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.
140 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
143 JH/02 The smtp transport option "multi_domain" is now expanded.
145 JH/03 The smtp transport now requests PRDR by default, if the server offers
148 JH/04 Certificate name checking on server certificates, when exim is a client,
149 is now done by default. The transport option tls_verify_cert_hostnames
150 can be used to disable this per-host. The build option
151 EXPERIMENTAL_CERTNAMES is withdrawn.
153 JH/05 The value of the tls_verify_certificates smtp transport and main options
154 default to the word "system" to access the system default CA bundle.
155 For GnuTLS, only version 3.0.20 or later.
157 JH/06 Verification of the server certificate for a TLS connection is now tried
158 (but not required) by default. The verification status is now logged by
159 default, for both outbound TLS and client-certificate supplying inbound
162 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
165 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
166 Status Notification (bounce) messages are now MIME format per RFC 3464.
167 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
168 under the control of the dsn_advertise_hosts option, and routers may
169 have a dsn_lasthop option.
171 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
172 default, modifiable by a malware= option. The list separator for
173 the options can now be changed in the usual way. Bug 68.
175 JH/10 The smtp_receive_timeout main option is now expanded before use.
177 JH/11 The incoming_interface log option now also enables logging of the
178 local interface on delivery outgoing connections.
180 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
181 if the interface and destination host and port all match.
183 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
186 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
187 Patch from Andrew Lewis.
189 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
190 now supports optional time-restrictions, weighting, and priority
191 modifiers per server. Patch originally by <rommer@active.by>.
193 JH/16 The spamd_address main option now supports a mixed list of local
194 and remote servers. Remote servers can be IPv6 addresses, and
195 specify a port-range.
197 JH/17 Bug 68: The spamd_address main option now supports an optional
198 timeout value per server.
200 JH/18 Bug 1581: Router and transport options headers_add/remove can
201 now have the list separator specified.
203 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
206 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
209 JH/21 Support for the A6 type of dns record is withdrawn.
211 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
212 rather than the verbs used.
214 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
215 from 255 to 1024 chars.
217 JH/24 Verification callouts now attempt to use TLS by default.
219 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
220 are generic router options now. The defaults didn't change.
222 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
223 Original patch from Alexander Shikoff, worked over by JH.
225 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
226 files if ZCAT_COMMAND is not executable.
228 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
230 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
232 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
233 Normally benign, it bites when the pair was led to by a CNAME;
234 modern usage is to not canoicalize the domain to a CNAME target
235 (and we were inconsistent anyway for A-only vs AAAA+A).
237 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
239 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
240 when evaluating $sender_host_dnssec.
242 JH/31 Check the HELO verification lookup for DNSSEC, adding new
243 $sender_helo_dnssec variable.
245 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
247 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
249 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
251 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
252 documented as working, but never had. Support all but $spam_report.
254 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
255 added for tls authenticator.
260 TL/01 When running the test suite, the README says that variables such as
261 no_msglog_check are global and can be placed anywhere in a specific
262 test's script, however it was observed that placement needed to be near
263 the beginning for it to behave that way. Changed the runtest perl
264 script to read through the entire script once to detect and set these
265 variables, reset to the beginning of the script, and then run through
266 the script parsing/test process like normal.
268 TL/02 The BSD's have an arc4random API. One of the functions to induce
269 adding randomness was arc4random_stir(), but it has been removed in
270 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
271 function when detected.
273 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
274 cause callback expansion.
276 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
277 syntax errors in an expansion can be treated as a string instead of
278 logging or causing an error, due to the internal use of bool_lax
279 instead of bool when processing it.
281 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
282 server certificates when making smtp deliveries.
284 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
286 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
288 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
290 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
291 Merged patch from Sebastian Wiedenroth.
293 JH/05 Fix results-pipe from transport process. Several recipients, combined
294 with certificate use, exposed issues where response data items split
295 over buffer boundaries were not parsed properly. This eventually
296 resulted in duplicates being sent. This issue only became common enough
297 to notice due to the introduction of conection certificate information,
298 the item size being so much larger. Found and fixed by Wolfgang Breyha.
300 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
301 size buffer was used, resulting in syntax errors when an expansion
304 JH/07 Add support for directories of certificates when compiled with a GnuTLS
305 version 3.3.6 or later.
307 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
308 is EXPERIMENTAL_EVENT, the main-configuration and transport options
309 both become "event_action", the variables become $event_name, $event_data
310 and $event_defer_errno. There is a new variable $verify_mode, usable in
311 routers, transports and related events. The tls:cert event is now also
312 raised for inbound connections, if the main configuration event_action
315 TL/06 In test suite, disable OCSP for old versions of openssl which contained
316 early OCSP support, but no stapling (appears to be less than 1.0.0).
318 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
319 server certificate names available under the smtp transport option
320 "tls_verify_cert_hostname" now do not permit multi-component wildcard
323 JH/10 Time-related extraction expansions from certificates now use the main
324 option "timezone" setting for output formatting, and are consistent
325 between OpenSSL and GnuTLS compilations. Bug 1541.
327 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
328 encoded parameter in the incoming message. Bug 1558.
330 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
331 include certificate info, eximon was claiming there were spoolfile
334 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
336 JH/14 Log delivery-related information more consistently, using the sequence
337 "H=<name> [<ip>]" wherever possible.
339 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
340 are problematic for Debian distribution, omit them from the release
343 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
345 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
348 JH/17 Fix a null-indirection in certextract expansions when a nondefault
349 output list separator was used.
354 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
355 checkers that were complaining about end of non-void function with no
358 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
359 This was a regression intruduced in 4.83 by another bugfix.
361 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
363 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
364 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
370 TF/01 Correctly close the server side of TLS when forking for delivery.
372 When a message was received over SMTP with TLS, Exim failed to clear up
373 the incoming connection properly after forking off the child process to
374 deliver the message. In some situations the subsequent outgoing
375 delivery connection happened to have the same fd number as the incoming
376 connection previously had. Exim would try to use TLS and fail, logging
377 a "Bad file descriptor" error.
379 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
380 utilities have not been installed.
382 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
383 temporary space as the ACL may create new global variables.
385 TL/01 LDAP support uses per connection or global context settings, depending
386 upon the detected version of the libraries at build time.
388 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
389 to extract and use the src ip:port in logging and expansions as if it
390 were a direct connection from the outside internet. PPv2 support was
391 updated based on HAProxy spec change in May 2014.
393 JH/02 Add ${listextract {number}{list}{success}{fail}}.
395 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
396 Properly escape header and check for NULL return.
398 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
401 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
403 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
404 characters in header names, implemented as a verify condition.
405 Contributed by Michael Fischer v. Mollard.
407 TL/05 Rename SPF condition results err_perm and err_temp to standardized
408 results permerror and temperror. Previous values are deprecated but
409 still accepted. In a future release, err_perm and err_temp will be
410 completely removed, which will be a backward incompatibility if the
411 ACL tests for either of these two old results. Patch contributed by
412 user bes-internal on the mailing list.
414 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
416 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
417 selectors, in both main and reject logs.
419 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
422 JH/07 Add malware type "sock" for talking to simple daemon.
424 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
426 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
427 routers/transports under cutthrough routing.
429 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
430 numbers. Touch up "bool" conditional to keep the same definition.
432 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
434 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
436 JH/12 Expand items in router/transport headers_add or headers_remove lists
437 individually rather than the list as a whole. Bug 1452.
439 Required for reasonable handling of multiple headers_ options when
440 they may be empty; requires that headers_remove items with embedded
441 colons must have them doubled (or the list-separator changed).
443 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
444 view the policy declared in the DMARC record. Currently, $dmarc_status
445 is a combined value of both the record presence and the result of the
448 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
450 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
451 dnslookup router and the smtp transport (applying to the forward
454 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
455 of ldap servers used for a specific lookup. Patch provided by Heiko
458 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
459 New variable $lookup_dnssec_authenticated for observability.
461 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
462 Patch submitted by Lars Timman.
464 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
466 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
467 Requires trusted mode and valid format message id, aborts otherwise.
468 Patch contributed by Heiko Schlichting.
470 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
471 certextract with support for various fields. Bug 1358.
473 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
474 is requested by default, modifiable by smtp transport option
477 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
478 operate on certificate variables to give certificate fingerprints
479 Also new ${sha256:cert_variable}.
481 JH/23 The PRDR feature is moved from being Experimental into the mainline.
483 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
484 Christian Aistleitner.
486 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
488 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
489 file. Patch from Wolfgang Breyha.
491 JH/25 Expand the coverage of the delivery $host and $host_address to
492 client authenticators run in verify callout. Bug 1476.
494 JH/26 Port service names are now accepted for tls_on_connect_ports, to
495 align with daemon_smtp_ports. Bug 72.
497 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
498 support and error reporting did not work properly.
500 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
501 and is readable. Patch from Andrew Colin Kissa.
503 TL/14 Enhance documentation of ${run expansion and how it parses the
504 commandline after expansion, particularly in the case when an
505 unquoted variable expansion results in an empty value.
507 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
509 PP/02 Fix internal collision of T_APL on systems which support RFC3123
510 by renaming away from it. Addresses GH issue 15, reported by
513 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
515 TL/15 SECURITY: prevent double expansion in math comparison functions
516 (can expand unsanitized data). Not remotely exploitable.
523 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
525 PP/02 Make -n do something, by making it not do something.
526 When combined with -bP, the name of an option is not output.
528 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
531 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
532 $sender_host_name and config options to manage this, and basic check
535 PP/05 DSCP support for outbound connections and control modifier for inbound.
537 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
538 (Only plugin which currently uses this is kerberos4, which nobody should
539 be using, but we should make it available and other future plugins might
540 conceivably use it, even though it would break NAT; stuff *should* be
541 using channel bindings instead).
543 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
544 name; added for Sendmail compatibility; requires admin caller.
545 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
546 just ignore it); requires trusted caller.
547 Also parse but ignore: -Ac -Am -X<logfile>
550 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
552 TL/02 Add +smtp_confirmation as a default logging option.
554 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
555 Patch by Magnus Holmgren from 2007-02-20.
557 TL/04 Bugzilla 1281 - Spec typo.
558 Bugzilla 1283 - Spec typo.
559 Bugzilla 1290 - Spec grammar fixes.
561 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
563 TL/06 Add Experimental DMARC support using libopendmarc libraries.
565 TL/07 Fix an out of order global option causing a segfault. Reported to dev
566 mailing list by by Dmitry Isaikin.
568 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
570 JH/02 Support "G" suffix to numbers in ${if comparisons.
572 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
574 NM/01 Bugzilla 1197 - Spec typo
575 Bugzilla 1196 - Spec examples corrections
577 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
579 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
580 gnutls_enable_pkcs11, but renamed to more accurately indicate its
583 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
584 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
586 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
587 "acl {{name}{arg}...}", and optional args on acl condition
590 JH/05 Permit multiple router/transport headers_add/remove lines.
592 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
594 JH/07 Avoid using a waiting database for a single-message-only transport.
595 Performance patch from Paul Fisher. Bugzilla 1262.
597 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
600 JH/09 Add $headers_added variable, with content from use of ACL modifier
601 add_header (but not yet added to the message). Bugzilla 199.
603 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
604 Pulled from Bugzilla 817 by Wolfgang Breyha.
606 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
608 (nb: this is the same fix as in Exim 4.80.1)
610 JH/11 Add A= logging on delivery lines, and a client_set_id option on
613 JH/12 Add optional authenticated_sender logging to A= and a log_selector
616 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
618 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
619 advertise SMTP AUTH mechanism to us, instead of a generic
620 protocol violation error. Also, make Exim more robust to bad
621 data from the Dovecot auth socket.
623 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
625 When a queue runner is handling a message, Exim first routes the
626 recipient addresses, during which it prunes them based on the retry
627 hints database. After that it attempts to deliver the message to
628 any remaining recipients. It then updates the hints database using
631 So if a recipient address works intermittently, it can get repeatedly
632 deferred at routing time. The retry hints record remains fresh so the
633 address never reaches the final cutoff time.
635 This is a fairly common occurrence when a user is bumping up against
636 their storage quota. Exim had some logic in its local delivery code
637 to deal with this. However it did not apply to per-recipient defers
638 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
640 This change adds a proper retry rule check during routing so that the
641 final cutoff time is checked against the message's age. We only do
642 this check if there is an address retry record and there is not a
643 domain retry record; this implies that previous attempts to handle
644 the address had the retry_use_local_parts option turned on. We use
645 this as an approximation for the destination being like a local
646 delivery, as in LMTP.
648 I suspect this new check makes the old local delivery cutoff check
649 redundant, but I have not verified this so I left the code in place.
651 TF/02 Correct gecos expansion when From: is a prefix of the username.
653 Test 0254 submits a message to Exim with the header
657 When I ran the test suite under the user fanf2, Exim expanded
658 the header to contain my full name, whereas it should have added
659 a Resent-Sender: header. It erroneously treats any prefix of the
660 username as equal to the username.
662 This change corrects that bug.
664 GF/01 DCC debug and logging tidyup
665 Error conditions log to paniclog rather than rejectlog.
666 Debug lines prefixed by "DCC: " to remove any ambiguity.
668 TF/03 Avoid unnecessary rebuilds of lookup-related code.
670 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
671 Bug spotted by Jeremy Harris; was flawed since initial commit.
672 Would have resulted in OCSP responses post-SNI triggering an Exim
673 NULL dereference and crash.
675 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
677 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
678 Bug detection, analysis and fix by Samuel Thibault.
679 Bugzilla 1331, Debian bug #698092.
681 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
683 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
684 Server implementation by Todd Lyons, client by JH.
685 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
686 config variable "prdr_enable" controls whether the server
687 advertises the facility. If the client requests PRDR a new
688 acl_data_smtp_prdr ACL is called once for each recipient, after
689 the body content is received and before the acl_smtp_data ACL.
690 The client is controlled by bolth of: a hosts_try_prdr option
691 on the smtp transport, and the server advertisement.
692 Default client logging of deliveries and rejections involving
693 PRDR are flagged with the string "PRDR".
695 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
696 fclose(). Diagnosis by Todd Lyons.
698 PP/17 Update configure.default to handle IPv6 localhost better.
699 Patch by Alain Williams (plus minor tweaks).
702 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
703 This is now consistent with GnuTLS, and is now documented: the
704 previous undocumented portable approach to treating the option as
705 unset was to force an expansion failure. That still works, and
706 an empty string is now equivalent.
708 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
709 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
710 not performing validation itself.
712 PP/20 Added force_command boolean option to pipe transport.
713 Patch from Nick Koston, of cPanel Inc.
715 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
718 TF/04 Added udpsend ACL modifer and hexquote expansion operator
720 PP/21 Fix eximon continuous updating with timestamped log-files.
721 Broken in a format-string cleanup in 4.80, missed when I repaired the
722 other false fix of the same issue.
723 Report and fix from Heiko Schlichting.
726 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
727 Report from Prashanth Katuri.
729 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
730 It's SecureTransport, so affects any MacOS clients which use the
731 system-integrated TLS libraries, including email clients.
733 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
734 using a MIME ACL for non-SMTP local injection.
735 Report and assistance in diagnosis by Warren Baker.
737 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
739 JH/16 Fix comparisons for 64b. Bugzilla 1385.
741 TL/09 Add expansion variable $authenticated_fail_id to keep track of
742 last id that failed so it may be referenced in subsequent ACL's.
744 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
747 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
748 ldap library initialization, allowing self-signed CA's to be
749 used. Also properly sets require_cert option later in code by
750 using NULL (global ldap config) instead of ldap handle (per
751 session). Bug diagnosis and testing by alxgomz.
753 TL/12 Enhanced documentation in the ratelimit.pl script provided in
754 the src/util/ subdirectory.
756 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
757 renamed to Transport Post Delivery Action by Jeremy Harris, as
760 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
761 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
762 redis_servers = needs to be configured which will be used by the redis
763 lookup. Patch from Warren Baker, of The Packet Hub.
765 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
767 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
768 hostname or reverse DNS when processing a host list. Used suggestions
769 from multiple comments on this bug.
771 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
773 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
774 Missed a few lines, added it to make the runtest require no keyboard
777 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
778 contains upper case chars. Make router use caseful_local_part.
780 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
781 support when GnuTLS has been built with p11-kit.
787 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
789 This, or similar/improved, will also be change PP/11 of 4.82.
795 PP/01 Handle short writes when writing local log-files.
796 In practice, only affects FreeBSD (8 onwards).
797 Bugzilla 1053, with thanks to Dmitry Isaikin.
799 NM/01 Bugzilla 949 - Documentation tweak
801 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
804 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
806 PP/02 Implemented gsasl authenticator.
808 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
810 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
811 `pkg-config foo` for cflags/libs.
813 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
814 with rest of GSASL and with heimdal_gssapi.
816 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
817 `pkg-config foo` for cflags/libs for the TLS implementation.
819 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
820 properties get this fed in as external SSF. A number of robustness
821 and debugging improvements to the cyrus_sasl authenticator.
823 PP/08 cyrus_sasl server now expands the server_realm option.
825 PP/09 Bugzilla 1214 - Log authentication information in reject log.
826 Patch by Jeremy Harris.
828 PP/10 Added dbmjz lookup type.
830 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
832 PP/12 MAIL args handles TAB as well as SP, for better interop with
833 non-compliant senders.
834 Analysis and variant patch by Todd Lyons.
836 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
837 Bug report from Lars Müller <lars@samba.org> (via SUSE),
838 Patch from Dirk Mueller <dmueller@suse.com>
840 PP/13 tls_peerdn now print-escaped for spool files.
841 Observed some $tls_peerdn in wild which contained \n, which resulted
842 in spool file corruption.
844 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
845 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
846 or write after TLS renegotiation, which otherwise led to messages
849 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
850 as a tracking header (ie: a signed header comes before the signature).
851 Patch from Wolfgang Breyha.
853 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
854 comma-sep list; embedded commas doubled.
856 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
858 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
860 Report and patch from Dmitry Banschikov.
862 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
863 Removed SSL_clear() after SSL_new() which led to protocol negotiation
864 failures. We appear to now support TLS1.1+ with Exim.
866 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
867 lets Exim select keys and certificates based upon TLS SNI from client.
868 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
869 before an outbound SMTP session. New log_selector, +tls_sni.
871 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
872 NULL dereference. Report and patch from Alun Jones.
874 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
875 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
876 Not seeing resolver debug output on NetBSD, but suspect this is a
877 resolver implementation change.
879 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
880 Left warnings. Added "eximon gdb" invocation mode.
882 PP/21 Defaulting "accept_8bitmime" to true, not false.
884 PP/22 Added -bw for inetd wait mode support.
886 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
887 locate the relevant includes and libraries. Made this the default.
889 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
890 Bugzilla 1246, report and most of solution from Tomasz Kusy.
892 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
893 This may cause build issues on older platforms.
895 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
896 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
897 gnutls_require_mac & gnutls_require_protocols (no longer supported).
898 Added SNI support via GnuTLS too.
899 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
901 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
903 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
904 Applied second patch from Janne, implementing suggestion to default
905 multiple-strings-in-record handling to match SPF spec.
907 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
909 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
910 read-only, out of scope).
911 Patch from Wolfgang Breyha, report from Stuart Northfield.
913 PP/29 Fix three issues highlighted by clang analyser static analysis.
914 Only crash-plausible issue would require the Cambridge-specific
915 iplookup router and a misconfiguration.
916 Report from Marcin Mirosław.
918 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
920 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
921 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
922 As part of this, removing so much warning spew let me fix some minor
923 real issues in debug logging.
925 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
926 assignment on my part. Fixed.
928 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
929 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
930 Janne Snabb (who went above and beyond: thank you).
932 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
933 string otherwise requires a connection and a bunch more work and it's
934 relatively easy to get wrong. Should also expose TLS library linkage
937 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
938 64-bit ${eval} (JH/03).
940 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
941 GNU libc to support some of the 64-bit stuff, should not lead to
942 conflicts. Defined before os.h is pulled in, so if a given platform
943 needs to override this, it can.
945 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
946 protection layer was required, which is not implemented.
947 Bugzilla 1254, patch from Wolfgang Breyha.
949 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
950 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
951 tls_dhparam take prime identifiers. Also unbreak combination of
952 OpenSSL+DH_params+TLSSNI.
954 PP/39 Disable SSLv2 by default in OpenSSL support.
960 PP/01 Solaris build fix for Oracle's LDAP libraries.
961 Bugzilla 1109, patch from Stephen Usher.
963 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
965 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
968 TF/02 Fix a couple more cases where we did not log the error message
969 when unlink() failed. See also change 4.74-TF/03.
971 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
972 lock up or crash if it happened to be inside a call to libc when it
973 got a SIGUSR1 from exiwhat.
975 The SIGUSR1 handler appends the current process status to the process
976 log which is later printed by exiwhat. It used to use the general
977 purpose logging code to do this, but several functions it calls are
978 not safe for signals.
980 The new output code in the SIGUSR1 handler is specific to the process
981 log, and simple enough that it's easy to inspect for signal safety.
982 Removing some special cases also simplifies the general logging code.
983 Removing the spurious timestamps from the process log simplifies
986 TF/04 Improved ratelimit ACL condition.
988 The /noupdate option has been deprecated in favour of /readonly which
989 has clearer semantics. The /leaky, /strict, and /readonly update modes
990 are mutually exclusive. The update mode is no longer included in the
991 database key; it just determines when the database is updated. (This
992 means that when you upgrde Exim will forget old rate measurements.)
994 Exim now checks that the per_* options are used with an update mode that
995 makes sense for the current ACL. For example, when Exim is processing a
996 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
997 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
998 must specify per_mail/readonly. If you omit the update mode it defaults to
999 /leaky where that makes sense (as before) or /readonly where required.
1001 The /noupdate option is now undocumented but still supported for
1002 backwards compatibility. It is equivalent to /readonly except that in
1003 ACLs where /readonly is required you may specify /leaky/noupdate or
1004 /strict/noupdate which are treated the same as /readonly.
1006 A useful new feature is the /count= option. This is a generalization
1007 of the per_byte option, so that you can measure the throughput of other
1008 aggregate values. For example, the per_byte option is now equivalent
1009 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1011 The per_rcpt option has been generalized using the /count= mechanism
1012 (though it's more complicated than the per_byte equivalence). When it is
1013 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1014 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1015 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1016 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1017 Note that using per_rcpt with a non-readonly update mode in more than
1018 one ACL will cause the recipients to be double-counted. (The per_mail
1019 and per_byte options don't have this problem.)
1021 The handling of very low rates has changed slightly. If the computed rate
1022 is less than the event's count (usually one) then this event is the first
1023 after a long gap. In this case the rate is set to the same as this event's
1024 count, so that the first message of a spam run is counted properly.
1026 The major new feature is a mechanism for counting the rate of unique
1027 events. The new per_addr option counts the number of different
1028 recipients that someone has sent messages to in the last time period. It
1029 behaves like per_rcpt if all the recipient addresses are different, but
1030 duplicate recipient addresses do not increase the measured rate. Like
1031 the /count= option this is a general mechanism, so the per_addr option
1032 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1033 example, measure the rate that a client uses different sender addresses
1034 with the options per_mail/unique=$sender_address. There are further
1035 details in the main documentation.
1037 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1039 TF/06 Removed a few PCRE remnants.
1041 TF/07 Automatically extract Exim's version number from tags in the git
1042 repository when doing development or release builds.
1044 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1045 Bugzilla 879. Patch from Paul Fisher.
1047 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1048 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1051 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1053 PP/05 Variable $av_failed, true if the AV scanner deferred.
1054 Bugzilla 1078. Patch from John Horne.
1056 PP/06 Stop make process more reliably on build failure.
1057 Bugzilla 1087. Patch from Heiko Schlittermann.
1059 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1060 Bugzilla 1089. Patch from Heiko Schlittermann.
1062 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1063 Bugzilla 1131. Patch from Holger Weiß.
1065 PP/09 Handle IPv6 addresses with SPF.
1066 Bugzilla 860. Patch from Wolfgang Breyha.
1068 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1070 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1073 PP/11 match_* no longer expand right-hand-side by default.
1074 New compile-time build option, EXPAND_LISTMATCH_RHS.
1075 New expansion conditions, "inlist", "inlisti".
1077 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1079 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1081 PP/14 fix log_write() format string regression from TF/03.
1082 Bugzilla 1152. Patch from Dmitry Isaikin.
1088 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1090 PP/02 Harmonised TLS library version reporting; only show if debugging.
1091 Layout now matches that introduced for other libraries in 4.74 PP/03.
1093 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1095 PP/04 New "dns_use_edns0" global option.
1097 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1100 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1101 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1103 TK/01 Updated PolarSSL code to 0.14.2.
1104 Bugzilla 1097. Patch from Andreas Metzler.
1106 PP/07 Catch divide-by-zero in ${eval:...}.
1107 Fixes bugzilla 1102.
1109 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1112 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1113 format-string attack -- SECURITY: remote arbitrary code execution.
1115 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1116 time unintentionally subject to list matching rules, letting the header
1117 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1118 arbitrary string expansion). This allowed for information disclosure.
1120 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1121 INT_MIN/-1 -- value coerced to INT_MAX.
1127 NM/01 Workround for PCRE version dependency in version reporting
1130 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1131 This fixes portability to compilers other than gcc, notably
1132 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1134 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1135 makefiles for portability to HP-UX and POSIX correctness.
1137 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1138 Also via indented variable definition in the Makefile.
1139 (Debugging by Oliver Heesakkers).
1141 PP/02 Restore caching of spamd results with expanded spamd_address.
1142 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1144 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1145 Improves build reliability. Fix from: Frank Elsner
1147 NM/02 Fix wide character breakage in the rfc2047 coding
1148 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1150 NM/03 Allow underscore in dnslist lookups
1151 Fixes bug 1026. Patch from Graeme Fowler
1153 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1154 Code patches from Adam Ciarcinski of NetBSD.
1156 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1159 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1160 is logged, to avoid truncation. Patch from John Horne.
1162 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1163 Patch from Jakob Hirsch.
1165 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1166 SQL string expansion failure details.
1167 Patch from Andrey Oktyabrski.
1169 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1170 Patch from Simon Arlott.
1172 PP/09 New lookups functionality failed to compile on old gcc which rejects
1173 extern declarations in function scope.
1174 Patch from Oliver Fleischmann
1176 PP/10 Use sig_atomic_t for flags set from signal handlers.
1177 Check getgroups() return and improve debugging.
1178 Fixed developed for diagnosis in bug 927 (which turned out to be
1181 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1182 Patch from Mark Zealey.
1184 PP/12 Bugzilla 1056: Improved spamd server selection.
1185 Patch from Mark Zealey.
1187 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1188 Based on patch from Heiko Schlittermann.
1190 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1191 Patch from Uwe Doering, sign-off by Michael Haardt.
1193 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1194 variable declaration deep within a block. Bug and patch from
1197 PP/15 lookups-Makefile IRIX compatibilty coercion.
1199 PP/16 Make DISABLE_DKIM build knob functional.
1201 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1202 Patch by Simon Arlott
1204 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1205 variable argument macros. Our copy now differs from upstream.
1211 TF/01 Failure to get a lock on a hints database can have serious
1212 consequences so log it to the panic log.
1214 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1215 controlled using the smtp_confirmation log selector.
1217 TF/03 Include the error message when we fail to unlink a spool file.
1219 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1220 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1221 for maintaining out-of-tree patches for some time.
1223 PP/01 Bugzilla 139: Documentation and portability issues.
1224 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1225 Handle per-OS dynamic-module compilation flags.
1227 PP/02 Let /dev/null have normal permissions.
1228 The 4.73 fixes were a little too stringent and complained about the
1229 permissions on /dev/null. Exempt it from some checks.
1230 Reported by Andreas M. Kirchwitz.
1232 PP/03 Report version information for many libraries, including
1233 Exim version information for dynamically loaded libraries. Created
1234 version.h, now support a version extension string for distributors
1235 who patch heavily. Dynamic module ABI change.
1237 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1238 privilege escalation vulnerability whereby the Exim run-time user
1239 can cause root to append content of the attacker's choosing to
1242 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1245 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1246 If dropping privileges for untrusted macros, we disabled normal logging
1247 on the basis that it would fail; for the Exim run-time user, this is not
1248 the case, and it resulted in successful deliveries going unlogged.
1249 Fixed. Reported by Andreas Metzler.
1255 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1256 only prepend for the Resent-* case. Fixes regression introduced in
1257 Exim 4.70 by NM/22 for Bugzilla 607.
1259 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1260 increasing numbers of administrators be bitten by this.
1262 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1264 PP/03 Bugzilla 994: added openssl_options main configuration option.
1266 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1268 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1270 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1272 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1273 without a peer certificate, leading to a segfault because of an
1274 assumption that peers always have certificates. Be a little more
1275 paranoid. Problem reported by Martin Tscholak.
1277 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1278 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1279 NB: ClamAV planning to remove STREAM in "middle of 2010".
1280 CL also introduces -bmalware, various -d+acl logging additions and
1281 more caution in buffer sizes.
1283 PP/09 Implemented reverse_ip expansion operator.
1285 PP/10 Bugzilla 937: provide a "debug" ACL control.
1287 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1289 PP/12 Bugzilla 973: Implement --version.
1291 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1293 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1295 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1297 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1298 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1299 ignore trailing whitespace.
1301 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1303 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1306 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1307 Notification from Dr Andrew Aitchison.
1309 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1310 ExtendedDetectionInfo response format.
1311 Notification from John Horne.
1313 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1316 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1317 XSL and documented dependency on system catalogs, with examples of how
1320 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1323 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1324 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1325 configuration file which is writeable by the Exim user or group.
1327 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1328 of configuration files to cover files specified with the -C option if
1329 they are going to be used with root privileges, not just the default
1332 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1333 option (effectively making it always true).
1335 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1336 files to be used while preserving root privileges.
1338 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1339 that rogue child processes cannot use them.
1341 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1342 run-time user, instead of root.
1344 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1345 Exim run-time user without dropping privileges.
1347 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1348 result string, instead of calling string_vformat() twice with the same
1351 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1352 for other users. Others should always drop root privileges if they use
1353 -C on the command line, even for a whitelisted configure file.
1355 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1357 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1363 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1364 $data_path, and $header_path variables; fixed documentation bugs and
1367 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1368 exipick to access non-standard spools, including the "frozen" queue
1371 NM/01 Bugzilla 965: Support mysql stored procedures.
1372 Patch from Alain Williams
1374 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1376 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1377 Patch from Andreas Metzler
1379 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1380 Patch from Kirill Miazine
1382 NM/05 Bugzilla 671: Added umask to procmail example.
1384 JJ/03 installed exipick 20100323.0, fixing doc bug
1386 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1387 directory. Notification and patch from Dan Rosenberg.
1389 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1391 TK/02 Improve log output when DKIM signing operation fails.
1393 MH/01 Treat the transport option dkim_domain as a colon separated
1394 list, not as a single string, and sign the message with each element,
1395 omitting multiple occurences of the same signer.
1397 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1398 Bugzilla 985, 986. Patch by Simon Arlott
1400 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1401 Patch by Simon Arlott
1403 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1404 MBX locking. Notification from Dan Rosenberg.
1410 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1412 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1414 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1416 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1418 NM/04 Bugzilla 915: Flag broken perl installation during build.
1424 TK/01 Added patch by Johannes Berg that expands the main option
1425 "spamd_address" if it starts with a dollar sign.
1427 TK/02 Write list of recipients to X-Envelope-Sender header when building
1428 the mbox-format spool file for content scanning (suggested by Jakob
1431 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1432 (http://www.dcc-servers.net/) support via dccifd. Activated by
1433 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1435 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1436 by Mark Daniel Reidel <mr@df.eu>.
1438 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1439 When building exim an external PCRE library is now needed -
1440 PCRE is a system library on the majority of modern systems.
1441 See entry on PCRE_LIBS in EDITME file.
1443 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1444 conversation. Added nologin parameter to request.
1445 Patch contributed by Kirill Miazine.
1447 TF/01 Do not log submission mode rewrites if they do not change the address.
1449 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1451 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1452 log files in place. Contributed by Roberto Lima.
1454 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1456 TF/03 Bugzilla 615: When checking the local_parts router precondition
1457 after a local_part_suffix or local_part_prefix option, Exim now
1458 does not use the address's named list lookup cache, since this
1459 contains cached lookups for the whole local part.
1461 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1462 Robert Millan. Documentation is in experimental-spec.txt.
1464 TF/04 Bugzilla 668: Fix parallel build (make -j).
1466 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1468 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1469 Patch provided by Jan Srzednicki.
1471 TF/05 Leading white space used to be stripped from $spam_report which
1472 wrecked the formatting. Now it is preserved.
1474 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1475 that they are available at delivery time.
1477 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1479 TF/08 TLS error reporting now respects the incoming_interface and
1480 incoming_port log selectors.
1482 TF/09 Produce a more useful error message if an SMTP transport's hosts
1483 setting expands to an empty string.
1485 NM/06 Bugzilla 744: EXPN did not work under TLS.
1486 Patch provided by Phil Pennock.
1488 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1489 Patch provided by Richard Godbee.
1491 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1492 acl_smtp_notquit, added index entry.
1494 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1495 Patch provided by Eugene Bujak.
1497 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1498 accept(). Patch provided by Maxim Dounin.
1500 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1501 Patch provided by Phil Pennock.
1503 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1505 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1506 Patch provided by Brad "anomie" Jorsch.
1508 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1509 Patch provided by Dean Brooks.
1511 TK/05 Add native DKIM support (does not depend on external libraries).
1513 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1514 Patch provided by Graeme Fowler.
1516 NM/16 Bugzilla 851: Documentation example syntax fix.
1518 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1520 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1523 NM/19 Bugzilla 745: TLS version reporting.
1524 Patch provided by Phil Pennock.
1526 NM/20 Bugzilla 167: bool: condition support.
1527 Patch provided by Phil Pennock.
1529 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1530 clients. Patch provided by Phil Pennock.
1532 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1533 Patch provided by Brad "anomie" Jorsch.
1535 NM/23 Bugzilla 687: Fix misparses in eximstats.
1536 Patch provided by Heiko Schlittermann.
1538 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1539 Patch provided by Heiko Schlittermann.
1541 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1542 plus update to original patch.
1544 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1546 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1547 Patch provided by David Brownlee.
1549 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1551 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1553 NM/30 Bugzilla 888: TLS documentation bugfixes.
1555 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1557 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1558 Unlike the original bugzilla I have changed all shell scripts in src tree.
1560 NM/33 Bugzilla 898: Transport filter timeout fix.
1561 Patch by Todd Rinaldo.
1563 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1564 Patch by Serge Demonchaux.
1566 NM/35 Bugzilla 39: Base64 decode bug fixes.
1567 Patch by Jakob Hirsch.
1569 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1571 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1573 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1575 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1581 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1582 ALT-N's libdkim that I have put here:
1583 http://duncanthrax.net/exim-experimental/
1585 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1586 were called 'true' and it seems that C99 defines that as a reserved
1587 keyword to be used with 'bool' variable types. That means you could
1588 not include C99-style headers which use bools without triggering
1589 build errors in sieve.c.
1591 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1592 as mailq or other aliases. Changed the --help handling significantly
1593 to do whats expected. exim_usage() emits usage/help information.
1595 SC/01 Added the -bylocaldomain option to eximstats.
1597 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1599 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1601 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1607 PH/01 Another patch from the Sieve maintainer.
1609 PH/02 When an IPv6 address is converted to a string for single-key lookup
1610 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1611 dots are used instead of colons so that keys in lsearch files need not
1612 contain colons. This was done some time before quoting was made available
1613 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1614 (notated using the quote facility) so as to distinguish them from IPv4
1615 keys. This meant that lookups for IP addresses in host lists did not work
1616 for iplsearch lookups.
1618 This has been fixed by arranging for IPv6 addresses to be expressed with
1619 colons if the lookup type is iplsearch. This is not incompatible, because
1620 previously such lookups could never work.
1622 The situation is now rather anomolous, since one *can* have colons in
1623 ordinary lsearch keys. However, making the change in all cases is
1624 incompatible and would probably break a number of configurations.
1626 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1629 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1630 conversion specification without a maximum field width, thereby enabling
1631 a rogue spamd server to cause a buffer overflow. While nobody in their
1632 right mind would setup Exim to query an untrusted spamd server, an
1633 attacker that gains access to a server running spamd could potentially
1634 exploit this vulnerability to run arbitrary code as the Exim user.
1636 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1637 $primary_hostname instead of what libspf2 thinks the hosts name is.
1639 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1640 a directory entry by the name of the lookup key. Previously, if a
1641 symlink pointed to a non-existing file or a file in a directory that
1642 Exim lacked permissions to read, a lookup for a key matching that
1643 symlink would fail. Now it is enough that a matching directory entry
1644 exists, symlink or not. (Bugzilla 503.)
1646 PH/03 The body_linecount and body_zerocount variables are now exported in the
1649 PH/04 Added the $dnslist_matched variable.
1651 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1652 This means they are set thereafter only if the connection becomes
1655 PH/06 Added the client_condition to authenticators so that some can be skipped
1656 by clients under certain conditions.
1658 PH/07 The error message for a badly-placed control=no_multiline_responses left
1659 "_responses" off the end of the name.
1661 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1663 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1664 (without spaces) instead of just copying the configuration text.
1666 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1668 PH/11 Added $max_received_linelength.
1670 PH/12 Added +ignore_defer and +include_defer to host lists.
1672 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1673 way in which PCRE > 7.0 is built.
1675 PH/14 Implemented queue_only_load_latch.
1677 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1678 MAIL command. The effect was to mangle the value on 64-bit systems.
1680 PH/16 Another patch from the Sieve maintainer.
1682 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1684 PH/18 If a system quota error occurred while trying to create the file for
1685 a maildir delivery, the message "Mailbox is full" was not appended to the
1686 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1687 only to a quota excession during the actual writing of the file.
1689 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1690 characters?) which causes problems in log lines. The DN values are now
1691 passed through string_printing() before being added to log lines.
1693 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1694 and InterBase are left for another time.)
1696 PH/21 Added message_body_newlines option.
1698 PH/22 Guard against possible overflow in moan_check_errorcopy().
1700 PH/23 POSIX allows open() to be a macro; guard against that.
1702 PH/24 If the recipient of an error message contained an @ in the local part
1703 (suitably quoted, of course), incorrect values were put in $domain and
1704 $local_part during the evaluation of errors_copy.
1710 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1711 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1714 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1715 issue a MAIL command.
1717 PH/02 In an ACL statement such as
1719 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1721 if a client was not listed at all, or was listed with a value other than
1722 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1723 the condition was not true (as it should be), so access was not denied.
1724 The bug was that the ! inversion was incorrectly passed on to the second
1725 item. This has been fixed.
1727 PH/03 Added additional dnslists conditions == and =& which are different from
1728 = and & when the dns lookup returns more than one IP address.
1730 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1731 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1733 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1734 FSYNC, which compiles an option called disable_fsync that allows for
1735 bypassing fsync(). The documentation is heavily laced with warnings.
1737 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1739 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1740 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1741 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1742 including adding "make clean"; (3) Added -fPIC when compiling the test
1743 dynamically loaded module, to get rid of a warning.
1745 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1746 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1747 = 0s. The bug is otherwise harmless.
1749 PH/07 There was a bug in the dovecot authenticator such that the value of
1750 $auth1 could be overwritten, and so not correctly preserved, after a
1751 successful authentication. This usually meant that the value preserved by
1752 the server_setid option was incorrect.
1754 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1756 PH/09 Installed PCRE release 7.0.
1758 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1759 run for batched SMTP input. It is now run at the start of every message
1760 in the batch. While fixing this I discovered that the process information
1761 (output by running exiwhat) was not always getting set for -bs and -bS
1762 input. This is fixed, and it now also says "batched" for BSMTP.
1764 PH/11 Added control=no_pipelining.
1766 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1767 patch, slightly modified), and move the expansion of helo_data till after
1768 the connection is made in the smtp transport (so it can use these
1771 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1773 PH/14 Added log_selector = +pid.
1775 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1777 PH/16 Add ${if forany and ${if forall.
1779 PH/17 Added dsn_from option to vary the From: line in DSNs.
1781 PH/18 Flush SMTP output before performing a callout, unless control =
1782 no_callout_flush is set.
1784 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1785 was true (the default) a successful delivery failed to delete the retry
1786 item, thus causing premature timeout of the address. The bug is now
1789 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1791 PH/21 Long custom messages for fakedefer and fakereject are now split up
1792 into multiline reponses in the same way that messages for "deny" and
1793 other ACL rejections are.
1795 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1796 with slight modification.
1798 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1799 draft, changing the syntax and factoring some duplicate code".
1801 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1802 for deliveries of the second and subsequent messages over the same SMTP
1805 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1806 ${reduce, with only minor "tidies".
1808 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1810 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1811 expansion side effects.
1813 PH/27 When a message times out after an over-quota error from an Exim-imposed
1814 quota, the bounce message says "mailbox is full". This message was not
1815 being given when it was a system quota that was exceeded. It now should
1818 MH/03 Made $recipients available in local_scan(). local_scan() already has
1819 better access to the recipient list through recipients_list[], but
1820 $recipients can be useful in postmaster-provided expansion strings.
1822 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1823 in the case of a MAIL command with additional options following the
1824 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1825 were accidentally chopped off.
1827 PH/29 SMTP synchronization checks are implemented when a command is read -
1828 there is a check that no more input is waiting when there shouldn't be
1829 any. However, for some commands, a delay in an ACL can mean that it is
1830 some time before the response is written. In this time, more input might
1831 arrive, invalidly. So now there are extra checks after an ACL has run for
1832 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1833 pipelining has not been advertised.
1835 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1837 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1838 returned in the SMTP response when smtp_return_error_details was false.
1839 This has been fixed.
1841 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1842 instead of the C I/O that was originally supplied, because problems were
1843 reported on Solaris.
1845 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1846 Exim which did not show up earlier: it was assuming that a call to
1847 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1848 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1849 was a macro that became an assignment, so it seemed to work. This has
1850 changed to a proper function call with a void return, hence the compile
1851 error. Exim's code has been fixed.
1853 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1856 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1858 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1860 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1861 because it turns out to be too big or there is a timeout) there is no
1862 "Completed" line in the log. When some messages of this type were
1863 selected by exigrep, they were listed as "not completed". Others were
1864 picked up by some special patterns. I have improved the selection
1865 criteria to be more general.
1867 PH/38 The host_find_failed option in the manualroute router can now be set
1868 to "ignore", to completely ignore a host whose IP address cannot be
1869 found. If all hosts are ignored, the behaviour is controlled by the new
1870 host_all_ignored option.
1872 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1873 homing or because of multiple MX records with /mx) generated more than
1874 one IP address, and the following item turned out to be the local host,
1875 all the secondary addresses of the first item were incorrectly removed
1876 from the list, along with the local host and any following hosts (which
1877 is what is supposed to happen).
1879 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1880 whoever called Exim into the -H file. In the case of the daemon it was
1881 behaving confusingly. When first started, it used values for whoever
1882 started the daemon, but after a SIGHUP it used the Exim user (because it
1883 calls itself on a restart). I have changed the code so that it now always
1886 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1887 message are rejected with the same error (e.g. no authentication or bad
1888 sender address), and a DATA command is nevertheless sent (as can happen
1889 with PIPELINING or a stupid MUA), the error message that was given to the
1890 RCPT commands is included in the rejection of the DATA command. This is
1891 intended to be helpful for MUAs that show only the final error to their
1894 PH/42 Another patch from the Sieve maintainer.
1896 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1897 Eximstats - Fixed some broken HTML links and added missing column headers
1899 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1900 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1902 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1904 PH/43 Yet another patch from the Sieve maintainer.
1906 PH/44 I found a way to check for a TCP/IP connection going away before sending
1907 the response to the final '.' that terminates a message, but only in the
1908 case where the client has not sent further data following the '.'
1909 (unfortunately, this is allowed). However, in many cases there won't be
1910 any further data because there won't be any more messages to send. A call
1911 to select() can be used: if it shows that the input is "ready", there is
1912 either input waiting, or the socket has been closed. An attempt to read
1913 the next input character can distinguish the two cases. Previously, Exim
1914 would have sent an OK response which the client would never have see.
1915 This could lead to message repetition. This fix should cure that, at
1916 least in a lot of common cases.
1918 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1919 advertised in response to EHLO.
1925 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1926 fixed by 4.65/MH/01 (is this a record?) are fixed:
1928 (i) An empty string was always treated as zero by the numeric comparison
1929 operators. This behaviour has been restored.
1931 (ii) It is documented that the numeric comparison operators always treat
1932 their arguments as decimal numbers. This was broken in that numbers
1933 starting with 0 were being interpreted as octal.
1935 While fixing these problems I realized that there was another issue that
1936 hadn't been noticed. Values of message_size_limit (both the global option
1937 and the transport option) were treated as octal if they started with 0.
1938 The documentation was vague. These values are now always treated as
1939 decimal, and I will make that clear in the documentation.
1945 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1946 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1949 MH/01 Don't check that the operands of numeric comparison operators are
1950 integers when their expansion is in "skipping" mode (fixes bug
1951 introduced by 4.64-PH/07).
1953 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1954 child addresses, Exim now panics and dies. Previously, because the count
1955 is held in a short int, deliveries were likely to be lost. As such a
1956 large number of recipients for a single message is ridiculous
1957 (performance will be very, very poor), I have chosen to impose a limit
1958 rather than extend the field.
1964 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1965 leftover -K file (the existence of which was triggered by #402).
1966 While we were at it, introduced process PID as part of the -K
1967 filename. This should rule out race conditions when creating
1970 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1971 processing considerably. Previous code took too long for large mails,
1972 triggering a timeout which in turn triggers #401.
1974 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1975 in the DK code in transports.c. sendfile() is not really portable,
1976 hence the _LINUX specificness.
1978 TF/01 In the add_headers option to the mail command in an Exim filter,
1979 there was a bug that Exim would claim a syntax error in any
1980 header after the first one which had an odd number of characters
1983 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1984 callout verification, Exim cached a "reject" for the entire domain. This
1985 is correct for most verifications, but it is not correct for a recipient
1986 verification with use_sender or use_postmaster set, because in that case
1987 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1988 case of MAIL FROM:<> rejection from other early rejections (e.g.
1989 rejection of HELO). When verifying a recipient using a non-null MAIL
1990 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1991 Whatever the result of the callout, the value of the domain cache is
1992 left unchanged (for any other kind of callout, getting as far as trying
1993 RCPT means that the domain itself is ok).
1995 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1998 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1999 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2000 session, the daemon ignores ECONNECT errors and logs others; it now
2001 ignores EPIPE as well.
2003 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2004 (quoted-printable decoding).
2006 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2007 later the small subsequent patch to fix an introduced bug.
2009 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2011 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2013 PH/08 An error is now given if message_size_limit is specified negative.
2015 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2016 to be given (somewhat) arbitrary names.
2018 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2021 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2022 miscellaneous code fixes
2024 PH/10 Added the log_reject_target ACL modifier to specify where to log
2027 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2028 hostname. This is wrong, because it relates to the incoming message (and
2029 probably the interface on which it is arriving) and not to the outgoing
2030 callout (which could be using a different interface). This has been
2031 changed to use the value of the helo_data option from the smtp transport
2032 instead - this is what is used when a message is actually being sent. If
2033 there is no remote transport (possible with a router that sets up host
2034 addresses), $smtp_active_hostname is used.
2036 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2037 tweaks were necessary in order to get it to work (see also 21 below):
2038 (a) The code assumed that strncpy() returns a negative number on buffer
2039 overflow, which isn't the case. Replaced with Exim's string_format()
2041 (b) There were several signed/unsigned issues. I just did the minimum
2042 hacking in of casts. There is scope for a larger refactoring.
2043 (c) The code used strcasecmp() which is not a standard C function.
2044 Replaced with Exim's strcmpic() function.
2045 (d) The code set only $1; it now sets $auth1 as well.
2046 (e) A simple test gave the error "authentication client didn't specify
2047 service in request". It would seem that Dovecot has changed its
2048 interface. Fortunately there's a specification; I followed it and
2049 changed what the client sends and it appears to be working now.
2051 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2054 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2055 address A is aliased to B and C, where B exists and C does not. Without
2056 -v the output is "A verified" because verification stops after a
2057 successful redirection if more than one address is generated. However,
2058 with -v the child addresses are also verified. Exim was outputting "A
2059 failed to verify" and then showing the successful verification for C,
2060 with its parentage. It now outputs "B failed to verify", showing B's
2061 parentage before showing the successful verification of C.
2063 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2064 look up a TXT record in a specific list after matching in a combined
2067 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2068 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2069 they consult the DNS. I had assumed they would set it the way they
2070 wanted; and indeed my experiments on Linux seem to show that in some
2071 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2072 To be on the safe side, however, I have now made the interface to
2073 host_find_byname() similar to host_find_bydns(), with an argument
2074 containing the DNS resolver options. The host_find_byname() function now
2075 sets these options at its start, just as host_find_bydns() does. The smtp
2076 transport options dns_qualify_single and dns_search_parents are passed to
2077 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2078 of host_find_byname() use the default settings of RES_DEFNAMES
2079 (qualify_single) but not RES_DNSRCH (search_parents).
2081 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2082 spool_read_header() do less string testing, by means of a preliminary
2083 switch on the second character of optional "-foo" lines. (This is
2084 overdue, caused by the large number of possibilities that now exist.
2085 Originally there were few.) While I was there, I also converted the
2086 str(n)cmp tests so they don't re-test the leading "-" and the first
2087 character, in the hope this might squeeze out yet more improvement.
2089 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2090 flag allowing group syntax was set by the header_syntax check but not
2091 turned off, possible causing trouble later; (2) The flag was not being
2092 set at all for the header_verify test, causing "group"-style headers to
2093 be rejected. I have now set it in this case, and also caused header_
2094 verify to ignore an empty address taken from a group. While doing this, I
2095 came across some other cases where the code for allowing group syntax
2096 while scanning a header line wasn't quite right (mostly, not resetting
2097 the flag correctly in the right place). These bugs could have caused
2098 trouble for malformed header lines. I hope it is now all correct.
2100 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2101 with the "reply" argument non-NULL. The code, however (which originally
2102 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2103 but it didn't always do it. This confused somebody who was copying the
2104 code for some other use. I have removed all the tests.
2106 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2107 feature that was used to support insecure browsers during the U.S. crypto
2108 embargo. It requires special client support, and Exim is probably the
2109 only MTA that supported it -- and would never use it because real RSA is
2110 always available. This code has been removed, because it had the bad
2111 effect of slowing Exim down by computing (never used) parameters for the
2112 RSA_EXPORT functionality.
2114 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2115 authenticator to fail if there's a tab character in the incoming data
2116 (there should never be unless someone is messing about, as it's supposed
2117 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2118 if the connection is using TLS or if the remote IP is the same as the
2119 local IP, and the "valid-client-cert option" if a client certificate has
2122 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2123 authenticators. This can be used for authorization after authentication
2124 succeeds. (In the case of plaintext, it servers for both authentication
2127 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2128 if any retry times were supplied.
2130 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2131 connection, where there is no client IP address to check. In this
2132 situation, the verify now always succeeds.
2134 PH/25 Applied John Jetmore's -Mset patch.
2136 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2138 PH/27 In a string expansion for a processed (not raw) header when multiple
2139 headers of the same name were present, leading whitespace was being
2140 removed from all of them, but trailing whitespace was being removed only
2141 from the last one. Now trailing whitespace is removed from each header
2142 before concatenation. Completely empty headers in a concatenation (as
2143 before) are ignored.
2145 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2146 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2148 PH/29 [Removed. This was a change that I later backed out, and forgot to
2149 correct the ChangeLog entry (that I had efficiently created) before
2150 committing the later change.]
2152 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2153 address errors (4xx response to RCPT) over the same connection as other
2154 messages routed to the same hosts. Such deliveries are always "forced",
2155 so retry times are not inspected. This resulted in far too many retries
2156 for the affected addresses. The effect occurred only when there were more
2157 hosts than the hosts_max_try setting in the smtp transport when it had
2158 the 4xx errors. Those hosts that it had tried were not added to the list
2159 of hosts for which the message was waiting, so if all were tried, there
2160 was no problem. Two fixes have been applied:
2162 (i) If there are any address or message errors in an SMTP delivery, none
2163 of the hosts (tried or untried) are now added to the list of hosts
2164 for which the message is waiting, so the message should not be a
2165 candidate for sending over the same connection that was used for a
2166 successful delivery of some other message. This seems entirely
2167 reasonable: after all the message is NOT "waiting for some host".
2168 This is so "obvious" that I'm not sure why it wasn't done
2169 previously. Hope I haven't missed anything, but it can't do any
2170 harm, as the worst effect is to miss an optimization.
2172 (ii) If, despite (i), such a delivery is accidentally attempted, the
2173 routing retry time is respected, so at least it doesn't keep
2174 hammering the server.
2176 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2177 in ${readsocket because some servers need this prod.
2179 PH/32 Added some extra debug output when updating a wait-xxx database.
2181 PH/33 The hint "could be header name not terminated by colon", which has been
2182 given for certain expansion errors for a long time, was not being given
2183 for the ${if def:h_colon_omitted{... case.
2185 PH/34 The spec says: "With one important exception, whenever a domain list is
2186 being scanned, $domain contains the subject domain." There was at least
2187 one case where this was not true.
2189 PH/35 The error "getsockname() failed: connection reset by peer" was being
2190 written to the panic log as well as the main log, but it isn't really
2191 panic-worthy as it just means the connection died rather early on. I have
2192 removed the panic log writing for the ECONNRESET error when getsockname()
2195 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2196 runs only) independently of the message's sender address. This meant
2197 that, if the 4xx error was in fact related to the sender, a different
2198 message to the same recipient with a different sender could confuse
2199 things. In particualar, this can happen when sending to a greylisting
2200 server, but other circumstances could also provoke similar problems.
2201 I have changed the default so that the retry time for these errors is now
2202 based a combination of the sender and recipient addresses. This change
2203 can be overridden by setting address_retry_include_sender=false in the
2206 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2207 remote server are returned as part of bounce messages. This was not
2208 happening for LMTP over a pipe (the lmtp transport), but now it is the
2209 same for both kinds of LMTP.
2211 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2212 in header lines that were in fact CNAMEs. This is no longer the case.
2214 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2215 and queue runs started by the daemon processed all messages. This has
2216 been fixed so that -R and -S can now usefully be given with -q<time>.
2218 PH/40 Import PCRE release 6.7 (fixes some bugs).
2220 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2222 PH/42 Give an error if -q is specified more than once.
2224 PH/43 Renamed the variables $interface_address and $interface_port as
2225 $received_ip_address and $received_port, to make it clear that these
2226 values apply to message reception, and not to the outgoing interface when
2227 a message is delivered. (The old names remain recognized, of course.)
2229 PH/44 There was no timeout on the connect() call when using a Unix domain
2230 socket in the ${readsocket expansion. There now is.
2232 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2233 be meaningful with "accept".
2235 SC/01 Eximstats V1.43
2236 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2238 SC/02 Eximstats V1.44
2239 Use a glob alias rather than an array ref in the generated
2240 parser. This improves both readability and performance.
2242 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2243 Collect SpamAssassin and rejection statistics.
2244 Don't display local sender or destination tables unless
2245 there is data to show.
2246 Added average volumes into the top table text output.
2248 SC/04 Eximstats V1.46
2249 Collect data on the number of addresses (recipients)
2250 as well as the number of messages.
2252 SC/05 Eximstats V1.47
2253 Added 'Message too big' to the list of mail rejection
2254 reasons (thanks to Marco Gaiarin).
2256 SC/06 Eximstats V1.48
2257 Mainlog lines which have GMT offsets and are too short to
2258 have a flag are now skipped.
2260 SC/07 Eximstats V1.49 (Alain Williams)
2261 Added the -emptyok flag.
2263 SC/08 Eximstats V1.50
2264 Fixes for obtaining the IP address from reject messages.
2266 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2267 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2268 whitesspace changes from 4.64-PH/27
2270 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2273 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2274 are found, allow negative numbers in numeric criteria)
2276 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2278 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2279 to match changes made in 4.64-PH/43
2281 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2283 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2284 "domain" is actually an IP address, and force a failure. This locks out
2285 those revolvers/nameservers that support "A-for-A" lookups, in
2286 contravention of the specifications.
2288 PH/48 When a host name was looked up from an IP address, and the subsequent
2289 forward lookup of the name timed out, the host name was left in
2290 $sender_host_name, contrary to the specification.
2292 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2293 restricted to single-key lookups, Exim was not diagnosing an error if
2294 * or *@ was used with a query-style lookup.
2296 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2298 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2299 long ago, but noone interested enough thought of it. Let's just say that
2300 the "1.1" means that there are some new functions that weren't there at
2301 some point in the past.
2303 PH/51 Error processing for expansion failure of helo_data from an smtp
2304 transport during callout processing was broken.
2306 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2307 tested/used via the -bh/-bhc/-bs options.
2309 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2310 bug, fixed in subsequent PCRE releases).
2312 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2313 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2315 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2321 SC/01 Use a glob alias rather than an array ref in eximstats generated
2322 parser. This improves both readability and performance.
2324 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2325 Don't display local sender or destination tables in eximstats unless
2326 there is data to show.
2327 Added average volumes into the eximstats top table text output.
2329 SC/03 Collect data on the number of addresses (recipients) as well
2330 as the number of messages in eximstats.
2332 TF/01 Correct an error in the documentation for the redirect router. Exim
2333 does (usually) call initgroups() when daemonizing.
2335 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2336 with consistent privilege compared to when running as a daemon.
2338 TF/03 Note in the spec that $authenticated_id is not set for local
2339 submissions from trusted users.
2341 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2342 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2344 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2345 by adding some example configuration directives to the default
2346 configuration file. A little bit of work is required to uncomment the
2347 directives and define how usernames and passwords are checked, but
2348 there is now a framework to start from.
2350 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2351 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2352 without this. I don't know how relevant this is to other LDAP libraries.
2354 PH/02 Add the verb name to the "unknown ACL verb" error.
2356 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2358 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2360 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2361 directory not expanded when it should be if an expanded home directory
2362 was set for the address (which is overridden by the transport).
2364 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2367 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2368 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2369 because it is too late at that time, and has no effect.
2371 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2372 security issue with \' (bugzilla #107). I could not use the
2373 PQescapeStringConn() function, because it needs a PGconn value as one of
2376 PH/08 When testing addresses using -bt, indicate those final addresses that
2377 are duplicates that would not cause an additional delivery. At least one
2378 person was confused, thinking that -bt output corresponded to deliveries.
2379 (Suppressing duplicates isn't a good idea as you lose the information
2380 about possibly different redirections that led to the duplicates.)
2382 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2383 systems where poll() doesn't work, in particular OS X.
2385 PH/10 Added more information to debugging output for retry time not reached.
2387 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2388 operations in malware.c.
2390 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2393 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2394 syslog_duplication set false, log lines that would normally be written
2395 both the the main log and to the reject log were not written to syslog at
2398 PH/14 In the default configuration, change the use of "message" in ACL warn
2399 statements to "add_header".
2401 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2402 not followed by a command (e.g. "seen endif").
2404 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2405 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2408 PH/17 Added extra conditions to the default value of delay_warning_condition
2412 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2413 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2414 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2417 The Auto-Submitted: and various List- headers are standardised, whereas I
2418 don't think Precedence: ever was.
2420 PH/18 Refactored debugging code in route_finduser() to show more information,
2421 in particular, the error code if getpwnam() issues one.
2423 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2424 This is apparently needed in addition to the PH/07 change above to avoid
2425 any possible encoding problems.
2427 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2428 but not after initializing Perl.
2430 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2431 output them only if debugging. By default they are written stderr,
2432 apparently, which is not desirable.
2434 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2437 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2440 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2442 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2443 authenticated or an ident call has been made. Suppress the default
2444 values for $authenticated_id and $authenticated_sender (but permit -oMai
2445 and -oMas) when testing with -bh.
2447 PH/24 Re-jigged the order of the tests in the default configuration so that the
2448 tests for valid domains and recipients precede the DNS black list and CSA
2449 tests, on the grounds that those ones are more expensive.
2451 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2452 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2453 This bug exists in every version of Exim that I still have, right back to
2456 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2457 However, an attempt to turn on case-sensitivity in a regex key by
2458 including (?-i) didn't work because the subject string was already
2459 lowercased, and the effects were non-intuitive. It turns out that a
2460 one-line patch can be used to allow (?-i) to work as expected.
2466 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2467 other effects) broke the use of negated acl sub-conditions.
2469 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2472 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2473 "Deny" causes Exim to reject the incoming connection with a 554 error.
2474 Unfortunately, if there is a major crisis, such as a disk failure,
2475 tcp-wrappers gives "deny", whereas what one would like would be some
2476 kind of temporary error. A kludge has been added to help with this.
2477 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2478 554 error is used if errno is still zero or contains ENOENT (which occurs
2479 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2482 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2484 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2485 errors. Otherwise a message that provokes a temporary error (when other
2486 messages do not) can cause a whole host to time out.
2488 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2489 addresses were routed directly to files or pipes from a redirect router.
2490 File deliveries just didn't batch; pipe deliveries might have suffered
2493 PH/06 A failure to get a lock for a hints database would erroneously always say
2494 "Failed to get write lock", even when it was really a read lock.
2496 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2497 of 0600. This has been changed to use the value of the lockfile_mode
2498 option (which defaults to 0600).
2500 PH/08 Applied small patch from the Sieve maintainer.
2502 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2503 folder from quota calculations, a direct delivery into this folder messed
2504 up the contents of the maildirsize file. This was because the regex was
2505 used only to exclude .Trash (or whatever) when the size of the mailbox
2506 was calculated. There was no check that a delivery was happening into an
2507 excluded directory. This bug has been fixed by ignoring all quota
2508 processing for deliveries into excluded directories.
2510 PH/10 Added the maildirfolder_create_regex option to appendfile.
2516 PH/01 The code for finding all the local interface addresses on a FreeBSD
2517 system running IPv6 was broken. This may well have applied to all BSD
2518 systems, as well as to others that have similar system calls. The broken
2519 code found IPv4 interfaces correctly, but gave incorrect values for the
2520 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2521 that it would not match correctly against @[] and not recognize the IPv6
2524 PH/02 The ipliteral router was not recognizing addresses of the form user@
2525 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2527 PH/03 Added disable_ipv6.
2529 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2530 decoded form, because it is most often used to construct To: headers
2531 lines in autoreplies, and the decoded form may well be syntactically
2532 invalid. However, $reply_address has leading white space removed, and all
2533 newlines turned into spaces so that the autoreply transport does not
2536 PH/05 If group was specified without a user on a router, and no group or user
2537 was specified on a transport, the group from the router was ignored.
2539 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2540 for visible compile-time settings that can be used to change these
2541 numbers, for those that want even more. Backwards compatibility with old
2542 spool files has been maintained. However, going back to a previous Exim
2543 release will lost any variables that are in spool files.
2545 PH/07 Two small changes when running in the test harness: increase delay when
2546 passing a TCP/IP connection to a new process, in case the original
2547 process has to generate a bounce, and remove special handling of
2548 127.0.0.2 (sic), which is no longer necessary.
2550 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2551 be the same on different OS.
2553 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2556 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2557 whether --show-vars was specified or not
2559 JJ/02 exipick: Added support for new ACL variable spool format introduced
2562 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2563 syntactically invalid From: or Reply-to: line, and a filter used this to
2564 generate an autoreply, and therefore failed to obtain an address for the
2565 autoreply, Exim could try to deliver to a non-existent relative file
2566 name, causing unrelated and misleading errors. What now happens is that
2567 it logs this as a hard delivery error, but does not attempt to create a
2570 PH/11 The exinext utility has a -C option for testing purposes, but although
2571 the given file was scanned by exinext itself; it wasn't being passed on
2572 when Exim was called.
2574 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2575 an end-of-file indication when reading a command response.
2577 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2578 compiled. In many other places in Exim, IPv6 addresses are always
2579 recognized, so I have changed this. It also means that IPv4 domain
2580 literals of the form [IPV4:n.n.n.n] are now always recognized.
2582 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2583 used if the router is not running as root, for example, when verifying at
2584 ACL time, or when using -bh. The debugging output from this situation was
2585 non-existent - all you got was a failure to exec. I have made two
2588 (a) Failures to set uid/gid, the current directory, or a process leader
2589 in a subprocess such as that created by queryprogram now generate
2590 suitable debugging ouput when -d is set.
2592 (b) The queryprogram router detects when it is not running as root,
2593 outputs suitable debugging information if -d is set, and then runs
2594 the subprocess without attempting to change uid/gid.
2596 PH/15 Minor change to Makefile for building test_host (undocumented testing
2599 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2600 additional section of a DNS packet that returns MX or SRV records.
2601 Instead, it always explicitly searches for A/AAAA records. This avoids
2602 major problems that occur when a DNS server includes only records of one
2603 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2604 fixed another bug: if SRV records were looked up and the corresponding
2605 address records were *not* found in the additional section, the port
2606 values from the SRV records were lost.
2608 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2609 using the correct key (the original address) when searching the retry
2610 rules in order to find which one to use for generating the retry hint.
2612 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2613 adding the default one. Similarly, if it contains a Reply-To: header, the
2614 errors_reply_to option, if set, is not used.
2616 PH/19 When calculating a retry time, Exim used to measure the "time since
2617 failure" by looking at the "first failed" field in the retry record. Now
2618 it does not use this if it is later than than the arrival time of the
2619 message. Instead it uses the arrival time. This makes for better
2620 behaviour in cases where some deliveries succeed, thus re-setting the
2621 "first failed" field. An example is a quota failure for a huge message
2622 when small messages continue to be delivered. Without this change, the
2623 "time since failure" will always be short, possible causing more frequent
2624 delivery attempts for the huge message than are intended.
2625 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2627 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2628 $1, $2, $3) because the numerical variables can be reset during some
2629 expansion items (e.g. "match"), thereby losing the authentication data.
2631 PH/21 Make -bV show the size of off_t variables so that the test suite can
2632 decide whether to run tests for quotas > 2G.
2634 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2635 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2636 filecount value is greater than 2G or if a quota value is greater than 2G
2637 on a system where the size of off_t is not greater than 4, a panic error
2640 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2641 never match. The debug and -bh output now contains an explicit error
2642 message indicating a malformed IPv4 address or mask.
2644 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2645 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2646 PH/23 above applies.
2648 PH/25 Do not write to syslog when running in the test harness. The only
2649 occasion when this arises is a failure to open the main or panic logs
2650 (for which there is an explicit test).
2652 PH/26 Added the /no_tell option to "control=freeze".
2654 PH/27 If a host name lookup failed very early in a connection, for example, if
2655 the IP address matched host_lookup and the reverse lookup yielded a name
2656 that did not have a forward lookup, an error message of the form "no IP
2657 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2658 could be logged. Now it outputs the IP address instead of "NULL".
2660 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2661 allows the sender and the authenticated sender to be set when
2662 submitting a message from within Exim. Since child_open_exim() is
2663 documented for local_scan(), the new function should be too.
2665 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2666 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2667 results in an empty string is now treated as unset.
2669 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2671 PH/31 Added sender_verify_fail logging option.
2673 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2674 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2675 tidied the source and removed it altogether.
2677 PH/33 When a queue run was abandoned because the load average was too high, a
2678 log line was always written; now it is written only if the queue_run log
2679 selector is set. In addition, the log line for abandonment now contains
2680 information about the queue run such as the pid. This is always present
2681 in "start" and "stop" lines but was omitted from the "abandon" line.
2683 PH/34 Omit spaces between a header name and the colon in the error message that
2684 is given when verify = headers_syntax fails (if there are lots of them,
2685 the message gets confusing).
2687 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2688 names, as there are now some PTR records that contain slashes. This check
2689 is only to protect against broken name servers that fall over on strange
2690 characters, so the fact that it applies to all lookups doesn't matter.
2692 PH/36 Now that the new test suite is complete, we can remove some of the
2693 special code in Exim that was needed for the old test suite. For example,
2694 sorting DNS records because real resolvers return them in an arbitrary
2695 order. The new test suite's fake resolver always returns records in the
2698 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2699 bounces) except when queue_only is set, to avoid logging races between
2700 the different processes.
2702 PH/38 Panic-die if .include specifies a non-absolute path.
2704 PH/39 A tweak to the "H" retry rule from its user.
2706 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2707 a label. They prevented compilation on older perls.
2709 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2710 a warning to be raised on newish perls.
2712 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2713 on queue. Changes to match documented behaviour of showing count of
2714 messages matching specified criteria.
2716 PH/40 Changed the default ident timeout from 30s to 5s.
2718 PH/41 Added support for the use of login_cap features, on those BSD systems
2719 that have them, for controlling the resources used by pipe deliveries.
2721 PH/42 The content-scanning code uses fopen() to create files in which to put
2722 message data. Previously it was not paying any attention to the mode of
2723 the files. Exim runs with umask(0) because the rest of the code creates
2724 files with open(), and sets the required mode explicitly. Thus, these
2725 files were ending up world-writeable. This was not a big issue, because,
2726 being within the spool directory, they were not world-accessible. I have
2727 created a function called modefopen, which takes an additional mode
2728 argument. It sets umask(777), creates the file, chmods it to the required
2729 mode, then resets the umask. All the relevant calls to fopen() in the
2730 content scanning code have been changed to use this function.
2732 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2733 to 24 hours. This avoids potential overflow problems when processing G
2734 and H retry rules. I suspect nobody ever tinkers with this value.
2736 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2738 PH/45 When the plaintext authenticator is running as a client, the server's
2739 challenges are checked to ensure they are valid base64 strings. By
2740 default, the authentication attempt is cancelled if an invalid string is
2741 received. Setting client_ignore_invalid_base64 true ignores these errors.
2742 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2743 they are received. Thus, the responses can be made to depend on the
2744 challenges. If an invalid string is ignored, an empty string is placed in
2747 PH/46 Messages that are created by the autoreply transport now contains a
2748 References: header, in accordance with RFCs 2822 and 3834.
2750 PH/47 Added authenticated_sender_force to the smtp transport.
2752 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2754 PH/49 Installed latest patch from the Sieve maintainer.
2756 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2757 was also set, the appendfile transport was unnecessarily scanning a
2758 directory of message files (e.g. for maildir delivery) to find the count
2759 of files (along with the size), even though it did not need this
2760 information. It now does the scan only if it needs to find either the
2761 size of the count of files.
2763 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2765 PH/52 Two bugs concerned with error handling when the smtp transport is
2768 (i) Exim was not creating retry information for temporary errors given
2769 for individual recipients after the DATA command when the smtp transport
2770 was used in LMTP mode. This meant that they could be retried too
2771 frequently, and not timed out correctly.
2773 (ii) Exim was setting the flag that allows error details to be returned
2774 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2775 recipients that were returned after the DATA command.
2777 PH/53 This is related to PH/52, but is more general: for any failing address,
2778 when detailed error information was permitted to be returned to the
2779 sender, but the error was temporary, then after the final timeout, only
2780 "retry timeout exceeded" was returned. Now it returns the full error as
2781 well as "retry timeout exceeded".
2783 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2784 do this, and (what is worse) MTAs that accept it.
2786 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2787 will now be deprecated.
2789 PH/56 New os.c-cygwin from the Cygwin maintainer.
2791 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2792 formats (previously only available in exim formats via -bpr, -bpru,
2793 and -bpra. Now also available in native and exiqgrep formats)
2795 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2796 with very large, slow to parse queues
2798 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2800 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2802 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2803 responses to authentication challenges, though it was showing the
2804 challenges; (ii) I've removed the CR characters from the debug output for
2807 PH/58 Allow for the insertion of a newline as well as a space when a string
2808 is turned into more than one encoded-word during RFC 2047 encoding. The
2809 Sieve code now uses this.
2811 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2812 data_4xx, lost_connection, tls_required.
2814 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2815 message was being sent as an SMTP response.
2817 PH/61 Add -l and -k options to exicyclog.
2819 PH/62 When verifying, if an address was redirected to one new address, so that
2820 verification continued, and the new address failed or deferred after
2821 having set something in $address_data, the value of $address_data was not
2822 passed back to the ACL. This was different to the case when no
2823 redirection occurred. The value is now passed back in both cases.
2825 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2826 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2827 use login_cap.h, so on its own it isn't the distinguishing feature. The
2828 new name refers directly to the setclassresources() function.
2830 PH/65 Added configuration files for NetBSD3.
2832 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2834 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2835 is preferred over IPv4.
2837 PH/68 The bounce_return_message and bounce_return_body options were not being
2838 honoured for bounces generated during the reception of non-SMTP messages.
2839 In particular, this applied to messages rejected by the ACL. This bug has
2840 been fixed. However, if bounce_return_message is true and bounce_return_
2841 body is false, the headers that are returned for a non-SMTP message
2842 include only those that have been read before the error was detected.
2843 (In the case of an ACL rejection, they have all been read.)
2845 PH/69 The HTML version of the specification is now built in a directory called
2846 spec_html instead of spec.html, because the latter looks like a path with
2847 a MIME-type, and this confuses some software.
2849 PH/70 Catch two compiler warnings in sieve.c.
2851 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2852 function verify_get_ident() calls ip_connect() to connect a socket, but
2853 if the "connect()" function timed out, ip_connect() used to close the
2854 socket. However, verify_get_ident() also closes the socket later, and in
2855 between Exim writes to the log, which may get opened at this point. When
2856 the socket was closed in ip_connect(), the log could get the same file
2857 descriptor number as the socket. This naturally causes chaos. The fix is
2858 not to close the socket in ip_connect(); the socket should be closed by
2859 the function that creates it. There was only one place in the code where
2860 this was missing, in the iplookup router, which I don't think anybody now
2861 uses, but I've fixed it anyway.
2863 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2864 well as to direct DNS lookups. Otherwise the handling of names in host
2865 lists is inconsistent and therefore confusing.
2871 PH/01 Two changes to the default runtime configuration:
2873 (1) Move the checks for relay_from_hosts and authenticated clients from
2874 after to before the (commented out) DNS black list checks.
2876 (2) Add control=submission to the relay_from_hosts and authenticated
2877 clients checks, on the grounds that messages accepted by these
2878 statements are most likely to be submissions.
2880 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2882 (1) Generate an error if the third argument for the ${prvs expansion is
2885 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2888 (3) Reset the variables that are obtained from the first argument of
2889 ${prvscheck and used in the second argument before leaving the code,
2890 because their memory is reclaimed, so using them afterwards may do
2893 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2894 one (it's much easier than Tom thought :-).
2896 (5) Because of (4), we can now allow for the use of $prvscheck_result
2897 inside the third argument.
2899 PH/03 For some reason, the default setting of PATH when running a command from
2900 a pipe transport was just "/usr/bin". I have changed it to
2903 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2904 anything to be listed in the output from -bV.
2906 PH/05 When a filter generated an autoreply, the entire To: header line was
2907 quoted in the delivery log line, like this:
2909 => >A.N.Other <ano@some.domain> <original@ddress> ...
2911 This has been changed so that it extracts the operative address. There
2912 may be more than one such address. If so, they are comma-separated, like
2915 => >ano@some.domain,ona@other.domain <original@ddress> ...
2917 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2918 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2919 not being looked up in the rDNS to get a host name, Exim was showing the
2920 IP address twice in Received: lines, even though the IP addresses were
2921 identical. For example:
2923 Received: from [1.2.3.4] (helo=[1.2.3.4])
2925 However, if the real host name was known, it was omitting the HELO data
2926 if it matched the actual IP address. This has been tidied up so that it
2927 doesn't show the same IP address twice.
2929 PH/07 When both +timestamp and +memory debugging was on, the value given by
2930 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2931 called by the debug printing, thereby overwriting the timestamp buffer.
2932 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2934 PH/08 When the original message was included in an autoreply transport, it
2935 always said "this is a copy of the message, including all the headers",
2936 even if body_only or headers_only was set. It now gives an appropriate
2939 PH/09 Applied a patch from the Sieve maintainer which:
2941 o fixes some comments
2942 o adds the (disabled) notify extension core
2943 o adds some debug output for the result of if/elsif tests
2944 o points to the current vacation draft in the documentation
2945 and documents the missing references header update
2949 o fixes a bug in processing the envelope test (when testing
2950 multiple envelope elements, the last element determinted the
2953 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2954 Electronic Mail") by including:
2956 Auto-submitted: auto-generated
2958 in the messages that it generates (bounce messages and others, such as
2959 warnings). In the case of bounce messages for non-SMTP mesages, there was
2960 also a typo: it was using "Auto_submitted" (underscore instead of
2961 hyphen). Since every message generated by Exim is necessarily in response
2962 to another message, thes have all been changed to:
2964 Auto-Submitted: auto-replied
2966 in accordance with these statements in the RFC:
2968 The auto-replied keyword:
2970 - SHOULD be used on messages sent in direct response to another
2971 message by an automatic process,
2973 - MUST NOT be used on manually-generated messages,
2975 - MAY be used on Delivery Status Notifications (DSNs) and Message
2976 Disposition Notifications (MDNs),
2978 - MUST NOT be used on messages generated by automatic or periodic
2979 processes, except for messages which are automatic responses to
2982 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2983 to the default Received: header definition.
2985 PH/12 Added log selector acl_warn_skipped (default on).
2987 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2988 variables because (a) they are in the wrong storage pool and (b) even if
2989 they were copied, it wouldn't work properly because of the caching.
2991 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2992 checking when decoding. Apparently there are clients that generate
2993 overlong encoded strings. Why am I not surprised?
2995 PH/15 If the first argument of "${if match_address" was not empty, but did not
2996 contain an "@" character, Exim crashed. Now it writes a panic log message
2997 and treats the condition as false.
2999 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3001 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3002 extension "envelope-auth". The code is finished and in agreement with
3003 other implementations, but there is no documentation so far and in fact,
3004 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3005 not changing the active code.
3007 Print executed "if" and "elsif" statements when debugging is used. This
3008 helps a great deal to understand what a filter does.
3010 Document more things not specified clearly in RFC3028. I had all this
3011 sorted out, when out of a sudden new issues came to my mind. Oops."
3013 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3016 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3017 canonical form (as documented). However, after a host name lookup from
3018 the IP address, check_host() was doing a simple string comparison with
3019 addresses acquired from the DNS when checking that the found name did
3020 have the original IP as one of its addresses. Since any found IPv6
3021 addresses are likely to be in abbreviated form, the comparison could
3022 fail. Luckily, there already exists a function for doing the comparison
3023 by converting both addresses to binary, so now that is used instead of
3024 the text comparison.
3026 PH/20 There was another similar case to PH/19, when a complete host name was
3027 given in a host list; looking up its IP address could give an abbreviated
3028 form, whereas the current host's name might or might not be abbreviated.
3029 The same fix has been applied.
3035 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3036 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3039 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3040 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3042 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3044 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3045 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3046 Also, for "Auto-Submitted", treat anything other than "no" as
3047 non-personal, in accordance with RFC 3834. (Previously it treated
3048 anything starting "auto-" as non-personal.)
3050 TF/01 The control=submission/name=... option had a problem with syntax
3051 errors if the name included a slash character. The /name= option
3052 now slurps the rest of the string, so it can include any characters
3053 but it must come last in the list of options (after /sender_retain
3056 PH/05 Some modifications to the interface to the fake nameserver for the new
3064 TK/01 Added the "success_on_redirect" address verification option. See
3065 NewStuff for rationale and an example.
3067 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3069 PH/02 Patch to exigrep to allow it to work on syslog lines.
3071 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3072 fread() to skip over the body file's header line, because in Cygwin the
3073 header line is locked and is inaccessible.
3075 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3076 co-exist for some time) to make it clear that it is the Exim ID that is
3077 referenced, not the Message-ID: header line.
3079 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3080 string_format() function, because snprintf() does not exist on all
3083 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3084 ${stat: expansion item.
3086 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3087 protocol synchronization error", to keep the pedants happy.
3089 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3090 well as for IRIX systems, when gcc is being used. See the host.c source
3093 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3095 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3098 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3099 transport and to the smtp transport in LMTP mode.
3101 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3103 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3104 run a filter in a subprocess. This could lead to confusion in subsequent
3105 lookups in the parent process. There should also be a search_tidyup() at
3106 the end of the subprocess.
3108 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3109 only if the host matched helo_try_verify_hosts, which caused the
3110 verification to occur when the EHLO/HELO command was issued. The ACL just
3111 tested the remembered result. Now, if a previous verification attempt has
3112 not happened, "verify = helo" does it there and then.
3114 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3116 TK/03 Fix log output including CR from clamd.
3118 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3119 exist provoked a memory error which could cause a segfault.
3121 PH/15 Installed PCRE 6.2
3123 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3125 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3126 of the problem. Specifically, suggested +O2 rather than +O1 for the
3129 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3131 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3132 record, the port was not correctly logged when the outgoing_port log
3133 selector was set (it logged the transort's default port).
3135 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3136 fallback_hosts, and "hosts" in the smtp transport.
3138 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3139 host errors such as "Connection refused".
3141 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3142 authentication with radiusclient 0.4.9:
3144 - Error returned from rc_read_config was caught wrongly
3145 - Username/password not passed on to radius server due to wrong length.
3147 The presumption is that some radiusclient API changes for 4.51/PH/17
3148 were not taken care of correctly. The code is still untested by me (my
3149 Linux distribution still has 0.3.2 of radiusclient), but it was
3150 contributed by a Radius user.
3152 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3153 expanding the "port" option of the smtp transport.
3155 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3156 while reading a MIME header. Thanks to Tom Hughes for a patch.
3158 PH/24 Include config.h inside local_scan.h so that configuration settings are
3161 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3162 that in an ACL for RCPT (for example), you can examine exactly what was
3165 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3166 commands, but it was not correctly comparing the address with the actual
3167 client host address. Thus, it would show the EHLO address in Received:
3168 header lines when this was not necessary.
3170 PH/27 Added the % operator to ${eval:}.
3172 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3173 it should be ignoring failures (because with -C, for example, it has lost
3174 privilege). It wasn't ignoring creation failures other than "already
3177 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3180 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3181 because an input line was too long, either on its own, or by virtue of
3182 too many continuations, the temporary file was not being removed, and the
3183 return code was incorrect.
3185 PH/31 Missing "BOOL" in function definition in filtertest.c.
3187 PH/32 Applied Sieve patches from the maintainer.
3189 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3191 PH/33 Added "verify = not_blind".
3193 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3194 Local/Makefile (with some defaults set). These are used in built scripts
3195 such as exicyclog, but they have never been used in the exim_install
3196 script (though there are many overriding facilities there). I have
3197 arranged that the exim_install script now takes note of these two
3200 PH/35 Installed configuration files for Dragonfly.
3202 PH/36 When a locally submitted message by a trusted user did not contain a
3203 From: header, and the sender address was obtained from -f or from an SMTP
3204 MAIL command, and the trusted user did not use -F to supply a sender
3205 name, $originator_name was incorrectly used when constructing a From:
3206 header. Furthermore, $originator_name was used for submission mode
3207 messages from external hosts without From: headers in a similar way,
3208 which is clearly wrong.
3210 PH/37 Added control=suppress_local_fixups.
3212 PH/38 When log_selector = +received_sender was set, and the addition of the
3213 sender made the log line's construction buffer exactly full, or one byte
3214 less than full, an overflow happened when the terminating "\n" was
3217 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3218 when the result of a list match is failure because a DNS lookup failed.
3220 PH/40 RM_COMMAND is now used in the building process.
3222 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3223 the "build-* directories that it finds.
3225 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3226 address in a domain literal was a prefix of an interface address.
3228 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3229 when verifying a sender address, unless rewrite_headers is false.
3231 PH/44 Wrote a long comment about why errors_to addresses are verified as
3232 recipients, not senders.
3234 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3235 the ratelimit ACL was added.
3237 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3239 PH/46 Added extra information about PostgreSQL errors to the error string.
3241 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3242 avoiding the need to install special zones in a real server. This is
3243 backwards compatible; if it can't find the fake resolver, it drops back.
3244 Thus, both old and new test suites can be run.
3246 TF/02 Added util/ratelimit.pl
3248 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3249 clock is set back in time.
3251 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3252 Candler <B.Candler@pobox.com>.
3254 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3255 OK for addresses that are the result of redirections.
3257 PH/48 A number of further additions for the benefit of the new test suite,
3258 including a fake gethostbyname() that interfaces to the fake DNS resolver
3261 TF/06 The fix for widen_domains has also been applied to qualify_single and
3262 search_parents which are the other dnslookup options that can cause
3265 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3268 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3270 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3271 service" if it encountered an MX record with an empty target hostname.
3272 The message is now "an MX or SRV record indicated no SMTP service".
3274 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3275 if the DNS of the sending site is misconfigured. This is quite a
3276 common situation. This change restores the behaviour of treating a
3277 helo verification defer as a failure.
3279 PH/51 If self=fail was set on a router, the bounce message did not include the
3280 actual error message.
3286 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3288 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3289 command itself ended in error, the underlying message about the transport
3290 filter timeout was being overwritten with the pipe command error. Now the
3291 underlying error message should be appended to the second error message.
3293 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3295 PH/02 Reduce the amount of output that "make" produces by default. Full output
3296 can still be requested.
3298 PH/03 The warning log line about a condition test deferring for a "warn" verb
3299 was being output only once per connection, rather than after each
3300 occurrence (because it was using the same function as for successful
3301 "warn" verbs). This seems wrong, so I have changed it.
3303 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3304 it should not have, which might have caused a crash in the right
3305 circumstances, but probably never did.
3307 PH/04 Installed a modified version of Tony Finch's patch to make submission
3308 mode fix the return path as well as the Sender: header line, and to
3309 add a /name= option so that you can make the user's friendly name appear
3312 TF/03 Added the control = fakedefer ACL modifier.
3314 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3315 Mark Lowes for thorough testing.
3317 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3319 TK/03 Merged latest SRS patch from Miles Wilton.
3321 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3322 with the definition in sysexits.h (which is #included earlier).
3323 Fortunately, Exim does not actually use EX_OK. The code used to try to
3324 preserve the sysexits.h value, by assumimg that macro definitions were
3325 scanned for macro replacements. I have been disabused of this notion,
3326 so now the code just undefines EX_OK before #including unistd.h.
3328 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3329 in the smtp transport. When a block could not be written in a single
3330 write() function, the timeout was being re-applied to each part-write.
3331 This seems wrong - if the receiver was accepting one byte at a time it
3332 would take for ever. The timeout is now adjusted when this happens. It
3333 doesn't have to be particularly precise.
3335 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3336 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3338 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3339 without a domain if the check to <postmaster@domain> fails.
3341 SC/01 Eximstats: added -xls and the ability to specify output files
3342 (patch written by Frank Heydlauf).
3344 SC/02 Eximstats: use FileHandles for outputing results.
3346 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3348 SC/04 Eximstats: fixed display of large numbers with -nvr option
3350 SC/05 Eximstats: fixed merging of reports with empty tables.
3352 SC/06 Eximstats: added the -include_original_destination flag
3354 SC/07 Eximstats: removed tabs and trailing whitespace.
3356 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3358 TK/06 MBOX spool code: Add real "From " MBOX separator line
3359 so the .eml file is really in mbox format (even though
3360 most programs do not really care). Patch from Alex Miller.
3362 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3363 The latter is generated from $received_to and is only set if the
3364 message has one envelope recipient. SA can use these headers,
3365 obviously out-of-the-box. Patch from Alex Miller.
3367 PH/08 The ${def test on a variable was returning false if the variable's
3368 value was "0", contrary to what the specification has always said!
3369 The result should be true unless the variable is empty.
3371 PH/09 The syntax error of a character other than { following "${if
3372 def:variable_name" (after optional whitespace) was not being diagnosed.
3373 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3374 accidental colon was present, for example, could give incorrect results.
3376 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3377 result is used (not including appendfile, where other changes are about
3380 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3381 This involved changing a lot of size variables from int to off_t. It
3382 should work with maildirs and everything.
3384 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3385 spamd dying while we are connected to it.
3387 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3390 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3391 function for BDB 4.3.
3393 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3395 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3396 the sources for PCRE have been re-arranged and more files are now
3399 PH/15 The code I had for printing potentially long long variables in PH/11
3400 above was not the best (it lost precision). The length of off_t variables
3401 is now inspected at build time, and an appropriate printing format (%ld
3402 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3403 to be "long long int" or "long int". This is needed for the internal
3404 formatting function string_vformat().
3406 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3407 the configuration file to be ":syslog", then the script "guesses" where
3408 the logs files are, rather than using the compiled in default. In our
3409 case the guess is not the same as the compiled default, so the script
3410 suddenly stopped working when I started to use syslog. The patch checks
3411 to see if log_file_path is "". If so, it attempts to read it from exim
3412 with no configuration file to get the compiled in version, before it
3413 falls back to the previous guessing code."
3415 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3416 implementing BATV in an Exim configuration. See NewStuff for the gory
3419 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3420 Makefile that are specific to HP-UX.
3422 PH/18 If the "use_postmaster" option was set for a recipient callout together
3423 with the "random" option, the postmaster address was used as the MAIL
3424 FROM address for the random test, but not for the subsequent recipient
3425 test. It is now used for both.
3427 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3428 patch removes a few documentation additions to RFC 3028, because the
3429 latest draft now contains them. It adds the new en;ascii-case comparator
3430 and a new error check for 8bit text in MIME parts. Comparator and
3431 require names are now matched exactly. I enabled the subaddress
3432 extension, but it is not well tested yet (read: it works for me)."
3434 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3435 rework some of the code of TK/09 above to avoid the hardwired use of
3436 "%lld" and "long long". Replaced the call to snprintf() with a call to
3439 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3440 records point to non-existent hosts", "retry timeout exceeded", and
3441 "retry time not reached for any host after a long failure period".
3443 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3444 experimental DomainKeys support:
3446 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3447 (2) On an error such as an illegally used "control", the wrong name for
3448 the control was given.
3450 These problems did NOT occur unless DomainKeys support was compiled.
3452 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3454 PH/24 Added ${if match_ip condition.
3456 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3457 databases so that it will be absolutely obvious if a crash occurs in the
3458 DB library. This is a regular occurrence (often caused by mis-matched
3461 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3462 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3463 picked up on a user's system that detects such things. There doesn't seem
3464 to be a gcc warning option for this - only an attribute that has to be
3465 put on the function's prototype. It seems that in Fedora Core 4 they have
3466 set this on a number of new functions. No doubt there will be more in due
3469 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3470 specify a transport. However, if an address that was verified by such a
3471 router was the subject of a callout, Exim crashed because it tried to
3472 read the rcpt_include_affixes from the non-existent transport. Now it
3473 just assumes that the setting of that option is false. This bug was
3474 introduced by 4.51/PH/31.
3476 PH/28 Changed -d+all to exclude +memory, because that information is very
3477 rarely of interest, but it makes the output a lot bigger. People tend to
3478 do -d+all out of habit.
3480 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3481 code in os-type was giving problems when libc.so lives in lib64, like on
3484 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3485 aren't the modern standard, and it seems that some systems' include files
3486 don't always have them. Exim was already checking for some of the newer
3487 ones like T_AAAA, and defining it itself. I've added checks for all the
3488 record types that Exim uses.
3490 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3491 not automatically generating a new one, as it is supposed to. This
3492 prevented TLS from working. If the file did exist, but contained invalid
3493 data, a new version was generated, as expected. It was only the case of a
3494 non-existent file that was broken.
3496 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3497 with a change in libdomainkeys > 0.64.
3499 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3500 from DNS. If the selector record carries the flag, it now has
3501 precedence over the domain-wide flag.
3503 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3505 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3506 the use of a port name that isn't defined in /etc/services) occurred, the
3507 message was deferred as in a normal delivery, and thus remained on the
3508 spool, instead of being failed because of the mua_wrapper setting. This
3509 is now fixed, and I tidied up some of the mua_wrapper messages at the
3512 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3513 the messages in a hash of arrays rather than using individual hashes.
3514 This is a bit cleaner and results in dramatic memory savings, albeit
3515 at a slight CPU cost.
3517 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3518 as requested by Marc Sherman.
3520 SC/10 Eximstats: added histograms for user specified patterns as requested
3523 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3525 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3526 fopen() in the content-scanning modules that did not already have it.
3532 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3533 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3535 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3537 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3539 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3540 to test Sieve filters that use "vacation".
3542 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3543 that changes the way the GnuTLS parameters are stored in the cache file.
3544 The new format can be generated externally. For backward compatibility,
3545 if the data in the cache doesn't make sense, Exim assumes it has read an
3546 old-format file, and it generates new data and writes a new file. This
3547 means that you can't go back to an older release without removing the
3550 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3551 work if there are any delivery delays because "one_time" forces the
3552 parent to be marked "delivered", so its unseen clone is never tried
3553 again. For this reason, Exim now forbids the simultaneous setting of
3556 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3557 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3558 entry if you want to know the details. The fix, however, neglected to
3559 consider the case when local delivery batching is involved. The test for
3560 "previously delivered" was not happening when checking to see if an
3561 address could be batched with a previous (undelivered) one; under
3562 certain circumstances this could lead to multiple deliveries to the same
3565 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3566 in its include files, and this causes problems building Exim.
3568 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3569 header_syntax) but Exim was just ignoring anything given after a slash.
3570 In particular, this caused confusion with an attempt to use "verify =
3571 reverse_host_lookup/defer_ok". An error is now given when options are
3572 supplied for verify items that do not have them. (Maybe reverse_host_
3573 lookup should have a defer_ok option, but that's a different point.)
3575 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3576 defined by RFC 821) to 2048, because there were problems with some AUTH
3577 commands, and RFC 1869 says the size should be increased for extended
3578 SMTP commands that take arguments.
3580 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3583 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3584 "unknown" error; now it says that the functionality isn't in the binary.
3586 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3587 an address' error message when a string expansion fails (syntax or
3588 whatever). Otherwise the password may appear in the log. Following change
3589 PH/42 below, there is no longer a chance of it appearing in a bounce
3592 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3594 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3595 of its addresses was ever tried. (Bugzilla bug #2.)
3597 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3598 the result incorrectly in the debug output. (It correctly added a newline
3599 to what was transported.)
3601 TF/01 Added $received_time.
3603 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3604 commented out examples of how to interface to a virus scanner and to
3605 SpamAssassin. Also added commented examples of av_scanner and
3606 spamd_address settings.
3608 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3609 and controls are allowed in which ACLs. There were a couple of minor
3610 errors. Some of the entries in the conditions table (which is a table of
3611 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3612 negation of where the condition IS allowed.
3614 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3616 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3617 header file does not have a version number, so I've had to invent a new
3618 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3619 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3620 radiusclient), but it was contributed by a Radius user.
3622 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3623 files or directories, for OpenSSL.
3625 PH/19 When an Exim process that is running as root has to create an Exim log
3626 file, it does so in a subprocess that runs as exim:exim so as to get the
3627 ownership right at creation (otherwise, other Exim processes might see
3628 the file with the wrong ownership). There was no test for failure of this
3629 fork() call, which would lead to the process getting stuck as it waited
3630 for a non-existent subprocess. Forks do occasionally fail when resources
3631 run out. I reviewed all the other calls to fork(); they all seem to check
3634 PH/20 When checking for unexpected SMTP input at connect time (before writing
3635 the banner), Exim was not dealing correctly with a non-positive return
3636 from the read() function. If the client had disconnected by this time,
3637 the result was a log entry for a synchronization error with an empty
3638 string after "input=" when read() returned zero. If read() returned -1
3639 (an event I could not check), uninitialized data bytes were printed.
3640 There were reports of junk text (parts of files, etc) appearing after
3643 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3645 PH/22 Added support for macro redefinition, and (re)definition in between
3646 driver and ACL definitions.
3648 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3649 forgetting to use the resulting value; it was using the unexpanded value.
3651 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3652 hadn't been configured. The fix is from Juergen Kreileder, who
3653 understands it better than I do:
3655 "Here's what I see happening with three configured cyrus_sasl
3656 authenticators configured (plain, login, cram-md5):
3658 On startup auth_cyrus_sasl_init() gets called for each of these.
3659 This means three calls to sasl_listmech() without a specified mech_list.
3660 => SASL tests which mechs of all available mechs actually work
3661 => three warnings about OTP not working
3662 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3664 With the patch, sasl_listmech() also gets called three times. But now
3665 SASL's mech_list option is set to the server_mech specified in the the
3666 authenticator. Or in other words, the answer from sasl_listmech()
3667 gets limited to just the mech you're testing for (which is different
3669 => the return list contains just 'plain' or 'login', 'cram-md5' or
3670 nothing depending on the value of ob->server_mech.
3672 I've just tested the patch: Authentication still works fine,
3673 unavailable mechs specified in the exim configuration are still
3674 caught, and the auth.log warnings about OTP are gone."
3676 PH/25 When debugging is enabled, the contents of the command line are added
3677 to the debugging output, even when log_selector=+arguments is not
3680 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3681 answer is "GNU", and only if the return is "GNU/something" is the answer
3684 PH/27 $acl_verify_message is now set immediately after the failure of a
3685 verification in an ACL, and so is available in subsequent modifiers. In
3686 particular, the message can be preserved by coding like this:
3688 warn !verify = sender
3689 set acl_m0 = $acl_verify_message
3691 Previously, $acl_verify_message was set only while expanding "message"
3692 and "log_message" when a very denied access.
3694 PH/28 Modified OS/os.c-Linux with
3696 -#ifndef OS_LOAD_AVERAGE
3697 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3699 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3700 nomenclature these days.)
3702 PH/29 Installed patch from the Sieve maintainer that adds the options
3703 sieve_useraddress and sieve_subaddress to the redirect router.
3705 PH/30 In these circumstances:
3706 . Two addresses routed to the same list of hosts;
3707 . First host does not offer TLS;
3708 . First host accepts first address;
3709 . First host gives temporary error to second address;
3710 . Second host offers TLS and a TLS session is established;
3711 . Second host accepts second address.
3712 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3713 and peerdn, if requested) that were in fact used only for the second
3716 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3717 attention to any local part prefix or suffix that was matched by the
3718 router that accepted the address. It now behaves in the same way as it
3719 does for delivery: the affixes are removed from the local part unless
3720 rcpt_include_affixes is set on the transport.
3722 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3723 timeout during the DATA phase of an incoming message.
3725 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3726 applied a patch sanctioned by the Sieve maintainer.
3728 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3729 the uid or gid is negative. A case of a negative gid caused this to be
3730 noticed. The fix allows for either to be negative.
3732 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3733 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3736 PH/36 The change PH/12 above was broken. Fixed it.
3738 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3739 the grounds that routing the same address twice would always produce the
3740 same answer. This might have been true once, but it is certainly no
3741 longer true now. Routing a child address may depend on the previous
3742 routing that produced that child. Some complicated redirection strategies
3743 went wrong when messages had multiple recipients, and made Exim's
3744 behaviour dependent on the order in which the addresses were given.
3746 I have moved the duplicate checking until after the routing is complete.
3747 Exim scans the addresses that are assigned to local and remote
3748 transports, and removes any duplicates. This means that more work will be
3749 done, as duplicates will always all be routed, but duplicates are
3750 presumably rare, so I don't expect this is of any significance.
3752 For deliveries to pipes, files, and autoreplies, the duplicate checking
3753 still happens during the routing process, since they are not going to be
3756 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3757 It corrects a timeout issue with spamd. This is Ian's comment: "The
3758 background is that sometimes spamd either never reads data from a
3759 connection it has accepted, or it never writes response data. The exiscan
3760 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3761 blindly assumes that writes won't block so it may never time out."
3763 PH/39 Allow G after quota size as well as K and M.
3765 PH/40 The value set for $authenticated_id in an authenticator may not contain
3766 binary zeroes or newlines because the value is written to log lines and
3767 to spool files. There was no check on this. Now the value is run through
3768 the string_printing() function so that such characters are converted to
3769 printable escape sequences.
3771 PH/41 $message_linecount is a new variable that contains the total number of
3772 lines in the message. Compare $body_linecount, which is the count for the
3775 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3776 bounce and delay warning messages, except in certain special cases, which
3779 (a) An SMTP error message from a remote host;
3780 (b) A message specified in a :fail: redirection;
3781 (c) A message specified in a "fail" command in a system filter;
3782 (d) A message specified in a FAIL return from the queryprogram router;
3783 (e) A message specified by the cannot_route_message router option.
3785 In these cases only, Exim does include the error details in bounce and
3786 warning messages. There are also a few cases where bland messages such
3787 as "unrouteable address" or "local delivery error" are given.
3789 PH/43 $value is now also set for the "else" part of a ${run expansion.
3791 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3792 being worked on, but at least Exim now implements the latest version to
3795 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3796 process to complete was treated as a delivery failure, a timeout while
3797 writing the message to the pipe was logged, but erroneously treated as a
3798 successful delivery. Such timeouts include transport filter timeouts. For
3799 consistency with the overall process timeout, these timeouts are now
3800 treated as errors, giving rise to delivery failures by default. However,
3801 there is now a new Boolean option for the pipe transport called
3802 timeout_defer, which, if set TRUE, converts the failures into defers for
3803 both kinds of timeout. A transport filter timeout is now identified in
3806 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3807 systems where "make" and "gmake" are different, calling "gmake" at top
3808 level broke things. I've arranged for the value of $(MAKE) to be passed
3809 from the Makefile to this script so that it can call the same version of
3813 A note about Exim versions 4.44 and 4.50
3814 ----------------------------------------
3816 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3817 changes of various kinds. As a consequence, a big documentation update was
3818 needed. This delayed the release for rather longer than seemed good, especially
3819 in the light of a couple of (minor) security issues. Therefore, the changes
3820 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3821 release. So 4.44 and 4.50 are in effect two different branches that both start
3824 I have left the 4.50 change log unchanged; it contains all the changes since
3825 4.43. The change log for 4.44 is below; many of its items are identical to
3826 those for 4.50. This seems to be the most sensible way to preserve the
3827 historical information.
3833 1. Minor wording change to the doc/README.SIEVE file.
3835 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3836 computation of the current number of files was incorrect.
3838 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3839 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3842 4. Give more explanation in the error message when the command for a transport
3843 filter fails to execute.
3845 5. There are several places where Exim runs a non-Exim command in a
3846 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3847 was being done only for the command run by the queryprogram router. It is
3848 now done for all such subprocesses. The other cases are: ${run, transport
3849 filters, and the commands run by the lmtp and pipe transports.
3851 6. Added CONFIGURE_GROUP build-time option.
3853 7. Some older OS have a limit of 256 on the maximum number of file
3854 descriptors. Exim was using setrlimit() to set 1000 as a large value
3855 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3856 systems. I've change it so that if it can't get 1000, it tries for 256.
3858 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3859 was an oversight, and furthermore, ever since the addition of extra
3860 controls (e.g. 4.43/32), the checks on when to allow different forms of
3861 "control" were broken. There should now be diagnostics for all cases when a
3862 control that does not make sense is encountered.
3864 9. Added the /retain_sender option to "control=submission".
3866 10. $recipients is now available in the predata ACL (oversight).
3868 11. Tidy the search cache before the fork to do a delivery from a message
3869 received from the command line. Otherwise the child will trigger a lookup
3870 failure and thereby defer the delivery if it tries to use (for example) a
3871 cached ldap connection that the parent has called unbind on.
3873 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3874 of $address_data from the recipient verification was clobbered by the
3875 sender verification.
3877 13. The value of address_data from a sender verification is now available in
3878 $sender_address_data in subsequent conditions in the ACL statement.
3880 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3882 15. Added a new option "connect=<time>" to callout options, to set a different
3885 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3886 was its contents. (It was OK if the option was not defined at all.)
3888 17. A "Completed" log line is now written for messages that are removed from
3889 the spool by the -Mrm option.
3891 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3892 information about exactly what failed.
3894 19. Added -dd to debug only the daemon process.
3896 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3897 handles timeouts, both on the server side and network timeouts. Renamed the
3898 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3900 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3901 It is now set to "smtps".
3903 22. $host_address is now set to the target address during the checking of
3904 ignore_target_hosts.
3906 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3907 being passed; this would have caused $sender_host_name to have been used if
3908 matching the list had actually called for a host name (not very likely,
3909 since this list is usually IP addresses). A host name is now passed as
3912 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3913 code that specifies a non-restarting handler (typically sigaction() in
3914 modern systems) in an attempt to fix a rare and obscure crash bug.
3916 25. Narrowed the window for a race in the daemon that could cause it to ignore
3917 SIGCHLD signals. This is not a major problem, because they are used only to
3918 wake it up if nothing else does.
3920 26. A malformed maildirsize file could cause Exim to calculate negative values
3921 for the mailbox size or file count. Odd effects could occur as a result.
3922 The maildirsize information is now recalculated if the size or filecount
3925 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3926 support for a long time. Removed HAVE_SYS_VFS_H.
3928 28. Installed the latest version of exipick from John Jetmore.
3930 29. In an address list, if the pattern was not a regular expression, an empty
3931 subject address (from a bounce message) matched only if the pattern was an
3932 empty string. Non-empty patterns were not even tested. This was the wrong
3933 because it is perfectly reasonable to use an empty address as part of a
3934 database query. An empty address is now tested by patterns that are
3935 lookups. However, all the other forms of pattern expect the subject to
3936 contain a local part and a domain, and therefore, for them, an empty
3937 address still always fails if the pattern is not itself empty.
3939 30. Exim went into a mad DNS loop when attempting to do a callout where the
3940 host was specified on an smtp transport, and looking it up yielded more
3941 than one IP address.
3943 31. Re-factored the code for checking spool and log partition space into a
3944 function that finds that data and another that does the check. The former
3945 is then used to implement four new variables: $spool_space, $log_space,
3946 $spool_inodes, and $log_inodes.
3948 32. The RFC2047 encoding function was originally intended for short strings
3949 such as real names; it was not keeping to the 75-character limit for
3950 encoded words that the RFC imposes. It now respects the limit, and
3951 generates multiple encoded words if necessary. To be on the safe side, I
3952 have increased the buffer size for the ${rfc2047: expansion operator from
3955 33. It is now permitted to omit both strings after an "if" condition; if the
3956 condition is true, the result is "true". As before, when the second string
3957 is omitted, a false condition yields an empty string. This makes it less
3958 cumbersome to write custom ACL and router conditions.
3960 34. Failure to deliver a bounce message always caused it to be frozen, even if
3961 there was an errors_to setting on the router. The errors_to setting is now
3964 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3965 canonical form (fully expanded) before being placed in
3966 $sender_host_address.
3968 36. The table in the code that translates DNS record types into text (T_A to
3969 "A" for instance) was missing entries for NS and CNAME. It is just possible
3970 that this could have caused confusion if both these types were looked up
3971 for the same domain, because the text type is used as part of Exim's
3972 per-process caching. But the chance of anyone hitting this buglet seems
3975 37. The dnsdb lookup has been extended in a number of ways.
3977 (1) There is a new type, "zns", which walks up the domain tree until it
3978 finds some nameserver records. It should be used with care.
3980 (2) There is a new type, "mxh", which is like "mx" except that it returns
3981 just the host names, not the priorities.
3983 (3) It is now possible to give a list of domains (or IP addresses) to be
3984 looked up. The behaviour when one of the lookups defers can be
3985 controlled by a keyword.
3987 (4) It is now possible to specify the separator character for use when
3988 multiple records are returned.
3990 38. The dnslists ACL condition has been extended: it is now possible to supply
3991 a list of IP addresses and/or domains to be looked up in a particular DNS
3994 39. Added log_selector=+queue_time_overall.
3996 40. When running the queue in the test harness, wait just a tad after forking a
3997 delivery process, to get repeatability of debugging output.
3999 41. Include certificate and key file names in error message when GnuTLS fails
4000 to set them up, because the GnuTLS error message doesn't include the name
4001 of the failing file when there is a problem reading it.
4003 42. Allow both -bf and -bF in the same test run.
4005 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4007 44. The "Exiscan patch" is now merged into the mainline Exim source.
4009 45. Sometimes the final signoff response after QUIT could fail to get
4010 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4011 < 0 before doing a fflush(). This bug looks as though it goes back to the
4012 introduction of TLS in release 3.20, but "sometimes" must have been rare
4013 because the tests only now provoked it.
4015 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4016 (this can affect the format of dates).
4018 47. exim_tidydb, when checking for the continued existence of a message for
4019 which it has found a message-specific retry record, was not finding
4020 messages that were in split spool directories. Consequently, it was
4021 deleting retry records that should have stayed in existence.
4023 48. Steve fixed some bugs in eximstats.
4025 49. The SPA authentication driver was not abandoning authentication and moving
4026 on to the next authenticator when an expansion was forced to fail,
4027 contradicting the general specification for all authenticators. Instead it
4028 was generating a temporary error. It now behaves as specified.
4030 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4031 (the order specifies the preference for clients). The order is now AES256,
4032 AES128, 3DES, ARCFOUR128.
4034 51. Small patch to Sieve code - explicitly set From: when generating an
4037 52. Exim crashed if a remote delivery caused a very long error message to be
4038 recorded - for instance if somebody sent an entire SpamAssassin report back
4039 as a large number of 550 error lines. This bug was coincidentally fixed by
4040 increasing the size of one of Exim's internal buffers (big_buffer) that
4041 happened as part of the Exiscan merge. However, to be on the safe side, I
4042 have made the code more robust (and fixed the comments that describe what
4045 53. Now that there can be additional text after "Completed" in log lines (if
4046 the queue_time_overall log selector is set), a one-byte patch to exigrep
4047 was needed to allow it to recognize "Completed" as not the last thing in
4050 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4051 patch that reportedly fixes this has been added. I am not expert enough to
4052 create a test for it. This is what the patch creator wrote:
4054 "I found a little strange behaviour of ldap code when working with
4055 Windows 2003 AD Domain, where users was placed in more than one
4056 Organization Units. When I tried to give exim partial DN, the exit code
4057 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4058 But simultaneously result of request was absolutely normal ldap result,
4059 so I produce this patch..."
4061 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4062 REFERENCE, so I have modified the code to exclude the patch when that macro
4065 55. Some experimental protocols are using DNS PTR records for new purposes. The
4066 keys for these records are domain names, not reversed IP addresses. The
4067 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4068 leaves it alone. Component reversal etc. now happens only for IP addresses.
4071 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4073 57. Double the size of the debug message buffer (to 2048) so that more of very
4074 long debug lines gets shown.
4076 58. The exicyclog utility now does better if the number of log files to keep
4077 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4079 59. Two changes related to the smtp_active_hostname option:
4081 (1) $smtp_active_hostname is now available as a variable.
4082 (2) The default for smtp_banner uses $smtp_active_hostname instead
4083 of $primary_hostname.
4085 60. The host_aton() function is supposed to be passed a string that is known
4086 to be a valid IP address. However, in the case of IPv6 addresses, it was
4087 not checking this. This is a hostage to fortune. Exim now panics and dies
4088 if the condition is not met. A case was found where this could be provoked
4089 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4090 components; fortuitously, this particular loophole had already been fixed
4091 by change 4.50/55 above.
4093 If there are any other similar loopholes, the new check in host_aton()
4094 itself should stop them being exploited. The report I received stated that
4095 data on the command line could provoke the exploit when Exim was running as
4096 exim, but did not say which command line option was involved. All I could
4097 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4098 running as the user.
4101 61. There was a buffer overflow vulnerability in the SPA authentication code
4102 (which came originally from the Samba project). I have added a test to the
4103 spa_base64_to_bits() function which I hope fixes it.
4106 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4107 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4109 63. The daemon start-up calls getloadavg() while still root for those OS that
4110 need the first call to be done as root, but it missed one case: when
4111 deliver_queue_load_max is set with deliver_drop_privilege. This is
4112 necessary for the benefit of the queue runner, because there is no re-exec
4113 when deliver_drop_privilege is set.
4115 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4116 This has been fixed.
4118 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4119 was in use, was not putting the data itself into the right store pool;
4120 consequently, it could be overwritten for a subsequent message in the same
4121 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4124 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4126 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4127 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4128 the function were treating the return as a boolean value, which happened to
4129 work because 0=false and not-0=true, but is not correct code.
4131 68. The host_aton() function was not handling scoped IPv6 addresses (those
4132 with, for example, "%eth0" on the end) correctly.
4134 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4135 negated items (that is, ~something) in unsigned ints. Some compilers
4136 apparently mutter when there is no cast.
4138 70. If an address verification called from an ACL failed, and did not produce a
4139 user-specific message (i.e. there was only a "system" message), nothing was
4140 put in $acl_verify_message. In this situation, it now puts the system
4143 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4144 session; change 4.31/43 added the unwanted input to the log line - except
4145 that it did not do this in the start of session case. It now does.
4147 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4148 This is wrong and can cause the other end to generate a synchronization
4149 error if it is another Exim or anything else that does the synchronization
4150 check. A QUIT command is no longer sent after a timeout.
4152 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4153 during host lookups.
4155 74. The defer_ok option of callout verification was not working if it was used
4156 when verifying addresses in header lines, that is, for this case:
4158 verify = header_sender/callout=defer_ok
4160 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4161 those file descriptors could be used for SMTP connections. If anything
4162 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4163 could be sent to the SMTP client, causing chaos. The daemon now opens
4164 stdin, stdout, and stderr to /dev/null when it puts itself into the
4167 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4168 log by default. The user can override this with suitable Perl magic.
4170 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4171 the log message when discard triggers, was not working for the DATA ACL or
4172 for the non-SMTP ACL.
4174 78. Error message wording change in sieve.c.
4176 79. If smtp_accept_max_per_host was set, the number of connections could be
4177 restricted to fewer than expected, because the daemon was trying to set up
4178 a new connection before checking whether the processes handling previous
4179 connections had finished. The check for completed processes is now done
4180 earlier. On busy systems, this bug wouldn't be noticed because something
4181 else would have woken the daemon, and it would have reaped the completed
4184 80. If a message was submitted locally by a user whose login name contained one
4185 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4186 It caused a spool format error. I have fixed the spool reading code. A
4187 related problem was that the "from" clause in the Received: line became
4188 illegal because of the space(s). It is now covered by ${quote_local_part.
4190 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4193 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4195 83. Patch from Sieve maintainer to fix unterminated string problem in
4196 "vacation" handling.
4198 84. Some minor changes to the Linux configuration files to help with other
4199 OS variants using glibc.
4201 85. One more patch for Sieve to update vacation handling to latest spec.
4204 ----------------------------------------------------
4205 See the note above about the 4.44 and 4.50 releases.
4206 ----------------------------------------------------
4212 1. Change 4.43/35 introduced a bug that caused file counts to be
4213 incorrectly computed when quota_filecount was set in an appendfile
4216 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4217 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4220 3. Give more explanation in the error message when the command for a transport
4221 filter fails to execute.
4223 4. There are several places where Exim runs a non-Exim command in a
4224 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4225 was being done only for the command run by the queryprogram router. It is
4226 now done for all such subprocesses. The other cases are: ${run, transport
4227 filters, and the commands run by the lmtp and pipe transports.
4229 5. Some older OS have a limit of 256 on the maximum number of file
4230 descriptors. Exim was using setrlimit() to set 1000 as a large value
4231 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4232 systems. I've change it so that if it can't get 1000, it tries for 256.
4234 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4235 was an oversight, and furthermore, ever since the addition of extra
4236 controls (e.g. 4.43/32), the checks on when to allow different forms of
4237 "control" were broken. There should now be diagnostics for all cases when a
4238 control that does not make sense is encountered.
4240 7. $recipients is now available in the predata ACL (oversight).
4242 8. Tidy the search cache before the fork to do a delivery from a message
4243 received from the command line. Otherwise the child will trigger a lookup
4244 failure and thereby defer the delivery if it tries to use (for example) a
4245 cached ldap connection that the parent has called unbind on.
4247 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4248 of $address_data from the recipient verification was clobbered by the
4249 sender verification.
4251 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4252 was its contents. (It was OK if the option was not defined at all.)
4254 11. A "Completed" log line is now written for messages that are removed from
4255 the spool by the -Mrm option.
4257 12. $host_address is now set to the target address during the checking of
4258 ignore_target_hosts.
4260 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4261 being passed; this would have caused $sender_host_name to have been used if
4262 matching the list had actually called for a host name (not very likely,
4263 since this list is usually IP addresses). A host name is now passed as
4266 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4267 code that specifies a non-restarting handler (typically sigaction() in
4268 modern systems) in an attempt to fix a rare and obscure crash bug.
4270 15. Narrowed the window for a race in the daemon that could cause it to ignore
4271 SIGCHLD signals. This is not a major problem, because they are used only to
4272 wake it up if nothing else does.
4274 16. A malformed maildirsize file could cause Exim to calculate negative values
4275 for the mailbox size or file count. Odd effects could occur as a result.
4276 The maildirsize information is now recalculated if the size or filecount
4279 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4280 support for a long time. Removed HAVE_SYS_VFS_H.
4282 18. Updated exipick to current release from John Jetmore.
4284 19. Allow an empty sender to be matched against a lookup in an address list.
4285 Previously the only cases considered were a regular expression, or an
4288 20. Exim went into a mad DNS lookup loop when doing a callout where the
4289 host was specified on the transport, if the DNS lookup yielded more than
4292 21. The RFC2047 encoding function was originally intended for short strings
4293 such as real names; it was not keeping to the 75-character limit for
4294 encoded words that the RFC imposes. It now respects the limit, and
4295 generates multiple encoded words if necessary. To be on the safe side, I
4296 have increased the buffer size for the ${rfc2047: expansion operator from
4299 22. Failure to deliver a bounce message always caused it to be frozen, even if
4300 there was an errors_to setting on the router. The errors_to setting is now
4303 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4304 canonical form (fully expanded) before being placed in
4305 $sender_host_address.
4307 24. Updated eximstats to version 1.33
4309 25. Include certificate and key file names in error message when GnuTLS fails
4310 to set them up, because the GnuTLS error message doesn't include the name
4311 of the failing file when there is a problem reading it.
4313 26. Expand error message when OpenSSL has problems setting up cert/key files.
4316 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4317 (this can affect the format of dates).
4319 28. exim_tidydb, when checking for the continued existence of a message for
4320 which it has found a message-specific retry record, was not finding
4321 messages that were in split spool directories. Consequently, it was
4322 deleting retry records that should have stayed in existence.
4324 29. eximstats updated to version 1.35
4325 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4326 1.35 - bugfix such that pie charts by volume are generated correctly
4328 30. The SPA authentication driver was not abandoning authentication and moving
4329 on to the next authenticator when an expansion was forced to fail,
4330 contradicting the general specification for all authenticators. Instead it
4331 was generating a temporary error. It now behaves as specified.
4333 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4334 (the order specifies the preference for clients). The order is now AES256,
4335 AES128, 3DES, ARCFOUR128.
4337 31. Small patch to Sieve code - explicitly set From: when generating an
4340 32. Exim crashed if a remote delivery caused a very long error message to be
4341 recorded - for instance if somebody sent an entire SpamAssassin report back
4342 as a large number of 550 error lines. This bug was coincidentally fixed by
4343 increasing the size of one of Exim's internal buffers (big_buffer) that
4344 happened as part of the Exiscan merge. However, to be on the safe side, I
4345 have made the code more robust (and fixed the comments that describe what
4348 33. Some experimental protocols are using DNS PTR records for new purposes. The
4349 keys for these records are domain names, not reversed IP addresses. The
4350 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4351 leaves it alone. Component reversal etc. now happens only for IP addresses.
4354 34. The host_aton() function is supposed to be passed a string that is known
4355 to be a valid IP address. However, in the case of IPv6 addresses, it was
4356 not checking this. This is a hostage to fortune. Exim now panics and dies
4357 if the condition is not met. A case was found where this could be provoked
4358 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4359 components; fortuitously, this particular loophole had already been fixed
4360 by change 4.50/55 or 4.44/33 above.
4362 If there are any other similar loopholes, the new check in host_aton()
4363 itself should stop them being exploited. The report I received stated that
4364 data on the command line could provoke the exploit when Exim was running as
4365 exim, but did not say which command line option was involved. All I could
4366 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4367 running as the user.
4370 35. There was a buffer overflow vulnerability in the SPA authentication code
4371 (which came originally from the Samba project). I have added a test to the
4372 spa_base64_to_bits() function which I hope fixes it.
4375 36. The daemon start-up calls getloadavg() while still root for those OS that
4376 need the first call to be done as root, but it missed one case: when
4377 deliver_queue_load_max is set with deliver_drop_privilege. This is
4378 necessary for the benefit of the queue runner, because there is no re-exec
4379 when deliver_drop_privilege is set.
4381 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4382 was in use, was not putting the data itself into the right store pool;
4383 consequently, it could be overwritten for a subsequent message in the same
4384 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4387 38. Sometimes the final signoff response after QUIT could fail to get
4388 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4389 < 0 before doing a fflush(). This bug looks as though it goes back to the
4390 introduction of TLS in release 3.20, but "sometimes" must have been rare
4391 because the tests only now provoked it.
4397 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4398 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4399 Now it returns an integer. A number of calls were still expecting a T/F
4400 return. Fortuitously, in all cases, the tests worked in OK situations,
4401 which is the norm. However, things would have gone wrong on any write
4402 failures on the smtp file descriptor. This function is used when sending
4403 messages over SMTP and also when doing verify callouts.
4405 2. When Exim is called to do synchronous delivery of a locally submitted
4406 message (the -odf or -odi options), it no longer closes stderr before doing
4409 3. Implemented the mua_wrapper option.
4411 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4413 5. Implemented the functions header_remove(), header_testname(),
4414 header_add_at_position(), and receive_remove_recipient(), and exported them
4417 6. If an ACL "warn" statement specified the addition of headers, Exim already
4418 inserted X-ACL-Warn: at the start if there was no header name. However, it
4419 was not making this test for the second and subsequent header lines if
4420 there were newlines in the string. This meant that an invalid header could
4421 be inserted if Exim was badly configured.
4423 7. Allow an ACL "warn" statement to add header lines at the start or after all
4424 the Received: headers, as well as at the end.
4426 8. Added the rcpt_4xx retry error code.
4428 9. Added postmaster_mailfrom=xxx to callout verification option.
4430 10. Added mailfrom=xxxx to the callout verification option, for verify=
4433 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4434 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4436 12. Inserted some casts to stop certain compilers warning when using pointer
4437 differences as field lengths or precisions in printf-type calls (mostly
4438 affecting debugging statements).
4440 13. Added optional readline() support for -be (dynamically loaded).
4442 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4443 same clock tick as a message's arrival, so that its received time was the
4444 same as the "first fail" time on the retry record, and that message
4445 remained on the queue past the ultimate address timeout, every queue runner
4446 would try a delivery (because it was past the ultimate address timeout) but
4447 after another failure, the ultimate address timeout, which should have then
4448 bounced the address, did not kick in. This was a "< instead of <=" error;
4449 in most cases the first failure would have been in the next clock tick
4450 after the received time, and all would be well.
4452 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4453 being recognized when the domain list was tested by the match_domain
4454 condition in an expansion string.
4456 16. Added the ${str2b64: operator.
4458 17. Exim was always calling setrlimit() to set a large limit for the number of
4459 processes, without checking whether the existing limit was already
4460 adequate. (It did check for the limit on file descriptors.) Furthermore,
4461 errors from getrlimit() and setrlimit() were being ignored. Now they are
4462 logged to the main and panic logs, but Exim does carry on, to try to do its
4463 job under whatever limits there are.
4465 18. Imported PCRE 5.0.
4467 19. Trivial typo in log message " temporarily refused connection" (the leading
4470 20. If the log selector return_path_on_delivery was set and an address was
4471 redirected to /dev/null, the delivery process crashed because it assumed
4472 that a return path would always be set for a "successful" delivery. In this
4473 case, the whole delivery is bypassed as an optimization, and therefore no
4476 21. Internal re-arrangement: the function for sending a challenge and reading
4477 a response while authentication was assuming a zero-terminated challenge
4478 string. It's now changed to take a pointer and a length, to allow for
4479 binary data in such strings.
4481 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4483 23. Exim was not respecting finduser_retries when seeking the login of the
4484 uid under which it was called; it was always trying 10 times. (The default
4485 setting of finduser_retries is zero.) Also, it was sleeping after the final
4486 failure, which is pointless.
4488 24. Implemented tls_on_connect_ports.
4490 25. Implemented acl_smtp_predata.
4492 26. If the domain in control=submission is set empty, Exim assumes that the
4493 authenticated id is a complete email address when it generates From: or
4494 Sender: header lines.
4496 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4497 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4498 chown and chgrp in /bin and hostname in /usr/bin.
4500 28. Exim was keeping the "process log" file open after each use, just as it
4501 does for the main log. This opens the possibility of it remaining open for
4502 long periods when the USR1 signal hits a daemon. Occasional processlog
4503 errors were reported, that could have been caused by this. Anyway, it seems
4504 much more sensible not to leave this file open at all, so that is what now
4507 29. The long-running daemon process does not normally write to the log once it
4508 has entered its main loop, and it closes the log before doing so. This is
4509 so that log files can straightforwardly be renamed and moved. However,
4510 there are a couple of unusual error situations where the daemon does write
4511 log entries, and I had neglected to close the log afterwards.
4513 30. The text of an SMTP error response that was received during a remote
4514 delivery was being truncated at 512 bytes. This is too short for some of
4515 the long messages that one sometimes sees. I've increased the limit to
4518 31. It is now possible to make retry rules that apply only when a message has a
4519 specific sender, in particular, an empty sender.
4521 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4522 it possible to be selective about when SMTP synchronization is enforced.
4524 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4526 32. Implemented hosts_connection_nolog.
4528 33. Added an ACL for QUIT.
4530 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4533 35. Added mailbox_size and mailbox_filecount to appendfile.
4535 36. Added control = no_multiline_responses to ACLs.
4537 37. There was a bug in the logic of the code that waits for the clock to tick
4538 in the case where the clock went backwards by a substantial amount such
4539 that the microsecond fraction of "now" was more than the microsecond
4540 fraction of "then" (but the whole seconds number was less).
4542 38. Added support for the libradius Radius client library this is found on
4543 FreeBSD (previously only the radiusclient library was supported).
4549 1. When certain lookups returned multiple values in the form name=value, the
4550 quoting of the values was not always being done properly. Specifically:
4551 (a) If the value started with a double quote, but contained no whitespace,
4553 (b) If the value contained whitespace other than a space character (i.e.
4554 tabs or newlines or carriage returns) it was not quoted.
4555 This fix has been applied to the mysql and pgsql lookups by writing a
4556 separate quoting function and calling it from the lookup code. The fix
4557 should probably also be applied to nisplus, ibase and oracle lookups, but
4558 since I cannot test any of those, I have not disturbed their existing code.
4560 2. A hit in the callout cache for a specific address caused a log line with no
4561 reason for rejecting RCPT. Now it says "Previous (cached) callout
4562 verification failure".
4564 3. There was an off-by-one bug in the queryprogram router. An over-long
4565 return line was truncated at 256 instead of 255 characters, thereby
4566 overflowing its buffer with the terminating zero. As well as fixing this, I
4567 have increased the buffer size to 1024 (and made a note to document this).
4569 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4570 when Exim is waiting for an SMTP response from a remote server, Exim
4571 restarts its select() call on the socket, thereby resetting its timeout.
4572 This is not a problem when such interrupts are rare. Somebody set up a cron
4573 job to run exiwhat every 2 minutes, which is less than the normal select()
4574 timeout (5 or 10 minutes). This meant that the select() timeout never
4575 kicked in because it was always reset. I have fixed this by comparing the
4576 time when an interrupt arrives with the time at the start of the first call
4577 to select(). If more time than the timeout has elapsed, the interrupt is
4578 treated as a timeout.
4580 5. Some internal re-factoring in preparation for the addition of Sieve
4581 extensions (by MH). In particular, the "personal" test is moved to a
4582 separate function, and given an option for scanning Cc: and Bcc: (which is
4583 not set for Exim filters).
4585 6. When Exim created an email address using the login of the caller as the
4586 local part (e.g. when creating a From: or Sender: header line), it was not
4587 quoting the local part when it contained special characters such as @.
4589 7. Installed new OpenBSD configuration files.
4591 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4592 try to make them clearer.
4594 9. Callout options, other than the timeout value, were being ignored when
4595 verifying sender addresses in header lines. For example, when using
4597 verify = header_sender/callout=no_cache
4599 the cache was (incorrectly) being used.
4601 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4602 only the Cygwin environment.
4604 11. When return_path_on_delivery was set as a log selector, if different remote
4605 addresses in the same message used different return paths and parallel
4606 remote delivery occurred, the wrong values would sometimes be logged.
4607 (Whenever a remote delivery process finished, the return path value from
4608 the most recently started remote delivery process was logged.)
4610 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4611 lines when AUTH and/or TLS are in use. This is the "received protocol"
4612 field. Exim used to use "asmtp" for authenticated SMTP, without any
4613 indication (in the protocol name) for TLS use. Now it follows the RFC and
4614 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4615 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4616 names appear in log lines as well as in Received: header lines.
4618 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4619 extensions, and comparison tests, and to fix some bugs.
4621 14. Changes to the "personal" filter test:
4623 (1) The test was buggy in that it was just doing the equivalent of
4624 "contains" tests on header lines. For example, if a user's address was
4625 anne@some.where, the "personal" test would incorrectly be true for
4627 To: susanne@some.where
4629 This test is now done by extracting each address from the header in turn,
4630 and checking the entire address. Other tests that are part of "personal"
4631 are now done using regular expressions (for example, to check local parts
4632 of addresses in From: header lines).
4634 (2) The list of non-personal local parts in From: addresses has been
4635 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4636 taken from the Sieve specification recommendations.
4638 (3) If the message contains any header line starting with "List-" it is
4639 treated as non-personal.
4641 (4) The test for "circular" in the Subject: header line has been removed
4642 because it now seems ill-conceived.
4644 15. Minor typos in src/EDITME comments corrected.
4646 16. Installed latest exipick from John Jetmore.
4648 17. If headers_add on a router specified a text string that was too long for
4649 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4650 of string_sprintf() is now avoided.
4652 18. $message_body_size was not set (it was always zero) when running the DATA
4653 ACL and the local_scan() function.
4655 19. For the "mail" command in an Exim filter, no default was being set for
4656 the once_repeat time, causing a random time value to be used if "once" was
4657 specified. (If the value happened to be <= 0, no repeat happened.) The
4658 default is now 0s, meaning "never repeat". The "vacation" command was OK
4659 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4660 (I found it when inspecting the code).
4662 20. There is now an overall timeout for performing a callout verification. It
4663 defaults to 4 times the callout timeout, which applies to individual SMTP
4664 commands during the callout. The overall timeout applies when there is more
4665 than one host that can be tried. The timeout is checked before trying the
4666 next host. This prevents very long delays if there are a large number of
4667 hosts and all are timing out (e.g. when the network connections are timing
4668 out). The value of the overall timeout can be changed by specifying an
4669 additional sub-option for "callout", called "maxwait". For example:
4671 verify = sender/callout=5s,maxwait=20s
4673 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4674 to the end before writing, but this should make it even safer).
4676 22. Exim was forgetting that it had advertised PIPELINING for the second and
4677 subsequent messages on an SMTP connection. It was also not resetting its
4678 memory on STARTTLS and an internal HELO.
4680 23. When Exim logs an SMTP synchronization error within a session, it now
4681 records whether PIPELINING has been advertised or not.
4683 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4684 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4685 rather than long int.
4687 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4689 26. Added the never_mail option to autoreply.
4695 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4696 crash if the getsockname() call failed; this can happen if a connection is
4697 closed very soon after it is established. The problem was simply in the
4698 order in which certain operations were done, causing Exim to try to write
4699 to the SMTP stream before it had set up the file descriptor. The bug has
4700 been fixed by making things happen in the correct order.
4706 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4707 before the connection was closed, thus losing the rejection response.
4709 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4710 some early Solaris releases, but causes trouble in current releases where
4711 socklen_t is defined.
4713 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4716 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4719 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4720 that releasing the top of it at the end releases what was used for sub-
4721 expansions (unless the block got too big). However, discard this block if
4722 the first thing is a variable or header, so that we can use its block when
4723 it is dynamic (useful for very large $message_headers, for example).
4725 6. Lookups now cache *every* query, not just the most recent. A new, separate
4726 store pool is used for this. It can be recovered when all lookup caches are
4727 flushed. Lookups now release memory at the end of their result strings.
4728 This has involved some general refactoring of the lookup sources.
4730 7. Some code has been added to the store_xxx() functions to reduce the amount
4731 of flapping under certain conditions.
4733 8. log_incoming_interface used to affect only the <= reception log lines. Now
4734 it causes the local interface and port to be added to several more SMTP log
4735 lines, for example "SMTP connection from", and rejection lines.
4737 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4739 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4741 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4742 could be overwritten at the end of the current message (or the start of a
4743 new message if it was set in a HELO ACL). The value is now preserved for
4744 the duration of the SMTP connection.
4746 12. If a transport had a headers_rewrite setting, and a matching header line
4747 contained an unqualified address, that address was qualified, even if it
4748 did not match any rewriting rules. The underlying bug was that the values
4749 of the flags that permit the existence of unqualified sender and recipient
4750 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4751 non-local messages, and by -bnq for local messages) were not being
4752 preserved with the message after it was received.
4754 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4755 "next input=" as part of the text comprising the host identity instead of
4756 the correct text. The code was using the same buffer for two different
4757 strings. However, depending on which order the printing function evaluated
4758 its arguments, the bug did not always show up. Under Linux, for example, my
4759 test suite worked just fine.
4761 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4762 doesn't work with some older versions of Perl. It has been changed to "my",
4763 which in any case is probably the better facility to use.
4765 15. A really picky compiler found some instances of statements for creating
4766 error messages that either had too many or two few arguments for the format
4769 16. The size of the buffer for calls to the DNS resolver has been increased
4770 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4771 for addresses that have a lot of PTR records. This alleviates a problem; it
4772 does not fully solve it.
4774 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4775 buffer now truncates the list and logs the incident, which is the same
4776 action as happens when Exim is looking up a host name and its aliases.
4777 Previously in this situation something unpredictable would happen;
4778 sometimes it was "internal error: store_reset failed".
4780 18. If a server dropped the connection unexpectedly when an Exim client was
4781 using GnuTLS and trying to read a response, the client delivery process
4782 crashed while trying to generate an error log message.
4784 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4785 string, for example:
4787 warn message = H1: something\nH2: something
4789 the text was added as a single header line from Exim's point of view
4790 though it ended up OK in the delivered message. However, searching for the
4791 second and subsequent header lines using $h_h2: did not work. This has been
4792 fixed. Similarly, if a system filter added multiple headers in this way,
4793 the routers could not see them.
4795 20. Expanded the error message when iplsearch is called with an invalid key to
4796 suggest using net-iplsearch in a host list.
4798 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4799 ACLs are no longer actually imposed (and a message to that effect is
4802 22. If a "gecos" field in a passwd entry contained escaped characters, in
4803 particular, if it contained a \" sequence, Exim got it wrong when building
4804 a From: or a Sender: header from that name. A second bug also caused
4805 incorrect handling when an unquoted " was present following a character
4806 that needed quoting.
4808 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4809 was not being matched caselessly.
4811 24. Arranged for all hyphens in the exim.8 source to be escaped with
4814 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4815 properly. Recipient callout cache records were still being keyed to include
4816 the sender, even when use_sender was set false. This led to far more
4817 callouts that were necessary. The sender is no longer included in the key
4818 when use_sender is false.
4820 26. Added "control = submission" modifier to ACLs.
4822 27. Added the ${base62d: operator to decode base 62 numbers.
4824 28. dnsdb lookups can now access SRV records.
4826 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4827 the configuration file.
4829 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4830 (-v) mode. This makes the output for a verbose queue run more intelligible.
4832 31. Added a use_postmaster feature to recipient callouts.
4834 32. Added the $body_zerocount variable, containing the number of binary zero
4835 bytes in the message body.
4837 33. The time of last modification of the "new" subdirectory is now used as the
4838 "mailbox time last read" when there is a quota error for a maildir
4841 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4843 35. Added +ignore_unknown as a special item in host lists.
4845 36. Code for decoding IPv6 addresses in host lists is now included, even if
4846 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4847 address was recognized as an IP address, but was then not correctly decoded
4848 into binary, causing unexpected and incorrect effects when compared with
4855 1. Very minor rewording of debugging text in manualroute to say "list of
4856 hosts" instead of "hostlist".
4858 2. If verify=header_syntax was set, and a header line with an unqualified
4859 address (no domain) and a large number of spaces between the end of the
4860 name and the colon was received, the reception process suffered a buffer
4861 overflow, and (when I tested it) crashed. This was caused by some obsolete
4862 code that should have been removed. The fix is to remove it!
4864 3. When running in the test harness, delay a bit after writing a bounce
4865 message to get a bit more predictability in the log output.
4867 4. Added a call to search_tidyup() just before forking a reception process. In
4868 theory, someone could use a lookup in the expansion of smtp_accept_max_
4869 per_host which, without the tidyup, could leave open a database connection.
4871 5. Added the variables $recipient_data and $sender_data which get set from a
4872 lookup success in an ACL "recipients" or "senders" condition, or a router
4873 "senders" option, similar to $domain_data and $local_part_data.
4875 6. Moved the writing of debug_print from before to after the "senders" test
4878 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4879 problems for message scanning, either using a data ACL, or using
4880 local_scan() because the Received: header was not generated till after they
4881 were called (in order to set the time as the time of reception completion).
4882 I have revised the way this works. The header is now generated after the
4883 body is received, but before the ACL or local_scan() are called. After they
4884 are run, the timestamp in the header is updated.
4890 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4891 before starting a queue runner without re-exec. This happened only when
4892 deliver_drop_privilege was set or when the Exim user was set to root. The
4893 effect of the bug was that timeouts during subsequent deliveries caused
4894 crashes instead of being properly handled. The handler is now left at its
4895 default (and expected) setting.
4897 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4898 message, again when deliver_drop_privilege is set or Exim is run as root.
4899 The bug described in (1) was not present in this case, but the tidying up
4900 of the other signals was missing. I have made the two cases consistent.
4902 3. The ignore_target_hosts setting on a manualroute router was being ignored
4903 for hosts that were looked up using the /MX notation.
4905 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4908 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4909 operated on the sender address. After changing the $sender_address to <>
4910 for the sender address verify, Exim was re-instated it as the original
4911 (before rewriting) address, but remembering that it had rewritten it, so it
4912 wasn't rewriting it again. This bug also had the effect of breaking the
4913 sender address verification caching when the sender address was rewritten.
4915 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4916 This has been changed so that if the ip literal address matches
4917 ignore_target_hosts, the router declines.
4919 7. Added expansion conditions match_domain, match_address, and match_local_
4920 part (NOT match_host).
4922 8. The placeholder for the Received: header didn't have a length field set.
4924 9. Added code to Exim itself and to exim_lock to test for a specific race
4925 condition that could lead to file corruption when using MBX delivery. The
4926 issue is with the lockfile that is created in /tmp. If this file is removed
4927 after a process has opened it but before that process has acquired a lock,
4928 there is the potential for a second process to recreate the file and also
4929 acquire a lock. This could lead to two Exim processes writing to the file
4930 at the same time. The added code performs the same test as UW imapd; it
4931 checks after acquiring the lock that its file descriptor still refers to
4932 the same named file.
4934 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4935 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4938 11. Added the smtp_active_hostname option. If used, this will typically be made
4939 to depend on the incoming interface address. Because $interface_address is
4940 not set up until the daemon has forked a reception process, error responses
4941 that can happen earlier (such as "too many connections") no longer contain
4944 12. If an expansion in a condition on a "warn" statement fails because a lookup
4945 defers, the "warn" statement is abandoned, and the next ACL statement is
4946 processed. Previously this caused the whole ACL to be aborted.
4948 13. Added the iplsearch lookup type.
4950 14. Added ident_timeout as a log selector.
4952 15. Added tls_certificate_verified as a log selector.
4954 16. Added a global option tls_require_ciphers (compare the smtp transport
4955 option of the same name). This controls incoming TLS connections.
4957 17. I finally figured out how to make tls_require_ciphers do a similar thing
4958 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4959 before starting the TLS session.
4961 18. Tabs are now shown as \t in -bP output.
4963 19. If the log selector return_path_on_delivery was set, Exim crashed when
4964 bouncing a message because it had too many Received: header lines.
4966 20. If two routers both had headers_remove settings, and the first one included
4967 a superfluous trailing colon, the final name in the first list and the
4968 first name in the second list were incorrectly joined into one item (with a
4969 colon in the middle).
4975 1. Added -C and -D options to the exinext utility, mainly to make it easier
4976 to include in the automated testing, but these could be helpful when
4977 multiple configurations are in use.
4979 2. The exinext utility was not formatting the output nicely when there was
4980 an alternate port involved in the retry record key, nor when there was a
4981 message id as well (for retries that were specific to a specific message
4982 and a specific host). It was also confused by IPv6 addresses, because of
4983 the additional colons they contain. I have fixed the IPv4 problem, and
4984 patched it up to do a reasonable job for IPv6.
4986 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4987 delivery, the log line now contains "pipelined" if PIPELINING was used.
4989 4. An SMTP transport process used to panic and die if the bind() call to set
4990 an explicit outgoing interface failed. This has been changed; it is now
4991 treated in the same way as a connect() failure.
4993 5. A reference to $sender_host_name in the part of a conditional expansion
4994 that was being skipped was still causing a DNS lookup. This no longer
4997 6. The def: expansion condition was not recognizing references to header lines
4998 that used bh_ and bheader_.
5000 7. Added the _cache feature to named lists.
5002 8. The code for checking quota_filecount in the appendfile transport was
5003 allowing one more file than it should have been.
5005 9. For compatibility with Sendmail, the command line option
5013 and sets the incoming protocol and host name (for trusted callers). The
5014 host name and its colon can be omitted when only the protocol is to be set.
5015 Note the Exim already has two private options, -pd and -ps, that refer to
5016 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5017 "s", but I don't think that's a major issue.
5019 10. A number of refactoring changes to the code, none of which should affect
5022 (a) The number of logging options was getting close to filling up the
5023 32-bit word that was used as a bit map. I have split them into two classes:
5024 those that are passed in the argument to log_write(), and those that are
5025 only ever tested independently outside of that function. These are now in
5026 separate 32-bit words, so there is plenty of room for expansion again.
5027 There is no change in the user interface or the logging behaviour.
5029 (b) When building, for example, log lines, the code previously used a
5030 macro that called string_cat() twice, in order to add two strings. This is
5031 not really sufficiently general. Furthermore, there was one instance where
5032 it was actually wrong because one of the argument was used twice, and in
5033 one call a function was used. (As it happened, calling the function twice
5034 did not affect the overall behaviour.) The macro has been replaced by a
5035 function that can join an arbitrary number of extra strings onto a growing
5038 (c) The code for expansion conditions now uses a table and a binary chop
5039 instead of a serial search (which was left over from when there were very
5040 few conditions). Also, it now recognizes conditions like "pam" even when
5041 the relevant support is not compiled in: a suitably worded error message is
5042 given if an attempt is made to use such a condition.
5044 11. Added ${time_interval:xxxxx}.
5046 12. A bug was causing one of the ddress fields not to be passed back correctly
5047 from remote delivery subprocesses. The field in question was not being
5048 subsequently used, so this caused to problems in practice.
5050 13. Added new log selectors queue_time and deliver_time.
5052 14. Might have fixed a bug in maildirsizefile handling that threw up
5053 "unexpected character" debug warnings, and recalculated the data
5054 unnecessarily. In any case, I expanded the warning message to give more
5057 15. Added the message "Restricted characters in address" to the statements in
5058 the default ACL that block characters like @ and % in local parts.
5060 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5061 Three changes have been made:
5063 (a) There was a serious bug; a negative response to MAIL caused the whole
5064 recipient domain to be cached as invalid, thereby blocking all messages
5065 to all local parts at the same domain, from all senders. This bug has
5066 been fixed. The domain is no longer cached after a negative response to
5067 MAIL if the sender used is not empty.
5069 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5072 (c) A new callout option, "use_sender" has been added for people who want
5073 the modified behaviour.
5079 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5082 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5083 indeed breaks things for older releases.
5085 3. Added additional logging to the case where there is a problem reading data
5086 from a filter that is running in a subprocess using a pipe, in order to
5087 try to track down a specific problem.
5089 4. Testing facility fudge: when running in the test harness and attempting
5090 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5091 getting "No route to host". Convert this to a timeout.
5093 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5096 6. Some OS don't have socklen_t but use size_t instead. This affects the
5097 fifth argument of getsockopt() amongst other things. This is now
5098 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5099 can be set for individual OS. I have set it for SunOS5, OSF1, and
5100 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5101 some earlier ones do not.
5103 7. Change 4.30/15 was not doing the test caselessly.
5105 8. The standard form for an IPv6 address literal was being rejected by address
5106 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5107 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5108 this, as well as the form without the "IPv6" on the front (but only when
5109 address literals are enabled, of course).
5111 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5113 10. Exim crashed if a message with an empty sender address specified by -f
5114 encountered a router with an errors_to setting. This could be provoked only
5115 by a command such as
5119 where an empty string was supplied; "<>" did not hit this bug.
5121 11. Installed PCRE release 4.5.
5123 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5124 remained set. It is now erased.
5126 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5127 times from message ids (which are base 36 rather than the normal 62).
5129 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5130 were being counted as actual protocol errors, and logged if the log
5131 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5132 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5133 and DATA following a set of rejected RCPTs do not count as protocol errors.
5134 In other words, Exim assumes they were pipelined, though this may not
5135 actually be the case. Of course, in all cases the client gets an
5136 appropriate error code.
5138 15. If a lookup fails in an ACL condition, a message about the failure may
5139 be available; it is used if testing the ACL cannot continue, because most
5140 such messages specify what the cause of the deferral is. However, some
5141 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5142 that caused an old message to be retained and used if a later statement
5143 caused a defer, replacing the real cause of the deferral.
5145 16. If an IP address had so many PTR records that the DNS lookup buffer
5146 was not large enough to hold them, Exim could crash while trying to process
5147 the truncated data. It now detects and logs this case.
5149 17. Further to 4.21/58, another change has been made: if (and only if) the
5150 first line of a message (the first header line) ends with CRLF, a bare LF
5151 in a subsequent header line has a space inserted after it, so as not to
5152 terminate the header.
5154 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5155 unnecessarily, used atoi() instead of strtol(), and didn't check the
5156 termination when getting file sizes from file names by regex.
5158 19. Completely re-implemented the support for maildirsize files, in the light
5159 of a number of problems with the previous contributed implementation
5160 (4.30/29). In particular:
5162 . If the quota is zero, the maildirsize file is maintained, but no quota is
5165 . If the maildir directory does not exist, it is created before any attempt
5166 to write a maildirsize file.
5168 . The quota value in the file is just a cache; if the quota is changed in
5169 the transport, the new value overrides.
5171 . A regular expression is available for excluding directories from the
5174 20. The autoreply transport checks the characters in options that define the
5175 message's headers; it allows continued headers, but it was checking with
5176 isspace() after an embedded newline instead of explicitly looking for a
5179 21. If all the "regular" hosts to which an address was routed had passed their
5180 expiry times, and had not reached their retry times, the address was
5181 bounced, even if fallback hosts were defined. Now Exim should go on to try
5184 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5185 equivalent code in the SMTP transport. Some hosts send humungous responses
5186 to HELO/EHLO, more than 1024 it seems.
5188 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5189 gives compiler warnings in some environments. I've now done it "properly",
5192 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5193 (because of problems with the built-in one) was declared to return uschar *
5194 instead of char *, causing compiler failure.
5196 25. Fixed a file descriptor leak when processing alias/forward files.
5198 26. Fixed a minor format string issue in dbfn.c.
5200 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5202 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5203 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5206 29. When a sender address is verified, it is cached, to save repeating the test
5207 when there is more than one recipient in a message. However, when the
5208 verification involves a callout, it is possible for different callout
5209 options to be set for different recipients. It is too complicated to keep
5210 track of this in the cache, so now Exim always runs a verification when a
5211 callout is required, relying on the callout cache for the optimization.
5212 The overhead is duplication of the address routing, but this should not be
5215 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5216 to be verified with callout=postmaster, and the main callout worked but the
5217 postmaster check failed, the verification correctly failed. However, if a
5218 subsequent RCPT command asked for sender verification *without* the
5219 postmaster check, incorrect caching caused this verification also to fail,
5222 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5223 it was not caching the DNS options (qualify_single, search_parents) that
5224 were used when the lookup failed. A subsequent lookup with different
5225 options therefore always gave the same answer, though there were cases
5226 where it should not have. (Example: a "domains = !$mx_any" option on a
5227 dnslookup router: the "domains" option is always processed without any
5228 widening, but the router might have qualify_single set.) Now Exim uses the
5229 cached value only when the same options are set.
5231 32. Added John Jetmore's "exipick" utility to the distribution.
5233 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5234 than a timeout (e.g. a certificate is required, and is not provided), an
5235 Exim server now closes the connection immediately. Previously it waited for
5236 the client to close - but if the client is SSL, it seems that they each
5237 wait for each other, leading to a delay before one of them times out.
5239 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5240 maintained 0.8.x compatibility because I don't think many are using it, and
5241 it is clearly obsolete.
5243 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5246 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5247 client certificate was expired. A simple patch fixes this, though I don't
5248 understand the full logic of why the verify callback is called multiple
5251 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5252 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5253 which causes problems with some clients (such as the Certicom SSL Plus
5254 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5255 disables the coutermeasure allowing Eudora to connect."
5257 38. Exim was not checking that a write() to a log file succeeded. This could
5258 lead to Bad Things if a log got too big, in particular if it hit a file
5259 size limit. Exim now panics and dies if it cannot write to a log file, just
5260 as it does if it cannot open a log file.
5262 39. Modified OS/Makefile-Linux so that it now contains
5264 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5266 The two -D definitions ensure that Exim is compiled with large file
5267 support, which makes it possible to handle log files that are bigger than
5270 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5271 instance) a domain was checked against a named list that involved a lookup,
5272 causing $domain_data to be set, then another domain was checked against the
5273 same list, then the first domain was re-checked, the value of $domain_data
5274 after the final check could be wrong. In particular, if the second check
5275 failed, it could be set empty. This bug probably also applied to
5278 41. The strip_trailing_dot option was not being applied to the address given
5279 with the -f command-line option.
5281 42. The code for reading a message's header from the spool was incrementing
5282 $received_count, but never initializing it. This meant that the value was
5283 incorrect (doubled) while delivering a message in the same process in which
5284 it was received. In the most common configuration of Exim, this never
5285 happens - a fresh exec is done - but it can happen when
5286 deliver_drop_privilege is set.
5288 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5289 it now includes up to 150 characters of the unexpected data in the log
5292 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5293 and building data strings. The size of both of these buffers was 10 000
5294 bytes - far larger than anybody would *ever* want, thought I. Needless to
5295 say, somebody hit the limit. I have increased the maximum line length to
5296 20 000 and the maximum data length of concatenated lines to 100 000. I have
5297 also fixed two bugs, because there was no checking on these buffers. Tsk,
5298 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5299 buffer is too small.
5301 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5302 lsearch lookups. Now it does.
5304 46. When parsing a route_list item in a manualroute router, a fixed-length
5305 buffer was used for the list of hosts. I made this 1024 bytes long,
5306 thinking that nobody would ever have a list of hosts that long. Wrong.
5307 Somebody had a whole pile of complicated expansion conditions, and the
5308 string was silently truncated, leading to an expansion error. It turns out
5309 that it is easier to change to an unlimited length (owing to other changes
5310 that have happened since this code was originally written) than to build
5311 structure for giving a limitation error. The length of the item that
5312 expands into the list of hosts is now unlimited.
5314 47. The lsearch lookup could not handle data where the length of text line was
5315 more than 4095 characters. Such lines were truncated, leading to shortened
5316 data being returned. It should now handle lines of any length.
5318 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5319 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5322 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5323 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5324 no longer happens in comment lines. A list of replacements is now placed
5325 at the head of all of the source files, except those whose only change is
5326 to replace PERL_COMMAND in the very first #! line.
5328 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5329 messages on the queue, with a bottom-up merge sort, using code contributed
5330 by Michael Haardt. This should make operations like -bp somewhat faster on
5331 large queues. It won't affect queue runners, except when queue_run_in_order
5334 51. Installed eximstats 1.31 in the distribution.
5336 52. Added support for SRV lookups to the dnslookup router.
5338 53. If an ACL referred to $message_body or $message_body_end, the value was not
5339 reset for any messages that followed in the same SMTP session.
5341 54. The store-handling optimization for building very long strings was not
5342 differentiating between the different store pools. I don't think this
5343 actually made any difference in practice, but I've tidied it.
5345 55. While running the routers to verify a sender address, $sender_address
5346 was still set to the sender address. This is wrong, because when routing to
5347 send a bounce to the sender, it would be empty. Therefore, I have changed
5348 it so that, while verifying a sender address, $sender_address is set to <>.
5349 (There is no change to what happens when verifying a recipient address.)
5351 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5352 target A or AAAA records (if not already returned) without resetting the
5353 qualify_single or search_parents options of the DNS resolver. These are
5354 inappropriate in this case because the targets of MX and SRV records must
5355 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5356 target that, when qualified, matched something in the local domain. These
5357 two options are now turned off when doing these lookups.
5359 57. It seems that at least some releases of Reiserfs (which does not have the
5360 concept of a fixed number of inodes) returns zero and not -1 for the
5361 number of available inodes. This interacted badly with check_spool_inodes,
5362 which assumed that -1 was the "no such thing" setting. What I have done is
5363 to check that the total number of inodes is greater than zero before doing
5364 the test of how many are available.
5366 58. When a "warn" ACL statement has a log_message modifier, the message is
5367 remembered, and not repeated. This is to avoid a lot of repetition when a
5368 message has many recipients that cause the same warning to be written.
5369 Howewer, Exim was preserving the list of already written lines for an
5370 entire SMTP session, which doesn't seem right. The memory is now reset if a
5371 new message is started.
5373 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5374 debugging output unless log_rewrite was also set.
5376 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5377 of (int)(handle) when we know that handle contains (void *)(-1).
5379 61. The Exim daemon panic-logs an error return when it closes the incoming
5380 connection. However "connection reset by peer" seems to be common, and
5381 isn't really an error worthy of noting specially, so that particular error
5384 62. When Exim is trying to find all the local interfaces, it used to panic and
5385 die if the ioctl to get the interface flags failed. However, it seems that
5386 on at least one OS (Solaris 9) it is possible to have an interface that is
5387 included in the list of interfaces, but for which you get a failure error
5388 for this call. This happens when the interface is not "plumbed" into a
5389 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5390 failure of the "get flags" call assumes that the interface is down.
5392 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5393 makes life easier for people who are doing arithmetic on fields extracted
5394 from dates, where you often get leading zeros that should not be
5395 interpreted as octal.
5397 64. Added qualify_domain to the redirect router, to override the global
5400 65. If a pathologically long header line contained very many addresses (the
5401 report of this problem mentioned 10 000) and each of them was rewritten,
5402 Exim could use up a very large amount of memory. (It kept on making new
5403 copies of the header line as it rewrote, and never released the old ones.)
5404 At the expense of a bit more processing, the header rewriting function has
5405 been changed so that it no longer eats memory in this way.
5407 66. The generation of the Received: header has been moved from the time that a
5408 message starts to be received, to the time that it finishes. The timestamp
5409 in the Received: header should now be very close to that of the <= log
5410 line. There are two side-effects of this change:
5412 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5413 logged header lines no longer include the local Received: line, because
5414 it has not yet been created. The same applies to a copy of the message
5415 that is returned to a non-SMTP sender when a message is rejected.
5417 (b) When a filter file is tested using -bf, no additional Received: header
5418 is added to the test message. After some thought, I decided that this
5421 This change does not affect the value of $received_for. It is still set
5422 after address rewriting, but before local_scan() is called.
5424 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5426 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5427 gave an unhelpful panic error message, and a defer error. I have managed to
5428 change this behaviour so that it now rejects any supplied certificate,
5429 which seems right, as the list of acceptable certificates is empty.
5431 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5432 gave an unhelpful defer error. I have not managed to make this reject any
5433 supplied certificates, but the error message it gives is "no certificate
5434 supplied", which is not helpful.
5436 70. exigrep's output now also includes lines that are not associated with any
5437 message, but which match the given pattern. Implemented by a patch from
5438 Martin Sluka, which also tidied up the Perl a bit.
5440 71. Recipient callout verification, like sender verification, was using <> in
5441 the MAIL FROM command. This isn't really the right thing, since the actual
5442 sender may affect whether the remote host accepts the recipient or not. I
5443 have changed it to use the actual sender in the callout; this means that
5444 the cache record is now keyed on a recipient/sender pair, not just the
5445 recipient address. There doesn't seem to be a real danger of callout loops,
5446 since a callout by the remote host to check the sender would use <>.
5447 [SEE ABOVE: changed after hitting problems.]
5449 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5450 temporary errors. However, in the case of such a code being given after
5451 the end of a data transmission (i.e. after ".") Exim was failing to write
5452 a retry record for the message. (Yes, there was some broken host that was
5453 actually sending 8xx at this point.)
5455 73. An unknown lookup type in a host list could cause Exim to panic-die when
5456 the list was checked. (An example that provoked this was putting <; in the
5457 middle of a list instead of at the start.) If this happened during a DATA
5458 ACL check, a -D file could be left lying around. This kind of configuration
5459 error no longer causes Exim to die; instead it causes a defer errror. The
5460 incident is still logged to the main and panic logs.
5462 74. Buglet left over from Exim 3 conversion. The message "too many messages
5463 in one connection" was written to the rejectlog but not the mainlog, except
5464 when address rewriting (yes!) was being logged.
5466 75. Added write_rejectlog option.
5468 76. When a system filter was run not as root (that is, when system_filter_user
5469 was set), the values of the $n variables were not being returned to the
5470 main process; thus, they were not subsequently available in the $sn
5473 77. Added +return_path_on_delivery log selector.
5475 78. A connection timeout was being treated differently from recipients deferred
5476 when testing hosts_max_try with a message that was older than the host's
5477 retry timeout. (The host should not be counted, thus allowing all hosts to
5478 be tried at least once before bouncing.) This may have been the cause of an
5479 occasionally reported bug whereby a message would remain on the queue
5480 longer than the retry timeout, but would be bounced if a delivery was
5481 forced. I say "may" because I never totally pinned down the problem;
5482 setting up timeout/retry tests is difficult. See also the next item.
5484 79. The ultimate address timeout was not being applied to errors that involved
5485 a combination of host plus message (for example, a timeout on a MAIL
5486 command). When an address resolved to a number of possible hosts, and they
5487 were not all tried for each delivery (e.g. because of hosts_max_try), a
5488 message could remain on the queue longer than the retry timeout.
5490 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5493 81. Fixed an obscure SMTP outgoing bug which required at least the following
5494 conditions: (a) there was another message waiting for the same server;
5495 (b) the server returned 5xx to all RCPT commands in the first message so
5496 that the message was not completed; (c) the server dropped the connection
5497 or gave a negative response to the RSET that Exim sends to abort the
5498 transaction. The observed case was a dropped connection after DATA that had
5499 been sent in pipelining mode. That is, the server had advertised PIPELINING
5500 but was not implementing it correctly. The effect of the bug was incorrect
5501 behaviour, such as trying another host, and this could lead to a crash.
5507 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5508 and daemon.c were passed as pointers to ints; they should have been
5509 pointers to socklen_t variables (which are typically unsigned ints).
5511 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5514 3. Fixed a really odd bug that affected only the testing scheme; patching a
5515 certain fixed string in the binary changed the value of another string that
5516 happened to be identical to the end of the original first string.
5518 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5519 name", it returns that address as the IP address. On some operating
5520 systems (e.g. Solaris), it also passes back the IP address string as the
5521 "host name". However, on others (e.g. Linux), it passes back an empty
5522 string. Exim wasn't checking for this, and was changing the host name to an
5523 empty string, assuming it had been canonicized.
5525 5. Although rare, it is permitted to have more than one PTR record for a given
5526 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5527 all the names associated with an address, because they do in Solaris.
5528 However, it seems that they do not in Linux for data that comes from the
5529 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5530 I found this out when I moved to a new Linux workstation and tried to run
5531 the Exim test suite.
5533 To get round this problem I have changed the code so that it now does its
5534 own call to the DNS to look up PTR records when searching for a host name.
5535 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5536 addresses that are only in /etc/hosts are still found.
5538 This behaviour is, however, controlled by an option called host_lookup_
5539 order, which defaults to "bydns:byaddr". If people want to use the other
5540 order, or indeed, just use one or the other means of lookup, they can
5541 specify it in this variable.
5543 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5544 some operating systems, this comes back from gethostbyaddr() as an empty
5545 string, and this is what Exim used to test for. However, it seems that in
5546 other systems, "." is yielded. Exim now tests for this case too.
5548 7. The values of check_spool_space and check_log_space are now held internally
5549 as a number of kilobytes instead of an absolute number of bytes. If a
5550 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5551 kilobyte. This means that much larger values can be stored.
5553 8. Exim monitor: an attempt to get the action menu when not actually pointing
5554 at a message produces an empty menu entitled "No message selected". This
5555 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5556 no entries in it ("Shell widget menu has zero width and/or height"). So I
5557 have added a single, blank menu entry in this case.
5559 9. Added ${quote_local_part.
5561 10. MIME decoding is now applied to the contents of Subject: header lines when
5564 11. Now that a reference to $sender_host_address automatically causes a reverse
5565 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5566 host lookup before query-style lookups in lists that might use this
5567 variable. This has therefore been abolished, and the "net-" prefix is no
5568 longer necessary for query-style lookups.
5570 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5571 have been a typo for LFLAGS, so it has been changed.
5573 13. The install script calls Exim with "-C /dev/null" in order to find the
5574 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5575 to be output. Howeve, since Exim outputs its version number before the
5576 error, it didn't break the script. It just looked ugly. I fixed this by
5577 always allowing "-C /dev/null" if the caller is root.
5579 14. Ignore overlarge ACL variable number when reading spool file - insurance
5580 against a later release with more variables having written the file.
5582 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5583 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5584 this, as well as the form without the "IPv6" on the front.
5586 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5587 OS/Makefile-Darwin file.
5589 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5590 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5592 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5593 variable. However, some OS use ints for inodes. I've added cast to long int
5594 to get rid of the compiler warning.
5596 19. I had forgotten to lock out "/../" in configuration file names when
5597 ALT_CONFIG_PREFIX was set.
5599 20. Routers used for verification do not need to specify transports. However,
5600 if such a router generated a host list, and callout was configured, Exim
5601 crashed, because it could not find a port number from the (non-existent)
5602 transport. It now assumes port 25 in this circumstance.
5604 21. Added the -t option to exigrep.
5606 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5607 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5608 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5609 them is set, LOOKUP_LSEARCH is forced.
5611 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5613 24. Added sender and host information to the "rejected by local_scan()" log
5614 line; previously there was no indication of these.
5616 25. Added .include_if_exists.
5618 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5619 Linux. It turns out that not all file systems support this. Apparently some
5620 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5621 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5622 not supported on the file descriptor, is now ignored when Exim is trying to
5623 sync a directory. This applies only to Linux.
5625 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5627 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5628 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5631 29. Added support for maildirsize files from supplied patch (modified a bit).
5633 30. The use of :fail: followed by an empty string could lead Exim to respond to
5634 sender verification failures with (e.g.):
5636 550 Verification failed for <xxx>
5637 550 Sender verify failed
5639 where the first response line was missing the '-' that indicates it is not
5640 the final line of the response.
5642 31. The loop for finding the name of the user that called Exim had a hardwired
5643 limit of 10; it now uses the value of finduser_retries, which is used for
5644 all other user lookups.
5646 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5649 33. Exim was neglecting to zero errno before one call of strtol() when
5650 expanding a string and expecting an integer value. On some systems this
5651 resulted in spurious "integer overflow" errors. Also, it was casting the
5652 result into an int without checking.
5654 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5655 did not work. The code looks as if it has *never* worked, though it appears
5656 to have been documented since at least releast 1.62. I have made it work.
5658 35. The "timeout_DNS" error in retry rules, also documented since at least
5659 1.62, also never worked. As it isn't clear exactly what this means, and
5660 clearly it isn't a major issue, I have abolished the feature by treating it
5661 as "timeout", and writing a warning to the main and panic logs.
5663 36. The display of retry rules for -brt wasn't always showing the error code
5666 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5667 timeout_connect_A, timeout_connect_MX.
5669 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5670 to the empty sender.
5672 39. The daemon was not analysing the content of -oX till after it had closed
5673 stderr and disconnected from the controlling terminal. This meant that any
5674 syntax errors were only noted on the panic log, and the return code from
5675 the command was 0. By re-arranging the code a little, I've made the
5676 decoding happen first, so such errors now appear on stderr, and the return
5677 code is 1. However, the actual setting up of the sockets still happens in
5678 the disconnected process, so errors there are still only recorded on the
5681 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5682 connections (as happens on some IP stacks) was logged at start up time as
5683 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5684 it from "IPv6 and IPv4", which means that two separate sockets are being
5687 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5688 says whether AF_INET or AF_INET6 was passed as an argument.
5690 42. Exiwhat output was messed up when time zones were included in log
5693 43. Exiwhat now gives more information about the daemon's listening ports,
5694 and whether -tls-on-connect was used.
5696 44. The "port" option of the smtp transport is now expanded.
5698 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5699 silently ignored. Now an error message is written to the main and panic
5702 46. There's a new ACL modifier called "logwrite" which writes to a log file
5703 as soon as it is encountered.
5705 47. Added $local_user_uid and $local_user_gid at routing time.
5707 48. Exim crashed when trying to verify a sender address that was being
5710 49. Exim was recognizing only a space character after ".include". It now also
5711 recognizes a tab character.
5713 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5714 extracting the relevant information from the specification. The man page no
5715 longer contains scrambled data for the -d option, and I've added a section
5716 at the front about calling Exim under different names.
5718 51. Added "extra_headers" argument to the "mail" command in filter files.
5720 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5723 53. Installed eximstats 1.29.
5725 54. Added transport_filter_timeout as a generic transport option.
5727 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5728 Cc: header lines. This was required by RFC 822, but it not required by RFC
5731 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5732 incoming messages that did not have them. Now it does so only if the
5733 message originates locally, that is, if there is no associated remote host
5734 address. When Resent- header lines are present, this applies to the Resent-
5735 lines rather than the non-Resent- lines.
5737 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5738 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5740 58. Messages for configuration errors now include the name of the main
5741 configuration file - useful now that there may be more than one file in a
5742 list (.included file names were always shown).
5744 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5745 for those rare installations that do not start the daemon as root or run it
5746 setuid root. I've cut out the call to initgroups() if the daemon is not
5749 60. The Exim user and group can now be bound into the binary as text strings
5750 that are looked up at the start of Exim's processing.
5752 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5754 62. Added $mailstore_basename variable.
5756 63. Installed patch to sieve.c from Michael Haardt.
5758 64. When Exim failed to open the panic log after failing to open the main log,
5759 the original message it was trying to log was written to stderr and debug
5760 output, but if they were not available (the usual case in production), it
5761 was lost. Now it is written to syslog before the two lines that record the
5762 failures to open the logs.
5764 65. Users' Exim filters run in subprocesses under the user's uid. It is
5765 possible for a "deliver" command or an alias in a "personal" command to
5766 provoke an address rewrite. If logging of address rewriting is configured,
5767 this fails because the process is not running as root or exim. There may be
5768 a better way of dealing with this, but for the moment (because 4.30 needs
5769 to be released), I have disabled address rewrite logging when running a
5770 filter in a non-root, non-exim process.
5776 1. The buildconfig auxiliary program wasn't quoting the value set for
5777 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5778 not defined. This bug was masked in 4.22 by the effect that was fixed in
5781 2. Some messages that were rejected after a message id was allocated were
5782 shown as "incomplete" by exigrep. It no longer does this for messages that
5783 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5785 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5786 have allow_domain_literals set, the ID did not get logged in the <= line.
5787 Domain literals are now always recognized in Message-ID: header lines.
5789 4. The first argument for a ${extract expansion item is the key name or field
5790 number. Leading and trailing spaces in this item were not being ignored,
5791 causing some misleading effects.
5793 5. When deliver_drop_privilege was set, single queue runner processes started
5794 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5795 same command in the process it spins off) were not dropping privilege.
5797 6. When the daemon running as "exim" started a queue runner, it always
5798 re-executed Exim in the spun-off process. This is a waste of effort when
5799 deliver_drop_privilege is set. The new process now just calls the
5800 queue-runner function directly.
5806 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5809 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5810 ignored. Though the use of -oP was forcing the writing of a pid file, it
5811 was always written to the default place.
5813 3. If the message "no IP address found for host xxxx" is generated during
5814 incoming verification, it is now followed by identification of the incoming
5815 connection (so you can more easily find what provoked it).
5817 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5819 5. Added some features to "harden" Exim a bit more against certain attacks:
5821 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5822 be put in Local/Makefile. This is like the never_users runtime option,
5823 but it cannot be overridden. The default setting is "root".
5825 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5826 prefix string with which any file named in a -C command line option
5829 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5830 is retained for -C and -D only if the caller of Exim is root. Without
5831 it, the exim user may also use -C and -D and retain privilege.
5833 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5834 command line option is disabled.
5836 6. Macro names set by the -D option must start with an upper case letter, just
5837 like macro names defined in the configuration file.
5839 7. Added "dereference=" facility to LDAP.
5841 8. Two instances of the typo "uknown" in the source files are fixed.
5843 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5844 the Configure-Makefile script screwed up while processing it.
5846 10. Incorporated PCRE 4.4.
5848 11. The SMTP synchronization check was not operating right at the start of an
5849 SMTP session. For example, it could not catch a HELO sent before the client
5850 waited for the greeting. There is now a check for outstanding input at the
5851 point when the greeting is written. Because of the duplex, asynchronous
5852 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5853 way, but not yet received, when the check is performed.
5855 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5856 on TCP/IP sockets, because this apparently causes some broken clients to
5859 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5860 unchanged) from the Cygwin maintainer.
5862 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5863 was supported instead of testing for mbx. Effectively a typo.
5865 15. The spa authenticator server code was not checking that the input it
5866 received was valid base64.
5868 16. The debug output line for the "set" modifier in ACLs was not showing the
5869 name of the variable that was being set.
5871 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5873 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5874 lookup on its own. Something else was needed to trigger the lookup. For
5875 example, a match in host_lookup or the need for a host name in a host list.
5876 Now, if $sender_host_name is referenced and the host name has not yet been
5877 looked up, a lookup is performed. If the lookup fails, the variable remains
5878 empty, and $host_lookup_failed is set to "1".
5880 19. Added "eqi" as a case-independent comparison operator.
5882 20. The saslauthd authentication condition could segfault if neither service
5883 nor realm was specified.
5885 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5886 error message that was logged was misleading, and incoming SMTP
5887 connections were dropped. The message is now more accurate, and temporary
5888 errors are given to SMTP connections.
5890 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5891 (except RSET) with a 421 error, until QUIT is received. However, it was
5892 failing to send a response to QUIT.
5894 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5895 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5896 ACL is now run after the helo_try_verify_hosts code.
5898 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5899 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5900 case-independent, so other case variants are also recognized). Apparently
5901 some systems use these upper case variants.
5903 25. If more than two messages were waiting for the same host, and a transport
5904 filter was specified for the transport, Exim sent two messages over the
5905 same TCP/IP connection, and then failed with "socket operation on non-
5906 socket" when it tried to send the third.
5908 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5910 27. The extern definition of crypt16() in expand.c was not being excluded when
5911 the OS had its own crypt16() function.
5913 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5914 as a preferred synonym for return_size_limit, both as an option and as an
5917 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5919 30. Changed the default configuration ACL to relax the local part checking rule
5920 for addresses that are not in any local domains. For these addresses,
5921 slashes and pipe symbols are allowed within local parts, but the sequence
5922 /../ is explicitly forbidden.
5924 31. SPA server authentication was not clearing the challenge buffer before
5927 32. log_message in a "warn" ACL statement was writing to the reject log as
5928 well as to the main log, which contradicts the documentation and doesn't
5929 seem right (because no rejection is happening). So I have stopped it.
5931 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5932 However, I am unable to do any testing of this.
5934 34. Fixed an infelicity in the appendfile transport. When checking directories
5935 for a mailbox, to see if any needed to be created, it was accidentally
5936 using path names with one or more superfluous leading slashes; tracing
5937 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5939 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5940 contents are added to the log line that is written for every discarded
5941 recipient. (Previously a log_message setting was ignored.)
5943 36. The ${quote: operator now quotes the string if it is empty.
5945 37. The install script runs exim in order to find its version number. If for
5946 some reason other than non-existence or emptiness, which it checks, it
5947 could not run './exim', it was installing it with an empty version number,
5948 i.e. as "exim-". This error state is now caught, and the installation is
5951 38. An argument was missing from the function that creates an error message
5952 when Exim fails to connect to the socket for saslauthd authentication.
5953 This could cause Exim to crash, or give a corrupted message.
5955 39. Added isip, isip4, and isip6 to ${if conditions.
5957 40. The ACL variables $acl_xx are now saved with the message, and can be
5958 accessed later in routers, transports, and filters.
5960 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5961 strings in the file are not string-expanded.
5963 42. If a MAIL command specified a SIZE value that was too large to fit into an
5964 int variable, the check against message_size_limit failed. Such values are
5965 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5966 day this will have to be increased, but I don't think I want to be around
5967 when emails are that large.
5974 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5975 iconv() is not standard in FreeBSD.
5977 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5978 IPv6 enabled. The observed symptom was a segmentation fault on return from
5979 the function os_common_find_running_interfaces() in src/os.c.
5981 3. In the check_special_case() function in daemon.c I had used "errno" as an
5982 argument name, which causes warnings on some systems. This was basically a
5983 typo, since it was named "eno" in the comments!
5985 4. The code that waits for the clock to tick (at a resolution of some fraction
5986 of a second) so as to ensure message-id uniqueness was always waiting for
5987 at least one whole tick, when it could have waited for less. [This is
5988 almost certainly not relevant at current processor speeds, where it is
5989 unlikely to ever wait at all. But we try to future-proof.]
5991 5. The function that sleeps for a time interval that includes fractions of a
5992 second contained a race. It did not block SIGALRM between setting the
5993 timer, and suspending (a couple of lines later). If the interval was short
5994 and the sigsuspend() was delayed until after it had expired, the suspension
5995 never ended. On busy systems this could lead to processes getting stuck for
5998 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5999 process, before it forks any delivery processes. The open lookup caching
6000 mechanism meant that the open file or database connection was passed into
6001 the delivery process. The problem was that delivery processes always tidy
6002 up cached lookup data. This could cause a problem for the next delivery
6003 process started by the queue runner, because the external queue runner
6004 process does not know about the closure. So the next delivery process
6005 still has data in the lookup cache. In the case of a file lookup, there was
6006 no problem because closing a file descriptor in a subprocess doesn't affect
6007 the parent. However, if the lookup was caching a connection to a database,
6008 the connection was closed, and the second delivery process was likely to
6009 see errors such as "PGSQL: query failed: server closed the connection
6010 unexpectedly". The problem has been fixed by closing all cached lookups
6011 in a queue runner before running a delivery process.
6013 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6014 seem to have the "const" qualifier which it has on other OS. I've
6017 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6018 *of the same type* (client or server) with the same public name that an
6019 error should be diagnosed.
6021 9. When Exim looked up a host name for an IP address, but failed to find the
6022 original IP address when looking up the host name (a safety check), it
6023 output the message "<ip address> does not match any IP for NULL", which was
6024 confusing, to say the least. The bug was that the host name should have
6025 appeared instead of "NULL".
6027 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6028 user that is built into the binary, and the -C or -D options is used, root
6029 privilege is dropped before the configuration file is read. In addition,
6030 logging is switched to stderr instead of the normal log files. If the
6031 configuration then re-defines the Exim user, the unprivileged environment
6032 is probably not what is expected, so Exim logs a panic warning message (but
6035 However, if deliver_drop_privilege is set, the unprivileged state may well
6036 be exactly what is intended, so the warning has been cut out in that case,
6037 and Exim is allowed to try to write to its normal log files.
6043 1. smtp_return_error_details was not giving details for temporary sender
6044 or receiver verification errors.
6046 2. Diagnose a configuration error if two authenticators have the same public
6049 3. Exim used not to create the message log file for a message until the first
6050 delivery attempt. This could be confusing when incoming messages were held
6051 for policy or load reasons. The message log file is now created at the time
6052 the message is received, and an initial "Received" line is written to it.
6054 4. The automatically generated man page for command line options had a minor
6055 bug that caused no ill effects; however, a more serious problem was that
6056 the procedure for building the man page automatically didn't always
6057 operate. Consequently, release 4.20 contains an out-of-date version. This
6058 shouldn't happen again.
6060 5. When building Exim with embedded Perl support, the script that builds the
6061 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6062 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6064 6. The freeze_tell option was not being used for messages that were frozen on
6065 arrival, either by an ACL or by local_scan().
6067 7. Added the smtp_incomplete_transaction log selector.
6069 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6070 was accepting AUTH without a new EHLO.
6072 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6073 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6076 10. Logging of TCP/IP connections (when configured) now happens in the main
6077 daemon process instead of the child process, so that the TCP/IP connection
6078 count is more accurate (but it can never be perfect).
6080 11. The use of "drop" in a nested ACL was not being handled correctly in the
6081 outer ACL. Now, if condition failure induced by the nested "drop" causes
6082 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6083 or "require"), the connection is dropped.
6085 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6086 that yield "discard" can now be used with an "accept" or a "discard" verb,
6087 but an error is generated for any others (because I can't see a useful way
6088 to define what should happen).
6090 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6091 are now processed in the same way as lines in the Exim configuration file.
6092 In particular, continuation lines are supported.
6094 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6096 15. Added -ti meaning -t -i.
6098 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6099 domains, and warn by logging if others are found.
6101 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6102 returned from GIFCONF when Exim is trying to find the list of interfaces on
6103 a host. The code in os.c has been modified to copy each ifreq to an aligned
6104 structure in all cases.
6106 Also, in some cases, the returned ifreq's were being copied to a 'struct
6107 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6108 means the last couple of bytes of an IPv6 address could be chopped if the
6109 ifreq contained only a normal sockaddr (14 bytes storage).
6111 18. Named domain lists were not supported in the hosts_treat_as_local option.
6112 An entry such as +xxxx was not recognized, and was treated as a literal
6115 19. Ensure that header lines added by a DATA ACL are included in the reject log
6116 if the ACL subsequently rejects the message.
6118 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6119 MD5 (which is deprecated).
6121 21. When testing a filter file using -bf, Exim was writing a message when it
6122 took the sender from a "From " line in the message, but it was not doing so
6123 when it took $return_path from a Return-Path: header line. It now does.
6125 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6126 with a valid header line field name (a series of printing characters
6127 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6129 23. Changed "disc" in the source to "disk" to conform to the documentation and
6130 the book and for uniformity.
6132 24. Ignore Sendmail's -Ooption=value command line item.
6134 25. When execve() failed while trying to run a command in a pipe transport,
6135 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6136 could be confused with a return value of 69 from the command itself. This
6137 has been changed to 127, the value the shell returns if it is asked to run
6138 a non-existent command. The wording for the related log line suggests a
6139 non-existent command as the problem.
6141 26. If received_header_text expands to an empty string, do not add a Received:
6142 header line to the message. (Well, it adds a token one on the spool, but
6143 marks it "old" so that it doesn't get used or transmitted.)
6145 27. Installed eximstats 1.28 (addition of -nt option).
6147 28. There was no check for failure on the call to getsockname() in the daemon
6148 code. This can fail if there is a shortage of resources on the system, with
6149 ENOMEM, for example. A temporary error is now given on failure.
6151 29. Contrary to the C standard, it seems that in some environments, the
6152 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6153 program. Exim now does this explicitly; it affects the formatting of
6154 timestamps using strftime().
6156 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6157 complaints. I've now initialized all the variables, to avoid this.
6159 32. Header lines added by a system filter were not being "seen" during
6160 transport-time rewrites.
6162 33. The info_callback() function passed to OpenSSL is set up with type void
6163 (*)(SSL *, int, int), as described somewhere. However, when calling the
6164 function (actually a macro) that sets it up, the type void(*)() is
6165 expected. I've put in a cast to prevent warnings from picky compilers.
6167 34. If a DNS black list lookup found a CNAME record, but there were no A
6168 records associated with the domain it pointed at, Exim crashed.
6170 35. If a DNS black list lookup returned more than one A record, Exim ignored
6171 all but the first. It now scans all returned addresses if a particular IP
6172 value is being sought. In this situation, the contents of the
6173 $dnslist_value variable are a list of all the addresses, separated by a
6176 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6177 to accept a host name and all its aliases if the forward lookup for any of
6178 them yielded the IP address of the incoming connection. Now it accepts only
6179 those names whose forward lookup yields the correct IP address. Any other
6180 names are discarded. This closes a loophole whereby a rogue DNS
6181 administrator could create reverse DNS records to break through a
6182 wildcarded host restriction in an ACL.
6184 37. If a user filter or a system filter that ran in a subprocess used any of
6185 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6186 the wrong values were passed to the pipe command ($thisaddress had the
6187 value of $0, $0 had the value of $1, etc). This bug was introduced by
6188 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6190 38. Improved the line breaking for long SMTP error messages from ACLs.
6191 Previously, if there was no break point between 40 and 75 characters, Exim
6192 left the rest of the message alone. Two changes have been made: (a) I've
6193 reduced the minimum length to 35 characters; (b) if it can't find a break
6194 point between 35 and 75 characters, it looks ahead and uses the first one
6195 that it finds. This may give the occasional overlong line, but at least the
6196 remaining text gets split now.
6198 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6199 file descriptors might be low, and that setting 1000 would always raise it.
6200 It turns out that in some environments, the limit is already over 1000 and
6201 that lowering it causes trouble. So now Exim takes care not to decrease it.
6203 40. When delivering a message, the value of $return_path is set to $sender_
6204 address at the start of routing (routers may change the value). By an
6205 oversight, this default was not being set up when an address was tested by
6206 -bt or -bv, which affected the outcome if any router or filter referred to
6209 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6210 log only when "message" or "log_message" are set. However, if one of the
6211 conditions was an address verification, or a call to a nested ACL, the
6212 messages generated by the underlying test were being passed through. This
6213 no longer happens. The underlying message is available in $acl_verify_
6214 message for both "message" and "log_message" expansions, so it can be
6215 passed through if needed.
6217 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6218 new expansion $bh_ to give the encoded byte string without charset
6219 translation. Translation happens only if iconv() is available; HAVE_ICONV
6220 indicates this at build time. HEADERS_CHARSET gives the charset to
6221 translate to; headers_charset can change it in the configuration, and
6222 "headers charset" can change it in an individual filter file.
6224 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6225 that creates RFC 2047 encoded "words" labels them as that charset instead
6226 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6227 expansion operator; (ii) when Exim creates a From: line for a local
6228 message; (iii) when a header line is rewritten to include a "phrase" part.
6230 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6231 buggy, causing it to skip the first lines of messages whose message ID
6232 ended in 'D'. This would not have bitten before Exim release 4.14, because
6233 message IDs were unlikely to end in 'D' before then. The effect was to have
6234 incorrect size information for certain domains.
6236 45. #include "config.h" was missing at the start of the crypt16.c module. This
6237 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6240 46. If there was a timeout during a "random" callout check, Exim treated it as
6241 a failure of the random address, and carried on sending RSET and the real
6242 address. If the delay was just some slowness somewhere, the response to the
6243 original RCPT would be taken as a response to RSET and so on, causing
6244 mayhem of various kinds.
6246 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6247 when I implemented it. It didn't allow for the fact that some option values
6248 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6249 the right test for positive values.
6251 48. Domain names in DNS records are case-independent. Exim always looks them up
6252 in lower case. Some resolvers return domain names in exactly the case they
6253 appear in the zone file, that is, they may contain uppercase letters. Not
6254 all resolvers do this - some return always lower case. Exim was treating a
6255 change of case by a resolver as a change of domain, similar to a widening
6256 of a domain abbreviation. This triggered its re-routing code and so it was
6257 trying to route what was effectively the same domain again. This normally
6258 caused routing to fail (because the router wouldn't handle the domain
6259 twice). Now Exim checks for this case specially, and just changes the
6260 casing of the domain that it ultimately uses when it transmits the message
6263 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6266 50. If a filter generated a file delivery with a non-absolute name (possible if
6267 no home directory exists for the router), the forbid_file option was not
6270 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6271 the existing equality matching.
6273 52. Exim was using ints instead of ino_t variables in some places where it was
6274 dealing with inode numbers.
6276 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6277 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6278 called TMPDIR, and if it finds it is different, it changes its value.
6280 54. The smtp_printf() function is now made available to local_scan() so
6281 additional output lines can be written before returning. There is also an
6282 smtp_fflush() function to enable the detection of a dropped connection.
6283 The variables smtp_input and smtp_batched_input are exported to
6286 55. Changed the default runtime configuration: the message "Unknown user"
6287 has been removed from the ACL, and instead placed on the localuser router,
6288 using the cannot_route_message feature. This means that any verification
6289 failures that generate their own messages won't get overridden. Similarly,
6290 the "Unrouteable address" message that was in the ACL for unverifiable
6291 relay addresses has also been removed.
6293 56. Added hosts_avoid_esmtp to the smtp transport.
6295 57. The exicyclog script was not checking for the esoteric option
6296 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6297 will work only if exicyclog is run under the appropriate euid.
6299 58. Following a discussion on the list, the rules by which Exim recognises line
6300 endings on incoming messages have been changed. The -dropcr and drop_cr
6301 options are now no-ops, retained only for backwards compatibility. The
6302 following line terminators are recognized: LF CRLF CR. However, special
6303 processing applies to CR:
6305 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6306 nor a local message in the state where . is a terminator.
6308 (ii) If a bare CR is encountered in a header line, an extra space is added
6309 after the line terminator so as not to end the header. The reasoning
6310 behind this is that bare CRs in header lines are most likely either
6311 to be mistakes, or people trying to play silly games.
6313 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6314 was being incorrectly given as 18 bytes larger than it should have been.
6315 This is a VOB (very old bug).
6317 60. This may never have affected anything current, but just in case it has:
6318 When the local host is found other than at the start of a list of hosts,
6319 the local host, those with the same MX, and any that follow, are discarded.
6320 When the list in question was part of a longer list of hosts, the following
6321 hosts (not currently being processed) were also being discarded. This no
6322 longer happens. I'm not sure if this situation could ever has previously
6325 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6328 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6329 header. This is something that is recommended in a new Internet Draft, and
6330 is something that is documented as being done by Sendmail. There are two
6331 possible values. For messages generated by the autoreply transport, Exim
6334 Auto-Submitted: auto-replied
6336 whereas for all other generated messages (e.g. bounces) it adds
6338 Auto-Submitted: auto-generated
6340 63. The "personal" condition in filters now includes a test for the
6341 Auto-Submitted: header. If it contains the string "auto-" the message it
6342 not considered personal.
6344 64. Added rcpt_include_affixes as a generic transport option.
6346 65. Added queue_only_override (default true).
6348 66. Added the syslog_duplication option.
6350 67. If what should have been the first header line of a message consisted of
6351 a space followed by a colon, Exim was mis-interpreting it as a header line.
6352 It isn't of course - it is syntactically invalid and should therefore be
6353 treated as the start of the message body. The misbehaviour could have
6354 caused a number of strange effects, including loss of data in subsequent
6355 header lines, and spool format errors.
6357 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6358 client host had authenticated. This control can now be exercised by an ACL
6359 for more flexibility.
6361 69. By default, callouts do not happen when testing with -bh. There is now a
6362 variant, -bhc, which does actually run the callout code, including
6363 consulting and updating the callout cache.
6365 70. Added support for saslauthd authentication, courtesy of Alexander
6368 71. If statvfs() failed on the spool or log directories while checking their
6369 size for availability, Exim confusingly gave the error "space shortage".
6370 Furthermore, in debugging mode it crashed with a floating point exception.
6371 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6372 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6373 serious problem, Exim now writes to the main and panic logs when this
6374 happens, with details of the failure. It then refuses to accept the
6375 incoming message, giving the message "spool directory problem" or "log
6376 directory problem" with a 421 code for SMTP messages.
6378 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6379 0, 1, and 2 exist, because some OS complain for execs without them (see
6380 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6381 descriptors. However, the code omitted to check that the open succeeded,
6382 causing mysterious errors if for some reason the permissions on /dev/null
6383 got screwed. Now Exim writes a message to the main and panic logs, and
6384 bombs out if it can't open /dev/null.
6386 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6387 interact so that it is all more flexible. It is supposed to remain
6388 backwards compatible. Also added extra_local_interfaces.
6390 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6391 to bomb out with an assertion failure - to the client this appears as a
6392 connection drop. This problem occurs in the part of the code that was taken
6393 from the Samba project. Fortunately, the assertion is in a very simple
6394 function, so I have fixed this by reproducing the function inline in the
6395 one place where it is called, and arranging for authentication to fail
6396 instead of killing the process with assert().
6398 75. The SPA client code was not working when the server requested OEM rather
6399 than Unicode encoding.
6401 76. Added code to make require_files with a specific uid setting more usable in
6402 the case where statting the file as root fails - usually a non-root-mounted
6403 NFS file system. When this happens and the failure is EACCES, Exim now
6404 forks a subprocess and does the per-uid checking as the relevant uid.
6406 77. Added process_log_path.
6408 78. If log_file_path was not explicitly set, a setting of check_log_space or
6409 check_log_inodes was ignored.
6411 79. If a space check for the spool or log partitions fails, the incident is now
6412 logged. Of course, in the latter case the data may get lost...
6414 80. Added the %p formatting code to string_format() so that it can be used to
6415 print addresses in debug_print(). Adjusted all the address printing in the
6416 debugging in store.c to use %p rather than %d.
6418 81. There was a concern that a line of code in smtp_in.c could overflow a
6419 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6420 initially expressed, the concern was not well-founded, because trailing
6421 spaces are removed early. However, if the trailing spaces were followed by
6422 a NULL, they did not get removed, so the overflow was possible. Two fixes
6425 (a) I re-wrote the offending code in a cleaner fashion.
6426 (b) If an incoming SMTP command contains a NULL character, it is rejected
6429 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6430 runs initgroups(), so that if the Exim user is in any additional groups,
6431 they will be used during message reception.
6437 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob