed IPv4 address or mask.
2650 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2651 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2652 PH/23 above applies.
2654 PH/25 Do not write to syslog when running in the test harness. The only
2655 occasion when this arises is a failure to open the main or panic logs
2656 (for which there is an explicit test).
2658 PH/26 Added the /no_tell option to "control=freeze".
2660 PH/27 If a host name lookup failed very early in a connection, for example, if
2661 the IP address matched host_lookup and the reverse lookup yielded a name
2662 that did not have a forward lookup, an error message of the form "no IP
2663 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2664 could be logged. Now it outputs the IP address instead of "NULL".
2666 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2667 allows the sender and the authenticated sender to be set when
2668 submitting a message from within Exim. Since child_open_exim() is
2669 documented for local_scan(), the new function should be too.
2671 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2672 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2673 results in an empty string is now treated as unset.
2675 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2677 PH/31 Added sender_verify_fail logging option.
2679 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2680 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2681 tidied the source and removed it altogether.
2683 PH/33 When a queue run was abandoned because the load average was too high, a
2684 log line was always written; now it is written only if the queue_run log
2685 selector is set. In addition, the log line for abandonment now contains
2686 information about the queue run such as the pid. This is always present
2687 in "start" and "stop" lines but was omitted from the "abandon" line.
2689 PH/34 Omit spaces between a header name and the colon in the error message that
2690 is given when verify = headers_syntax fails (if there are lots of them,
2691 the message gets confusing).
2693 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2694 names, as there are now some PTR records that contain slashes. This check
2695 is only to protect against broken name servers that fall over on strange
2696 characters, so the fact that it applies to all lookups doesn't matter.
2698 PH/36 Now that the new test suite is complete, we can remove some of the
2699 special code in Exim that was needed for the old test suite. For example,
2700 sorting DNS records because real resolvers return them in an arbitrary
2701 order. The new test suite's fake resolver always returns records in the
2704 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2705 bounces) except when queue_only is set, to avoid logging races between
2706 the different processes.
2708 PH/38 Panic-die if .include specifies a non-absolute path.
2710 PH/39 A tweak to the "H" retry rule from its user.
2712 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2713 a label. They prevented compilation on older perls.
2715 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2716 a warning to be raised on newish perls.
2718 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2719 on queue. Changes to match documented behaviour of showing count of
2720 messages matching specified criteria.
2722 PH/40 Changed the default ident timeout from 30s to 5s.
2724 PH/41 Added support for the use of login_cap features, on those BSD systems
2725 that have them, for controlling the resources used by pipe deliveries.
2727 PH/42 The content-scanning code uses fopen() to create files in which to put
2728 message data. Previously it was not paying any attention to the mode of
2729 the files. Exim runs with umask(0) because the rest of the code creates
2730 files with open(), and sets the required mode explicitly. Thus, these
2731 files were ending up world-writeable. This was not a big issue, because,
2732 being within the spool directory, they were not world-accessible. I have
2733 created a function called modefopen, which takes an additional mode
2734 argument. It sets umask(777), creates the file, chmods it to the required
2735 mode, then resets the umask. All the relevant calls to fopen() in the
2736 content scanning code have been changed to use this function.
2738 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2739 to 24 hours. This avoids potential overflow problems when processing G
2740 and H retry rules. I suspect nobody ever tinkers with this value.
2742 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2744 PH/45 When the plaintext authenticator is running as a client, the server's
2745 challenges are checked to ensure they are valid base64 strings. By
2746 default, the authentication attempt is cancelled if an invalid string is
2747 received. Setting client_ignore_invalid_base64 true ignores these errors.
2748 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2749 they are received. Thus, the responses can be made to depend on the
2750 challenges. If an invalid string is ignored, an empty string is placed in
2753 PH/46 Messages that are created by the autoreply transport now contains a
2754 References: header, in accordance with RFCs 2822 and 3834.
2756 PH/47 Added authenticated_sender_force to the smtp transport.
2758 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2760 PH/49 Installed latest patch from the Sieve maintainer.
2762 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2763 was also set, the appendfile transport was unnecessarily scanning a
2764 directory of message files (e.g. for maildir delivery) to find the count
2765 of files (along with the size), even though it did not need this
2766 information. It now does the scan only if it needs to find either the
2767 size of the count of files.
2769 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2771 PH/52 Two bugs concerned with error handling when the smtp transport is
2774 (i) Exim was not creating retry information for temporary errors given
2775 for individual recipients after the DATA command when the smtp transport
2776 was used in LMTP mode. This meant that they could be retried too
2777 frequently, and not timed out correctly.
2779 (ii) Exim was setting the flag that allows error details to be returned
2780 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2781 recipients that were returned after the DATA command.
2783 PH/53 This is related to PH/52, but is more general: for any failing address,
2784 when detailed error information was permitted to be returned to the
2785 sender, but the error was temporary, then after the final timeout, only
2786 "retry timeout exceeded" was returned. Now it returns the full error as
2787 well as "retry timeout exceeded".
2789 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2790 do this, and (what is worse) MTAs that accept it.
2792 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2793 will now be deprecated.
2795 PH/56 New os.c-cygwin from the Cygwin maintainer.
2797 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2798 formats (previously only available in exim formats via -bpr, -bpru,
2799 and -bpra. Now also available in native and exiqgrep formats)
2801 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2802 with very large, slow to parse queues
2804 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2806 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2808 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2809 responses to authentication challenges, though it was showing the
2810 challenges; (ii) I've removed the CR characters from the debug output for
2813 PH/58 Allow for the insertion of a newline as well as a space when a string
2814 is turned into more than one encoded-word during RFC 2047 encoding. The
2815 Sieve code now uses this.
2817 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2818 data_4xx, lost_connection, tls_required.
2820 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2821 message was being sent as an SMTP response.
2823 PH/61 Add -l and -k options to exicyclog.
2825 PH/62 When verifying, if an address was redirected to one new address, so that
2826 verification continued, and the new address failed or deferred after
2827 having set something in $address_data, the value of $address_data was not
2828 passed back to the ACL. This was different to the case when no
2829 redirection occurred. The value is now passed back in both cases.
2831 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2832 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2833 use login_cap.h, so on its own it isn't the distinguishing feature. The
2834 new name refers directly to the setclassresources() function.
2836 PH/65 Added configuration files for NetBSD3.
2838 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2840 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2841 is preferred over IPv4.
2843 PH/68 The bounce_return_message and bounce_return_body options were not being
2844 honoured for bounces generated during the reception of non-SMTP messages.
2845 In particular, this applied to messages rejected by the ACL. This bug has
2846 been fixed. However, if bounce_return_message is true and bounce_return_
2847 body is false, the headers that are returned for a non-SMTP message
2848 include only those that have been read before the error was detected.
2849 (In the case of an ACL rejection, they have all been read.)
2851 PH/69 The HTML version of the specification is now built in a directory called
2852 spec_html instead of spec.html, because the latter looks like a path with
2853 a MIME-type, and this confuses some software.
2855 PH/70 Catch two compiler warnings in sieve.c.
2857 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2858 function verify_get_ident() calls ip_connect() to connect a socket, but
2859 if the "connect()" function timed out, ip_connect() used to close the
2860 socket. However, verify_get_ident() also closes the socket later, and in
2861 between Exim writes to the log, which may get opened at this point. When
2862 the socket was closed in ip_connect(), the log could get the same file
2863 descriptor number as the socket. This naturally causes chaos. The fix is
2864 not to close the socket in ip_connect(); the socket should be closed by
2865 the function that creates it. There was only one place in the code where
2866 this was missing, in the iplookup router, which I don't think anybody now
2867 uses, but I've fixed it anyway.
2869 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2870 well as to direct DNS lookups. Otherwise the handling of names in host
2871 lists is inconsistent and therefore confusing.
2877 PH/01 Two changes to the default runtime configuration:
2879 (1) Move the checks for relay_from_hosts and authenticated clients from
2880 after to before the (commented out) DNS black list checks.
2882 (2) Add control=submission to the relay_from_hosts and authenticated
2883 clients checks, on the grounds that messages accepted by these
2884 statements are most likely to be submissions.
2886 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2888 (1) Generate an error if the third argument for the ${prvs expansion is
2891 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2894 (3) Reset the variables that are obtained from the first argument of
2895 ${prvscheck and used in the second argument before leaving the code,
2896 because their memory is reclaimed, so using them afterwards may do
2899 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2900 one (it's much easier than Tom thought :-).
2902 (5) Because of (4), we can now allow for the use of $prvscheck_result
2903 inside the third argument.
2905 PH/03 For some reason, the default setting of PATH when running a command from
2906 a pipe transport was just "/usr/bin". I have changed it to
2909 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2910 anything to be listed in the output from -bV.
2912 PH/05 When a filter generated an autoreply, the entire To: header line was
2913 quoted in the delivery log line, like this:
2915 => >A.N.Other <ano@some.domain> <original@ddress> ...
2917 This has been changed so that it extracts the operative address. There
2918 may be more than one such address. If so, they are comma-separated, like
2921 => >ano@some.domain,ona@other.domain <original@ddress> ...
2923 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2924 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2925 not being looked up in the rDNS to get a host name, Exim was showing the
2926 IP address twice in Received: lines, even though the IP addresses were
2927 identical. For example:
2929 Received: from [1.2.3.4] (helo=[1.2.3.4])
2931 However, if the real host name was known, it was omitting the HELO data
2932 if it matched the actual IP address. This has been tidied up so that it
2933 doesn't show the same IP address twice.
2935 PH/07 When both +timestamp and +memory debugging was on, the value given by
2936 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2937 called by the debug printing, thereby overwriting the timestamp buffer.
2938 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2940 PH/08 When the original message was included in an autoreply transport, it
2941 always said "this is a copy of the message, including all the headers",
2942 even if body_only or headers_only was set. It now gives an appropriate
2945 PH/09 Applied a patch from the Sieve maintainer which:
2947 o fixes some comments
2948 o adds the (disabled) notify extension core
2949 o adds some debug output for the result of if/elsif tests
2950 o points to the current vacation draft in the documentation
2951 and documents the missing references header update
2955 o fixes a bug in processing the envelope test (when testing
2956 multiple envelope elements, the last element determinted the
2959 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2960 Electronic Mail") by including:
2962 Auto-submitted: auto-generated
2964 in the messages that it generates (bounce messages and others, such as
2965 warnings). In the case of bounce messages for non-SMTP mesages, there was
2966 also a typo: it was using "Auto_submitted" (underscore instead of
2967 hyphen). Since every message generated by Exim is necessarily in response
2968 to another message, thes have all been changed to:
2970 Auto-Submitted: auto-replied
2972 in accordance with these statements in the RFC:
2974 The auto-replied keyword:
2976 - SHOULD be used on messages sent in direct response to another
2977 message by an automatic process,
2979 - MUST NOT be used on manually-generated messages,
2981 - MAY be used on Delivery Status Notifications (DSNs) and Message
2982 Disposition Notifications (MDNs),
2984 - MUST NOT be used on messages generated by automatic or periodic
2985 processes, except for messages which are automatic responses to
2988 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2989 to the default Received: header definition.
2991 PH/12 Added log selector acl_warn_skipped (default on).
2993 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2994 variables because (a) they are in the wrong storage pool and (b) even if
2995 they were copied, it wouldn't work properly because of the caching.
2997 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2998 checking when decoding. Apparently there are clients that generate
2999 overlong encoded strings. Why am I not surprised?
3001 PH/15 If the first argument of "${if match_address" was not empty, but did not
3002 contain an "@" character, Exim crashed. Now it writes a panic log message
3003 and treats the condition as false.
3005 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3007 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3008 extension "envelope-auth". The code is finished and in agreement with
3009 other implementations, but there is no documentation so far and in fact,
3010 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3011 not changing the active code.
3013 Print executed "if" and "elsif" statements when debugging is used. This
3014 helps a great deal to understand what a filter does.
3016 Document more things not specified clearly in RFC3028. I had all this
3017 sorted out, when out of a sudden new issues came to my mind. Oops."
3019 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3022 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3023 canonical form (as documented). However, after a host name lookup from
3024 the IP address, check_host() was doing a simple string comparison with
3025 addresses acquired from the DNS when checking that the found name did
3026 have the original IP as one of its addresses. Since any found IPv6
3027 addresses are likely to be in abbreviated form, the comparison could
3028 fail. Luckily, there already exists a function for doing the comparison
3029 by converting both addresses to binary, so now that is used instead of
3030 the text comparison.
3032 PH/20 There was another similar case to PH/19, when a complete host name was
3033 given in a host list; looking up its IP address could give an abbreviated
3034 form, whereas the current host's name might or might not be abbreviated.
3035 The same fix has been applied.
3041 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3042 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3045 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3046 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3048 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3050 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3051 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3052 Also, for "Auto-Submitted", treat anything other than "no" as
3053 non-personal, in accordance with RFC 3834. (Previously it treated
3054 anything starting "auto-" as non-personal.)
3056 TF/01 The control=submission/name=... option had a problem with syntax
3057 errors if the name included a slash character. The /name= option
3058 now slurps the rest of the string, so it can include any characters
3059 but it must come last in the list of options (after /sender_retain
3062 PH/05 Some modifications to the interface to the fake nameserver for the new
3070 TK/01 Added the "success_on_redirect" address verification option. See
3071 NewStuff for rationale and an example.
3073 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3075 PH/02 Patch to exigrep to allow it to work on syslog lines.
3077 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3078 fread() to skip over the body file's header line, because in Cygwin the
3079 header line is locked and is inaccessible.
3081 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3082 co-exist for some time) to make it clear that it is the Exim ID that is
3083 referenced, not the Message-ID: header line.
3085 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3086 string_format() function, because snprintf() does not exist on all
3089 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3090 ${stat: expansion item.
3092 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3093 protocol synchronization error", to keep the pedants happy.
3095 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3096 well as for IRIX systems, when gcc is being used. See the host.c source
3099 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3101 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3104 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3105 transport and to the smtp transport in LMTP mode.
3107 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3109 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3110 run a filter in a subprocess. This could lead to confusion in subsequent
3111 lookups in the parent process. There should also be a search_tidyup() at
3112 the end of the subprocess.
3114 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3115 only if the host matched helo_try_verify_hosts, which caused the
3116 verification to occur when the EHLO/HELO command was issued. The ACL just
3117 tested the remembered result. Now, if a previous verification attempt has
3118 not happened, "verify = helo" does it there and then.
3120 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3122 TK/03 Fix log output including CR from clamd.
3124 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3125 exist provoked a memory error which could cause a segfault.
3127 PH/15 Installed PCRE 6.2
3129 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3131 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3132 of the problem. Specifically, suggested +O2 rather than +O1 for the
3135 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3137 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3138 record, the port was not correctly logged when the outgoing_port log
3139 selector was set (it logged the transort's default port).
3141 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3142 fallback_hosts, and "hosts" in the smtp transport.
3144 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3145 host errors such as "Connection refused".
3147 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3148 authentication with radiusclient 0.4.9:
3150 - Error returned from rc_read_config was caught wrongly
3151 - Username/password not passed on to radius server due to wrong length.
3153 The presumption is that some radiusclient API changes for 4.51/PH/17
3154 were not taken care of correctly. The code is still untested by me (my
3155 Linux distribution still has 0.3.2 of radiusclient), but it was
3156 contributed by a Radius user.
3158 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3159 expanding the "port" option of the smtp transport.
3161 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3162 while reading a MIME header. Thanks to Tom Hughes for a patch.
3164 PH/24 Include config.h inside local_scan.h so that configuration settings are
3167 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3168 that in an ACL for RCPT (for example), you can examine exactly what was
3171 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3172 commands, but it was not correctly comparing the address with the actual
3173 client host address. Thus, it would show the EHLO address in Received:
3174 header lines when this was not necessary.
3176 PH/27 Added the % operator to ${eval:}.
3178 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3179 it should be ignoring failures (because with -C, for example, it has lost
3180 privilege). It wasn't ignoring creation failures other than "already
3183 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3186 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3187 because an input line was too long, either on its own, or by virtue of
3188 too many continuations, the temporary file was not being removed, and the
3189 return code was incorrect.
3191 PH/31 Missing "BOOL" in function definition in filtertest.c.
3193 PH/32 Applied Sieve patches from the maintainer.
3195 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3197 PH/33 Added "verify = not_blind".
3199 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3200 Local/Makefile (with some defaults set). These are used in built scripts
3201 such as exicyclog, but they have never been used in the exim_install
3202 script (though there are many overriding facilities there). I have
3203 arranged that the exim_install script now takes note of these two
3206 PH/35 Installed configuration files for Dragonfly.
3208 PH/36 When a locally submitted message by a trusted user did not contain a
3209 From: header, and the sender address was obtained from -f or from an SMTP
3210 MAIL command, and the trusted user did not use -F to supply a sender
3211 name, $originator_name was incorrectly used when constructing a From:
3212 header. Furthermore, $originator_name was used for submission mode
3213 messages from external hosts without From: headers in a similar way,
3214 which is clearly wrong.
3216 PH/37 Added control=suppress_local_fixups.
3218 PH/38 When log_selector = +received_sender was set, and the addition of the
3219 sender made the log line's construction buffer exactly full, or one byte
3220 less than full, an overflow happened when the terminating "\n" was
3223 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3224 when the result of a list match is failure because a DNS lookup failed.
3226 PH/40 RM_COMMAND is now used in the building process.
3228 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3229 the "build-* directories that it finds.
3231 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3232 address in a domain literal was a prefix of an interface address.
3234 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3235 when verifying a sender address, unless rewrite_headers is false.
3237 PH/44 Wrote a long comment about why errors_to addresses are verified as
3238 recipients, not senders.
3240 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3241 the ratelimit ACL was added.
3243 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3245 PH/46 Added extra information about PostgreSQL errors to the error string.
3247 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3248 avoiding the need to install special zones in a real server. This is
3249 backwards compatible; if it can't find the fake resolver, it drops back.
3250 Thus, both old and new test suites can be run.
3252 TF/02 Added util/ratelimit.pl
3254 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3255 clock is set back in time.
3257 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3258 Candler <B.Candler@pobox.com>.
3260 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3261 OK for addresses that are the result of redirections.
3263 PH/48 A number of further additions for the benefit of the new test suite,
3264 including a fake gethostbyname() that interfaces to the fake DNS resolver
3267 TF/06 The fix for widen_domains has also been applied to qualify_single and
3268 search_parents which are the other dnslookup options that can cause
3271 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3274 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3276 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3277 service" if it encountered an MX record with an empty target hostname.
3278 The message is now "an MX or SRV record indicated no SMTP service".
3280 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3281 if the DNS of the sending site is misconfigured. This is quite a
3282 common situation. This change restores the behaviour of treating a
3283 helo verification defer as a failure.
3285 PH/51 If self=fail was set on a router, the bounce message did not include the
3286 actual error message.
3292 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3294 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3295 command itself ended in error, the underlying message about the transport
3296 filter timeout was being overwritten with the pipe command error. Now the
3297 underlying error message should be appended to the second error message.
3299 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3301 PH/02 Reduce the amount of output that "make" produces by default. Full output
3302 can still be requested.
3304 PH/03 The warning log line about a condition test deferring for a "warn" verb
3305 was being output only once per connection, rather than after each
3306 occurrence (because it was using the same function as for successful
3307 "warn" verbs). This seems wrong, so I have changed it.
3309 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3310 it should not have, which might have caused a crash in the right
3311 circumstances, but probably never did.
3313 PH/04 Installed a modified version of Tony Finch's patch to make submission
3314 mode fix the return path as well as the Sender: header line, and to
3315 add a /name= option so that you can make the user's friendly name appear
3318 TF/03 Added the control = fakedefer ACL modifier.
3320 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3321 Mark Lowes for thorough testing.
3323 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3325 TK/03 Merged latest SRS patch from Miles Wilton.
3327 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3328 with the definition in sysexits.h (which is #included earlier).
3329 Fortunately, Exim does not actually use EX_OK. The code used to try to
3330 preserve the sysexits.h value, by assumimg that macro definitions were
3331 scanned for macro replacements. I have been disabused of this notion,
3332 so now the code just undefines EX_OK before #including unistd.h.
3334 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3335 in the smtp transport. When a block could not be written in a single
3336 write() function, the timeout was being re-applied to each part-write.
3337 This seems wrong - if the receiver was accepting one byte at a time it
3338 would take for ever. The timeout is now adjusted when this happens. It
3339 doesn't have to be particularly precise.
3341 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3342 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3344 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3345 without a domain if the check to <postmaster@domain> fails.
3347 SC/01 Eximstats: added -xls and the ability to specify output files
3348 (patch written by Frank Heydlauf).
3350 SC/02 Eximstats: use FileHandles for outputing results.
3352 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3354 SC/04 Eximstats: fixed display of large numbers with -nvr option
3356 SC/05 Eximstats: fixed merging of reports with empty tables.
3358 SC/06 Eximstats: added the -include_original_destination flag
3360 SC/07 Eximstats: removed tabs and trailing whitespace.
3362 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3364 TK/06 MBOX spool code: Add real "From " MBOX separator line
3365 so the .eml file is really in mbox format (even though
3366 most programs do not really care). Patch from Alex Miller.
3368 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3369 The latter is generated from $received_to and is only set if the
3370 message has one envelope recipient. SA can use these headers,
3371 obviously out-of-the-box. Patch from Alex Miller.
3373 PH/08 The ${def test on a variable was returning false if the variable's
3374 value was "0", contrary to what the specification has always said!
3375 The result should be true unless the variable is empty.
3377 PH/09 The syntax error of a character other than { following "${if
3378 def:variable_name" (after optional whitespace) was not being diagnosed.
3379 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3380 accidental colon was present, for example, could give incorrect results.
3382 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3383 result is used (not including appendfile, where other changes are about
3386 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3387 This involved changing a lot of size variables from int to off_t. It
3388 should work with maildirs and everything.
3390 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3391 spamd dying while we are connected to it.
3393 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3396 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3397 function for BDB 4.3.
3399 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3401 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3402 the sources for PCRE have been re-arranged and more files are now
3405 PH/15 The code I had for printing potentially long long variables in PH/11
3406 above was not the best (it lost precision). The length of off_t variables
3407 is now inspected at build time, and an appropriate printing format (%ld
3408 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3409 to be "long long int" or "long int". This is needed for the internal
3410 formatting function string_vformat().
3412 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3413 the configuration file to be ":syslog", then the script "guesses" where
3414 the logs files are, rather than using the compiled in default. In our
3415 case the guess is not the same as the compiled default, so the script
3416 suddenly stopped working when I started to use syslog. The patch checks
3417 to see if log_file_path is "". If so, it attempts to read it from exim
3418 with no configuration file to get the compiled in version, before it
3419 falls back to the previous guessing code."
3421 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3422 implementing BATV in an Exim configuration. See NewStuff for the gory
3425 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3426 Makefile that are specific to HP-UX.
3428 PH/18 If the "use_postmaster" option was set for a recipient callout together
3429 with the "random" option, the postmaster address was used as the MAIL
3430 FROM address for the random test, but not for the subsequent recipient
3431 test. It is now used for both.
3433 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3434 patch removes a few documentation additions to RFC 3028, because the
3435 latest draft now contains them. It adds the new en;ascii-case comparator
3436 and a new error check for 8bit text in MIME parts. Comparator and
3437 require names are now matched exactly. I enabled the subaddress
3438 extension, but it is not well tested yet (read: it works for me)."
3440 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3441 rework some of the code of TK/09 above to avoid the hardwired use of
3442 "%lld" and "long long". Replaced the call to snprintf() with a call to
3445 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3446 records point to non-existent hosts", "retry timeout exceeded", and
3447 "retry time not reached for any host after a long failure period".
3449 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3450 experimental DomainKeys support:
3452 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3453 (2) On an error such as an illegally used "control", the wrong name for
3454 the control was given.
3456 These problems did NOT occur unless DomainKeys support was compiled.
3458 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3460 PH/24 Added ${if match_ip condition.
3462 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3463 databases so that it will be absolutely obvious if a crash occurs in the
3464 DB library. This is a regular occurrence (often caused by mis-matched
3467 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3468 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3469 picked up on a user's system that detects such things. There doesn't seem
3470 to be a gcc warning option for this - only an attribute that has to be
3471 put on the function's prototype. It seems that in Fedora Core 4 they have
3472 set this on a number of new functions. No doubt there will be more in due
3475 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3476 specify a transport. However, if an address that was verified by such a
3477 router was the subject of a callout, Exim crashed because it tried to
3478 read the rcpt_include_affixes from the non-existent transport. Now it
3479 just assumes that the setting of that option is false. This bug was
3480 introduced by 4.51/PH/31.
3482 PH/28 Changed -d+all to exclude +memory, because that information is very
3483 rarely of interest, but it makes the output a lot bigger. People tend to
3484 do -d+all out of habit.
3486 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3487 code in os-type was giving problems when libc.so lives in lib64, like on
3490 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3491 aren't the modern standard, and it seems that some systems' include files
3492 don't always have them. Exim was already checking for some of the newer
3493 ones like T_AAAA, and defining it itself. I've added checks for all the
3494 record types that Exim uses.
3496 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3497 not automatically generating a new one, as it is supposed to. This
3498 prevented TLS from working. If the file did exist, but contained invalid
3499 data, a new version was generated, as expected. It was only the case of a
3500 non-existent file that was broken.
3502 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3503 with a change in libdomainkeys > 0.64.
3505 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3506 from DNS. If the selector record carries the flag, it now has
3507 precedence over the domain-wide flag.
3509 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3511 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3512 the use of a port name that isn't defined in /etc/services) occurred, the
3513 message was deferred as in a normal delivery, and thus remained on the
3514 spool, instead of being failed because of the mua_wrapper setting. This
3515 is now fixed, and I tidied up some of the mua_wrapper messages at the
3518 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3519 the messages in a hash of arrays rather than using individual hashes.
3520 This is a bit cleaner and results in dramatic memory savings, albeit
3521 at a slight CPU cost.
3523 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3524 as requested by Marc Sherman.
3526 SC/10 Eximstats: added histograms for user specified patterns as requested
3529 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3531 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3532 fopen() in the content-scanning modules that did not already have it.
3538 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3539 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3541 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3543 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3545 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3546 to test Sieve filters that use "vacation".
3548 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3549 that changes the way the GnuTLS parameters are stored in the cache file.
3550 The new format can be generated externally. For backward compatibility,
3551 if the data in the cache doesn't make sense, Exim assumes it has read an
3552 old-format file, and it generates new data and writes a new file. This
3553 means that you can't go back to an older release without removing the
3556 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3557 work if there are any delivery delays because "one_time" forces the
3558 parent to be marked "delivered", so its unseen clone is never tried
3559 again. For this reason, Exim now forbids the simultaneous setting of
3562 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3563 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3564 entry if you want to know the details. The fix, however, neglected to
3565 consider the case when local delivery batching is involved. The test for
3566 "previously delivered" was not happening when checking to see if an
3567 address could be batched with a previous (undelivered) one; under
3568 certain circumstances this could lead to multiple deliveries to the same
3571 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3572 in its include files, and this causes problems building Exim.
3574 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3575 header_syntax) but Exim was just ignoring anything given after a slash.
3576 In particular, this caused confusion with an attempt to use "verify =
3577 reverse_host_lookup/defer_ok". An error is now given when options are
3578 supplied for verify items that do not have them. (Maybe reverse_host_
3579 lookup should have a defer_ok option, but that's a different point.)
3581 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3582 defined by RFC 821) to 2048, because there were problems with some AUTH
3583 commands, and RFC 1869 says the size should be increased for extended
3584 SMTP commands that take arguments.
3586 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3589 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3590 "unknown" error; now it says that the functionality isn't in the binary.
3592 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3593 an address' error message when a string expansion fails (syntax or
3594 whatever). Otherwise the password may appear in the log. Following change
3595 PH/42 below, there is no longer a chance of it appearing in a bounce
3598 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3600 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3601 of its addresses was ever tried. (Bugzilla bug #2.)
3603 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3604 the result incorrectly in the debug output. (It correctly added a newline
3605 to what was transported.)
3607 TF/01 Added $received_time.
3609 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3610 commented out examples of how to interface to a virus scanner and to
3611 SpamAssassin. Also added commented examples of av_scanner and
3612 spamd_address settings.
3614 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3615 and controls are allowed in which ACLs. There were a couple of minor
3616 errors. Some of the entries in the conditions table (which is a table of
3617 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3618 negation of where the condition IS allowed.
3620 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3622 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3623 header file does not have a version number, so I've had to invent a new
3624 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3625 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3626 radiusclient), but it was contributed by a Radius user.
3628 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3629 files or directories, for OpenSSL.
3631 PH/19 When an Exim process that is running as root has to create an Exim log
3632 file, it does so in a subprocess that runs as exim:exim so as to get the
3633 ownership right at creation (otherwise, other Exim processes might see
3634 the file with the wrong ownership). There was no test for failure of this
3635 fork() call, which would lead to the process getting stuck as it waited
3636 for a non-existent subprocess. Forks do occasionally fail when resources
3637 run out. I reviewed all the other calls to fork(); they all seem to check
3640 PH/20 When checking for unexpected SMTP input at connect time (before writing
3641 the banner), Exim was not dealing correctly with a non-positive return
3642 from the read() function. If the client had disconnected by this time,
3643 the result was a log entry for a synchronization error with an empty
3644 string after "input=" when read() returned zero. If read() returned -1
3645 (an event I could not check), uninitialized data bytes were printed.
3646 There were reports of junk text (parts of files, etc) appearing after
3649 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3651 PH/22 Added support for macro redefinition, and (re)definition in between
3652 driver and ACL definitions.
3654 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3655 forgetting to use the resulting value; it was using the unexpanded value.
3657 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3658 hadn't been configured. The fix is from Juergen Kreileder, who
3659 understands it better than I do:
3661 "Here's what I see happening with three configured cyrus_sasl
3662 authenticators configured (plain, login, cram-md5):
3664 On startup auth_cyrus_sasl_init() gets called for each of these.
3665 This means three calls to sasl_listmech() without a specified mech_list.
3666 => SASL tests which mechs of all available mechs actually work
3667 => three warnings about OTP not working
3668 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3670 With the patch, sasl_listmech() also gets called three times. But now
3671 SASL's mech_list option is set to the server_mech specified in the the
3672 authenticator. Or in other words, the answer from sasl_listmech()
3673 gets limited to just the mech you're testing for (which is different
3675 => the return list contains just 'plain' or 'login', 'cram-md5' or
3676 nothing depending on the value of ob->server_mech.
3678 I've just tested the patch: Authentication still works fine,
3679 unavailable mechs specified in the exim configuration are still
3680 caught, and the auth.log warnings about OTP are gone."
3682 PH/25 When debugging is enabled, the contents of the command line are added
3683 to the debugging output, even when log_selector=+arguments is not
3686 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3687 answer is "GNU", and only if the return is "GNU/something" is the answer
3690 PH/27 $acl_verify_message is now set immediately after the failure of a
3691 verification in an ACL, and so is available in subsequent modifiers. In
3692 particular, the message can be preserved by coding like this:
3694 warn !verify = sender
3695 set acl_m0 = $acl_verify_message
3697 Previously, $acl_verify_message was set only while expanding "message"
3698 and "log_message" when a very denied access.
3700 PH/28 Modified OS/os.c-Linux with
3702 -#ifndef OS_LOAD_AVERAGE
3703 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3705 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3706 nomenclature these days.)
3708 PH/29 Installed patch from the Sieve maintainer that adds the options
3709 sieve_useraddress and sieve_subaddress to the redirect router.
3711 PH/30 In these circumstances:
3712 . Two addresses routed to the same list of hosts;
3713 . First host does not offer TLS;
3714 . First host accepts first address;
3715 . First host gives temporary error to second address;
3716 . Second host offers TLS and a TLS session is established;
3717 . Second host accepts second address.
3718 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3719 and peerdn, if requested) that were in fact used only for the second
3722 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3723 attention to any local part prefix or suffix that was matched by the
3724 router that accepted the address. It now behaves in the same way as it
3725 does for delivery: the affixes are removed from the local part unless
3726 rcpt_include_affixes is set on the transport.
3728 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3729 timeout during the DATA phase of an incoming message.
3731 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3732 applied a patch sanctioned by the Sieve maintainer.
3734 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3735 the uid or gid is negative. A case of a negative gid caused this to be
3736 noticed. The fix allows for either to be negative.
3738 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3739 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3742 PH/36 The change PH/12 above was broken. Fixed it.
3744 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3745 the grounds that routing the same address twice would always produce the
3746 same answer. This might have been true once, but it is certainly no
3747 longer true now. Routing a child address may depend on the previous
3748 routing that produced that child. Some complicated redirection strategies
3749 went wrong when messages had multiple recipients, and made Exim's
3750 behaviour dependent on the order in which the addresses were given.
3752 I have moved the duplicate checking until after the routing is complete.
3753 Exim scans the addresses that are assigned to local and remote
3754 transports, and removes any duplicates. This means that more work will be
3755 done, as duplicates will always all be routed, but duplicates are
3756 presumably rare, so I don't expect this is of any significance.
3758 For deliveries to pipes, files, and autoreplies, the duplicate checking
3759 still happens during the routing process, since they are not going to be
3762 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3763 It corrects a timeout issue with spamd. This is Ian's comment: "The
3764 background is that sometimes spamd either never reads data from a
3765 connection it has accepted, or it never writes response data. The exiscan
3766 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3767 blindly assumes that writes won't block so it may never time out."
3769 PH/39 Allow G after quota size as well as K and M.
3771 PH/40 The value set for $authenticated_id in an authenticator may not contain
3772 binary zeroes or newlines because the value is written to log lines and
3773 to spool files. There was no check on this. Now the value is run through
3774 the string_printing() function so that such characters are converted to
3775 printable escape sequences.
3777 PH/41 $message_linecount is a new variable that contains the total number of
3778 lines in the message. Compare $body_linecount, which is the count for the
3781 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3782 bounce and delay warning messages, except in certain special cases, which
3785 (a) An SMTP error message from a remote host;
3786 (b) A message specified in a :fail: redirection;
3787 (c) A message specified in a "fail" command in a system filter;
3788 (d) A message specified in a FAIL return from the queryprogram router;
3789 (e) A message specified by the cannot_route_message router option.
3791 In these cases only, Exim does include the error details in bounce and
3792 warning messages. There are also a few cases where bland messages such
3793 as "unrouteable address" or "local delivery error" are given.
3795 PH/43 $value is now also set for the "else" part of a ${run expansion.
3797 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3798 being worked on, but at least Exim now implements the latest version to
3801 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3802 process to complete was treated as a delivery failure, a timeout while
3803 writing the message to the pipe was logged, but erroneously treated as a
3804 successful delivery. Such timeouts include transport filter timeouts. For
3805 consistency with the overall process timeout, these timeouts are now
3806 treated as errors, giving rise to delivery failures by default. However,
3807 there is now a new Boolean option for the pipe transport called
3808 timeout_defer, which, if set TRUE, converts the failures into defers for
3809 both kinds of timeout. A transport filter timeout is now identified in
3812 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3813 systems where "make" and "gmake" are different, calling "gmake" at top
3814 level broke things. I've arranged for the value of $(MAKE) to be passed
3815 from the Makefile to this script so that it can call the same version of
3819 A note about Exim versions 4.44 and 4.50
3820 ----------------------------------------
3822 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3823 changes of various kinds. As a consequence, a big documentation update was
3824 needed. This delayed the release for rather longer than seemed good, especially
3825 in the light of a couple of (minor) security issues. Therefore, the changes
3826 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3827 release. So 4.44 and 4.50 are in effect two different branches that both start
3830 I have left the 4.50 change log unchanged; it contains all the changes since
3831 4.43. The change log for 4.44 is below; many of its items are identical to
3832 those for 4.50. This seems to be the most sensible way to preserve the
3833 historical information.
3839 1. Minor wording change to the doc/README.SIEVE file.
3841 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3842 computation of the current number of files was incorrect.
3844 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3845 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3848 4. Give more explanation in the error message when the command for a transport
3849 filter fails to execute.
3851 5. There are several places where Exim runs a non-Exim command in a
3852 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3853 was being done only for the command run by the queryprogram router. It is
3854 now done for all such subprocesses. The other cases are: ${run, transport
3855 filters, and the commands run by the lmtp and pipe transports.
3857 6. Added CONFIGURE_GROUP build-time option.
3859 7. Some older OS have a limit of 256 on the maximum number of file
3860 descriptors. Exim was using setrlimit() to set 1000 as a large value
3861 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3862 systems. I've change it so that if it can't get 1000, it tries for 256.
3864 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3865 was an oversight, and furthermore, ever since the addition of extra
3866 controls (e.g. 4.43/32), the checks on when to allow different forms of
3867 "control" were broken. There should now be diagnostics for all cases when a
3868 control that does not make sense is encountered.
3870 9. Added the /retain_sender option to "control=submission".
3872 10. $recipients is now available in the predata ACL (oversight).
3874 11. Tidy the search cache before the fork to do a delivery from a message
3875 received from the command line. Otherwise the child will trigger a lookup
3876 failure and thereby defer the delivery if it tries to use (for example) a
3877 cached ldap connection that the parent has called unbind on.
3879 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3880 of $address_data from the recipient verification was clobbered by the
3881 sender verification.
3883 13. The value of address_data from a sender verification is now available in
3884 $sender_address_data in subsequent conditions in the ACL statement.
3886 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3888 15. Added a new option "connect=<time>" to callout options, to set a different
3891 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3892 was its contents. (It was OK if the option was not defined at all.)
3894 17. A "Completed" log line is now written for messages that are removed from
3895 the spool by the -Mrm option.
3897 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3898 information about exactly what failed.
3900 19. Added -dd to debug only the daemon process.
3902 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3903 handles timeouts, both on the server side and network timeouts. Renamed the
3904 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3906 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3907 It is now set to "smtps".
3909 22. $host_address is now set to the target address during the checking of
3910 ignore_target_hosts.
3912 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3913 being passed; this would have caused $sender_host_name to have been used if
3914 matching the list had actually called for a host name (not very likely,
3915 since this list is usually IP addresses). A host name is now passed as
3918 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3919 code that specifies a non-restarting handler (typically sigaction() in
3920 modern systems) in an attempt to fix a rare and obscure crash bug.
3922 25. Narrowed the window for a race in the daemon that could cause it to ignore
3923 SIGCHLD signals. This is not a major problem, because they are used only to
3924 wake it up if nothing else does.
3926 26. A malformed maildirsize file could cause Exim to calculate negative values
3927 for the mailbox size or file count. Odd effects could occur as a result.
3928 The maildirsize information is now recalculated if the size or filecount
3931 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3932 support for a long time. Removed HAVE_SYS_VFS_H.
3934 28. Installed the latest version of exipick from John Jetmore.
3936 29. In an address list, if the pattern was not a regular expression, an empty
3937 subject address (from a bounce message) matched only if the pattern was an
3938 empty string. Non-empty patterns were not even tested. This was the wrong
3939 because it is perfectly reasonable to use an empty address as part of a
3940 database query. An empty address is now tested by patterns that are
3941 lookups. However, all the other forms of pattern expect the subject to
3942 contain a local part and a domain, and therefore, for them, an empty
3943 address still always fails if the pattern is not itself empty.
3945 30. Exim went into a mad DNS loop when attempting to do a callout where the
3946 host was specified on an smtp transport, and looking it up yielded more
3947 than one IP address.
3949 31. Re-factored the code for checking spool and log partition space into a
3950 function that finds that data and another that does the check. The former
3951 is then used to implement four new variables: $spool_space, $log_space,
3952 $spool_inodes, and $log_inodes.
3954 32. The RFC2047 encoding function was originally intended for short strings
3955 such as real names; it was not keeping to the 75-character limit for
3956 encoded words that the RFC imposes. It now respects the limit, and
3957 generates multiple encoded words if necessary. To be on the safe side, I
3958 have increased the buffer size for the ${rfc2047: expansion operator from
3961 33. It is now permitted to omit both strings after an "if" condition; if the
3962 condition is true, the result is "true". As before, when the second string
3963 is omitted, a false condition yields an empty string. This makes it less
3964 cumbersome to write custom ACL and router conditions.
3966 34. Failure to deliver a bounce message always caused it to be frozen, even if
3967 there was an errors_to setting on the router. The errors_to setting is now
3970 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3971 canonical form (fully expanded) before being placed in
3972 $sender_host_address.
3974 36. The table in the code that translates DNS record types into text (T_A to
3975 "A" for instance) was missing entries for NS and CNAME. It is just possible
3976 that this could have caused confusion if both these types were looked up
3977 for the same domain, because the text type is used as part of Exim's
3978 per-process caching. But the chance of anyone hitting this buglet seems
3981 37. The dnsdb lookup has been extended in a number of ways.
3983 (1) There is a new type, "zns", which walks up the domain tree until it
3984 finds some nameserver records. It should be used with care.
3986 (2) There is a new type, "mxh", which is like "mx" except that it returns
3987 just the host names, not the priorities.
3989 (3) It is now possible to give a list of domains (or IP addresses) to be
3990 looked up. The behaviour when one of the lookups defers can be
3991 controlled by a keyword.
3993 (4) It is now possible to specify the separator character for use when
3994 multiple records are returned.
3996 38. The dnslists ACL condition has been extended: it is now possible to supply
3997 a list of IP addresses and/or domains to be looked up in a particular DNS
4000 39. Added log_selector=+queue_time_overall.
4002 40. When running the queue in the test harness, wait just a tad after forking a
4003 delivery process, to get repeatability of debugging output.
4005 41. Include certificate and key file names in error message when GnuTLS fails
4006 to set them up, because the GnuTLS error message doesn't include the name
4007 of the failing file when there is a problem reading it.
4009 42. Allow both -bf and -bF in the same test run.
4011 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4013 44. The "Exiscan patch" is now merged into the mainline Exim source.
4015 45. Sometimes the final signoff response after QUIT could fail to get
4016 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4017 < 0 before doing a fflush(). This bug looks as though it goes back to the
4018 introduction of TLS in release 3.20, but "sometimes" must have been rare
4019 because the tests only now provoked it.
4021 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4022 (this can affect the format of dates).
4024 47. exim_tidydb, when checking for the continued existence of a message for
4025 which it has found a message-specific retry record, was not finding
4026 messages that were in split spool directories. Consequently, it was
4027 deleting retry records that should have stayed in existence.
4029 48. Steve fixed some bugs in eximstats.
4031 49. The SPA authentication driver was not abandoning authentication and moving
4032 on to the next authenticator when an expansion was forced to fail,
4033 contradicting the general specification for all authenticators. Instead it
4034 was generating a temporary error. It now behaves as specified.
4036 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4037 (the order specifies the preference for clients). The order is now AES256,
4038 AES128, 3DES, ARCFOUR128.
4040 51. Small patch to Sieve code - explicitly set From: when generating an
4043 52. Exim crashed if a remote delivery caused a very long error message to be
4044 recorded - for instance if somebody sent an entire SpamAssassin report back
4045 as a large number of 550 error lines. This bug was coincidentally fixed by
4046 increasing the size of one of Exim's internal buffers (big_buffer) that
4047 happened as part of the Exiscan merge. However, to be on the safe side, I
4048 have made the code more robust (and fixed the comments that describe what
4051 53. Now that there can be additional text after "Completed" in log lines (if
4052 the queue_time_overall log selector is set), a one-byte patch to exigrep
4053 was needed to allow it to recognize "Completed" as not the last thing in
4056 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4057 patch that reportedly fixes this has been added. I am not expert enough to
4058 create a test for it. This is what the patch creator wrote:
4060 "I found a little strange behaviour of ldap code when working with
4061 Windows 2003 AD Domain, where users was placed in more than one
4062 Organization Units. When I tried to give exim partial DN, the exit code
4063 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4064 But simultaneously result of request was absolutely normal ldap result,
4065 so I produce this patch..."
4067 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4068 REFERENCE, so I have modified the code to exclude the patch when that macro
4071 55. Some experimental protocols are using DNS PTR records for new purposes. The
4072 keys for these records are domain names, not reversed IP addresses. The
4073 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4074 leaves it alone. Component reversal etc. now happens only for IP addresses.
4077 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4079 57. Double the size of the debug message buffer (to 2048) so that more of very
4080 long debug lines gets shown.
4082 58. The exicyclog utility now does better if the number of log files to keep
4083 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4085 59. Two changes related to the smtp_active_hostname option:
4087 (1) $smtp_active_hostname is now available as a variable.
4088 (2) The default for smtp_banner uses $smtp_active_hostname instead
4089 of $primary_hostname.
4091 60. The host_aton() function is supposed to be passed a string that is known
4092 to be a valid IP address. However, in the case of IPv6 addresses, it was
4093 not checking this. This is a hostage to fortune. Exim now panics and dies
4094 if the condition is not met. A case was found where this could be provoked
4095 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4096 components; fortuitously, this particular loophole had already been fixed
4097 by change 4.50/55 above.
4099 If there are any other similar loopholes, the new check in host_aton()
4100 itself should stop them being exploited. The report I received stated that
4101 data on the command line could provoke the exploit when Exim was running as
4102 exim, but did not say which command line option was involved. All I could
4103 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4104 running as the user.
4107 61. There was a buffer overflow vulnerability in the SPA authentication code
4108 (which came originally from the Samba project). I have added a test to the
4109 spa_base64_to_bits() function which I hope fixes it.
4112 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4113 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4115 63. The daemon start-up calls getloadavg() while still root for those OS that
4116 need the first call to be done as root, but it missed one case: when
4117 deliver_queue_load_max is set with deliver_drop_privilege. This is
4118 necessary for the benefit of the queue runner, because there is no re-exec
4119 when deliver_drop_privilege is set.
4121 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4122 This has been fixed.
4124 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4125 was in use, was not putting the data itself into the right store pool;
4126 consequently, it could be overwritten for a subsequent message in the same
4127 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4130 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4132 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4133 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4134 the function were treating the return as a boolean value, which happened to
4135 work because 0=false and not-0=true, but is not correct code.
4137 68. The host_aton() function was not handling scoped IPv6 addresses (those
4138 with, for example, "%eth0" on the end) correctly.
4140 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4141 negated items (that is, ~something) in unsigned ints. Some compilers
4142 apparently mutter when there is no cast.
4144 70. If an address verification called from an ACL failed, and did not produce a
4145 user-specific message (i.e. there was only a "system" message), nothing was
4146 put in $acl_verify_message. In this situation, it now puts the system
4149 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4150 session; change 4.31/43 added the unwanted input to the log line - except
4151 that it did not do this in the start of session case. It now does.
4153 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4154 This is wrong and can cause the other end to generate a synchronization
4155 error if it is another Exim or anything else that does the synchronization
4156 check. A QUIT command is no longer sent after a timeout.
4158 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4159 during host lookups.
4161 74. The defer_ok option of callout verification was not working if it was used
4162 when verifying addresses in header lines, that is, for this case:
4164 verify = header_sender/callout=defer_ok
4166 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4167 those file descriptors could be used for SMTP connections. If anything
4168 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4169 could be sent to the SMTP client, causing chaos. The daemon now opens
4170 stdin, stdout, and stderr to /dev/null when it puts itself into the
4173 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4174 log by default. The user can override this with suitable Perl magic.
4176 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4177 the log message when discard triggers, was not working for the DATA ACL or
4178 for the non-SMTP ACL.
4180 78. Error message wording change in sieve.c.
4182 79. If smtp_accept_max_per_host was set, the number of connections could be
4183 restricted to fewer than expected, because the daemon was trying to set up
4184 a new connection before checking whether the processes handling previous
4185 connections had finished. The check for completed processes is now done
4186 earlier. On busy systems, this bug wouldn't be noticed because something
4187 else would have woken the daemon, and it would have reaped the completed
4190 80. If a message was submitted locally by a user whose login name contained one
4191 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4192 It caused a spool format error. I have fixed the spool reading code. A
4193 related problem was that the "from" clause in the Received: line became
4194 illegal because of the space(s). It is now covered by ${quote_local_part.
4196 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4199 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4201 83. Patch from Sieve maintainer to fix unterminated string problem in
4202 "vacation" handling.
4204 84. Some minor changes to the Linux configuration files to help with other
4205 OS variants using glibc.
4207 85. One more patch for Sieve to update vacation handling to latest spec.
4210 ----------------------------------------------------
4211 See the note above about the 4.44 and 4.50 releases.
4212 ----------------------------------------------------
4218 1. Change 4.43/35 introduced a bug that caused file counts to be
4219 incorrectly computed when quota_filecount was set in an appendfile
4222 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4223 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4226 3. Give more explanation in the error message when the command for a transport
4227 filter fails to execute.
4229 4. There are several places where Exim runs a non-Exim command in a
4230 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4231 was being done only for the command run by the queryprogram router. It is
4232 now done for all such subprocesses. The other cases are: ${run, transport
4233 filters, and the commands run by the lmtp and pipe transports.
4235 5. Some older OS have a limit of 256 on the maximum number of file
4236 descriptors. Exim was using setrlimit() to set 1000 as a large value
4237 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4238 systems. I've change it so that if it can't get 1000, it tries for 256.
4240 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4241 was an oversight, and furthermore, ever since the addition of extra
4242 controls (e.g. 4.43/32), the checks on when to allow different forms of
4243 "control" were broken. There should now be diagnostics for all cases when a
4244 control that does not make sense is encountered.
4246 7. $recipients is now available in the predata ACL (oversight).
4248 8. Tidy the search cache before the fork to do a delivery from a message
4249 received from the command line. Otherwise the child will trigger a lookup
4250 failure and thereby defer the delivery if it tries to use (for example) a
4251 cached ldap connection that the parent has called unbind on.
4253 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4254 of $address_data from the recipient verification was clobbered by the
4255 sender verification.
4257 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4258 was its contents. (It was OK if the option was not defined at all.)
4260 11. A "Completed" log line is now written for messages that are removed from
4261 the spool by the -Mrm option.
4263 12. $host_address is now set to the target address during the checking of
4264 ignore_target_hosts.
4266 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4267 being passed; this would have caused $sender_host_name to have been used if
4268 matching the list had actually called for a host name (not very likely,
4269 since this list is usually IP addresses). A host name is now passed as
4272 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4273 code that specifies a non-restarting handler (typically sigaction() in
4274 modern systems) in an attempt to fix a rare and obscure crash bug.
4276 15. Narrowed the window for a race in the daemon that could cause it to ignore
4277 SIGCHLD signals. This is not a major problem, because they are used only to
4278 wake it up if nothing else does.
4280 16. A malformed maildirsize file could cause Exim to calculate negative values
4281 for the mailbox size or file count. Odd effects could occur as a result.
4282 The maildirsize information is now recalculated if the size or filecount
4285 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4286 support for a long time. Removed HAVE_SYS_VFS_H.
4288 18. Updated exipick to current release from John Jetmore.
4290 19. Allow an empty sender to be matched against a lookup in an address list.
4291 Previously the only cases considered were a regular expression, or an
4294 20. Exim went into a mad DNS lookup loop when doing a callout where the
4295 host was specified on the transport, if the DNS lookup yielded more than
4298 21. The RFC2047 encoding function was originally intended for short strings
4299 such as real names; it was not keeping to the 75-character limit for
4300 encoded words that the RFC imposes. It now respects the limit, and
4301 generates multiple encoded words if necessary. To be on the safe side, I
4302 have increased the buffer size for the ${rfc2047: expansion operator from
4305 22. Failure to deliver a bounce message always caused it to be frozen, even if
4306 there was an errors_to setting on the router. The errors_to setting is now
4309 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4310 canonical form (fully expanded) before being placed in
4311 $sender_host_address.
4313 24. Updated eximstats to version 1.33
4315 25. Include certificate and key file names in error message when GnuTLS fails
4316 to set them up, because the GnuTLS error message doesn't include the name
4317 of the failing file when there is a problem reading it.
4319 26. Expand error message when OpenSSL has problems setting up cert/key files.
4322 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4323 (this can affect the format of dates).
4325 28. exim_tidydb, when checking for the continued existence of a message for
4326 which it has found a message-specific retry record, was not finding
4327 messages that were in split spool directories. Consequently, it was
4328 deleting retry records that should have stayed in existence.
4330 29. eximstats updated to version 1.35
4331 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4332 1.35 - bugfix such that pie charts by volume are generated correctly
4334 30. The SPA authentication driver was not abandoning authentication and moving
4335 on to the next authenticator when an expansion was forced to fail,
4336 contradicting the general specification for all authenticators. Instead it
4337 was generating a temporary error. It now behaves as specified.
4339 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4340 (the order specifies the preference for clients). The order is now AES256,
4341 AES128, 3DES, ARCFOUR128.
4343 31. Small patch to Sieve code - explicitly set From: when generating an
4346 32. Exim crashed if a remote delivery caused a very long error message to be
4347 recorded - for instance if somebody sent an entire SpamAssassin report back
4348 as a large number of 550 error lines. This bug was coincidentally fixed by
4349 increasing the size of one of Exim's internal buffers (big_buffer) that
4350 happened as part of the Exiscan merge. However, to be on the safe side, I
4351 have made the code more robust (and fixed the comments that describe what
4354 33. Some experimental protocols are using DNS PTR records for new purposes. The
4355 keys for these records are domain names, not reversed IP addresses. The
4356 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4357 leaves it alone. Component reversal etc. now happens only for IP addresses.
4360 34. The host_aton() function is supposed to be passed a string that is known
4361 to be a valid IP address. However, in the case of IPv6 addresses, it was
4362 not checking this. This is a hostage to fortune. Exim now panics and dies
4363 if the condition is not met. A case was found where this could be provoked
4364 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4365 components; fortuitously, this particular loophole had already been fixed
4366 by change 4.50/55 or 4.44/33 above.
4368 If there are any other similar loopholes, the new check in host_aton()
4369 itself should stop them being exploited. The report I received stated that
4370 data on the command line could provoke the exploit when Exim was running as
4371 exim, but did not say which command line option was involved. All I could
4372 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4373 running as the user.
4376 35. There was a buffer overflow vulnerability in the SPA authentication code
4377 (which came originally from the Samba project). I have added a test to the
4378 spa_base64_to_bits() function which I hope fixes it.
4381 36. The daemon start-up calls getloadavg() while still root for those OS that
4382 need the first call to be done as root, but it missed one case: when
4383 deliver_queue_load_max is set with deliver_drop_privilege. This is
4384 necessary for the benefit of the queue runner, because there is no re-exec
4385 when deliver_drop_privilege is set.
4387 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4388 was in use, was not putting the data itself into the right store pool;
4389 consequently, it could be overwritten for a subsequent message in the same
4390 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4393 38. Sometimes the final signoff response after QUIT could fail to get
4394 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4395 < 0 before doing a fflush(). This bug looks as though it goes back to the
4396 introduction of TLS in release 3.20, but "sometimes" must have been rare
4397 because the tests only now provoked it.
4403 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4404 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4405 Now it returns an integer. A number of calls were still expecting a T/F
4406 return. Fortuitously, in all cases, the tests worked in OK situations,
4407 which is the norm. However, things would have gone wrong on any write
4408 failures on the smtp file descriptor. This function is used when sending
4409 messages over SMTP and also when doing verify callouts.
4411 2. When Exim is called to do synchronous delivery of a locally submitted
4412 message (the -odf or -odi options), it no longer closes stderr before doing
4415 3. Implemented the mua_wrapper option.
4417 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4419 5. Implemented the functions header_remove(), header_testname(),
4420 header_add_at_position(), and receive_remove_recipient(), and exported them
4423 6. If an ACL "warn" statement specified the addition of headers, Exim already
4424 inserted X-ACL-Warn: at the start if there was no header name. However, it
4425 was not making this test for the second and subsequent header lines if
4426 there were newlines in the string. This meant that an invalid header could
4427 be inserted if Exim was badly configured.
4429 7. Allow an ACL "warn" statement to add header lines at the start or after all
4430 the Received: headers, as well as at the end.
4432 8. Added the rcpt_4xx retry error code.
4434 9. Added postmaster_mailfrom=xxx to callout verification option.
4436 10. Added mailfrom=xxxx to the callout verification option, for verify=
4439 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4440 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4442 12. Inserted some casts to stop certain compilers warning when using pointer
4443 differences as field lengths or precisions in printf-type calls (mostly
4444 affecting debugging statements).
4446 13. Added optional readline() support for -be (dynamically loaded).
4448 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4449 same clock tick as a message's arrival, so that its received time was the
4450 same as the "first fail" time on the retry record, and that message
4451 remained on the queue past the ultimate address timeout, every queue runner
4452 would try a delivery (because it was past the ultimate address timeout) but
4453 after another failure, the ultimate address timeout, which should have then
4454 bounced the address, did not kick in. This was a "< instead of <=" error;
4455 in most cases the first failure would have been in the next clock tick
4456 after the received time, and all would be well.
4458 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4459 being recognized when the domain list was tested by the match_domain
4460 condition in an expansion string.
4462 16. Added the ${str2b64: operator.
4464 17. Exim was always calling setrlimit() to set a large limit for the number of
4465 processes, without checking whether the existing limit was already
4466 adequate. (It did check for the limit on file descriptors.) Furthermore,
4467 errors from getrlimit() and setrlimit() were being ignored. Now they are
4468 logged to the main and panic logs, but Exim does carry on, to try to do its
4469 job under whatever limits there are.
4471 18. Imported PCRE 5.0.
4473 19. Trivial typo in log message " temporarily refused connection" (the leading
4476 20. If the log selector return_path_on_delivery was set and an address was
4477 redirected to /dev/null, the delivery process crashed because it assumed
4478 that a return path would always be set for a "successful" delivery. In this
4479 case, the whole delivery is bypassed as an optimization, and therefore no
4482 21. Internal re-arrangement: the function for sending a challenge and reading
4483 a response while authentication was assuming a zero-terminated challenge
4484 string. It's now changed to take a pointer and a length, to allow for
4485 binary data in such strings.
4487 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4489 23. Exim was not respecting finduser_retries when seeking the login of the
4490 uid under which it was called; it was always trying 10 times. (The default
4491 setting of finduser_retries is zero.) Also, it was sleeping after the final
4492 failure, which is pointless.
4494 24. Implemented tls_on_connect_ports.
4496 25. Implemented acl_smtp_predata.
4498 26. If the domain in control=submission is set empty, Exim assumes that the
4499 authenticated id is a complete email address when it generates From: or
4500 Sender: header lines.
4502 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4503 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4504 chown and chgrp in /bin and hostname in /usr/bin.
4506 28. Exim was keeping the "process log" file open after each use, just as it
4507 does for the main log. This opens the possibility of it remaining open for
4508 long periods when the USR1 signal hits a daemon. Occasional processlog
4509 errors were reported, that could have been caused by this. Anyway, it seems
4510 much more sensible not to leave this file open at all, so that is what now
4513 29. The long-running daemon process does not normally write to the log once it
4514 has entered its main loop, and it closes the log before doing so. This is
4515 so that log files can straightforwardly be renamed and moved. However,
4516 there are a couple of unusual error situations where the daemon does write
4517 log entries, and I had neglected to close the log afterwards.
4519 30. The text of an SMTP error response that was received during a remote
4520 delivery was being truncated at 512 bytes. This is too short for some of
4521 the long messages that one sometimes sees. I've increased the limit to
4524 31. It is now possible to make retry rules that apply only when a message has a
4525 specific sender, in particular, an empty sender.
4527 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4528 it possible to be selective about when SMTP synchronization is enforced.
4530 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4532 32. Implemented hosts_connection_nolog.
4534 33. Added an ACL for QUIT.
4536 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4539 35. Added mailbox_size and mailbox_filecount to appendfile.
4541 36. Added control = no_multiline_responses to ACLs.
4543 37. There was a bug in the logic of the code that waits for the clock to tick
4544 in the case where the clock went backwards by a substantial amount such
4545 that the microsecond fraction of "now" was more than the microsecond
4546 fraction of "then" (but the whole seconds number was less).
4548 38. Added support for the libradius Radius client library this is found on
4549 FreeBSD (previously only the radiusclient library was supported).
4555 1. When certain lookups returned multiple values in the form name=value, the
4556 quoting of the values was not always being done properly. Specifically:
4557 (a) If the value started with a double quote, but contained no whitespace,
4559 (b) If the value contained whitespace other than a space character (i.e.
4560 tabs or newlines or carriage returns) it was not quoted.
4561 This fix has been applied to the mysql and pgsql lookups by writing a
4562 separate quoting function and calling it from the lookup code. The fix
4563 should probably also be applied to nisplus, ibase and oracle lookups, but
4564 since I cannot test any of those, I have not disturbed their existing code.
4566 2. A hit in the callout cache for a specific address caused a log line with no
4567 reason for rejecting RCPT. Now it says "Previous (cached) callout
4568 verification failure".
4570 3. There was an off-by-one bug in the queryprogram router. An over-long
4571 return line was truncated at 256 instead of 255 characters, thereby
4572 overflowing its buffer with the terminating zero. As well as fixing this, I
4573 have increased the buffer size to 1024 (and made a note to document this).
4575 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4576 when Exim is waiting for an SMTP response from a remote server, Exim
4577 restarts its select() call on the socket, thereby resetting its timeout.
4578 This is not a problem when such interrupts are rare. Somebody set up a cron
4579 job to run exiwhat every 2 minutes, which is less than the normal select()
4580 timeout (5 or 10 minutes). This meant that the select() timeout never
4581 kicked in because it was always reset. I have fixed this by comparing the
4582 time when an interrupt arrives with the time at the start of the first call
4583 to select(). If more time than the timeout has elapsed, the interrupt is
4584 treated as a timeout.
4586 5. Some internal re-factoring in preparation for the addition of Sieve
4587 extensions (by MH). In particular, the "personal" test is moved to a
4588 separate function, and given an option for scanning Cc: and Bcc: (which is
4589 not set for Exim filters).
4591 6. When Exim created an email address using the login of the caller as the
4592 local part (e.g. when creating a From: or Sender: header line), it was not
4593 quoting the local part when it contained special characters such as @.
4595 7. Installed new OpenBSD configuration files.
4597 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4598 try to make them clearer.
4600 9. Callout options, other than the timeout value, were being ignored when
4601 verifying sender addresses in header lines. For example, when using
4603 verify = header_sender/callout=no_cache
4605 the cache was (incorrectly) being used.
4607 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4608 only the Cygwin environment.
4610 11. When return_path_on_delivery was set as a log selector, if different remote
4611 addresses in the same message used different return paths and parallel
4612 remote delivery occurred, the wrong values would sometimes be logged.
4613 (Whenever a remote delivery process finished, the return path value from
4614 the most recently started remote delivery process was logged.)
4616 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4617 lines when AUTH and/or TLS are in use. This is the "received protocol"
4618 field. Exim used to use "asmtp" for authenticated SMTP, without any
4619 indication (in the protocol name) for TLS use. Now it follows the RFC and
4620 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4621 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4622 names appear in log lines as well as in Received: header lines.
4624 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4625 extensions, and comparison tests, and to fix some bugs.
4627 14. Changes to the "personal" filter test:
4629 (1) The test was buggy in that it was just doing the equivalent of
4630 "contains" tests on header lines. For example, if a user's address was
4631 anne@some.where, the "personal" test would incorrectly be true for
4633 To: susanne@some.where
4635 This test is now done by extracting each address from the header in turn,
4636 and checking the entire address. Other tests that are part of "personal"
4637 are now done using regular expressions (for example, to check local parts
4638 of addresses in From: header lines).
4640 (2) The list of non-personal local parts in From: addresses has been
4641 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4642 taken from the Sieve specification recommendations.
4644 (3) If the message contains any header line starting with "List-" it is
4645 treated as non-personal.
4647 (4) The test for "circular" in the Subject: header line has been removed
4648 because it now seems ill-conceived.
4650 15. Minor typos in src/EDITME comments corrected.
4652 16. Installed latest exipick from John Jetmore.
4654 17. If headers_add on a router specified a text string that was too long for
4655 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4656 of string_sprintf() is now avoided.
4658 18. $message_body_size was not set (it was always zero) when running the DATA
4659 ACL and the local_scan() function.
4661 19. For the "mail" command in an Exim filter, no default was being set for
4662 the once_repeat time, causing a random time value to be used if "once" was
4663 specified. (If the value happened to be <= 0, no repeat happened.) The
4664 default is now 0s, meaning "never repeat". The "vacation" command was OK
4665 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4666 (I found it when inspecting the code).
4668 20. There is now an overall timeout for performing a callout verification. It
4669 defaults to 4 times the callout timeout, which applies to individual SMTP
4670 commands during the callout. The overall timeout applies when there is more
4671 than one host that can be tried. The timeout is checked before trying the
4672 next host. This prevents very long delays if there are a large number of
4673 hosts and all are timing out (e.g. when the network connections are timing
4674 out). The value of the overall timeout can be changed by specifying an
4675 additional sub-option for "callout", called "maxwait". For example:
4677 verify = sender/callout=5s,maxwait=20s
4679 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4680 to the end before writing, but this should make it even safer).
4682 22. Exim was forgetting that it had advertised PIPELINING for the second and
4683 subsequent messages on an SMTP connection. It was also not resetting its
4684 memory on STARTTLS and an internal HELO.
4686 23. When Exim logs an SMTP synchronization error within a session, it now
4687 records whether PIPELINING has been advertised or not.
4689 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4690 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4691 rather than long int.
4693 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4695 26. Added the never_mail option to autoreply.
4701 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4702 crash if the getsockname() call failed; this can happen if a connection is
4703 closed very soon after it is established. The problem was simply in the
4704 order in which certain operations were done, causing Exim to try to write
4705 to the SMTP stream before it had set up the file descriptor. The bug has
4706 been fixed by making things happen in the correct order.
4712 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4713 before the connection was closed, thus losing the rejection response.
4715 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4716 some early Solaris releases, but causes trouble in current releases where
4717 socklen_t is defined.
4719 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4722 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4725 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4726 that releasing the top of it at the end releases what was used for sub-
4727 expansions (unless the block got too big). However, discard this block if
4728 the first thing is a variable or header, so that we can use its block when
4729 it is dynamic (useful for very large $message_headers, for example).
4731 6. Lookups now cache *every* query, not just the most recent. A new, separate
4732 store pool is used for this. It can be recovered when all lookup caches are
4733 flushed. Lookups now release memory at the end of their result strings.
4734 This has involved some general refactoring of the lookup sources.
4736 7. Some code has been added to the store_xxx() functions to reduce the amount
4737 of flapping under certain conditions.
4739 8. log_incoming_interface used to affect only the <= reception log lines. Now
4740 it causes the local interface and port to be added to several more SMTP log
4741 lines, for example "SMTP connection from", and rejection lines.
4743 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4745 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4747 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4748 could be overwritten at the end of the current message (or the start of a
4749 new message if it was set in a HELO ACL). The value is now preserved for
4750 the duration of the SMTP connection.
4752 12. If a transport had a headers_rewrite setting, and a matching header line
4753 contained an unqualified address, that address was qualified, even if it
4754 did not match any rewriting rules. The underlying bug was that the values
4755 of the flags that permit the existence of unqualified sender and recipient
4756 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4757 non-local messages, and by -bnq for local messages) were not being
4758 preserved with the message after it was received.
4760 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4761 "next input=" as part of the text comprising the host identity instead of
4762 the correct text. The code was using the same buffer for two different
4763 strings. However, depending on which order the printing function evaluated
4764 its arguments, the bug did not always show up. Under Linux, for example, my
4765 test suite worked just fine.
4767 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4768 doesn't work with some older versions of Perl. It has been changed to "my",
4769 which in any case is probably the better facility to use.
4771 15. A really picky compiler found some instances of statements for creating
4772 error messages that either had too many or two few arguments for the format
4775 16. The size of the buffer for calls to the DNS resolver has been increased
4776 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4777 for addresses that have a lot of PTR records. This alleviates a problem; it
4778 does not fully solve it.
4780 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4781 buffer now truncates the list and logs the incident, which is the same
4782 action as happens when Exim is looking up a host name and its aliases.
4783 Previously in this situation something unpredictable would happen;
4784 sometimes it was "internal error: store_reset failed".
4786 18. If a server dropped the connection unexpectedly when an Exim client was
4787 using GnuTLS and trying to read a response, the client delivery process
4788 crashed while trying to generate an error log message.
4790 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4791 string, for example:
4793 warn message = H1: something\nH2: something
4795 the text was added as a single header line from Exim's point of view
4796 though it ended up OK in the delivered message. However, searching for the
4797 second and subsequent header lines using $h_h2: did not work. This has been
4798 fixed. Similarly, if a system filter added multiple headers in this way,
4799 the routers could not see them.
4801 20. Expanded the error message when iplsearch is called with an invalid key to
4802 suggest using net-iplsearch in a host list.
4804 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4805 ACLs are no longer actually imposed (and a message to that effect is
4808 22. If a "gecos" field in a passwd entry contained escaped characters, in
4809 particular, if it contained a \" sequence, Exim got it wrong when building
4810 a From: or a Sender: header from that name. A second bug also caused
4811 incorrect handling when an unquoted " was present following a character
4812 that needed quoting.
4814 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4815 was not being matched caselessly.
4817 24. Arranged for all hyphens in the exim.8 source to be escaped with
4820 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4821 properly. Recipient callout cache records were still being keyed to include
4822 the sender, even when use_sender was set false. This led to far more
4823 callouts that were necessary. The sender is no longer included in the key
4824 when use_sender is false.
4826 26. Added "control = submission" modifier to ACLs.
4828 27. Added the ${base62d: operator to decode base 62 numbers.
4830 28. dnsdb lookups can now access SRV records.
4832 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4833 the configuration file.
4835 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4836 (-v) mode. This makes the output for a verbose queue run more intelligible.
4838 31. Added a use_postmaster feature to recipient callouts.
4840 32. Added the $body_zerocount variable, containing the number of binary zero
4841 bytes in the message body.
4843 33. The time of last modification of the "new" subdirectory is now used as the
4844 "mailbox time last read" when there is a quota error for a maildir
4847 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4849 35. Added +ignore_unknown as a special item in host lists.
4851 36. Code for decoding IPv6 addresses in host lists is now included, even if
4852 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4853 address was recognized as an IP address, but was then not correctly decoded
4854 into binary, causing unexpected and incorrect effects when compared with
4861 1. Very minor rewording of debugging text in manualroute to say "list of
4862 hosts" instead of "hostlist".
4864 2. If verify=header_syntax was set, and a header line with an unqualified
4865 address (no domain) and a large number of spaces between the end of the
4866 name and the colon was received, the reception process suffered a buffer
4867 overflow, and (when I tested it) crashed. This was caused by some obsolete
4868 code that should have been removed. The fix is to remove it!
4870 3. When running in the test harness, delay a bit after writing a bounce
4871 message to get a bit more predictability in the log output.
4873 4. Added a call to search_tidyup() just before forking a reception process. In
4874 theory, someone could use a lookup in the expansion of smtp_accept_max_
4875 per_host which, without the tidyup, could leave open a database connection.
4877 5. Added the variables $recipient_data and $sender_data which get set from a
4878 lookup success in an ACL "recipients" or "senders" condition, or a router
4879 "senders" option, similar to $domain_data and $local_part_data.
4881 6. Moved the writing of debug_print from before to after the "senders" test
4884 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4885 problems for message scanning, either using a data ACL, or using
4886 local_scan() because the Received: header was not generated till after they
4887 were called (in order to set the time as the time of reception completion).
4888 I have revised the way this works. The header is now generated after the
4889 body is received, but before the ACL or local_scan() are called. After they
4890 are run, the timestamp in the header is updated.
4896 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4897 before starting a queue runner without re-exec. This happened only when
4898 deliver_drop_privilege was set or when the Exim user was set to root. The
4899 effect of the bug was that timeouts during subsequent deliveries caused
4900 crashes instead of being properly handled. The handler is now left at its
4901 default (and expected) setting.
4903 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4904 message, again when deliver_drop_privilege is set or Exim is run as root.
4905 The bug described in (1) was not present in this case, but the tidying up
4906 of the other signals was missing. I have made the two cases consistent.
4908 3. The ignore_target_hosts setting on a manualroute router was being ignored
4909 for hosts that were looked up using the /MX notation.
4911 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4914 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4915 operated on the sender address. After changing the $sender_address to <>
4916 for the sender address verify, Exim was re-instated it as the original
4917 (before rewriting) address, but remembering that it had rewritten it, so it
4918 wasn't rewriting it again. This bug also had the effect of breaking the
4919 sender address verification caching when the sender address was rewritten.
4921 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4922 This has been changed so that if the ip literal address matches
4923 ignore_target_hosts, the router declines.
4925 7. Added expansion conditions match_domain, match_address, and match_local_
4926 part (NOT match_host).
4928 8. The placeholder for the Received: header didn't have a length field set.
4930 9. Added code to Exim itself and to exim_lock to test for a specific race
4931 condition that could lead to file corruption when using MBX delivery. The
4932 issue is with the lockfile that is created in /tmp. If this file is removed
4933 after a process has opened it but before that process has acquired a lock,
4934 there is the potential for a second process to recreate the file and also
4935 acquire a lock. This could lead to two Exim processes writing to the file
4936 at the same time. The added code performs the same test as UW imapd; it
4937 checks after acquiring the lock that its file descriptor still refers to
4938 the same named file.
4940 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4941 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4944 11. Added the smtp_active_hostname option. If used, this will typically be made
4945 to depend on the incoming interface address. Because $interface_address is
4946 not set up until the daemon has forked a reception process, error responses
4947 that can happen earlier (such as "too many connections") no longer contain
4950 12. If an expansion in a condition on a "warn" statement fails because a lookup
4951 defers, the "warn" statement is abandoned, and the next ACL statement is
4952 processed. Previously this caused the whole ACL to be aborted.
4954 13. Added the iplsearch lookup type.
4956 14. Added ident_timeout as a log selector.
4958 15. Added tls_certificate_verified as a log selector.
4960 16. Added a global option tls_require_ciphers (compare the smtp transport
4961 option of the same name). This controls incoming TLS connections.
4963 17. I finally figured out how to make tls_require_ciphers do a similar thing
4964 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4965 before starting the TLS session.
4967 18. Tabs are now shown as \t in -bP output.
4969 19. If the log selector return_path_on_delivery was set, Exim crashed when
4970 bouncing a message because it had too many Received: header lines.
4972 20. If two routers both had headers_remove settings, and the first one included
4973 a superfluous trailing colon, the final name in the first list and the
4974 first name in the second list were incorrectly joined into one item (with a
4975 colon in the middle).
4981 1. Added -C and -D options to the exinext utility, mainly to make it easier
4982 to include in the automated testing, but these could be helpful when
4983 multiple configurations are in use.
4985 2. The exinext utility was not formatting the output nicely when there was
4986 an alternate port involved in the retry record key, nor when there was a
4987 message id as well (for retries that were specific to a specific message
4988 and a specific host). It was also confused by IPv6 addresses, because of
4989 the additional colons they contain. I have fixed the IPv4 problem, and
4990 patched it up to do a reasonable job for IPv6.
4992 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4993 delivery, the log line now contains "pipelined" if PIPELINING was used.
4995 4. An SMTP transport process used to panic and die if the bind() call to set
4996 an explicit outgoing interface failed. This has been changed; it is now
4997 treated in the same way as a connect() failure.
4999 5. A reference to $sender_host_name in the part of a conditional expansion
5000 that was being skipped was still causing a DNS lookup. This no longer
5003 6. The def: expansion condition was not recognizing references to header lines
5004 that used bh_ and bheader_.
5006 7. Added the _cache feature to named lists.
5008 8. The code for checking quota_filecount in the appendfile transport was
5009 allowing one more file than it should have been.
5011 9. For compatibility with Sendmail, the command line option
5019 and sets the incoming protocol and host name (for trusted callers). The
5020 host name and its colon can be omitted when only the protocol is to be set.
5021 Note the Exim already has two private options, -pd and -ps, that refer to
5022 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5023 "s", but I don't think that's a major issue.
5025 10. A number of refactoring changes to the code, none of which should affect
5028 (a) The number of logging options was getting close to filling up the
5029 32-bit word that was used as a bit map. I have split them into two classes:
5030 those that are passed in the argument to log_write(), and those that are
5031 only ever tested independently outside of that function. These are now in
5032 separate 32-bit words, so there is plenty of room for expansion again.
5033 There is no change in the user interface or the logging behaviour.
5035 (b) When building, for example, log lines, the code previously used a
5036 macro that called string_cat() twice, in order to add two strings. This is
5037 not really sufficiently general. Furthermore, there was one instance where
5038 it was actually wrong because one of the argument was used twice, and in
5039 one call a function was used. (As it happened, calling the function twice
5040 did not affect the overall behaviour.) The macro has been replaced by a
5041 function that can join an arbitrary number of extra strings onto a growing
5044 (c) The code for expansion conditions now uses a table and a binary chop
5045 instead of a serial search (which was left over from when there were very
5046 few conditions). Also, it now recognizes conditions like "pam" even when
5047 the relevant support is not compiled in: a suitably worded error message is
5048 given if an attempt is made to use such a condition.
5050 11. Added ${time_interval:xxxxx}.
5052 12. A bug was causing one of the ddress fields not to be passed back correctly
5053 from remote delivery subprocesses. The field in question was not being
5054 subsequently used, so this caused to problems in practice.
5056 13. Added new log selectors queue_time and deliver_time.
5058 14. Might have fixed a bug in maildirsizefile handling that threw up
5059 "unexpected character" debug warnings, and recalculated the data
5060 unnecessarily. In any case, I expanded the warning message to give more
5063 15. Added the message "Restricted characters in address" to the statements in
5064 the default ACL that block characters like @ and % in local parts.
5066 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5067 Three changes have been made:
5069 (a) There was a serious bug; a negative response to MAIL caused the whole
5070 recipient domain to be cached as invalid, thereby blocking all messages
5071 to all local parts at the same domain, from all senders. This bug has
5072 been fixed. The domain is no longer cached after a negative response to
5073 MAIL if the sender used is not empty.
5075 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5078 (c) A new callout option, "use_sender" has been added for people who want
5079 the modified behaviour.
5085 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5088 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5089 indeed breaks things for older releases.
5091 3. Added additional logging to the case where there is a problem reading data
5092 from a filter that is running in a subprocess using a pipe, in order to
5093 try to track down a specific problem.
5095 4. Testing facility fudge: when running in the test harness and attempting
5096 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5097 getting "No route to host". Convert this to a timeout.
5099 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5102 6. Some OS don't have socklen_t but use size_t instead. This affects the
5103 fifth argument of getsockopt() amongst other things. This is now
5104 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5105 can be set for individual OS. I have set it for SunOS5, OSF1, and
5106 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5107 some earlier ones do not.
5109 7. Change 4.30/15 was not doing the test caselessly.
5111 8. The standard form for an IPv6 address literal was being rejected by address
5112 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5113 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5114 this, as well as the form without the "IPv6" on the front (but only when
5115 address literals are enabled, of course).
5117 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5119 10. Exim crashed if a message with an empty sender address specified by -f
5120 encountered a router with an errors_to setting. This could be provoked only
5121 by a command such as
5125 where an empty string was supplied; "<>" did not hit this bug.
5127 11. Installed PCRE release 4.5.
5129 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5130 remained set. It is now erased.
5132 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5133 times from message ids (which are base 36 rather than the normal 62).
5135 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5136 were being counted as actual protocol errors, and logged if the log
5137 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5138 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5139 and DATA following a set of rejected RCPTs do not count as protocol errors.
5140 In other words, Exim assumes they were pipelined, though this may not
5141 actually be the case. Of course, in all cases the client gets an
5142 appropriate error code.
5144 15. If a lookup fails in an ACL condition, a message about the failure may
5145 be available; it is used if testing the ACL cannot continue, because most
5146 such messages specify what the cause of the deferral is. However, some
5147 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5148 that caused an old message to be retained and used if a later statement
5149 caused a defer, replacing the real cause of the deferral.
5151 16. If an IP address had so many PTR records that the DNS lookup buffer
5152 was not large enough to hold them, Exim could crash while trying to process
5153 the truncated data. It now detects and logs this case.
5155 17. Further to 4.21/58, another change has been made: if (and only if) the
5156 first line of a message (the first header line) ends with CRLF, a bare LF
5157 in a subsequent header line has a space inserted after it, so as not to
5158 terminate the header.
5160 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5161 unnecessarily, used atoi() instead of strtol(), and didn't check the
5162 termination when getting file sizes from file names by regex.
5164 19. Completely re-implemented the support for maildirsize files, in the light
5165 of a number of problems with the previous contributed implementation
5166 (4.30/29). In particular:
5168 . If the quota is zero, the maildirsize file is maintained, but no quota is
5171 . If the maildir directory does not exist, it is created before any attempt
5172 to write a maildirsize file.
5174 . The quota value in the file is just a cache; if the quota is changed in
5175 the transport, the new value overrides.
5177 . A regular expression is available for excluding directories from the
5180 20. The autoreply transport checks the characters in options that define the
5181 message's headers; it allows continued headers, but it was checking with
5182 isspace() after an embedded newline instead of explicitly looking for a
5185 21. If all the "regular" hosts to which an address was routed had passed their
5186 expiry times, and had not reached their retry times, the address was
5187 bounced, even if fallback hosts were defined. Now Exim should go on to try
5190 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5191 equivalent code in the SMTP transport. Some hosts send humungous responses
5192 to HELO/EHLO, more than 1024 it seems.
5194 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5195 gives compiler warnings in some environments. I've now done it "properly",
5198 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5199 (because of problems with the built-in one) was declared to return uschar *
5200 instead of char *, causing compiler failure.
5202 25. Fixed a file descriptor leak when processing alias/forward files.
5204 26. Fixed a minor format string issue in dbfn.c.
5206 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5208 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5209 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5212 29. When a sender address is verified, it is cached, to save repeating the test
5213 when there is more than one recipient in a message. However, when the
5214 verification involves a callout, it is possible for different callout
5215 options to be set for different recipients. It is too complicated to keep
5216 track of this in the cache, so now Exim always runs a verification when a
5217 callout is required, relying on the callout cache for the optimization.
5218 The overhead is duplication of the address routing, but this should not be
5221 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5222 to be verified with callout=postmaster, and the main callout worked but the
5223 postmaster check failed, the verification correctly failed. However, if a
5224 subsequent RCPT command asked for sender verification *without* the
5225 postmaster check, incorrect caching caused this verification also to fail,
5228 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5229 it was not caching the DNS options (qualify_single, search_parents) that
5230 were used when the lookup failed. A subsequent lookup with different
5231 options therefore always gave the same answer, though there were cases
5232 where it should not have. (Example: a "domains = !$mx_any" option on a
5233 dnslookup router: the "domains" option is always processed without any
5234 widening, but the router might have qualify_single set.) Now Exim uses the
5235 cached value only when the same options are set.
5237 32. Added John Jetmore's "exipick" utility to the distribution.
5239 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5240 than a timeout (e.g. a certificate is required, and is not provided), an
5241 Exim server now closes the connection immediately. Previously it waited for
5242 the client to close - but if the client is SSL, it seems that they each
5243 wait for each other, leading to a delay before one of them times out.
5245 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5246 maintained 0.8.x compatibility because I don't think many are using it, and
5247 it is clearly obsolete.
5249 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5252 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5253 client certificate was expired. A simple patch fixes this, though I don't
5254 understand the full logic of why the verify callback is called multiple
5257 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5258 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5259 which causes problems with some clients (such as the Certicom SSL Plus
5260 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5261 disables the coutermeasure allowing Eudora to connect."
5263 38. Exim was not checking that a write() to a log file succeeded. This could
5264 lead to Bad Things if a log got too big, in particular if it hit a file
5265 size limit. Exim now panics and dies if it cannot write to a log file, just
5266 as it does if it cannot open a log file.
5268 39. Modified OS/Makefile-Linux so that it now contains
5270 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5272 The two -D definitions ensure that Exim is compiled with large file
5273 support, which makes it possible to handle log files that are bigger than
5276 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5277 instance) a domain was checked against a named list that involved a lookup,
5278 causing $domain_data to be set, then another domain was checked against the
5279 same list, then the first domain was re-checked, the value of $domain_data
5280 after the final check could be wrong. In particular, if the second check
5281 failed, it could be set empty. This bug probably also applied to
5284 41. The strip_trailing_dot option was not being applied to the address given
5285 with the -f command-line option.
5287 42. The code for reading a message's header from the spool was incrementing
5288 $received_count, but never initializing it. This meant that the value was
5289 incorrect (doubled) while delivering a message in the same process in which
5290 it was received. In the most common configuration of Exim, this never
5291 happens - a fresh exec is done - but it can happen when
5292 deliver_drop_privilege is set.
5294 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5295 it now includes up to 150 characters of the unexpected data in the log
5298 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5299 and building data strings. The size of both of these buffers was 10 000
5300 bytes - far larger than anybody would *ever* want, thought I. Needless to
5301 say, somebody hit the limit. I have increased the maximum line length to
5302 20 000 and the maximum data length of concatenated lines to 100 000. I have
5303 also fixed two bugs, because there was no checking on these buffers. Tsk,
5304 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5305 buffer is too small.
5307 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5308 lsearch lookups. Now it does.
5310 46. When parsing a route_list item in a manualroute router, a fixed-length
5311 buffer was used for the list of hosts. I made this 1024 bytes long,
5312 thinking that nobody would ever have a list of hosts that long. Wrong.
5313 Somebody had a whole pile of complicated expansion conditions, and the
5314 string was silently truncated, leading to an expansion error. It turns out
5315 that it is easier to change to an unlimited length (owing to other changes
5316 that have happened since this code was originally written) than to build
5317 structure for giving a limitation error. The length of the item that
5318 expands into the list of hosts is now unlimited.
5320 47. The lsearch lookup could not handle data where the length of text line was
5321 more than 4095 characters. Such lines were truncated, leading to shortened
5322 data being returned. It should now handle lines of any length.
5324 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5325 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5328 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5329 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5330 no longer happens in comment lines. A list of replacements is now placed
5331 at the head of all of the source files, except those whose only change is
5332 to replace PERL_COMMAND in the very first #! line.
5334 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5335 messages on the queue, with a bottom-up merge sort, using code contributed
5336 by Michael Haardt. This should make operations like -bp somewhat faster on
5337 large queues. It won't affect queue runners, except when queue_run_in_order
5340 51. Installed eximstats 1.31 in the distribution.
5342 52. Added support for SRV lookups to the dnslookup router.
5344 53. If an ACL referred to $message_body or $message_body_end, the value was not
5345 reset for any messages that followed in the same SMTP session.
5347 54. The store-handling optimization for building very long strings was not
5348 differentiating between the different store pools. I don't think this
5349 actually made any difference in practice, but I've tidied it.
5351 55. While running the routers to verify a sender address, $sender_address
5352 was still set to the sender address. This is wrong, because when routing to
5353 send a bounce to the sender, it would be empty. Therefore, I have changed
5354 it so that, while verifying a sender address, $sender_address is set to <>.
5355 (There is no change to what happens when verifying a recipient address.)
5357 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5358 target A or AAAA records (if not already returned) without resetting the
5359 qualify_single or search_parents options of the DNS resolver. These are
5360 inappropriate in this case because the targets of MX and SRV records must
5361 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5362 target that, when qualified, matched something in the local domain. These
5363 two options are now turned off when doing these lookups.
5365 57. It seems that at least some releases of Reiserfs (which does not have the
5366 concept of a fixed number of inodes) returns zero and not -1 for the
5367 number of available inodes. This interacted badly with check_spool_inodes,
5368 which assumed that -1 was the "no such thing" setting. What I have done is
5369 to check that the total number of inodes is greater than zero before doing
5370 the test of how many are available.
5372 58. When a "warn" ACL statement has a log_message modifier, the message is
5373 remembered, and not repeated. This is to avoid a lot of repetition when a
5374 message has many recipients that cause the same warning to be written.
5375 Howewer, Exim was preserving the list of already written lines for an
5376 entire SMTP session, which doesn't seem right. The memory is now reset if a
5377 new message is started.
5379 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5380 debugging output unless log_rewrite was also set.
5382 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5383 of (int)(handle) when we know that handle contains (void *)(-1).
5385 61. The Exim daemon panic-logs an error return when it closes the incoming
5386 connection. However "connection reset by peer" seems to be common, and
5387 isn't really an error worthy of noting specially, so that particular error
5390 62. When Exim is trying to find all the local interfaces, it used to panic and
5391 die if the ioctl to get the interface flags failed. However, it seems that
5392 on at least one OS (Solaris 9) it is possible to have an interface that is
5393 included in the list of interfaces, but for which you get a failure error
5394 for this call. This happens when the interface is not "plumbed" into a
5395 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5396 failure of the "get flags" call assumes that the interface is down.
5398 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5399 makes life easier for people who are doing arithmetic on fields extracted
5400 from dates, where you often get leading zeros that should not be
5401 interpreted as octal.
5403 64. Added qualify_domain to the redirect router, to override the global
5406 65. If a pathologically long header line contained very many addresses (the
5407 report of this problem mentioned 10 000) and each of them was rewritten,
5408 Exim could use up a very large amount of memory. (It kept on making new
5409 copies of the header line as it rewrote, and never released the old ones.)
5410 At the expense of a bit more processing, the header rewriting function has
5411 been changed so that it no longer eats memory in this way.
5413 66. The generation of the Received: header has been moved from the time that a
5414 message starts to be received, to the time that it finishes. The timestamp
5415 in the Received: header should now be very close to that of the <= log
5416 line. There are two side-effects of this change:
5418 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5419 logged header lines no longer include the local Received: line, because
5420 it has not yet been created. The same applies to a copy of the message
5421 that is returned to a non-SMTP sender when a message is rejected.
5423 (b) When a filter file is tested using -bf, no additional Received: header
5424 is added to the test message. After some thought, I decided that this
5427 This change does not affect the value of $received_for. It is still set
5428 after address rewriting, but before local_scan() is called.
5430 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5432 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5433 gave an unhelpful panic error message, and a defer error. I have managed to
5434 change this behaviour so that it now rejects any supplied certificate,
5435 which seems right, as the list of acceptable certificates is empty.
5437 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5438 gave an unhelpful defer error. I have not managed to make this reject any
5439 supplied certificates, but the error message it gives is "no certificate
5440 supplied", which is not helpful.
5442 70. exigrep's output now also includes lines that are not associated with any
5443 message, but which match the given pattern. Implemented by a patch from
5444 Martin Sluka, which also tidied up the Perl a bit.
5446 71. Recipient callout verification, like sender verification, was using <> in
5447 the MAIL FROM command. This isn't really the right thing, since the actual
5448 sender may affect whether the remote host accepts the recipient or not. I
5449 have changed it to use the actual sender in the callout; this means that
5450 the cache record is now keyed on a recipient/sender pair, not just the
5451 recipient address. There doesn't seem to be a real danger of callout loops,
5452 since a callout by the remote host to check the sender would use <>.
5453 [SEE ABOVE: changed after hitting problems.]
5455 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5456 temporary errors. However, in the case of such a code being given after
5457 the end of a data transmission (i.e. after ".") Exim was failing to write
5458 a retry record for the message. (Yes, there was some broken host that was
5459 actually sending 8xx at this point.)
5461 73. An unknown lookup type in a host list could cause Exim to panic-die when
5462 the list was checked. (An example that provoked this was putting <; in the
5463 middle of a list instead of at the start.) If this happened during a DATA
5464 ACL check, a -D file could be left lying around. This kind of configuration
5465 error no longer causes Exim to die; instead it causes a defer errror. The
5466 incident is still logged to the main and panic logs.
5468 74. Buglet left over from Exim 3 conversion. The message "too many messages
5469 in one connection" was written to the rejectlog but not the mainlog, except
5470 when address rewriting (yes!) was being logged.
5472 75. Added write_rejectlog option.
5474 76. When a system filter was run not as root (that is, when system_filter_user
5475 was set), the values of the $n variables were not being returned to the
5476 main process; thus, they were not subsequently available in the $sn
5479 77. Added +return_path_on_delivery log selector.
5481 78. A connection timeout was being treated differently from recipients deferred
5482 when testing hosts_max_try with a message that was older than the host's
5483 retry timeout. (The host should not be counted, thus allowing all hosts to
5484 be tried at least once before bouncing.) This may have been the cause of an
5485 occasionally reported bug whereby a message would remain on the queue
5486 longer than the retry timeout, but would be bounced if a delivery was
5487 forced. I say "may" because I never totally pinned down the problem;
5488 setting up timeout/retry tests is difficult. See also the next item.
5490 79. The ultimate address timeout was not being applied to errors that involved
5491 a combination of host plus message (for example, a timeout on a MAIL
5492 command). When an address resolved to a number of possible hosts, and they
5493 were not all tried for each delivery (e.g. because of hosts_max_try), a
5494 message could remain on the queue longer than the retry timeout.
5496 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5499 81. Fixed an obscure SMTP outgoing bug which required at least the following
5500 conditions: (a) there was another message waiting for the same server;
5501 (b) the server returned 5xx to all RCPT commands in the first message so
5502 that the message was not completed; (c) the server dropped the connection
5503 or gave a negative response to the RSET that Exim sends to abort the
5504 transaction. The observed case was a dropped connection after DATA that had
5505 been sent in pipelining mode. That is, the server had advertised PIPELINING
5506 but was not implementing it correctly. The effect of the bug was incorrect
5507 behaviour, such as trying another host, and this could lead to a crash.
5513 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5514 and daemon.c were passed as pointers to ints; they should have been
5515 pointers to socklen_t variables (which are typically unsigned ints).
5517 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5520 3. Fixed a really odd bug that affected only the testing scheme; patching a
5521 certain fixed string in the binary changed the value of another string that
5522 happened to be identical to the end of the original first string.
5524 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5525 name", it returns that address as the IP address. On some operating
5526 systems (e.g. Solaris), it also passes back the IP address string as the
5527 "host name". However, on others (e.g. Linux), it passes back an empty
5528 string. Exim wasn't checking for this, and was changing the host name to an
5529 empty string, assuming it had been canonicized.
5531 5. Although rare, it is permitted to have more than one PTR record for a given
5532 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5533 all the names associated with an address, because they do in Solaris.
5534 However, it seems that they do not in Linux for data that comes from the
5535 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5536 I found this out when I moved to a new Linux workstation and tried to run
5537 the Exim test suite.
5539 To get round this problem I have changed the code so that it now does its
5540 own call to the DNS to look up PTR records when searching for a host name.
5541 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5542 addresses that are only in /etc/hosts are still found.
5544 This behaviour is, however, controlled by an option called host_lookup_
5545 order, which defaults to "bydns:byaddr". If people want to use the other
5546 order, or indeed, just use one or the other means of lookup, they can
5547 specify it in this variable.
5549 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5550 some operating systems, this comes back from gethostbyaddr() as an empty
5551 string, and this is what Exim used to test for. However, it seems that in
5552 other systems, "." is yielded. Exim now tests for this case too.
5554 7. The values of check_spool_space and check_log_space are now held internally
5555 as a number of kilobytes instead of an absolute number of bytes. If a
5556 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5557 kilobyte. This means that much larger values can be stored.
5559 8. Exim monitor: an attempt to get the action menu when not actually pointing
5560 at a message produces an empty menu entitled "No message selected". This
5561 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5562 no entries in it ("Shell widget menu has zero width and/or height"). So I
5563 have added a single, blank menu entry in this case.
5565 9. Added ${quote_local_part.
5567 10. MIME decoding is now applied to the contents of Subject: header lines when
5570 11. Now that a reference to $sender_host_address automatically causes a reverse
5571 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5572 host lookup before query-style lookups in lists that might use this
5573 variable. This has therefore been abolished, and the "net-" prefix is no
5574 longer necessary for query-style lookups.
5576 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5577 have been a typo for LFLAGS, so it has been changed.
5579 13. The install script calls Exim with "-C /dev/null" in order to find the
5580 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5581 to be output. Howeve, since Exim outputs its version number before the
5582 error, it didn't break the script. It just looked ugly. I fixed this by
5583 always allowing "-C /dev/null" if the caller is root.
5585 14. Ignore overlarge ACL variable number when reading spool file - insurance
5586 against a later release with more variables having written the file.
5588 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5589 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5590 this, as well as the form without the "IPv6" on the front.
5592 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5593 OS/Makefile-Darwin file.
5595 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5596 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5598 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5599 variable. However, some OS use ints for inodes. I've added cast to long int
5600 to get rid of the compiler warning.
5602 19. I had forgotten to lock out "/../" in configuration file names when
5603 ALT_CONFIG_PREFIX was set.
5605 20. Routers used for verification do not need to specify transports. However,
5606 if such a router generated a host list, and callout was configured, Exim
5607 crashed, because it could not find a port number from the (non-existent)
5608 transport. It now assumes port 25 in this circumstance.
5610 21. Added the -t option to exigrep.
5612 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5613 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5614 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5615 them is set, LOOKUP_LSEARCH is forced.
5617 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5619 24. Added sender and host information to the "rejected by local_scan()" log
5620 line; previously there was no indication of these.
5622 25. Added .include_if_exists.
5624 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5625 Linux. It turns out that not all file systems support this. Apparently some
5626 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5627 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5628 not supported on the file descriptor, is now ignored when Exim is trying to
5629 sync a directory. This applies only to Linux.
5631 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5633 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5634 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5637 29. Added support for maildirsize files from supplied patch (modified a bit).
5639 30. The use of :fail: followed by an empty string could lead Exim to respond to
5640 sender verification failures with (e.g.):
5642 550 Verification failed for <xxx>
5643 550 Sender verify failed
5645 where the first response line was missing the '-' that indicates it is not
5646 the final line of the response.
5648 31. The loop for finding the name of the user that called Exim had a hardwired
5649 limit of 10; it now uses the value of finduser_retries, which is used for
5650 all other user lookups.
5652 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5655 33. Exim was neglecting to zero errno before one call of strtol() when
5656 expanding a string and expecting an integer value. On some systems this
5657 resulted in spurious "integer overflow" errors. Also, it was casting the
5658 result into an int without checking.
5660 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5661 did not work. The code looks as if it has *never* worked, though it appears
5662 to have been documented since at least releast 1.62. I have made it work.
5664 35. The "timeout_DNS" error in retry rules, also documented since at least
5665 1.62, also never worked. As it isn't clear exactly what this means, and
5666 clearly it isn't a major issue, I have abolished the feature by treating it
5667 as "timeout", and writing a warning to the main and panic logs.
5669 36. The display of retry rules for -brt wasn't always showing the error code
5672 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5673 timeout_connect_A, timeout_connect_MX.
5675 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5676 to the empty sender.
5678 39. The daemon was not analysing the content of -oX till after it had closed
5679 stderr and disconnected from the controlling terminal. This meant that any
5680 syntax errors were only noted on the panic log, and the return code from
5681 the command was 0. By re-arranging the code a little, I've made the
5682 decoding happen first, so such errors now appear on stderr, and the return
5683 code is 1. However, the actual setting up of the sockets still happens in
5684 the disconnected process, so errors there are still only recorded on the
5687 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5688 connections (as happens on some IP stacks) was logged at start up time as
5689 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5690 it from "IPv6 and IPv4", which means that two separate sockets are being
5693 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5694 says whether AF_INET or AF_INET6 was passed as an argument.
5696 42. Exiwhat output was messed up when time zones were included in log
5699 43. Exiwhat now gives more information about the daemon's listening ports,
5700 and whether -tls-on-connect was used.
5702 44. The "port" option of the smtp transport is now expanded.
5704 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5705 silently ignored. Now an error message is written to the main and panic
5708 46. There's a new ACL modifier called "logwrite" which writes to a log file
5709 as soon as it is encountered.
5711 47. Added $local_user_uid and $local_user_gid at routing time.
5713 48. Exim crashed when trying to verify a sender address that was being
5716 49. Exim was recognizing only a space character after ".include". It now also
5717 recognizes a tab character.
5719 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5720 extracting the relevant information from the specification. The man page no
5721 longer contains scrambled data for the -d option, and I've added a section
5722 at the front about calling Exim under different names.
5724 51. Added "extra_headers" argument to the "mail" command in filter files.
5726 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5729 53. Installed eximstats 1.29.
5731 54. Added transport_filter_timeout as a generic transport option.
5733 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5734 Cc: header lines. This was required by RFC 822, but it not required by RFC
5737 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5738 incoming messages that did not have them. Now it does so only if the
5739 message originates locally, that is, if there is no associated remote host
5740 address. When Resent- header lines are present, this applies to the Resent-
5741 lines rather than the non-Resent- lines.
5743 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5744 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5746 58. Messages for configuration errors now include the name of the main
5747 configuration file - useful now that there may be more than one file in a
5748 list (.included file names were always shown).
5750 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5751 for those rare installations that do not start the daemon as root or run it
5752 setuid root. I've cut out the call to initgroups() if the daemon is not
5755 60. The Exim user and group can now be bound into the binary as text strings
5756 that are looked up at the start of Exim's processing.
5758 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5760 62. Added $mailstore_basename variable.
5762 63. Installed patch to sieve.c from Michael Haardt.
5764 64. When Exim failed to open the panic log after failing to open the main log,
5765 the original message it was trying to log was written to stderr and debug
5766 output, but if they were not available (the usual case in production), it
5767 was lost. Now it is written to syslog before the two lines that record the
5768 failures to open the logs.
5770 65. Users' Exim filters run in subprocesses under the user's uid. It is
5771 possible for a "deliver" command or an alias in a "personal" command to
5772 provoke an address rewrite. If logging of address rewriting is configured,
5773 this fails because the process is not running as root or exim. There may be
5774 a better way of dealing with this, but for the moment (because 4.30 needs
5775 to be released), I have disabled address rewrite logging when running a
5776 filter in a non-root, non-exim process.
5782 1. The buildconfig auxiliary program wasn't quoting the value set for
5783 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5784 not defined. This bug was masked in 4.22 by the effect that was fixed in
5787 2. Some messages that were rejected after a message id was allocated were
5788 shown as "incomplete" by exigrep. It no longer does this for messages that
5789 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5791 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5792 have allow_domain_literals set, the ID did not get logged in the <= line.
5793 Domain literals are now always recognized in Message-ID: header lines.
5795 4. The first argument for a ${extract expansion item is the key name or field
5796 number. Leading and trailing spaces in this item were not being ignored,
5797 causing some misleading effects.
5799 5. When deliver_drop_privilege was set, single queue runner processes started
5800 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5801 same command in the process it spins off) were not dropping privilege.
5803 6. When the daemon running as "exim" started a queue runner, it always
5804 re-executed Exim in the spun-off process. This is a waste of effort when
5805 deliver_drop_privilege is set. The new process now just calls the
5806 queue-runner function directly.
5812 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5815 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5816 ignored. Though the use of -oP was forcing the writing of a pid file, it
5817 was always written to the default place.
5819 3. If the message "no IP address found for host xxxx" is generated during
5820 incoming verification, it is now followed by identification of the incoming
5821 connection (so you can more easily find what provoked it).
5823 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5825 5. Added some features to "harden" Exim a bit more against certain attacks:
5827 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5828 be put in Local/Makefile. This is like the never_users runtime option,
5829 but it cannot be overridden. The default setting is "root".
5831 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5832 prefix string with which any file named in a -C command line option
5835 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5836 is retained for -C and -D only if the caller of Exim is root. Without
5837 it, the exim user may also use -C and -D and retain privilege.
5839 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5840 command line option is disabled.
5842 6. Macro names set by the -D option must start with an upper case letter, just
5843 like macro names defined in the configuration file.
5845 7. Added "dereference=" facility to LDAP.
5847 8. Two instances of the typo "uknown" in the source files are fixed.
5849 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5850 the Configure-Makefile script screwed up while processing it.
5852 10. Incorporated PCRE 4.4.
5854 11. The SMTP synchronization check was not operating right at the start of an
5855 SMTP session. For example, it could not catch a HELO sent before the client
5856 waited for the greeting. There is now a check for outstanding input at the
5857 point when the greeting is written. Because of the duplex, asynchronous
5858 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5859 way, but not yet received, when the check is performed.
5861 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5862 on TCP/IP sockets, because this apparently causes some broken clients to
5865 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5866 unchanged) from the Cygwin maintainer.
5868 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5869 was supported instead of testing for mbx. Effectively a typo.
5871 15. The spa authenticator server code was not checking that the input it
5872 received was valid base64.
5874 16. The debug output line for the "set" modifier in ACLs was not showing the
5875 name of the variable that was being set.
5877 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5879 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5880 lookup on its own. Something else was needed to trigger the lookup. For
5881 example, a match in host_lookup or the need for a host name in a host list.
5882 Now, if $sender_host_name is referenced and the host name has not yet been
5883 looked up, a lookup is performed. If the lookup fails, the variable remains
5884 empty, and $host_lookup_failed is set to "1".
5886 19. Added "eqi" as a case-independent comparison operator.
5888 20. The saslauthd authentication condition could segfault if neither service
5889 nor realm was specified.
5891 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5892 error message that was logged was misleading, and incoming SMTP
5893 connections were dropped. The message is now more accurate, and temporary
5894 errors are given to SMTP connections.
5896 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5897 (except RSET) with a 421 error, until QUIT is received. However, it was
5898 failing to send a response to QUIT.
5900 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5901 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5902 ACL is now run after the helo_try_verify_hosts code.
5904 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5905 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5906 case-independent, so other case variants are also recognized). Apparently
5907 some systems use these upper case variants.
5909 25. If more than two messages were waiting for the same host, and a transport
5910 filter was specified for the transport, Exim sent two messages over the
5911 same TCP/IP connection, and then failed with "socket operation on non-
5912 socket" when it tried to send the third.
5914 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5916 27. The extern definition of crypt16() in expand.c was not being excluded when
5917 the OS had its own crypt16() function.
5919 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5920 as a preferred synonym for return_size_limit, both as an option and as an
5923 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5925 30. Changed the default configuration ACL to relax the local part checking rule
5926 for addresses that are not in any local domains. For these addresses,
5927 slashes and pipe symbols are allowed within local parts, but the sequence
5928 /../ is explicitly forbidden.
5930 31. SPA server authentication was not clearing the challenge buffer before
5933 32. log_message in a "warn" ACL statement was writing to the reject log as
5934 well as to the main log, which contradicts the documentation and doesn't
5935 seem right (because no rejection is happening). So I have stopped it.
5937 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5938 However, I am unable to do any testing of this.
5940 34. Fixed an infelicity in the appendfile transport. When checking directories
5941 for a mailbox, to see if any needed to be created, it was accidentally
5942 using path names with one or more superfluous leading slashes; tracing
5943 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5945 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5946 contents are added to the log line that is written for every discarded
5947 recipient. (Previously a log_message setting was ignored.)
5949 36. The ${quote: operator now quotes the string if it is empty.
5951 37. The install script runs exim in order to find its version number. If for
5952 some reason other than non-existence or emptiness, which it checks, it
5953 could not run './exim', it was installing it with an empty version number,
5954 i.e. as "exim-". This error state is now caught, and the installation is
5957 38. An argument was missing from the function that creates an error message
5958 when Exim fails to connect to the socket for saslauthd authentication.
5959 This could cause Exim to crash, or give a corrupted message.
5961 39. Added isip, isip4, and isip6 to ${if conditions.
5963 40. The ACL variables $acl_xx are now saved with the message, and can be
5964 accessed later in routers, transports, and filters.
5966 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5967 strings in the file are not string-expanded.
5969 42. If a MAIL command specified a SIZE value that was too large to fit into an
5970 int variable, the check against message_size_limit failed. Such values are
5971 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5972 day this will have to be increased, but I don't think I want to be around
5973 when emails are that large.
5980 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5981 iconv() is not standard in FreeBSD.
5983 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5984 IPv6 enabled. The observed symptom was a segmentation fault on return from
5985 the function os_common_find_running_interfaces() in src/os.c.
5987 3. In the check_special_case() function in daemon.c I had used "errno" as an
5988 argument name, which causes warnings on some systems. This was basically a
5989 typo, since it was named "eno" in the comments!
5991 4. The code that waits for the clock to tick (at a resolution of some fraction
5992 of a second) so as to ensure message-id uniqueness was always waiting for
5993 at least one whole tick, when it could have waited for less. [This is
5994 almost certainly not relevant at current processor speeds, where it is
5995 unlikely to ever wait at all. But we try to future-proof.]
5997 5. The function that sleeps for a time interval that includes fractions of a
5998 second contained a race. It did not block SIGALRM between setting the
5999 timer, and suspending (a couple of lines later). If the interval was short
6000 and the sigsuspend() was delayed until after it had expired, the suspension
6001 never ended. On busy systems this could lead to processes getting stuck for
6004 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6005 process, before it forks any delivery processes. The open lookup caching
6006 mechanism meant that the open file or database connection was passed into
6007 the delivery process. The problem was that delivery processes always tidy
6008 up cached lookup data. This could cause a problem for the next delivery
6009 process started by the queue runner, because the external queue runner
6010 process does not know about the closure. So the next delivery process
6011 still has data in the lookup cache. In the case of a file lookup, there was
6012 no problem because closing a file descriptor in a subprocess doesn't affect
6013 the parent. However, if the lookup was caching a connection to a database,
6014 the connection was closed, and the second delivery process was likely to
6015 see errors such as "PGSQL: query failed: server closed the connection
6016 unexpectedly". The problem has been fixed by closing all cached lookups
6017 in a queue runner before running a delivery process.
6019 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6020 seem to have the "const" qualifier which it has on other OS. I've
6023 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6024 *of the same type* (client or server) with the same public name that an
6025 error should be diagnosed.
6027 9. When Exim looked up a host name for an IP address, but failed to find the
6028 original IP address when looking up the host name (a safety check), it
6029 output the message "<ip address> does not match any IP for NULL", which was
6030 confusing, to say the least. The bug was that the host name should have
6031 appeared instead of "NULL".
6033 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6034 user that is built into the binary, and the -C or -D options is used, root
6035 privilege is dropped before the configuration file is read. In addition,
6036 logging is switched to stderr instead of the normal log files. If the
6037 configuration then re-defines the Exim user, the unprivileged environment
6038 is probably not what is expected, so Exim logs a panic warning message (but
6041 However, if deliver_drop_privilege is set, the unprivileged state may well
6042 be exactly what is intended, so the warning has been cut out in that case,
6043 and Exim is allowed to try to write to its normal log files.
6049 1. smtp_return_error_details was not giving details for temporary sender
6050 or receiver verification errors.
6052 2. Diagnose a configuration error if two authenticators have the same public
6055 3. Exim used not to create the message log file for a message until the first
6056 delivery attempt. This could be confusing when incoming messages were held
6057 for policy or load reasons. The message log file is now created at the time
6058 the message is received, and an initial "Received" line is written to it.
6060 4. The automatically generated man page for command line options had a minor
6061 bug that caused no ill effects; however, a more serious problem was that
6062 the procedure for building the man page automatically didn't always
6063 operate. Consequently, release 4.20 contains an out-of-date version. This
6064 shouldn't happen again.
6066 5. When building Exim with embedded Perl support, the script that builds the
6067 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6068 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6070 6. The freeze_tell option was not being used for messages that were frozen on
6071 arrival, either by an ACL or by local_scan().
6073 7. Added the smtp_incomplete_transaction log selector.
6075 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6076 was accepting AUTH without a new EHLO.
6078 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6079 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6082 10. Logging of TCP/IP connections (when configured) now happens in the main
6083 daemon process instead of the child process, so that the TCP/IP connection
6084 count is more accurate (but it can never be perfect).
6086 11. The use of "drop" in a nested ACL was not being handled correctly in the
6087 outer ACL. Now, if condition failure induced by the nested "drop" causes
6088 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6089 or "require"), the connection is dropped.
6091 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6092 that yield "discard" can now be used with an "accept" or a "discard" verb,
6093 but an error is generated for any others (because I can't see a useful way
6094 to define what should happen).
6096 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6097 are now processed in the same way as lines in the Exim configuration file.
6098 In particular, continuation lines are supported.
6100 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6102 15. Added -ti meaning -t -i.
6104 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6105 domains, and warn by logging if others are found.
6107 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6108 returned from GIFCONF when Exim is trying to find the list of interfaces on
6109 a host. The code in os.c has been modified to copy each ifreq to an aligned
6110 structure in all cases.
6112 Also, in some cases, the returned ifreq's were being copied to a 'struct
6113 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6114 means the last couple of bytes of an IPv6 address could be chopped if the
6115 ifreq contained only a normal sockaddr (14 bytes storage).
6117 18. Named domain lists were not supported in the hosts_treat_as_local option.
6118 An entry such as +xxxx was not recognized, and was treated as a literal
6121 19. Ensure that header lines added by a DATA ACL are included in the reject log
6122 if the ACL subsequently rejects the message.
6124 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6125 MD5 (which is deprecated).
6127 21. When testing a filter file using -bf, Exim was writing a message when it
6128 took the sender from a "From " line in the message, but it was not doing so
6129 when it took $return_path from a Return-Path: header line. It now does.
6131 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6132 with a valid header line field name (a series of printing characters
6133 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6135 23. Changed "disc" in the source to "disk" to conform to the documentation and
6136 the book and for uniformity.
6138 24. Ignore Sendmail's -Ooption=value command line item.
6140 25. When execve() failed while trying to run a command in a pipe transport,
6141 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6142 could be confused with a return value of 69 from the command itself. This
6143 has been changed to 127, the value the shell returns if it is asked to run
6144 a non-existent command. The wording for the related log line suggests a
6145 non-existent command as the problem.
6147 26. If received_header_text expands to an empty string, do not add a Received:
6148 header line to the message. (Well, it adds a token one on the spool, but
6149 marks it "old" so that it doesn't get used or transmitted.)
6151 27. Installed eximstats 1.28 (addition of -nt option).
6153 28. There was no check for failure on the call to getsockname() in the daemon
6154 code. This can fail if there is a shortage of resources on the system, with
6155 ENOMEM, for example. A temporary error is now given on failure.
6157 29. Contrary to the C standard, it seems that in some environments, the
6158 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6159 program. Exim now does this explicitly; it affects the formatting of
6160 timestamps using strftime().
6162 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6163 complaints. I've now initialized all the variables, to avoid this.
6165 32. Header lines added by a system filter were not being "seen" during
6166 transport-time rewrites.
6168 33. The info_callback() function passed to OpenSSL is set up with type void
6169 (*)(SSL *, int, int), as described somewhere. However, when calling the
6170 function (actually a macro) that sets it up, the type void(*)() is
6171 expected. I've put in a cast to prevent warnings from picky compilers.
6173 34. If a DNS black list lookup found a CNAME record, but there were no A
6174 records associated with the domain it pointed at, Exim crashed.
6176 35. If a DNS black list lookup returned more than one A record, Exim ignored
6177 all but the first. It now scans all returned addresses if a particular IP
6178 value is being sought. In this situation, the contents of the
6179 $dnslist_value variable are a list of all the addresses, separated by a
6182 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6183 to accept a host name and all its aliases if the forward lookup for any of
6184 them yielded the IP address of the incoming connection. Now it accepts only
6185 those names whose forward lookup yields the correct IP address. Any other
6186 names are discarded. This closes a loophole whereby a rogue DNS
6187 administrator could create reverse DNS records to break through a
6188 wildcarded host restriction in an ACL.
6190 37. If a user filter or a system filter that ran in a subprocess used any of
6191 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6192 the wrong values were passed to the pipe command ($thisaddress had the
6193 value of $0, $0 had the value of $1, etc). This bug was introduced by
6194 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6196 38. Improved the line breaking for long SMTP error messages from ACLs.
6197 Previously, if there was no break point between 40 and 75 characters, Exim
6198 left the rest of the message alone. Two changes have been made: (a) I've
6199 reduced the minimum length to 35 characters; (b) if it can't find a break
6200 point between 35 and 75 characters, it looks ahead and uses the first one
6201 that it finds. This may give the occasional overlong line, but at least the
6202 remaining text gets split now.
6204 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6205 file descriptors might be low, and that setting 1000 would always raise it.
6206 It turns out that in some environments, the limit is already over 1000 and
6207 that lowering it causes trouble. So now Exim takes care not to decrease it.
6209 40. When delivering a message, the value of $return_path is set to $sender_
6210 address at the start of routing (routers may change the value). By an
6211 oversight, this default was not being set up when an address was tested by
6212 -bt or -bv, which affected the outcome if any router or filter referred to
6215 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6216 log only when "message" or "log_message" are set. However, if one of the
6217 conditions was an address verification, or a call to a nested ACL, the
6218 messages generated by the underlying test were being passed through. This
6219 no longer happens. The underlying message is available in $acl_verify_
6220 message for both "message" and "log_message" expansions, so it can be
6221 passed through if needed.
6223 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6224 new expansion $bh_ to give the encoded byte string without charset
6225 translation. Translation happens only if iconv() is available; HAVE_ICONV
6226 indicates this at build time. HEADERS_CHARSET gives the charset to
6227 translate to; headers_charset can change it in the configuration, and
6228 "headers charset" can change it in an individual filter file.
6230 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6231 that creates RFC 2047 encoded "words" labels them as that charset instead
6232 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6233 expansion operator; (ii) when Exim creates a From: line for a local
6234 message; (iii) when a header line is rewritten to include a "phrase" part.
6236 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6237 buggy, causing it to skip the first lines of messages whose message ID
6238 ended in 'D'. This would not have bitten before Exim release 4.14, because
6239 message IDs were unlikely to end in 'D' before then. The effect was to have
6240 incorrect size information for certain domains.
6242 45. #include "config.h" was missing at the start of the crypt16.c module. This
6243 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6246 46. If there was a timeout during a "random" callout check, Exim treated it as
6247 a failure of the random address, and carried on sending RSET and the real
6248 address. If the delay was just some slowness somewhere, the response to the
6249 original RCPT would be taken as a response to RSET and so on, causing
6250 mayhem of various kinds.
6252 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6253 when I implemented it. It didn't allow for the fact that some option values
6254 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6255 the right test for positive values.
6257 48. Domain names in DNS records are case-independent. Exim always looks them up
6258 in lower case. Some resolvers return domain names in exactly the case they
6259 appear in the zone file, that is, they may contain uppercase letters. Not
6260 all resolvers do this - some return always lower case. Exim was treating a
6261 change of case by a resolver as a change of domain, similar to a widening
6262 of a domain abbreviation. This triggered its re-routing code and so it was
6263 trying to route what was effectively the same domain again. This normally
6264 caused routing to fail (because the router wouldn't handle the domain
6265 twice). Now Exim checks for this case specially, and just changes the
6266 casing of the domain that it ultimately uses when it transmits the message
6269 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6272 50. If a filter generated a file delivery with a non-absolute name (possible if
6273 no home directory exists for the router), the forbid_file option was not
6276 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6277 the existing equality matching.
6279 52. Exim was using ints instead of ino_t variables in some places where it was
6280 dealing with inode numbers.
6282 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6283 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6284 called TMPDIR, and if it finds it is different, it changes its value.
6286 54. The smtp_printf() function is now made available to local_scan() so
6287 additional output lines can be written before returning. There is also an
6288 smtp_fflush() function to enable the detection of a dropped connection.
6289 The variables smtp_input and smtp_batched_input are exported to
6292 55. Changed the default runtime configuration: the message "Unknown user"
6293 has been removed from the ACL, and instead placed on the localuser router,
6294 using the cannot_route_message feature. This means that any verification
6295 failures that generate their own messages won't get overridden. Similarly,
6296 the "Unrouteable address" message that was in the ACL for unverifiable
6297 relay addresses has also been removed.
6299 56. Added hosts_avoid_esmtp to the smtp transport.
6301 57. The exicyclog script was not checking for the esoteric option
6302 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6303 will work only if exicyclog is run under the appropriate euid.
6305 58. Following a discussion on the list, the rules by which Exim recognises line
6306 endings on incoming messages have been changed. The -dropcr and drop_cr
6307 options are now no-ops, retained only for backwards compatibility. The
6308 following line terminators are recognized: LF CRLF CR. However, special
6309 processing applies to CR:
6311 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6312 nor a local message in the state where . is a terminator.
6314 (ii) If a bare CR is encountered in a header line, an extra space is added
6315 after the line terminator so as not to end the header. The reasoning
6316 behind this is that bare CRs in header lines are most likely either
6317 to be mistakes, or people trying to play silly games.
6319 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6320 was being incorrectly given as 18 bytes larger than it should have been.
6321 This is a VOB (very old bug).
6323 60. This may never have affected anything current, but just in case it has:
6324 When the local host is found other than at the start of a list of hosts,
6325 the local host, those with the same MX, and any that follow, are discarded.
6326 When the list in question was part of a longer list of hosts, the following
6327 hosts (not currently being processed) were also being discarded. This no
6328 longer happens. I'm not sure if this situation could ever has previously
6331 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6334 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6335 header. This is something that is recommended in a new Internet Draft, and
6336 is something that is documented as being done by Sendmail. There are two
6337 possible values. For messages generated by the autoreply transport, Exim
6340 Auto-Submitted: auto-replied
6342 whereas for all other generated messages (e.g. bounces) it adds
6344 Auto-Submitted: auto-generated
6346 63. The "personal" condition in filters now includes a test for the
6347 Auto-Submitted: header. If it contains the string "auto-" the message it
6348 not considered personal.
6350 64. Added rcpt_include_affixes as a generic transport option.
6352 65. Added queue_only_override (default true).
6354 66. Added the syslog_duplication option.
6356 67. If what should have been the first header line of a message consisted of
6357 a space followed by a colon, Exim was mis-interpreting it as a header line.
6358 It isn't of course - it is syntactically invalid and should therefore be
6359 treated as the start of the message body. The misbehaviour could have
6360 caused a number of strange effects, including loss of data in subsequent
6361 header lines, and spool format errors.
6363 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6364 client host had authenticated. This control can now be exercised by an ACL
6365 for more flexibility.
6367 69. By default, callouts do not happen when testing with -bh. There is now a
6368 variant, -bhc, which does actually run the callout code, including
6369 consulting and updating the callout cache.
6371 70. Added support for saslauthd authentication, courtesy of Alexander
6374 71. If statvfs() failed on the spool or log directories while checking their
6375 size for availability, Exim confusingly gave the error "space shortage".
6376 Furthermore, in debugging mode it crashed with a floating point exception.
6377 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6378 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6379 serious problem, Exim now writes to the main and panic logs when this
6380 happens, with details of the failure. It then refuses to accept the
6381 incoming message, giving the message "spool directory problem" or "log
6382 directory problem" with a 421 code for SMTP messages.
6384 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6385 0, 1, and 2 exist, because some OS complain for execs without them (see
6386 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6387 descriptors. However, the code omitted to check that the open succeeded,
6388 causing mysterious errors if for some reason the permissions on /dev/null
6389 got screwed. Now Exim writes a message to the main and panic logs, and
6390 bombs out if it can't open /dev/null.
6392 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6393 interact so that it is all more flexible. It is supposed to remain
6394 backwards compatible. Also added extra_local_interfaces.
6396 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6397 to bomb out with an assertion failure - to the client this appears as a
6398 connection drop. This problem occurs in the part of the code that was taken
6399 from the Samba project. Fortunately, the assertion is in a very simple
6400 function, so I have fixed this by reproducing the function inline in the
6401 one place where it is called, and arranging for authentication to fail
6402 instead of killing the process with assert().
6404 75. The SPA client code was not working when the server requested OEM rather
6405 than Unicode encoding.
6407 76. Added code to make require_files with a specific uid setting more usable in
6408 the case where statting the file as root fails - usually a non-root-mounted
6409 NFS file system. When this happens and the failure is EACCES, Exim now
6410 forks a subprocess and does the per-uid checking as the relevant uid.
6412 77. Added process_log_path.
6414 78. If log_file_path was not explicitly set, a setting of check_log_space or
6415 check_log_inodes was ignored.
6417 79. If a space check for the spool or log partitions fails, the incident is now
6418 logged. Of course, in the latter case the data may get lost...
6420 80. Added the %p formatting code to string_format() so that it can be used to
6421 print addresses in debug_print(). Adjusted all the address printing in the
6422 debugging in store.c to use %p rather than %d.
6424 81. There was a concern that a line of code in smtp_in.c could overflow a
6425 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6426 initially expressed, the concern was not well-founded, because trailing
6427 spaces are removed early. However, if the trailing spaces were followed by
6428 a NULL, they did not get removed, so the overflow was possible. Two fixes
6431 (a) I re-wrote the offending code in a cleaner fashion.
6432 (b) If an incoming SMTP command contains a NULL character, it is rejected
6435 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6436 runs initgroups(), so that if the Exim user is in any additional groups,
6437 they will be used during message reception.
6443 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / jgh / exim.git / blob