1 Change log file for Exim from version 4.21
2 -------------------------------------------
7 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
8 and 3.4.4 - once the server is enabled to respond to an OCSP request
9 it does even when not requested, resulting in a stapling non-aware
10 client dropping the TLS connection.
12 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
13 support variable-length bit vectors. No functional change.
15 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
16 The I= interface field on outgoing lines is now after the H= remote
17 host field, same as incoming lines. There is a separate
18 outgoing_interface log selector which allows you to disable the
21 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
22 If not running log_selector +smtp_connection the mainlog would be held
23 open indefinitely after a "too many connections" event, including to a
24 deleted file after a log rotate. Leave the per net connection logging
25 leaving it open for efficiency as that will be quickly detected by the
26 check on the next write.
28 HS/01 Bug 1671: Fix post transport crash.
29 Processing the wait-<transport> messages could crash the delivery
30 process if the message IDs didn't exist for some reason. When
31 using 'split_spool_directory=yes' the construction of the spool
32 file name failed already, exposing the same netto behaviour.
34 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
35 mime_regex ACL conditions.
37 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
38 to DSN fail messages (bounces): remote IP, remote greeting, remote response
39 to HELO, local diagnostic string.
41 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
42 log line to debug. Even when configured with a tls authenticator many
43 client connections are expected to not authenticate in this way, so
44 an authenticate fail is not an error.
46 HS/02 Add the Exim version string to the process info. This way exiwhat
47 gives some more detail about the running daemon.
49 JH/06 Bug 1395: time-limit cacheing of DNS lookups, to the TTL value. This may
50 matter for fast-change records such as DNSBLs.
52 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
53 retry record, even if constant. There may be multiple transports with
54 different interface settings and the retry behaviour needs to be kept
57 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
59 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
61 JH/10 Bug 840: fix log_defer_output option of pipe transport
63 JH/11 Bug 830: use same host for all RCPTS of a message, even under
64 hosts_randomize. This matters a lot when combined with mua_wrapper.
66 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
67 ${quote_pgsql:<string>} operator.
69 JH/13 Bug 1708: avoid misaligned access in cached lookup.
71 JH/14 Change header file name for freeradius-client. Relevant if compiling
72 with Radius support; from the Gentoo tree and checked under Fedora.
74 JH/15 Bug 1712: Introduce $prdr_requested flag variable
76 JH/16 Bug 1714: Permit an empty string as expansion result for transport
77 option transport_filter, meaning no filtering.
79 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
81 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
82 defaults to "*" (all hosts). The variable is now available when not built
83 with TLS, default unset, mainly to enable keeping the testuite sane.
84 If a server certificate is not supplied (via tls_certificate) an error is
85 logged, and clients will find TLS connections fail on startup. Presumably
86 they will retry in-clear.
87 Packagers of Exim are strongly encouraged to create a server certificate
90 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
91 with the $config_file variable.
93 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
94 in transport context, after the attempt, and per-recipient. The latter type
95 is per host attempted. The event data is the error message, and the errno
96 information encodes the lookup type (A vs. MX) used for the (first) host,
97 and the trailing two digits of the smtp 4xx reponse.
99 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
100 to write to mainlog (or rejectlog, paniclog) in the window between file
101 creation and permissions/ownership being changed. Particularly affects
102 installations where exicyclog is run as root, rather than exim user;
103 result is that the running daemon panics and dies.
105 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
107 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
108 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
109 "pri" and "weight". Note that the previous implicit priority given by the
110 list order is no longer honoured.
112 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalisation
115 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
116 by defining SUPPORT_SOCKS.
118 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
119 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
120 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
121 variables are renamed to proxy_{local,external}_{address,port}.
123 JH/27 Move Internationalisation support from Experimental to mainline, enabled
124 for a build by defining SUPPORT_I18N
126 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
127 of the query string, and make ${quote_redis:} do that quoting.
129 JH/29 Move Events support from Experimental to mainline, enabled by default
130 and removable for a build by defining DISABLE_EVENT.
132 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
134 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
135 cached by the daemon.
137 JH/32 Move Redis support from Experimental to mainline, enabled for a build
138 by defining LOOKUP_REDIS. The libhiredis library is required.
140 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
141 keys are given for lookup.
143 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
144 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
145 only supported when built with TLS support.
147 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
148 openssl_options), for security. OpenSSL forces this from version 1.1.0
149 server-side so match that on older versions.
154 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
157 JH/02 The smtp transport option "multi_domain" is now expanded.
159 JH/03 The smtp transport now requests PRDR by default, if the server offers
162 JH/04 Certificate name checking on server certificates, when exim is a client,
163 is now done by default. The transport option tls_verify_cert_hostnames
164 can be used to disable this per-host. The build option
165 EXPERIMENTAL_CERTNAMES is withdrawn.
167 JH/05 The value of the tls_verify_certificates smtp transport and main options
168 default to the word "system" to access the system default CA bundle.
169 For GnuTLS, only version 3.0.20 or later.
171 JH/06 Verification of the server certificate for a TLS connection is now tried
172 (but not required) by default. The verification status is now logged by
173 default, for both outbound TLS and client-certificate supplying inbound
176 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
179 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
180 Status Notification (bounce) messages are now MIME format per RFC 3464.
181 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
182 under the control of the dsn_advertise_hosts option, and routers may
183 have a dsn_lasthop option.
185 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
186 default, modifiable by a malware= option. The list separator for
187 the options can now be changed in the usual way. Bug 68.
189 JH/10 The smtp_receive_timeout main option is now expanded before use.
191 JH/11 The incoming_interface log option now also enables logging of the
192 local interface on delivery outgoing connections.
194 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
195 if the interface and destination host and port all match.
197 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
200 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
201 Patch from Andrew Lewis.
203 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
204 now supports optional time-restrictions, weighting, and priority
205 modifiers per server. Patch originally by <rommer@active.by>.
207 JH/16 The spamd_address main option now supports a mixed list of local
208 and remote servers. Remote servers can be IPv6 addresses, and
209 specify a port-range.
211 JH/17 Bug 68: The spamd_address main option now supports an optional
212 timeout value per server.
214 JH/18 Bug 1581: Router and transport options headers_add/remove can
215 now have the list separator specified.
217 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
220 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
223 JH/21 Support for the A6 type of dns record is withdrawn.
225 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
226 rather than the verbs used.
228 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
229 from 255 to 1024 chars.
231 JH/24 Verification callouts now attempt to use TLS by default.
233 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
234 are generic router options now. The defaults didn't change.
236 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
237 Original patch from Alexander Shikoff, worked over by JH.
239 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
240 files if ZCAT_COMMAND is not executable.
242 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
244 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
246 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
247 Normally benign, it bites when the pair was led to by a CNAME;
248 modern usage is to not canoicalize the domain to a CNAME target
249 (and we were inconsistent anyway for A-only vs AAAA+A).
251 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
253 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
254 when evaluating $sender_host_dnssec.
256 JH/31 Check the HELO verification lookup for DNSSEC, adding new
257 $sender_helo_dnssec variable.
259 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
261 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
263 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
265 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
266 documented as working, but never had. Support all but $spam_report.
268 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
269 added for tls authenticator.
274 TL/01 When running the test suite, the README says that variables such as
275 no_msglog_check are global and can be placed anywhere in a specific
276 test's script, however it was observed that placement needed to be near
277 the beginning for it to behave that way. Changed the runtest perl
278 script to read through the entire script once to detect and set these
279 variables, reset to the beginning of the script, and then run through
280 the script parsing/test process like normal.
282 TL/02 The BSD's have an arc4random API. One of the functions to induce
283 adding randomness was arc4random_stir(), but it has been removed in
284 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
285 function when detected.
287 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
288 cause callback expansion.
290 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
291 syntax errors in an expansion can be treated as a string instead of
292 logging or causing an error, due to the internal use of bool_lax
293 instead of bool when processing it.
295 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
296 server certificates when making smtp deliveries.
298 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
300 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
302 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
304 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
305 Merged patch from Sebastian Wiedenroth.
307 JH/05 Fix results-pipe from transport process. Several recipients, combined
308 with certificate use, exposed issues where response data items split
309 over buffer boundaries were not parsed properly. This eventually
310 resulted in duplicates being sent. This issue only became common enough
311 to notice due to the introduction of conection certificate information,
312 the item size being so much larger. Found and fixed by Wolfgang Breyha.
314 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
315 size buffer was used, resulting in syntax errors when an expansion
318 JH/07 Add support for directories of certificates when compiled with a GnuTLS
319 version 3.3.6 or later.
321 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
322 is EXPERIMENTAL_EVENT, the main-configuration and transport options
323 both become "event_action", the variables become $event_name, $event_data
324 and $event_defer_errno. There is a new variable $verify_mode, usable in
325 routers, transports and related events. The tls:cert event is now also
326 raised for inbound connections, if the main configuration event_action
329 TL/06 In test suite, disable OCSP for old versions of openssl which contained
330 early OCSP support, but no stapling (appears to be less than 1.0.0).
332 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
333 server certificate names available under the smtp transport option
334 "tls_verify_cert_hostname" now do not permit multi-component wildcard
337 JH/10 Time-related extraction expansions from certificates now use the main
338 option "timezone" setting for output formatting, and are consistent
339 between OpenSSL and GnuTLS compilations. Bug 1541.
341 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
342 encoded parameter in the incoming message. Bug 1558.
344 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
345 include certificate info, eximon was claiming there were spoolfile
348 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
350 JH/14 Log delivery-related information more consistently, using the sequence
351 "H=<name> [<ip>]" wherever possible.
353 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
354 are problematic for Debian distribution, omit them from the release
357 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
359 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
362 JH/17 Fix a null-indirection in certextract expansions when a nondefault
363 output list separator was used.
368 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
369 checkers that were complaining about end of non-void function with no
372 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
373 This was a regression intruduced in 4.83 by another bugfix.
375 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
377 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
378 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
384 TF/01 Correctly close the server side of TLS when forking for delivery.
386 When a message was received over SMTP with TLS, Exim failed to clear up
387 the incoming connection properly after forking off the child process to
388 deliver the message. In some situations the subsequent outgoing
389 delivery connection happened to have the same fd number as the incoming
390 connection previously had. Exim would try to use TLS and fail, logging
391 a "Bad file descriptor" error.
393 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
394 utilities have not been installed.
396 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
397 temporary space as the ACL may create new global variables.
399 TL/01 LDAP support uses per connection or global context settings, depending
400 upon the detected version of the libraries at build time.
402 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
403 to extract and use the src ip:port in logging and expansions as if it
404 were a direct connection from the outside internet. PPv2 support was
405 updated based on HAProxy spec change in May 2014.
407 JH/02 Add ${listextract {number}{list}{success}{fail}}.
409 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
410 Properly escape header and check for NULL return.
412 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
415 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
417 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
418 characters in header names, implemented as a verify condition.
419 Contributed by Michael Fischer v. Mollard.
421 TL/05 Rename SPF condition results err_perm and err_temp to standardized
422 results permerror and temperror. Previous values are deprecated but
423 still accepted. In a future release, err_perm and err_temp will be
424 completely removed, which will be a backward incompatibility if the
425 ACL tests for either of these two old results. Patch contributed by
426 user bes-internal on the mailing list.
428 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
430 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
431 selectors, in both main and reject logs.
433 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
436 JH/07 Add malware type "sock" for talking to simple daemon.
438 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
440 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
441 routers/transports under cutthrough routing.
443 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
444 numbers. Touch up "bool" conditional to keep the same definition.
446 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
448 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
450 JH/12 Expand items in router/transport headers_add or headers_remove lists
451 individually rather than the list as a whole. Bug 1452.
453 Required for reasonable handling of multiple headers_ options when
454 they may be empty; requires that headers_remove items with embedded
455 colons must have them doubled (or the list-separator changed).
457 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
458 view the policy declared in the DMARC record. Currently, $dmarc_status
459 is a combined value of both the record presence and the result of the
462 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
464 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
465 dnslookup router and the smtp transport (applying to the forward
468 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
469 of ldap servers used for a specific lookup. Patch provided by Heiko
472 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
473 New variable $lookup_dnssec_authenticated for observability.
475 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
476 Patch submitted by Lars Timman.
478 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
480 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
481 Requires trusted mode and valid format message id, aborts otherwise.
482 Patch contributed by Heiko Schlichting.
484 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
485 certextract with support for various fields. Bug 1358.
487 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
488 is requested by default, modifiable by smtp transport option
491 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
492 operate on certificate variables to give certificate fingerprints
493 Also new ${sha256:cert_variable}.
495 JH/23 The PRDR feature is moved from being Experimental into the mainline.
497 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
498 Christian Aistleitner.
500 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
502 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
503 file. Patch from Wolfgang Breyha.
505 JH/25 Expand the coverage of the delivery $host and $host_address to
506 client authenticators run in verify callout. Bug 1476.
508 JH/26 Port service names are now accepted for tls_on_connect_ports, to
509 align with daemon_smtp_ports. Bug 72.
511 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
512 support and error reporting did not work properly.
514 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
515 and is readable. Patch from Andrew Colin Kissa.
517 TL/14 Enhance documentation of ${run expansion and how it parses the
518 commandline after expansion, particularly in the case when an
519 unquoted variable expansion results in an empty value.
521 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
523 PP/02 Fix internal collision of T_APL on systems which support RFC3123
524 by renaming away from it. Addresses GH issue 15, reported by
527 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
529 TL/15 SECURITY: prevent double expansion in math comparison functions
530 (can expand unsanitized data). Not remotely exploitable.
537 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
539 PP/02 Make -n do something, by making it not do something.
540 When combined with -bP, the name of an option is not output.
542 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
545 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
546 $sender_host_name and config options to manage this, and basic check
549 PP/05 DSCP support for outbound connections and control modifier for inbound.
551 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
552 (Only plugin which currently uses this is kerberos4, which nobody should
553 be using, but we should make it available and other future plugins might
554 conceivably use it, even though it would break NAT; stuff *should* be
555 using channel bindings instead).
557 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
558 name; added for Sendmail compatibility; requires admin caller.
559 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
560 just ignore it); requires trusted caller.
561 Also parse but ignore: -Ac -Am -X<logfile>
564 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
566 TL/02 Add +smtp_confirmation as a default logging option.
568 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
569 Patch by Magnus Holmgren from 2007-02-20.
571 TL/04 Bugzilla 1281 - Spec typo.
572 Bugzilla 1283 - Spec typo.
573 Bugzilla 1290 - Spec grammar fixes.
575 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
577 TL/06 Add Experimental DMARC support using libopendmarc libraries.
579 TL/07 Fix an out of order global option causing a segfault. Reported to dev
580 mailing list by by Dmitry Isaikin.
582 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
584 JH/02 Support "G" suffix to numbers in ${if comparisons.
586 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
588 NM/01 Bugzilla 1197 - Spec typo
589 Bugzilla 1196 - Spec examples corrections
591 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
593 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
594 gnutls_enable_pkcs11, but renamed to more accurately indicate its
597 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
598 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
600 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
601 "acl {{name}{arg}...}", and optional args on acl condition
604 JH/05 Permit multiple router/transport headers_add/remove lines.
606 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
608 JH/07 Avoid using a waiting database for a single-message-only transport.
609 Performance patch from Paul Fisher. Bugzilla 1262.
611 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
614 JH/09 Add $headers_added variable, with content from use of ACL modifier
615 add_header (but not yet added to the message). Bugzilla 199.
617 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
618 Pulled from Bugzilla 817 by Wolfgang Breyha.
620 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
622 (nb: this is the same fix as in Exim 4.80.1)
624 JH/11 Add A= logging on delivery lines, and a client_set_id option on
627 JH/12 Add optional authenticated_sender logging to A= and a log_selector
630 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
632 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
633 advertise SMTP AUTH mechanism to us, instead of a generic
634 protocol violation error. Also, make Exim more robust to bad
635 data from the Dovecot auth socket.
637 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
639 When a queue runner is handling a message, Exim first routes the
640 recipient addresses, during which it prunes them based on the retry
641 hints database. After that it attempts to deliver the message to
642 any remaining recipients. It then updates the hints database using
645 So if a recipient address works intermittently, it can get repeatedly
646 deferred at routing time. The retry hints record remains fresh so the
647 address never reaches the final cutoff time.
649 This is a fairly common occurrence when a user is bumping up against
650 their storage quota. Exim had some logic in its local delivery code
651 to deal with this. However it did not apply to per-recipient defers
652 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
654 This change adds a proper retry rule check during routing so that the
655 final cutoff time is checked against the message's age. We only do
656 this check if there is an address retry record and there is not a
657 domain retry record; this implies that previous attempts to handle
658 the address had the retry_use_local_parts option turned on. We use
659 this as an approximation for the destination being like a local
660 delivery, as in LMTP.
662 I suspect this new check makes the old local delivery cutoff check
663 redundant, but I have not verified this so I left the code in place.
665 TF/02 Correct gecos expansion when From: is a prefix of the username.
667 Test 0254 submits a message to Exim with the header
671 When I ran the test suite under the user fanf2, Exim expanded
672 the header to contain my full name, whereas it should have added
673 a Resent-Sender: header. It erroneously treats any prefix of the
674 username as equal to the username.
676 This change corrects that bug.
678 GF/01 DCC debug and logging tidyup
679 Error conditions log to paniclog rather than rejectlog.
680 Debug lines prefixed by "DCC: " to remove any ambiguity.
682 TF/03 Avoid unnecessary rebuilds of lookup-related code.
684 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
685 Bug spotted by Jeremy Harris; was flawed since initial commit.
686 Would have resulted in OCSP responses post-SNI triggering an Exim
687 NULL dereference and crash.
689 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
691 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
692 Bug detection, analysis and fix by Samuel Thibault.
693 Bugzilla 1331, Debian bug #698092.
695 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
697 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
698 Server implementation by Todd Lyons, client by JH.
699 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
700 config variable "prdr_enable" controls whether the server
701 advertises the facility. If the client requests PRDR a new
702 acl_data_smtp_prdr ACL is called once for each recipient, after
703 the body content is received and before the acl_smtp_data ACL.
704 The client is controlled by bolth of: a hosts_try_prdr option
705 on the smtp transport, and the server advertisement.
706 Default client logging of deliveries and rejections involving
707 PRDR are flagged with the string "PRDR".
709 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
710 fclose(). Diagnosis by Todd Lyons.
712 PP/17 Update configure.default to handle IPv6 localhost better.
713 Patch by Alain Williams (plus minor tweaks).
716 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
717 This is now consistent with GnuTLS, and is now documented: the
718 previous undocumented portable approach to treating the option as
719 unset was to force an expansion failure. That still works, and
720 an empty string is now equivalent.
722 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
723 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
724 not performing validation itself.
726 PP/20 Added force_command boolean option to pipe transport.
727 Patch from Nick Koston, of cPanel Inc.
729 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
732 TF/04 Added udpsend ACL modifer and hexquote expansion operator
734 PP/21 Fix eximon continuous updating with timestamped log-files.
735 Broken in a format-string cleanup in 4.80, missed when I repaired the
736 other false fix of the same issue.
737 Report and fix from Heiko Schlichting.
740 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
741 Report from Prashanth Katuri.
743 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
744 It's SecureTransport, so affects any MacOS clients which use the
745 system-integrated TLS libraries, including email clients.
747 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
748 using a MIME ACL for non-SMTP local injection.
749 Report and assistance in diagnosis by Warren Baker.
751 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
753 JH/16 Fix comparisons for 64b. Bugzilla 1385.
755 TL/09 Add expansion variable $authenticated_fail_id to keep track of
756 last id that failed so it may be referenced in subsequent ACL's.
758 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
761 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
762 ldap library initialization, allowing self-signed CA's to be
763 used. Also properly sets require_cert option later in code by
764 using NULL (global ldap config) instead of ldap handle (per
765 session). Bug diagnosis and testing by alxgomz.
767 TL/12 Enhanced documentation in the ratelimit.pl script provided in
768 the src/util/ subdirectory.
770 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
771 renamed to Transport Post Delivery Action by Jeremy Harris, as
774 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
775 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
776 redis_servers = needs to be configured which will be used by the redis
777 lookup. Patch from Warren Baker, of The Packet Hub.
779 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
781 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
782 hostname or reverse DNS when processing a host list. Used suggestions
783 from multiple comments on this bug.
785 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
787 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
788 Missed a few lines, added it to make the runtest require no keyboard
791 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
792 contains upper case chars. Make router use caseful_local_part.
794 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
795 support when GnuTLS has been built with p11-kit.
801 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
803 This, or similar/improved, will also be change PP/11 of 4.82.
809 PP/01 Handle short writes when writing local log-files.
810 In practice, only affects FreeBSD (8 onwards).
811 Bugzilla 1053, with thanks to Dmitry Isaikin.
813 NM/01 Bugzilla 949 - Documentation tweak
815 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
818 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
820 PP/02 Implemented gsasl authenticator.
822 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
824 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
825 `pkg-config foo` for cflags/libs.
827 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
828 with rest of GSASL and with heimdal_gssapi.
830 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
831 `pkg-config foo` for cflags/libs for the TLS implementation.
833 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
834 properties get this fed in as external SSF. A number of robustness
835 and debugging improvements to the cyrus_sasl authenticator.
837 PP/08 cyrus_sasl server now expands the server_realm option.
839 PP/09 Bugzilla 1214 - Log authentication information in reject log.
840 Patch by Jeremy Harris.
842 PP/10 Added dbmjz lookup type.
844 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
846 PP/12 MAIL args handles TAB as well as SP, for better interop with
847 non-compliant senders.
848 Analysis and variant patch by Todd Lyons.
850 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
851 Bug report from Lars Müller <lars@samba.org> (via SUSE),
852 Patch from Dirk Mueller <dmueller@suse.com>
854 PP/13 tls_peerdn now print-escaped for spool files.
855 Observed some $tls_peerdn in wild which contained \n, which resulted
856 in spool file corruption.
858 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
859 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
860 or write after TLS renegotiation, which otherwise led to messages
863 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
864 as a tracking header (ie: a signed header comes before the signature).
865 Patch from Wolfgang Breyha.
867 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
868 comma-sep list; embedded commas doubled.
870 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
872 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
874 Report and patch from Dmitry Banschikov.
876 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
877 Removed SSL_clear() after SSL_new() which led to protocol negotiation
878 failures. We appear to now support TLS1.1+ with Exim.
880 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
881 lets Exim select keys and certificates based upon TLS SNI from client.
882 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
883 before an outbound SMTP session. New log_selector, +tls_sni.
885 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
886 NULL dereference. Report and patch from Alun Jones.
888 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
889 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
890 Not seeing resolver debug output on NetBSD, but suspect this is a
891 resolver implementation change.
893 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
894 Left warnings. Added "eximon gdb" invocation mode.
896 PP/21 Defaulting "accept_8bitmime" to true, not false.
898 PP/22 Added -bw for inetd wait mode support.
900 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
901 locate the relevant includes and libraries. Made this the default.
903 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
904 Bugzilla 1246, report and most of solution from Tomasz Kusy.
906 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
907 This may cause build issues on older platforms.
909 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
910 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
911 gnutls_require_mac & gnutls_require_protocols (no longer supported).
912 Added SNI support via GnuTLS too.
913 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
915 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
917 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
918 Applied second patch from Janne, implementing suggestion to default
919 multiple-strings-in-record handling to match SPF spec.
921 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
923 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
924 read-only, out of scope).
925 Patch from Wolfgang Breyha, report from Stuart Northfield.
927 PP/29 Fix three issues highlighted by clang analyser static analysis.
928 Only crash-plausible issue would require the Cambridge-specific
929 iplookup router and a misconfiguration.
930 Report from Marcin Mirosław.
932 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
934 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
935 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
936 As part of this, removing so much warning spew let me fix some minor
937 real issues in debug logging.
939 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
940 assignment on my part. Fixed.
942 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
943 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
944 Janne Snabb (who went above and beyond: thank you).
946 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
947 string otherwise requires a connection and a bunch more work and it's
948 relatively easy to get wrong. Should also expose TLS library linkage
951 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
952 64-bit ${eval} (JH/03).
954 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
955 GNU libc to support some of the 64-bit stuff, should not lead to
956 conflicts. Defined before os.h is pulled in, so if a given platform
957 needs to override this, it can.
959 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
960 protection layer was required, which is not implemented.
961 Bugzilla 1254, patch from Wolfgang Breyha.
963 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
964 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
965 tls_dhparam take prime identifiers. Also unbreak combination of
966 OpenSSL+DH_params+TLSSNI.
968 PP/39 Disable SSLv2 by default in OpenSSL support.
974 PP/01 Solaris build fix for Oracle's LDAP libraries.
975 Bugzilla 1109, patch from Stephen Usher.
977 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
979 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
982 TF/02 Fix a couple more cases where we did not log the error message
983 when unlink() failed. See also change 4.74-TF/03.
985 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
986 lock up or crash if it happened to be inside a call to libc when it
987 got a SIGUSR1 from exiwhat.
989 The SIGUSR1 handler appends the current process status to the process
990 log which is later printed by exiwhat. It used to use the general
991 purpose logging code to do this, but several functions it calls are
992 not safe for signals.
994 The new output code in the SIGUSR1 handler is specific to the process
995 log, and simple enough that it's easy to inspect for signal safety.
996 Removing some special cases also simplifies the general logging code.
997 Removing the spurious timestamps from the process log simplifies
1000 TF/04 Improved ratelimit ACL condition.
1002 The /noupdate option has been deprecated in favour of /readonly which
1003 has clearer semantics. The /leaky, /strict, and /readonly update modes
1004 are mutually exclusive. The update mode is no longer included in the
1005 database key; it just determines when the database is updated. (This
1006 means that when you upgrde Exim will forget old rate measurements.)
1008 Exim now checks that the per_* options are used with an update mode that
1009 makes sense for the current ACL. For example, when Exim is processing a
1010 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1011 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1012 must specify per_mail/readonly. If you omit the update mode it defaults to
1013 /leaky where that makes sense (as before) or /readonly where required.
1015 The /noupdate option is now undocumented but still supported for
1016 backwards compatibility. It is equivalent to /readonly except that in
1017 ACLs where /readonly is required you may specify /leaky/noupdate or
1018 /strict/noupdate which are treated the same as /readonly.
1020 A useful new feature is the /count= option. This is a generalization
1021 of the per_byte option, so that you can measure the throughput of other
1022 aggregate values. For example, the per_byte option is now equivalent
1023 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1025 The per_rcpt option has been generalized using the /count= mechanism
1026 (though it's more complicated than the per_byte equivalence). When it is
1027 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1028 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1029 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1030 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1031 Note that using per_rcpt with a non-readonly update mode in more than
1032 one ACL will cause the recipients to be double-counted. (The per_mail
1033 and per_byte options don't have this problem.)
1035 The handling of very low rates has changed slightly. If the computed rate
1036 is less than the event's count (usually one) then this event is the first
1037 after a long gap. In this case the rate is set to the same as this event's
1038 count, so that the first message of a spam run is counted properly.
1040 The major new feature is a mechanism for counting the rate of unique
1041 events. The new per_addr option counts the number of different
1042 recipients that someone has sent messages to in the last time period. It
1043 behaves like per_rcpt if all the recipient addresses are different, but
1044 duplicate recipient addresses do not increase the measured rate. Like
1045 the /count= option this is a general mechanism, so the per_addr option
1046 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1047 example, measure the rate that a client uses different sender addresses
1048 with the options per_mail/unique=$sender_address. There are further
1049 details in the main documentation.
1051 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1053 TF/06 Removed a few PCRE remnants.
1055 TF/07 Automatically extract Exim's version number from tags in the git
1056 repository when doing development or release builds.
1058 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1059 Bugzilla 879. Patch from Paul Fisher.
1061 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1062 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1065 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1067 PP/05 Variable $av_failed, true if the AV scanner deferred.
1068 Bugzilla 1078. Patch from John Horne.
1070 PP/06 Stop make process more reliably on build failure.
1071 Bugzilla 1087. Patch from Heiko Schlittermann.
1073 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1074 Bugzilla 1089. Patch from Heiko Schlittermann.
1076 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1077 Bugzilla 1131. Patch from Holger Weiß.
1079 PP/09 Handle IPv6 addresses with SPF.
1080 Bugzilla 860. Patch from Wolfgang Breyha.
1082 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1084 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1087 PP/11 match_* no longer expand right-hand-side by default.
1088 New compile-time build option, EXPAND_LISTMATCH_RHS.
1089 New expansion conditions, "inlist", "inlisti".
1091 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1093 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1095 PP/14 fix log_write() format string regression from TF/03.
1096 Bugzilla 1152. Patch from Dmitry Isaikin.
1102 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1104 PP/02 Harmonised TLS library version reporting; only show if debugging.
1105 Layout now matches that introduced for other libraries in 4.74 PP/03.
1107 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1109 PP/04 New "dns_use_edns0" global option.
1111 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1114 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1115 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1117 TK/01 Updated PolarSSL code to 0.14.2.
1118 Bugzilla 1097. Patch from Andreas Metzler.
1120 PP/07 Catch divide-by-zero in ${eval:...}.
1121 Fixes bugzilla 1102.
1123 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1126 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1127 format-string attack -- SECURITY: remote arbitrary code execution.
1129 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1130 time unintentionally subject to list matching rules, letting the header
1131 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1132 arbitrary string expansion). This allowed for information disclosure.
1134 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1135 INT_MIN/-1 -- value coerced to INT_MAX.
1141 NM/01 Workround for PCRE version dependency in version reporting
1144 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1145 This fixes portability to compilers other than gcc, notably
1146 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1148 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1149 makefiles for portability to HP-UX and POSIX correctness.
1151 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1152 Also via indented variable definition in the Makefile.
1153 (Debugging by Oliver Heesakkers).
1155 PP/02 Restore caching of spamd results with expanded spamd_address.
1156 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1158 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1159 Improves build reliability. Fix from: Frank Elsner
1161 NM/02 Fix wide character breakage in the rfc2047 coding
1162 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1164 NM/03 Allow underscore in dnslist lookups
1165 Fixes bug 1026. Patch from Graeme Fowler
1167 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1168 Code patches from Adam Ciarcinski of NetBSD.
1170 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1173 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1174 is logged, to avoid truncation. Patch from John Horne.
1176 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1177 Patch from Jakob Hirsch.
1179 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1180 SQL string expansion failure details.
1181 Patch from Andrey Oktyabrski.
1183 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1184 Patch from Simon Arlott.
1186 PP/09 New lookups functionality failed to compile on old gcc which rejects
1187 extern declarations in function scope.
1188 Patch from Oliver Fleischmann
1190 PP/10 Use sig_atomic_t for flags set from signal handlers.
1191 Check getgroups() return and improve debugging.
1192 Fixed developed for diagnosis in bug 927 (which turned out to be
1195 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1196 Patch from Mark Zealey.
1198 PP/12 Bugzilla 1056: Improved spamd server selection.
1199 Patch from Mark Zealey.
1201 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1202 Based on patch from Heiko Schlittermann.
1204 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1205 Patch from Uwe Doering, sign-off by Michael Haardt.
1207 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1208 variable declaration deep within a block. Bug and patch from
1211 PP/15 lookups-Makefile IRIX compatibilty coercion.
1213 PP/16 Make DISABLE_DKIM build knob functional.
1215 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1216 Patch by Simon Arlott
1218 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1219 variable argument macros. Our copy now differs from upstream.
1225 TF/01 Failure to get a lock on a hints database can have serious
1226 consequences so log it to the panic log.
1228 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1229 controlled using the smtp_confirmation log selector.
1231 TF/03 Include the error message when we fail to unlink a spool file.
1233 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1234 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1235 for maintaining out-of-tree patches for some time.
1237 PP/01 Bugzilla 139: Documentation and portability issues.
1238 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1239 Handle per-OS dynamic-module compilation flags.
1241 PP/02 Let /dev/null have normal permissions.
1242 The 4.73 fixes were a little too stringent and complained about the
1243 permissions on /dev/null. Exempt it from some checks.
1244 Reported by Andreas M. Kirchwitz.
1246 PP/03 Report version information for many libraries, including
1247 Exim version information for dynamically loaded libraries. Created
1248 version.h, now support a version extension string for distributors
1249 who patch heavily. Dynamic module ABI change.
1251 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1252 privilege escalation vulnerability whereby the Exim run-time user
1253 can cause root to append content of the attacker's choosing to
1256 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1259 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1260 If dropping privileges for untrusted macros, we disabled normal logging
1261 on the basis that it would fail; for the Exim run-time user, this is not
1262 the case, and it resulted in successful deliveries going unlogged.
1263 Fixed. Reported by Andreas Metzler.
1269 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1270 only prepend for the Resent-* case. Fixes regression introduced in
1271 Exim 4.70 by NM/22 for Bugzilla 607.
1273 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1274 increasing numbers of administrators be bitten by this.
1276 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1278 PP/03 Bugzilla 994: added openssl_options main configuration option.
1280 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1282 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1284 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1286 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1287 without a peer certificate, leading to a segfault because of an
1288 assumption that peers always have certificates. Be a little more
1289 paranoid. Problem reported by Martin Tscholak.
1291 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1292 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1293 NB: ClamAV planning to remove STREAM in "middle of 2010".
1294 CL also introduces -bmalware, various -d+acl logging additions and
1295 more caution in buffer sizes.
1297 PP/09 Implemented reverse_ip expansion operator.
1299 PP/10 Bugzilla 937: provide a "debug" ACL control.
1301 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1303 PP/12 Bugzilla 973: Implement --version.
1305 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1307 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1309 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1311 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1312 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1313 ignore trailing whitespace.
1315 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1317 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1320 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1321 Notification from Dr Andrew Aitchison.
1323 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1324 ExtendedDetectionInfo response format.
1325 Notification from John Horne.
1327 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1330 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1331 XSL and documented dependency on system catalogs, with examples of how
1334 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1337 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1338 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1339 configuration file which is writeable by the Exim user or group.
1341 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1342 of configuration files to cover files specified with the -C option if
1343 they are going to be used with root privileges, not just the default
1346 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1347 option (effectively making it always true).
1349 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1350 files to be used while preserving root privileges.
1352 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1353 that rogue child processes cannot use them.
1355 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1356 run-time user, instead of root.
1358 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1359 Exim run-time user without dropping privileges.
1361 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1362 result string, instead of calling string_vformat() twice with the same
1365 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1366 for other users. Others should always drop root privileges if they use
1367 -C on the command line, even for a whitelisted configure file.
1369 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1371 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1377 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1378 $data_path, and $header_path variables; fixed documentation bugs and
1381 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1382 exipick to access non-standard spools, including the "frozen" queue
1385 NM/01 Bugzilla 965: Support mysql stored procedures.
1386 Patch from Alain Williams
1388 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1390 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1391 Patch from Andreas Metzler
1393 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1394 Patch from Kirill Miazine
1396 NM/05 Bugzilla 671: Added umask to procmail example.
1398 JJ/03 installed exipick 20100323.0, fixing doc bug
1400 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1401 directory. Notification and patch from Dan Rosenberg.
1403 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1405 TK/02 Improve log output when DKIM signing operation fails.
1407 MH/01 Treat the transport option dkim_domain as a colon separated
1408 list, not as a single string, and sign the message with each element,
1409 omitting multiple occurences of the same signer.
1411 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1412 Bugzilla 985, 986. Patch by Simon Arlott
1414 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1415 Patch by Simon Arlott
1417 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1418 MBX locking. Notification from Dan Rosenberg.
1424 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1426 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1428 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1430 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1432 NM/04 Bugzilla 915: Flag broken perl installation during build.
1438 TK/01 Added patch by Johannes Berg that expands the main option
1439 "spamd_address" if it starts with a dollar sign.
1441 TK/02 Write list of recipients to X-Envelope-Sender header when building
1442 the mbox-format spool file for content scanning (suggested by Jakob
1445 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1446 (http://www.dcc-servers.net/) support via dccifd. Activated by
1447 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1449 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1450 by Mark Daniel Reidel <mr@df.eu>.
1452 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1453 When building exim an external PCRE library is now needed -
1454 PCRE is a system library on the majority of modern systems.
1455 See entry on PCRE_LIBS in EDITME file.
1457 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1458 conversation. Added nologin parameter to request.
1459 Patch contributed by Kirill Miazine.
1461 TF/01 Do not log submission mode rewrites if they do not change the address.
1463 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1465 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1466 log files in place. Contributed by Roberto Lima.
1468 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1470 TF/03 Bugzilla 615: When checking the local_parts router precondition
1471 after a local_part_suffix or local_part_prefix option, Exim now
1472 does not use the address's named list lookup cache, since this
1473 contains cached lookups for the whole local part.
1475 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1476 Robert Millan. Documentation is in experimental-spec.txt.
1478 TF/04 Bugzilla 668: Fix parallel build (make -j).
1480 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1482 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1483 Patch provided by Jan Srzednicki.
1485 TF/05 Leading white space used to be stripped from $spam_report which
1486 wrecked the formatting. Now it is preserved.
1488 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1489 that they are available at delivery time.
1491 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1493 TF/08 TLS error reporting now respects the incoming_interface and
1494 incoming_port log selectors.
1496 TF/09 Produce a more useful error message if an SMTP transport's hosts
1497 setting expands to an empty string.
1499 NM/06 Bugzilla 744: EXPN did not work under TLS.
1500 Patch provided by Phil Pennock.
1502 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1503 Patch provided by Richard Godbee.
1505 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1506 acl_smtp_notquit, added index entry.
1508 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1509 Patch provided by Eugene Bujak.
1511 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1512 accept(). Patch provided by Maxim Dounin.
1514 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1515 Patch provided by Phil Pennock.
1517 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1519 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1520 Patch provided by Brad "anomie" Jorsch.
1522 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1523 Patch provided by Dean Brooks.
1525 TK/05 Add native DKIM support (does not depend on external libraries).
1527 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1528 Patch provided by Graeme Fowler.
1530 NM/16 Bugzilla 851: Documentation example syntax fix.
1532 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1534 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1537 NM/19 Bugzilla 745: TLS version reporting.
1538 Patch provided by Phil Pennock.
1540 NM/20 Bugzilla 167: bool: condition support.
1541 Patch provided by Phil Pennock.
1543 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1544 clients. Patch provided by Phil Pennock.
1546 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1547 Patch provided by Brad "anomie" Jorsch.
1549 NM/23 Bugzilla 687: Fix misparses in eximstats.
1550 Patch provided by Heiko Schlittermann.
1552 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1553 Patch provided by Heiko Schlittermann.
1555 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1556 plus update to original patch.
1558 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1560 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1561 Patch provided by David Brownlee.
1563 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1565 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1567 NM/30 Bugzilla 888: TLS documentation bugfixes.
1569 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1571 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1572 Unlike the original bugzilla I have changed all shell scripts in src tree.
1574 NM/33 Bugzilla 898: Transport filter timeout fix.
1575 Patch by Todd Rinaldo.
1577 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1578 Patch by Serge Demonchaux.
1580 NM/35 Bugzilla 39: Base64 decode bug fixes.
1581 Patch by Jakob Hirsch.
1583 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1585 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1587 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1589 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1595 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1596 ALT-N's libdkim that I have put here:
1597 http://duncanthrax.net/exim-experimental/
1599 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1600 were called 'true' and it seems that C99 defines that as a reserved
1601 keyword to be used with 'bool' variable types. That means you could
1602 not include C99-style headers which use bools without triggering
1603 build errors in sieve.c.
1605 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1606 as mailq or other aliases. Changed the --help handling significantly
1607 to do whats expected. exim_usage() emits usage/help information.
1609 SC/01 Added the -bylocaldomain option to eximstats.
1611 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1613 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1615 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1621 PH/01 Another patch from the Sieve maintainer.
1623 PH/02 When an IPv6 address is converted to a string for single-key lookup
1624 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1625 dots are used instead of colons so that keys in lsearch files need not
1626 contain colons. This was done some time before quoting was made available
1627 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1628 (notated using the quote facility) so as to distinguish them from IPv4
1629 keys. This meant that lookups for IP addresses in host lists did not work
1630 for iplsearch lookups.
1632 This has been fixed by arranging for IPv6 addresses to be expressed with
1633 colons if the lookup type is iplsearch. This is not incompatible, because
1634 previously such lookups could never work.
1636 The situation is now rather anomolous, since one *can* have colons in
1637 ordinary lsearch keys. However, making the change in all cases is
1638 incompatible and would probably break a number of configurations.
1640 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1643 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1644 conversion specification without a maximum field width, thereby enabling
1645 a rogue spamd server to cause a buffer overflow. While nobody in their
1646 right mind would setup Exim to query an untrusted spamd server, an
1647 attacker that gains access to a server running spamd could potentially
1648 exploit this vulnerability to run arbitrary code as the Exim user.
1650 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1651 $primary_hostname instead of what libspf2 thinks the hosts name is.
1653 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1654 a directory entry by the name of the lookup key. Previously, if a
1655 symlink pointed to a non-existing file or a file in a directory that
1656 Exim lacked permissions to read, a lookup for a key matching that
1657 symlink would fail. Now it is enough that a matching directory entry
1658 exists, symlink or not. (Bugzilla 503.)
1660 PH/03 The body_linecount and body_zerocount variables are now exported in the
1663 PH/04 Added the $dnslist_matched variable.
1665 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1666 This means they are set thereafter only if the connection becomes
1669 PH/06 Added the client_condition to authenticators so that some can be skipped
1670 by clients under certain conditions.
1672 PH/07 The error message for a badly-placed control=no_multiline_responses left
1673 "_responses" off the end of the name.
1675 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1677 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1678 (without spaces) instead of just copying the configuration text.
1680 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1682 PH/11 Added $max_received_linelength.
1684 PH/12 Added +ignore_defer and +include_defer to host lists.
1686 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1687 way in which PCRE > 7.0 is built.
1689 PH/14 Implemented queue_only_load_latch.
1691 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1692 MAIL command. The effect was to mangle the value on 64-bit systems.
1694 PH/16 Another patch from the Sieve maintainer.
1696 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1698 PH/18 If a system quota error occurred while trying to create the file for
1699 a maildir delivery, the message "Mailbox is full" was not appended to the
1700 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1701 only to a quota excession during the actual writing of the file.
1703 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1704 characters?) which causes problems in log lines. The DN values are now
1705 passed through string_printing() before being added to log lines.
1707 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1708 and InterBase are left for another time.)
1710 PH/21 Added message_body_newlines option.
1712 PH/22 Guard against possible overflow in moan_check_errorcopy().
1714 PH/23 POSIX allows open() to be a macro; guard against that.
1716 PH/24 If the recipient of an error message contained an @ in the local part
1717 (suitably quoted, of course), incorrect values were put in $domain and
1718 $local_part during the evaluation of errors_copy.
1724 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1725 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1728 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1729 issue a MAIL command.
1731 PH/02 In an ACL statement such as
1733 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1735 if a client was not listed at all, or was listed with a value other than
1736 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1737 the condition was not true (as it should be), so access was not denied.
1738 The bug was that the ! inversion was incorrectly passed on to the second
1739 item. This has been fixed.
1741 PH/03 Added additional dnslists conditions == and =& which are different from
1742 = and & when the dns lookup returns more than one IP address.
1744 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1745 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1747 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1748 FSYNC, which compiles an option called disable_fsync that allows for
1749 bypassing fsync(). The documentation is heavily laced with warnings.
1751 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1753 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1754 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1755 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1756 including adding "make clean"; (3) Added -fPIC when compiling the test
1757 dynamically loaded module, to get rid of a warning.
1759 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1760 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1761 = 0s. The bug is otherwise harmless.
1763 PH/07 There was a bug in the dovecot authenticator such that the value of
1764 $auth1 could be overwritten, and so not correctly preserved, after a
1765 successful authentication. This usually meant that the value preserved by
1766 the server_setid option was incorrect.
1768 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1770 PH/09 Installed PCRE release 7.0.
1772 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1773 run for batched SMTP input. It is now run at the start of every message
1774 in the batch. While fixing this I discovered that the process information
1775 (output by running exiwhat) was not always getting set for -bs and -bS
1776 input. This is fixed, and it now also says "batched" for BSMTP.
1778 PH/11 Added control=no_pipelining.
1780 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1781 patch, slightly modified), and move the expansion of helo_data till after
1782 the connection is made in the smtp transport (so it can use these
1785 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1787 PH/14 Added log_selector = +pid.
1789 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1791 PH/16 Add ${if forany and ${if forall.
1793 PH/17 Added dsn_from option to vary the From: line in DSNs.
1795 PH/18 Flush SMTP output before performing a callout, unless control =
1796 no_callout_flush is set.
1798 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1799 was true (the default) a successful delivery failed to delete the retry
1800 item, thus causing premature timeout of the address. The bug is now
1803 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1805 PH/21 Long custom messages for fakedefer and fakereject are now split up
1806 into multiline reponses in the same way that messages for "deny" and
1807 other ACL rejections are.
1809 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1810 with slight modification.
1812 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1813 draft, changing the syntax and factoring some duplicate code".
1815 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1816 for deliveries of the second and subsequent messages over the same SMTP
1819 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1820 ${reduce, with only minor "tidies".
1822 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1824 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1825 expansion side effects.
1827 PH/27 When a message times out after an over-quota error from an Exim-imposed
1828 quota, the bounce message says "mailbox is full". This message was not
1829 being given when it was a system quota that was exceeded. It now should
1832 MH/03 Made $recipients available in local_scan(). local_scan() already has
1833 better access to the recipient list through recipients_list[], but
1834 $recipients can be useful in postmaster-provided expansion strings.
1836 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1837 in the case of a MAIL command with additional options following the
1838 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1839 were accidentally chopped off.
1841 PH/29 SMTP synchronization checks are implemented when a command is read -
1842 there is a check that no more input is waiting when there shouldn't be
1843 any. However, for some commands, a delay in an ACL can mean that it is
1844 some time before the response is written. In this time, more input might
1845 arrive, invalidly. So now there are extra checks after an ACL has run for
1846 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1847 pipelining has not been advertised.
1849 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1851 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1852 returned in the SMTP response when smtp_return_error_details was false.
1853 This has been fixed.
1855 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1856 instead of the C I/O that was originally supplied, because problems were
1857 reported on Solaris.
1859 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1860 Exim which did not show up earlier: it was assuming that a call to
1861 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1862 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1863 was a macro that became an assignment, so it seemed to work. This has
1864 changed to a proper function call with a void return, hence the compile
1865 error. Exim's code has been fixed.
1867 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1870 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1872 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1874 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1875 because it turns out to be too big or there is a timeout) there is no
1876 "Completed" line in the log. When some messages of this type were
1877 selected by exigrep, they were listed as "not completed". Others were
1878 picked up by some special patterns. I have improved the selection
1879 criteria to be more general.
1881 PH/38 The host_find_failed option in the manualroute router can now be set
1882 to "ignore", to completely ignore a host whose IP address cannot be
1883 found. If all hosts are ignored, the behaviour is controlled by the new
1884 host_all_ignored option.
1886 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1887 homing or because of multiple MX records with /mx) generated more than
1888 one IP address, and the following item turned out to be the local host,
1889 all the secondary addresses of the first item were incorrectly removed
1890 from the list, along with the local host and any following hosts (which
1891 is what is supposed to happen).
1893 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1894 whoever called Exim into the -H file. In the case of the daemon it was
1895 behaving confusingly. When first started, it used values for whoever
1896 started the daemon, but after a SIGHUP it used the Exim user (because it
1897 calls itself on a restart). I have changed the code so that it now always
1900 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1901 message are rejected with the same error (e.g. no authentication or bad
1902 sender address), and a DATA command is nevertheless sent (as can happen
1903 with PIPELINING or a stupid MUA), the error message that was given to the
1904 RCPT commands is included in the rejection of the DATA command. This is
1905 intended to be helpful for MUAs that show only the final error to their
1908 PH/42 Another patch from the Sieve maintainer.
1910 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1911 Eximstats - Fixed some broken HTML links and added missing column headers
1913 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1914 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1916 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1918 PH/43 Yet another patch from the Sieve maintainer.
1920 PH/44 I found a way to check for a TCP/IP connection going away before sending
1921 the response to the final '.' that terminates a message, but only in the
1922 case where the client has not sent further data following the '.'
1923 (unfortunately, this is allowed). However, in many cases there won't be
1924 any further data because there won't be any more messages to send. A call
1925 to select() can be used: if it shows that the input is "ready", there is
1926 either input waiting, or the socket has been closed. An attempt to read
1927 the next input character can distinguish the two cases. Previously, Exim
1928 would have sent an OK response which the client would never have see.
1929 This could lead to message repetition. This fix should cure that, at
1930 least in a lot of common cases.
1932 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1933 advertised in response to EHLO.
1939 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1940 fixed by 4.65/MH/01 (is this a record?) are fixed:
1942 (i) An empty string was always treated as zero by the numeric comparison
1943 operators. This behaviour has been restored.
1945 (ii) It is documented that the numeric comparison operators always treat
1946 their arguments as decimal numbers. This was broken in that numbers
1947 starting with 0 were being interpreted as octal.
1949 While fixing these problems I realized that there was another issue that
1950 hadn't been noticed. Values of message_size_limit (both the global option
1951 and the transport option) were treated as octal if they started with 0.
1952 The documentation was vague. These values are now always treated as
1953 decimal, and I will make that clear in the documentation.
1959 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1960 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1963 MH/01 Don't check that the operands of numeric comparison operators are
1964 integers when their expansion is in "skipping" mode (fixes bug
1965 introduced by 4.64-PH/07).
1967 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1968 child addresses, Exim now panics and dies. Previously, because the count
1969 is held in a short int, deliveries were likely to be lost. As such a
1970 large number of recipients for a single message is ridiculous
1971 (performance will be very, very poor), I have chosen to impose a limit
1972 rather than extend the field.
1978 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1979 leftover -K file (the existence of which was triggered by #402).
1980 While we were at it, introduced process PID as part of the -K
1981 filename. This should rule out race conditions when creating
1984 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1985 processing considerably. Previous code took too long for large mails,
1986 triggering a timeout which in turn triggers #401.
1988 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1989 in the DK code in transports.c. sendfile() is not really portable,
1990 hence the _LINUX specificness.
1992 TF/01 In the add_headers option to the mail command in an Exim filter,
1993 there was a bug that Exim would claim a syntax error in any
1994 header after the first one which had an odd number of characters
1997 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1998 callout verification, Exim cached a "reject" for the entire domain. This
1999 is correct for most verifications, but it is not correct for a recipient
2000 verification with use_sender or use_postmaster set, because in that case
2001 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2002 case of MAIL FROM:<> rejection from other early rejections (e.g.
2003 rejection of HELO). When verifying a recipient using a non-null MAIL
2004 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2005 Whatever the result of the callout, the value of the domain cache is
2006 left unchanged (for any other kind of callout, getting as far as trying
2007 RCPT means that the domain itself is ok).
2009 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2012 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2013 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2014 session, the daemon ignores ECONNECT errors and logs others; it now
2015 ignores EPIPE as well.
2017 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2018 (quoted-printable decoding).
2020 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2021 later the small subsequent patch to fix an introduced bug.
2023 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2025 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2027 PH/08 An error is now given if message_size_limit is specified negative.
2029 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2030 to be given (somewhat) arbitrary names.
2032 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2035 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2036 miscellaneous code fixes
2038 PH/10 Added the log_reject_target ACL modifier to specify where to log
2041 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2042 hostname. This is wrong, because it relates to the incoming message (and
2043 probably the interface on which it is arriving) and not to the outgoing
2044 callout (which could be using a different interface). This has been
2045 changed to use the value of the helo_data option from the smtp transport
2046 instead - this is what is used when a message is actually being sent. If
2047 there is no remote transport (possible with a router that sets up host
2048 addresses), $smtp_active_hostname is used.
2050 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2051 tweaks were necessary in order to get it to work (see also 21 below):
2052 (a) The code assumed that strncpy() returns a negative number on buffer
2053 overflow, which isn't the case. Replaced with Exim's string_format()
2055 (b) There were several signed/unsigned issues. I just did the minimum
2056 hacking in of casts. There is scope for a larger refactoring.
2057 (c) The code used strcasecmp() which is not a standard C function.
2058 Replaced with Exim's strcmpic() function.
2059 (d) The code set only $1; it now sets $auth1 as well.
2060 (e) A simple test gave the error "authentication client didn't specify
2061 service in request". It would seem that Dovecot has changed its
2062 interface. Fortunately there's a specification; I followed it and
2063 changed what the client sends and it appears to be working now.
2065 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2068 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2069 address A is aliased to B and C, where B exists and C does not. Without
2070 -v the output is "A verified" because verification stops after a
2071 successful redirection if more than one address is generated. However,
2072 with -v the child addresses are also verified. Exim was outputting "A
2073 failed to verify" and then showing the successful verification for C,
2074 with its parentage. It now outputs "B failed to verify", showing B's
2075 parentage before showing the successful verification of C.
2077 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2078 look up a TXT record in a specific list after matching in a combined
2081 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2082 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2083 they consult the DNS. I had assumed they would set it the way they
2084 wanted; and indeed my experiments on Linux seem to show that in some
2085 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2086 To be on the safe side, however, I have now made the interface to
2087 host_find_byname() similar to host_find_bydns(), with an argument
2088 containing the DNS resolver options. The host_find_byname() function now
2089 sets these options at its start, just as host_find_bydns() does. The smtp
2090 transport options dns_qualify_single and dns_search_parents are passed to
2091 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2092 of host_find_byname() use the default settings of RES_DEFNAMES
2093 (qualify_single) but not RES_DNSRCH (search_parents).
2095 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2096 spool_read_header() do less string testing, by means of a preliminary
2097 switch on the second character of optional "-foo" lines. (This is
2098 overdue, caused by the large number of possibilities that now exist.
2099 Originally there were few.) While I was there, I also converted the
2100 str(n)cmp tests so they don't re-test the leading "-" and the first
2101 character, in the hope this might squeeze out yet more improvement.
2103 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2104 flag allowing group syntax was set by the header_syntax check but not
2105 turned off, possible causing trouble later; (2) The flag was not being
2106 set at all for the header_verify test, causing "group"-style headers to
2107 be rejected. I have now set it in this case, and also caused header_
2108 verify to ignore an empty address taken from a group. While doing this, I
2109 came across some other cases where the code for allowing group syntax
2110 while scanning a header line wasn't quite right (mostly, not resetting
2111 the flag correctly in the right place). These bugs could have caused
2112 trouble for malformed header lines. I hope it is now all correct.
2114 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2115 with the "reply" argument non-NULL. The code, however (which originally
2116 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2117 but it didn't always do it. This confused somebody who was copying the
2118 code for some other use. I have removed all the tests.
2120 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2121 feature that was used to support insecure browsers during the U.S. crypto
2122 embargo. It requires special client support, and Exim is probably the
2123 only MTA that supported it -- and would never use it because real RSA is
2124 always available. This code has been removed, because it had the bad
2125 effect of slowing Exim down by computing (never used) parameters for the
2126 RSA_EXPORT functionality.
2128 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2129 authenticator to fail if there's a tab character in the incoming data
2130 (there should never be unless someone is messing about, as it's supposed
2131 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2132 if the connection is using TLS or if the remote IP is the same as the
2133 local IP, and the "valid-client-cert option" if a client certificate has
2136 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2137 authenticators. This can be used for authorization after authentication
2138 succeeds. (In the case of plaintext, it servers for both authentication
2141 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2142 if any retry times were supplied.
2144 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2145 connection, where there is no client IP address to check. In this
2146 situation, the verify now always succeeds.
2148 PH/25 Applied John Jetmore's -Mset patch.
2150 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2152 PH/27 In a string expansion for a processed (not raw) header when multiple
2153 headers of the same name were present, leading whitespace was being
2154 removed from all of them, but trailing whitespace was being removed only
2155 from the last one. Now trailing whitespace is removed from each header
2156 before concatenation. Completely empty headers in a concatenation (as
2157 before) are ignored.
2159 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2160 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2162 PH/29 [Removed. This was a change that I later backed out, and forgot to
2163 correct the ChangeLog entry (that I had efficiently created) before
2164 committing the later change.]
2166 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2167 address errors (4xx response to RCPT) over the same connection as other
2168 messages routed to the same hosts. Such deliveries are always "forced",
2169 so retry times are not inspected. This resulted in far too many retries
2170 for the affected addresses. The effect occurred only when there were more
2171 hosts than the hosts_max_try setting in the smtp transport when it had
2172 the 4xx errors. Those hosts that it had tried were not added to the list
2173 of hosts for which the message was waiting, so if all were tried, there
2174 was no problem. Two fixes have been applied:
2176 (i) If there are any address or message errors in an SMTP delivery, none
2177 of the hosts (tried or untried) are now added to the list of hosts
2178 for which the message is waiting, so the message should not be a
2179 candidate for sending over the same connection that was used for a
2180 successful delivery of some other message. This seems entirely
2181 reasonable: after all the message is NOT "waiting for some host".
2182 This is so "obvious" that I'm not sure why it wasn't done
2183 previously. Hope I haven't missed anything, but it can't do any
2184 harm, as the worst effect is to miss an optimization.
2186 (ii) If, despite (i), such a delivery is accidentally attempted, the
2187 routing retry time is respected, so at least it doesn't keep
2188 hammering the server.
2190 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2191 in ${readsocket because some servers need this prod.
2193 PH/32 Added some extra debug output when updating a wait-xxx database.
2195 PH/33 The hint "could be header name not terminated by colon", which has been
2196 given for certain expansion errors for a long time, was not being given
2197 for the ${if def:h_colon_omitted{... case.
2199 PH/34 The spec says: "With one important exception, whenever a domain list is
2200 being scanned, $domain contains the subject domain." There was at least
2201 one case where this was not true.
2203 PH/35 The error "getsockname() failed: connection reset by peer" was being
2204 written to the panic log as well as the main log, but it isn't really
2205 panic-worthy as it just means the connection died rather early on. I have
2206 removed the panic log writing for the ECONNRESET error when getsockname()
2209 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2210 runs only) independently of the message's sender address. This meant
2211 that, if the 4xx error was in fact related to the sender, a different
2212 message to the same recipient with a different sender could confuse
2213 things. In particualar, this can happen when sending to a greylisting
2214 server, but other circumstances could also provoke similar problems.
2215 I have changed the default so that the retry time for these errors is now
2216 based a combination of the sender and recipient addresses. This change
2217 can be overridden by setting address_retry_include_sender=false in the
2220 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2221 remote server are returned as part of bounce messages. This was not
2222 happening for LMTP over a pipe (the lmtp transport), but now it is the
2223 same for both kinds of LMTP.
2225 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2226 in header lines that were in fact CNAMEs. This is no longer the case.
2228 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2229 and queue runs started by the daemon processed all messages. This has
2230 been fixed so that -R and -S can now usefully be given with -q<time>.
2232 PH/40 Import PCRE release 6.7 (fixes some bugs).
2234 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2236 PH/42 Give an error if -q is specified more than once.
2238 PH/43 Renamed the variables $interface_address and $interface_port as
2239 $received_ip_address and $received_port, to make it clear that these
2240 values apply to message reception, and not to the outgoing interface when
2241 a message is delivered. (The old names remain recognized, of course.)
2243 PH/44 There was no timeout on the connect() call when using a Unix domain
2244 socket in the ${readsocket expansion. There now is.
2246 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2247 be meaningful with "accept".
2249 SC/01 Eximstats V1.43
2250 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2252 SC/02 Eximstats V1.44
2253 Use a glob alias rather than an array ref in the generated
2254 parser. This improves both readability and performance.
2256 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2257 Collect SpamAssassin and rejection statistics.
2258 Don't display local sender or destination tables unless
2259 there is data to show.
2260 Added average volumes into the top table text output.
2262 SC/04 Eximstats V1.46
2263 Collect data on the number of addresses (recipients)
2264 as well as the number of messages.
2266 SC/05 Eximstats V1.47
2267 Added 'Message too big' to the list of mail rejection
2268 reasons (thanks to Marco Gaiarin).
2270 SC/06 Eximstats V1.48
2271 Mainlog lines which have GMT offsets and are too short to
2272 have a flag are now skipped.
2274 SC/07 Eximstats V1.49 (Alain Williams)
2275 Added the -emptyok flag.
2277 SC/08 Eximstats V1.50
2278 Fixes for obtaining the IP address from reject messages.
2280 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2281 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2282 whitesspace changes from 4.64-PH/27
2284 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2287 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2288 are found, allow negative numbers in numeric criteria)
2290 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2292 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2293 to match changes made in 4.64-PH/43
2295 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2297 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2298 "domain" is actually an IP address, and force a failure. This locks out
2299 those revolvers/nameservers that support "A-for-A" lookups, in
2300 contravention of the specifications.
2302 PH/48 When a host name was looked up from an IP address, and the subsequent
2303 forward lookup of the name timed out, the host name was left in
2304 $sender_host_name, contrary to the specification.
2306 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2307 restricted to single-key lookups, Exim was not diagnosing an error if
2308 * or *@ was used with a query-style lookup.
2310 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2312 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2313 long ago, but noone interested enough thought of it. Let's just say that
2314 the "1.1" means that there are some new functions that weren't there at
2315 some point in the past.
2317 PH/51 Error processing for expansion failure of helo_data from an smtp
2318 transport during callout processing was broken.
2320 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2321 tested/used via the -bh/-bhc/-bs options.
2323 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2324 bug, fixed in subsequent PCRE releases).
2326 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2327 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2329 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2335 SC/01 Use a glob alias rather than an array ref in eximstats generated
2336 parser. This improves both readability and performance.
2338 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2339 Don't display local sender or destination tables in eximstats unless
2340 there is data to show.
2341 Added average volumes into the eximstats top table text output.
2343 SC/03 Collect data on the number of addresses (recipients) as well
2344 as the number of messages in eximstats.
2346 TF/01 Correct an error in the documentation for the redirect router. Exim
2347 does (usually) call initgroups() when daemonizing.
2349 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2350 with consistent privilege compared to when running as a daemon.
2352 TF/03 Note in the spec that $authenticated_id is not set for local
2353 submissions from trusted users.
2355 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2356 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2358 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2359 by adding some example configuration directives to the default
2360 configuration file. A little bit of work is required to uncomment the
2361 directives and define how usernames and passwords are checked, but
2362 there is now a framework to start from.
2364 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2365 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2366 without this. I don't know how relevant this is to other LDAP libraries.
2368 PH/02 Add the verb name to the "unknown ACL verb" error.
2370 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2372 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2374 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2375 directory not expanded when it should be if an expanded home directory
2376 was set for the address (which is overridden by the transport).
2378 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2381 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2382 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2383 because it is too late at that time, and has no effect.
2385 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2386 security issue with \' (bugzilla #107). I could not use the
2387 PQescapeStringConn() function, because it needs a PGconn value as one of
2390 PH/08 When testing addresses using -bt, indicate those final addresses that
2391 are duplicates that would not cause an additional delivery. At least one
2392 person was confused, thinking that -bt output corresponded to deliveries.
2393 (Suppressing duplicates isn't a good idea as you lose the information
2394 about possibly different redirections that led to the duplicates.)
2396 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2397 systems where poll() doesn't work, in particular OS X.
2399 PH/10 Added more information to debugging output for retry time not reached.
2401 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2402 operations in malware.c.
2404 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2407 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2408 syslog_duplication set false, log lines that would normally be written
2409 both the the main log and to the reject log were not written to syslog at
2412 PH/14 In the default configuration, change the use of "message" in ACL warn
2413 statements to "add_header".
2415 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2416 not followed by a command (e.g. "seen endif").
2418 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2419 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2422 PH/17 Added extra conditions to the default value of delay_warning_condition
2426 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2427 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2428 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2431 The Auto-Submitted: and various List- headers are standardised, whereas I
2432 don't think Precedence: ever was.
2434 PH/18 Refactored debugging code in route_finduser() to show more information,
2435 in particular, the error code if getpwnam() issues one.
2437 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2438 This is apparently needed in addition to the PH/07 change above to avoid
2439 any possible encoding problems.
2441 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2442 but not after initializing Perl.
2444 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2445 output them only if debugging. By default they are written stderr,
2446 apparently, which is not desirable.
2448 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2451 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2454 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2456 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2457 authenticated or an ident call has been made. Suppress the default
2458 values for $authenticated_id and $authenticated_sender (but permit -oMai
2459 and -oMas) when testing with -bh.
2461 PH/24 Re-jigged the order of the tests in the default configuration so that the
2462 tests for valid domains and recipients precede the DNS black list and CSA
2463 tests, on the grounds that those ones are more expensive.
2465 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2466 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2467 This bug exists in every version of Exim that I still have, right back to
2470 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2471 However, an attempt to turn on case-sensitivity in a regex key by
2472 including (?-i) didn't work because the subject string was already
2473 lowercased, and the effects were non-intuitive. It turns out that a
2474 one-line patch can be used to allow (?-i) to work as expected.
2480 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2481 other effects) broke the use of negated acl sub-conditions.
2483 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2486 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2487 "Deny" causes Exim to reject the incoming connection with a 554 error.
2488 Unfortunately, if there is a major crisis, such as a disk failure,
2489 tcp-wrappers gives "deny", whereas what one would like would be some
2490 kind of temporary error. A kludge has been added to help with this.
2491 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2492 554 error is used if errno is still zero or contains ENOENT (which occurs
2493 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2496 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2498 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2499 errors. Otherwise a message that provokes a temporary error (when other
2500 messages do not) can cause a whole host to time out.
2502 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2503 addresses were routed directly to files or pipes from a redirect router.
2504 File deliveries just didn't batch; pipe deliveries might have suffered
2507 PH/06 A failure to get a lock for a hints database would erroneously always say
2508 "Failed to get write lock", even when it was really a read lock.
2510 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2511 of 0600. This has been changed to use the value of the lockfile_mode
2512 option (which defaults to 0600).
2514 PH/08 Applied small patch from the Sieve maintainer.
2516 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2517 folder from quota calculations, a direct delivery into this folder messed
2518 up the contents of the maildirsize file. This was because the regex was
2519 used only to exclude .Trash (or whatever) when the size of the mailbox
2520 was calculated. There was no check that a delivery was happening into an
2521 excluded directory. This bug has been fixed by ignoring all quota
2522 processing for deliveries into excluded directories.
2524 PH/10 Added the maildirfolder_create_regex option to appendfile.
2530 PH/01 The code for finding all the local interface addresses on a FreeBSD
2531 system running IPv6 was broken. This may well have applied to all BSD
2532 systems, as well as to others that have similar system calls. The broken
2533 code found IPv4 interfaces correctly, but gave incorrect values for the
2534 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2535 that it would not match correctly against @[] and not recognize the IPv6
2538 PH/02 The ipliteral router was not recognizing addresses of the form user@
2539 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2541 PH/03 Added disable_ipv6.
2543 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2544 decoded form, because it is most often used to construct To: headers
2545 lines in autoreplies, and the decoded form may well be syntactically
2546 invalid. However, $reply_address has leading white space removed, and all
2547 newlines turned into spaces so that the autoreply transport does not
2550 PH/05 If group was specified without a user on a router, and no group or user
2551 was specified on a transport, the group from the router was ignored.
2553 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2554 for visible compile-time settings that can be used to change these
2555 numbers, for those that want even more. Backwards compatibility with old
2556 spool files has been maintained. However, going back to a previous Exim
2557 release will lost any variables that are in spool files.
2559 PH/07 Two small changes when running in the test harness: increase delay when
2560 passing a TCP/IP connection to a new process, in case the original
2561 process has to generate a bounce, and remove special handling of
2562 127.0.0.2 (sic), which is no longer necessary.
2564 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2565 be the same on different OS.
2567 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2570 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2571 whether --show-vars was specified or not
2573 JJ/02 exipick: Added support for new ACL variable spool format introduced
2576 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2577 syntactically invalid From: or Reply-to: line, and a filter used this to
2578 generate an autoreply, and therefore failed to obtain an address for the
2579 autoreply, Exim could try to deliver to a non-existent relative file
2580 name, causing unrelated and misleading errors. What now happens is that
2581 it logs this as a hard delivery error, but does not attempt to create a
2584 PH/11 The exinext utility has a -C option for testing purposes, but although
2585 the given file was scanned by exinext itself; it wasn't being passed on
2586 when Exim was called.
2588 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2589 an end-of-file indication when reading a command response.
2591 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2592 compiled. In many other places in Exim, IPv6 addresses are always
2593 recognized, so I have changed this. It also means that IPv4 domain
2594 literals of the form [IPV4:n.n.n.n] are now always recognized.
2596 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2597 used if the router is not running as root, for example, when verifying at
2598 ACL time, or when using -bh. The debugging output from this situation was
2599 non-existent - all you got was a failure to exec. I have made two
2602 (a) Failures to set uid/gid, the current directory, or a process leader
2603 in a subprocess such as that created by queryprogram now generate
2604 suitable debugging ouput when -d is set.
2606 (b) The queryprogram router detects when it is not running as root,
2607 outputs suitable debugging information if -d is set, and then runs
2608 the subprocess without attempting to change uid/gid.
2610 PH/15 Minor change to Makefile for building test_host (undocumented testing
2613 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2614 additional section of a DNS packet that returns MX or SRV records.
2615 Instead, it always explicitly searches for A/AAAA records. This avoids
2616 major problems that occur when a DNS server includes only records of one
2617 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2618 fixed another bug: if SRV records were looked up and the corresponding
2619 address records were *not* found in the additional section, the port
2620 values from the SRV records were lost.
2622 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2623 using the correct key (the original address) when searching the retry
2624 rules in order to find which one to use for generating the retry hint.
2626 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2627 adding the default one. Similarly, if it contains a Reply-To: header, the
2628 errors_reply_to option, if set, is not used.
2630 PH/19 When calculating a retry time, Exim used to measure the "time since
2631 failure" by looking at the "first failed" field in the retry record. Now
2632 it does not use this if it is later than than the arrival time of the
2633 message. Instead it uses the arrival time. This makes for better
2634 behaviour in cases where some deliveries succeed, thus re-setting the
2635 "first failed" field. An example is a quota failure for a huge message
2636 when small messages continue to be delivered. Without this change, the
2637 "time since failure" will always be short, possible causing more frequent
2638 delivery attempts for the huge message than are intended.
2639 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2641 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2642 $1, $2, $3) because the numerical variables can be reset during some
2643 expansion items (e.g. "match"), thereby losing the authentication data.
2645 PH/21 Make -bV show the size of off_t variables so that the test suite can
2646 decide whether to run tests for quotas > 2G.
2648 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2649 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2650 filecount value is greater than 2G or if a quota value is greater than 2G
2651 on a system where the size of off_t is not greater than 4, a panic error
2654 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2655 never match. The debug and -bh output now contains an explicit error
2656 message indicating a malformed IPv4 address or mask.
2658 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2659 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2660 PH/23 above applies.
2662 PH/25 Do not write to syslog when running in the test harness. The only
2663 occasion when this arises is a failure to open the main or panic logs
2664 (for which there is an explicit test).
2666 PH/26 Added the /no_tell option to "control=freeze".
2668 PH/27 If a host name lookup failed very early in a connection, for example, if
2669 the IP address matched host_lookup and the reverse lookup yielded a name
2670 that did not have a forward lookup, an error message of the form "no IP
2671 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2672 could be logged. Now it outputs the IP address instead of "NULL".
2674 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2675 allows the sender and the authenticated sender to be set when
2676 submitting a message from within Exim. Since child_open_exim() is
2677 documented for local_scan(), the new function should be too.
2679 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2680 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2681 results in an empty string is now treated as unset.
2683 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2685 PH/31 Added sender_verify_fail logging option.
2687 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2688 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2689 tidied the source and removed it altogether.
2691 PH/33 When a queue run was abandoned because the load average was too high, a
2692 log line was always written; now it is written only if the queue_run log
2693 selector is set. In addition, the log line for abandonment now contains
2694 information about the queue run such as the pid. This is always present
2695 in "start" and "stop" lines but was omitted from the "abandon" line.
2697 PH/34 Omit spaces between a header name and the colon in the error message that
2698 is given when verify = headers_syntax fails (if there are lots of them,
2699 the message gets confusing).
2701 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2702 names, as there are now some PTR records that contain slashes. This check
2703 is only to protect against broken name servers that fall over on strange
2704 characters, so the fact that it applies to all lookups doesn't matter.
2706 PH/36 Now that the new test suite is complete, we can remove some of the
2707 special code in Exim that was needed for the old test suite. For example,
2708 sorting DNS records because real resolvers return them in an arbitrary
2709 order. The new test suite's fake resolver always returns records in the
2712 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2713 bounces) except when queue_only is set, to avoid logging races between
2714 the different processes.
2716 PH/38 Panic-die if .include specifies a non-absolute path.
2718 PH/39 A tweak to the "H" retry rule from its user.
2720 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2721 a label. They prevented compilation on older perls.
2723 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2724 a warning to be raised on newish perls.
2726 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2727 on queue. Changes to match documented behaviour of showing count of
2728 messages matching specified criteria.
2730 PH/40 Changed the default ident timeout from 30s to 5s.
2732 PH/41 Added support for the use of login_cap features, on those BSD systems
2733 that have them, for controlling the resources used by pipe deliveries.
2735 PH/42 The content-scanning code uses fopen() to create files in which to put
2736 message data. Previously it was not paying any attention to the mode of
2737 the files. Exim runs with umask(0) because the rest of the code creates
2738 files with open(), and sets the required mode explicitly. Thus, these
2739 files were ending up world-writeable. This was not a big issue, because,
2740 being within the spool directory, they were not world-accessible. I have
2741 created a function called modefopen, which takes an additional mode
2742 argument. It sets umask(777), creates the file, chmods it to the required
2743 mode, then resets the umask. All the relevant calls to fopen() in the
2744 content scanning code have been changed to use this function.
2746 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2747 to 24 hours. This avoids potential overflow problems when processing G
2748 and H retry rules. I suspect nobody ever tinkers with this value.
2750 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2752 PH/45 When the plaintext authenticator is running as a client, the server's
2753 challenges are checked to ensure they are valid base64 strings. By
2754 default, the authentication attempt is cancelled if an invalid string is
2755 received. Setting client_ignore_invalid_base64 true ignores these errors.
2756 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2757 they are received. Thus, the responses can be made to depend on the
2758 challenges. If an invalid string is ignored, an empty string is placed in
2761 PH/46 Messages that are created by the autoreply transport now contains a
2762 References: header, in accordance with RFCs 2822 and 3834.
2764 PH/47 Added authenticated_sender_force to the smtp transport.
2766 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2768 PH/49 Installed latest patch from the Sieve maintainer.
2770 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2771 was also set, the appendfile transport was unnecessarily scanning a
2772 directory of message files (e.g. for maildir delivery) to find the count
2773 of files (along with the size), even though it did not need this
2774 information. It now does the scan only if it needs to find either the
2775 size of the count of files.
2777 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2779 PH/52 Two bugs concerned with error handling when the smtp transport is
2782 (i) Exim was not creating retry information for temporary errors given
2783 for individual recipients after the DATA command when the smtp transport
2784 was used in LMTP mode. This meant that they could be retried too
2785 frequently, and not timed out correctly.
2787 (ii) Exim was setting the flag that allows error details to be returned
2788 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2789 recipients that were returned after the DATA command.
2791 PH/53 This is related to PH/52, but is more general: for any failing address,
2792 when detailed error information was permitted to be returned to the
2793 sender, but the error was temporary, then after the final timeout, only
2794 "retry timeout exceeded" was returned. Now it returns the full error as
2795 well as "retry timeout exceeded".
2797 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2798 do this, and (what is worse) MTAs that accept it.
2800 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2801 will now be deprecated.
2803 PH/56 New os.c-cygwin from the Cygwin maintainer.
2805 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2806 formats (previously only available in exim formats via -bpr, -bpru,
2807 and -bpra. Now also available in native and exiqgrep formats)
2809 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2810 with very large, slow to parse queues
2812 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2814 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2816 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2817 responses to authentication challenges, though it was showing the
2818 challenges; (ii) I've removed the CR characters from the debug output for
2821 PH/58 Allow for the insertion of a newline as well as a space when a string
2822 is turned into more than one encoded-word during RFC 2047 encoding. The
2823 Sieve code now uses this.
2825 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2826 data_4xx, lost_connection, tls_required.
2828 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2829 message was being sent as an SMTP response.
2831 PH/61 Add -l and -k options to exicyclog.
2833 PH/62 When verifying, if an address was redirected to one new address, so that
2834 verification continued, and the new address failed or deferred after
2835 having set something in $address_data, the value of $address_data was not
2836 passed back to the ACL. This was different to the case when no
2837 redirection occurred. The value is now passed back in both cases.
2839 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2840 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2841 use login_cap.h, so on its own it isn't the distinguishing feature. The
2842 new name refers directly to the setclassresources() function.
2844 PH/65 Added configuration files for NetBSD3.
2846 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2848 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2849 is preferred over IPv4.
2851 PH/68 The bounce_return_message and bounce_return_body options were not being
2852 honoured for bounces generated during the reception of non-SMTP messages.
2853 In particular, this applied to messages rejected by the ACL. This bug has
2854 been fixed. However, if bounce_return_message is true and bounce_return_
2855 body is false, the headers that are returned for a non-SMTP message
2856 include only those that have been read before the error was detected.
2857 (In the case of an ACL rejection, they have all been read.)
2859 PH/69 The HTML version of the specification is now built in a directory called
2860 spec_html instead of spec.html, because the latter looks like a path with
2861 a MIME-type, and this confuses some software.
2863 PH/70 Catch two compiler warnings in sieve.c.
2865 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2866 function verify_get_ident() calls ip_connect() to connect a socket, but
2867 if the "connect()" function timed out, ip_connect() used to close the
2868 socket. However, verify_get_ident() also closes the socket later, and in
2869 between Exim writes to the log, which may get opened at this point. When
2870 the socket was closed in ip_connect(), the log could get the same file
2871 descriptor number as the socket. This naturally causes chaos. The fix is
2872 not to close the socket in ip_connect(); the socket should be closed by
2873 the function that creates it. There was only one place in the code where
2874 this was missing, in the iplookup router, which I don't think anybody now
2875 uses, but I've fixed it anyway.
2877 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2878 well as to direct DNS lookups. Otherwise the handling of names in host
2879 lists is inconsistent and therefore confusing.
2885 PH/01 Two changes to the default runtime configuration:
2887 (1) Move the checks for relay_from_hosts and authenticated clients from
2888 after to before the (commented out) DNS black list checks.
2890 (2) Add control=submission to the relay_from_hosts and authenticated
2891 clients checks, on the grounds that messages accepted by these
2892 statements are most likely to be submissions.
2894 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2896 (1) Generate an error if the third argument for the ${prvs expansion is
2899 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2902 (3) Reset the variables that are obtained from the first argument of
2903 ${prvscheck and used in the second argument before leaving the code,
2904 because their memory is reclaimed, so using them afterwards may do
2907 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2908 one (it's much easier than Tom thought :-).
2910 (5) Because of (4), we can now allow for the use of $prvscheck_result
2911 inside the third argument.
2913 PH/03 For some reason, the default setting of PATH when running a command from
2914 a pipe transport was just "/usr/bin". I have changed it to
2917 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2918 anything to be listed in the output from -bV.
2920 PH/05 When a filter generated an autoreply, the entire To: header line was
2921 quoted in the delivery log line, like this:
2923 => >A.N.Other <ano@some.domain> <original@ddress> ...
2925 This has been changed so that it extracts the operative address. There
2926 may be more than one such address. If so, they are comma-separated, like
2929 => >ano@some.domain,ona@other.domain <original@ddress> ...
2931 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2932 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2933 not being looked up in the rDNS to get a host name, Exim was showing the
2934 IP address twice in Received: lines, even though the IP addresses were
2935 identical. For example:
2937 Received: from [1.2.3.4] (helo=[1.2.3.4])
2939 However, if the real host name was known, it was omitting the HELO data
2940 if it matched the actual IP address. This has been tidied up so that it
2941 doesn't show the same IP address twice.
2943 PH/07 When both +timestamp and +memory debugging was on, the value given by
2944 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2945 called by the debug printing, thereby overwriting the timestamp buffer.
2946 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2948 PH/08 When the original message was included in an autoreply transport, it
2949 always said "this is a copy of the message, including all the headers",
2950 even if body_only or headers_only was set. It now gives an appropriate
2953 PH/09 Applied a patch from the Sieve maintainer which:
2955 o fixes some comments
2956 o adds the (disabled) notify extension core
2957 o adds some debug output for the result of if/elsif tests
2958 o points to the current vacation draft in the documentation
2959 and documents the missing references header update
2963 o fixes a bug in processing the envelope test (when testing
2964 multiple envelope elements, the last element determinted the
2967 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2968 Electronic Mail") by including:
2970 Auto-submitted: auto-generated
2972 in the messages that it generates (bounce messages and others, such as
2973 warnings). In the case of bounce messages for non-SMTP mesages, there was
2974 also a typo: it was using "Auto_submitted" (underscore instead of
2975 hyphen). Since every message generated by Exim is necessarily in response
2976 to another message, thes have all been changed to:
2978 Auto-Submitted: auto-replied
2980 in accordance with these statements in the RFC:
2982 The auto-replied keyword:
2984 - SHOULD be used on messages sent in direct response to another
2985 message by an automatic process,
2987 - MUST NOT be used on manually-generated messages,
2989 - MAY be used on Delivery Status Notifications (DSNs) and Message
2990 Disposition Notifications (MDNs),
2992 - MUST NOT be used on messages generated by automatic or periodic
2993 processes, except for messages which are automatic responses to
2996 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2997 to the default Received: header definition.
2999 PH/12 Added log selector acl_warn_skipped (default on).
3001 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3002 variables because (a) they are in the wrong storage pool and (b) even if
3003 they were copied, it wouldn't work properly because of the caching.
3005 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3006 checking when decoding. Apparently there are clients that generate
3007 overlong encoded strings. Why am I not surprised?
3009 PH/15 If the first argument of "${if match_address" was not empty, but did not
3010 contain an "@" character, Exim crashed. Now it writes a panic log message
3011 and treats the condition as false.
3013 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3015 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3016 extension "envelope-auth". The code is finished and in agreement with
3017 other implementations, but there is no documentation so far and in fact,
3018 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3019 not changing the active code.
3021 Print executed "if" and "elsif" statements when debugging is used. This
3022 helps a great deal to understand what a filter does.
3024 Document more things not specified clearly in RFC3028. I had all this
3025 sorted out, when out of a sudden new issues came to my mind. Oops."
3027 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3030 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3031 canonical form (as documented). However, after a host name lookup from
3032 the IP address, check_host() was doing a simple string comparison with
3033 addresses acquired from the DNS when checking that the found name did
3034 have the original IP as one of its addresses. Since any found IPv6
3035 addresses are likely to be in abbreviated form, the comparison could
3036 fail. Luckily, there already exists a function for doing the comparison
3037 by converting both addresses to binary, so now that is used instead of
3038 the text comparison.
3040 PH/20 There was another similar case to PH/19, when a complete host name was
3041 given in a host list; looking up its IP address could give an abbreviated
3042 form, whereas the current host's name might or might not be abbreviated.
3043 The same fix has been applied.
3049 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3050 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3053 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3054 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3056 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3058 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3059 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3060 Also, for "Auto-Submitted", treat anything other than "no" as
3061 non-personal, in accordance with RFC 3834. (Previously it treated
3062 anything starting "auto-" as non-personal.)
3064 TF/01 The control=submission/name=... option had a problem with syntax
3065 errors if the name included a slash character. The /name= option
3066 now slurps the rest of the string, so it can include any characters
3067 but it must come last in the list of options (after /sender_retain
3070 PH/05 Some modifications to the interface to the fake nameserver for the new
3078 TK/01 Added the "success_on_redirect" address verification option. See
3079 NewStuff for rationale and an example.
3081 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3083 PH/02 Patch to exigrep to allow it to work on syslog lines.
3085 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3086 fread() to skip over the body file's header line, because in Cygwin the
3087 header line is locked and is inaccessible.
3089 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3090 co-exist for some time) to make it clear that it is the Exim ID that is
3091 referenced, not the Message-ID: header line.
3093 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3094 string_format() function, because snprintf() does not exist on all
3097 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3098 ${stat: expansion item.
3100 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3101 protocol synchronization error", to keep the pedants happy.
3103 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3104 well as for IRIX systems, when gcc is being used. See the host.c source
3107 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3109 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3112 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3113 transport and to the smtp transport in LMTP mode.
3115 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3117 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3118 run a filter in a subprocess. This could lead to confusion in subsequent
3119 lookups in the parent process. There should also be a search_tidyup() at
3120 the end of the subprocess.
3122 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3123 only if the host matched helo_try_verify_hosts, which caused the
3124 verification to occur when the EHLO/HELO command was issued. The ACL just
3125 tested the remembered result. Now, if a previous verification attempt has
3126 not happened, "verify = helo" does it there and then.
3128 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3130 TK/03 Fix log output including CR from clamd.
3132 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3133 exist provoked a memory error which could cause a segfault.
3135 PH/15 Installed PCRE 6.2
3137 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3139 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3140 of the problem. Specifically, suggested +O2 rather than +O1 for the
3143 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3145 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3146 record, the port was not correctly logged when the outgoing_port log
3147 selector was set (it logged the transort's default port).
3149 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3150 fallback_hosts, and "hosts" in the smtp transport.
3152 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3153 host errors such as "Connection refused".
3155 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3156 authentication with radiusclient 0.4.9:
3158 - Error returned from rc_read_config was caught wrongly
3159 - Username/password not passed on to radius server due to wrong length.
3161 The presumption is that some radiusclient API changes for 4.51/PH/17
3162 were not taken care of correctly. The code is still untested by me (my
3163 Linux distribution still has 0.3.2 of radiusclient), but it was
3164 contributed by a Radius user.
3166 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3167 expanding the "port" option of the smtp transport.
3169 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3170 while reading a MIME header. Thanks to Tom Hughes for a patch.
3172 PH/24 Include config.h inside local_scan.h so that configuration settings are
3175 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3176 that in an ACL for RCPT (for example), you can examine exactly what was
3179 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3180 commands, but it was not correctly comparing the address with the actual
3181 client host address. Thus, it would show the EHLO address in Received:
3182 header lines when this was not necessary.
3184 PH/27 Added the % operator to ${eval:}.
3186 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3187 it should be ignoring failures (because with -C, for example, it has lost
3188 privilege). It wasn't ignoring creation failures other than "already
3191 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3194 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3195 because an input line was too long, either on its own, or by virtue of
3196 too many continuations, the temporary file was not being removed, and the
3197 return code was incorrect.
3199 PH/31 Missing "BOOL" in function definition in filtertest.c.
3201 PH/32 Applied Sieve patches from the maintainer.
3203 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3205 PH/33 Added "verify = not_blind".
3207 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3208 Local/Makefile (with some defaults set). These are used in built scripts
3209 such as exicyclog, but they have never been used in the exim_install
3210 script (though there are many overriding facilities there). I have
3211 arranged that the exim_install script now takes note of these two
3214 PH/35 Installed configuration files for Dragonfly.
3216 PH/36 When a locally submitted message by a trusted user did not contain a
3217 From: header, and the sender address was obtained from -f or from an SMTP
3218 MAIL command, and the trusted user did not use -F to supply a sender
3219 name, $originator_name was incorrectly used when constructing a From:
3220 header. Furthermore, $originator_name was used for submission mode
3221 messages from external hosts without From: headers in a similar way,
3222 which is clearly wrong.
3224 PH/37 Added control=suppress_local_fixups.
3226 PH/38 When log_selector = +received_sender was set, and the addition of the
3227 sender made the log line's construction buffer exactly full, or one byte
3228 less than full, an overflow happened when the terminating "\n" was
3231 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3232 when the result of a list match is failure because a DNS lookup failed.
3234 PH/40 RM_COMMAND is now used in the building process.
3236 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3237 the "build-* directories that it finds.
3239 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3240 address in a domain literal was a prefix of an interface address.
3242 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3243 when verifying a sender address, unless rewrite_headers is false.
3245 PH/44 Wrote a long comment about why errors_to addresses are verified as
3246 recipients, not senders.
3248 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3249 the ratelimit ACL was added.
3251 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3253 PH/46 Added extra information about PostgreSQL errors to the error string.
3255 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3256 avoiding the need to install special zones in a real server. This is
3257 backwards compatible; if it can't find the fake resolver, it drops back.
3258 Thus, both old and new test suites can be run.
3260 TF/02 Added util/ratelimit.pl
3262 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3263 clock is set back in time.
3265 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3266 Candler <B.Candler@pobox.com>.
3268 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3269 OK for addresses that are the result of redirections.
3271 PH/48 A number of further additions for the benefit of the new test suite,
3272 including a fake gethostbyname() that interfaces to the fake DNS resolver
3275 TF/06 The fix for widen_domains has also been applied to qualify_single and
3276 search_parents which are the other dnslookup options that can cause
3279 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3282 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3284 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3285 service" if it encountered an MX record with an empty target hostname.
3286 The message is now "an MX or SRV record indicated no SMTP service".
3288 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3289 if the DNS of the sending site is misconfigured. This is quite a
3290 common situation. This change restores the behaviour of treating a
3291 helo verification defer as a failure.
3293 PH/51 If self=fail was set on a router, the bounce message did not include the
3294 actual error message.
3300 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3302 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3303 command itself ended in error, the underlying message about the transport
3304 filter timeout was being overwritten with the pipe command error. Now the
3305 underlying error message should be appended to the second error message.
3307 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3309 PH/02 Reduce the amount of output that "make" produces by default. Full output
3310 can still be requested.
3312 PH/03 The warning log line about a condition test deferring for a "warn" verb
3313 was being output only once per connection, rather than after each
3314 occurrence (because it was using the same function as for successful
3315 "warn" verbs). This seems wrong, so I have changed it.
3317 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3318 it should not have, which might have caused a crash in the right
3319 circumstances, but probably never did.
3321 PH/04 Installed a modified version of Tony Finch's patch to make submission
3322 mode fix the return path as well as the Sender: header line, and to
3323 add a /name= option so that you can make the user's friendly name appear
3326 TF/03 Added the control = fakedefer ACL modifier.
3328 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3329 Mark Lowes for thorough testing.
3331 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3333 TK/03 Merged latest SRS patch from Miles Wilton.
3335 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3336 with the definition in sysexits.h (which is #included earlier).
3337 Fortunately, Exim does not actually use EX_OK. The code used to try to
3338 preserve the sysexits.h value, by assumimg that macro definitions were
3339 scanned for macro replacements. I have been disabused of this notion,
3340 so now the code just undefines EX_OK before #including unistd.h.
3342 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3343 in the smtp transport. When a block could not be written in a single
3344 write() function, the timeout was being re-applied to each part-write.
3345 This seems wrong - if the receiver was accepting one byte at a time it
3346 would take for ever. The timeout is now adjusted when this happens. It
3347 doesn't have to be particularly precise.
3349 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3350 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3352 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3353 without a domain if the check to <postmaster@domain> fails.
3355 SC/01 Eximstats: added -xls and the ability to specify output files
3356 (patch written by Frank Heydlauf).
3358 SC/02 Eximstats: use FileHandles for outputing results.
3360 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3362 SC/04 Eximstats: fixed display of large numbers with -nvr option
3364 SC/05 Eximstats: fixed merging of reports with empty tables.
3366 SC/06 Eximstats: added the -include_original_destination flag
3368 SC/07 Eximstats: removed tabs and trailing whitespace.
3370 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3372 TK/06 MBOX spool code: Add real "From " MBOX separator line
3373 so the .eml file is really in mbox format (even though
3374 most programs do not really care). Patch from Alex Miller.
3376 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3377 The latter is generated from $received_to and is only set if the
3378 message has one envelope recipient. SA can use these headers,
3379 obviously out-of-the-box. Patch from Alex Miller.
3381 PH/08 The ${def test on a variable was returning false if the variable's
3382 value was "0", contrary to what the specification has always said!
3383 The result should be true unless the variable is empty.
3385 PH/09 The syntax error of a character other than { following "${if
3386 def:variable_name" (after optional whitespace) was not being diagnosed.
3387 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3388 accidental colon was present, for example, could give incorrect results.
3390 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3391 result is used (not including appendfile, where other changes are about
3394 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3395 This involved changing a lot of size variables from int to off_t. It
3396 should work with maildirs and everything.
3398 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3399 spamd dying while we are connected to it.
3401 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3404 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3405 function for BDB 4.3.
3407 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3409 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3410 the sources for PCRE have been re-arranged and more files are now
3413 PH/15 The code I had for printing potentially long long variables in PH/11
3414 above was not the best (it lost precision). The length of off_t variables
3415 is now inspected at build time, and an appropriate printing format (%ld
3416 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3417 to be "long long int" or "long int". This is needed for the internal
3418 formatting function string_vformat().
3420 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3421 the configuration file to be ":syslog", then the script "guesses" where
3422 the logs files are, rather than using the compiled in default. In our
3423 case the guess is not the same as the compiled default, so the script
3424 suddenly stopped working when I started to use syslog. The patch checks
3425 to see if log_file_path is "". If so, it attempts to read it from exim
3426 with no configuration file to get the compiled in version, before it
3427 falls back to the previous guessing code."
3429 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3430 implementing BATV in an Exim configuration. See NewStuff for the gory
3433 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3434 Makefile that are specific to HP-UX.
3436 PH/18 If the "use_postmaster" option was set for a recipient callout together
3437 with the "random" option, the postmaster address was used as the MAIL
3438 FROM address for the random test, but not for the subsequent recipient
3439 test. It is now used for both.
3441 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3442 patch removes a few documentation additions to RFC 3028, because the
3443 latest draft now contains them. It adds the new en;ascii-case comparator
3444 and a new error check for 8bit text in MIME parts. Comparator and
3445 require names are now matched exactly. I enabled the subaddress
3446 extension, but it is not well tested yet (read: it works for me)."
3448 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3449 rework some of the code of TK/09 above to avoid the hardwired use of
3450 "%lld" and "long long". Replaced the call to snprintf() with a call to
3453 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3454 records point to non-existent hosts", "retry timeout exceeded", and
3455 "retry time not reached for any host after a long failure period".
3457 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3458 experimental DomainKeys support:
3460 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3461 (2) On an error such as an illegally used "control", the wrong name for
3462 the control was given.
3464 These problems did NOT occur unless DomainKeys support was compiled.
3466 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3468 PH/24 Added ${if match_ip condition.
3470 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3471 databases so that it will be absolutely obvious if a crash occurs in the
3472 DB library. This is a regular occurrence (often caused by mis-matched
3475 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3476 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3477 picked up on a user's system that detects such things. There doesn't seem
3478 to be a gcc warning option for this - only an attribute that has to be
3479 put on the function's prototype. It seems that in Fedora Core 4 they have
3480 set this on a number of new functions. No doubt there will be more in due
3483 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3484 specify a transport. However, if an address that was verified by such a
3485 router was the subject of a callout, Exim crashed because it tried to
3486 read the rcpt_include_affixes from the non-existent transport. Now it
3487 just assumes that the setting of that option is false. This bug was
3488 introduced by 4.51/PH/31.
3490 PH/28 Changed -d+all to exclude +memory, because that information is very
3491 rarely of interest, but it makes the output a lot bigger. People tend to
3492 do -d+all out of habit.
3494 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3495 code in os-type was giving problems when libc.so lives in lib64, like on
3498 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3499 aren't the modern standard, and it seems that some systems' include files
3500 don't always have them. Exim was already checking for some of the newer
3501 ones like T_AAAA, and defining it itself. I've added checks for all the
3502 record types that Exim uses.
3504 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3505 not automatically generating a new one, as it is supposed to. This
3506 prevented TLS from working. If the file did exist, but contained invalid
3507 data, a new version was generated, as expected. It was only the case of a
3508 non-existent file that was broken.
3510 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3511 with a change in libdomainkeys > 0.64.
3513 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3514 from DNS. If the selector record carries the flag, it now has
3515 precedence over the domain-wide flag.
3517 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3519 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3520 the use of a port name that isn't defined in /etc/services) occurred, the
3521 message was deferred as in a normal delivery, and thus remained on the
3522 spool, instead of being failed because of the mua_wrapper setting. This
3523 is now fixed, and I tidied up some of the mua_wrapper messages at the
3526 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3527 the messages in a hash of arrays rather than using individual hashes.
3528 This is a bit cleaner and results in dramatic memory savings, albeit
3529 at a slight CPU cost.
3531 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3532 as requested by Marc Sherman.
3534 SC/10 Eximstats: added histograms for user specified patterns as requested
3537 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3539 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3540 fopen() in the content-scanning modules that did not already have it.
3546 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3547 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3549 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3551 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3553 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3554 to test Sieve filters that use "vacation".
3556 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3557 that changes the way the GnuTLS parameters are stored in the cache file.
3558 The new format can be generated externally. For backward compatibility,
3559 if the data in the cache doesn't make sense, Exim assumes it has read an
3560 old-format file, and it generates new data and writes a new file. This
3561 means that you can't go back to an older release without removing the
3564 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3565 work if there are any delivery delays because "one_time" forces the
3566 parent to be marked "delivered", so its unseen clone is never tried
3567 again. For this reason, Exim now forbids the simultaneous setting of
3570 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3571 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3572 entry if you want to know the details. The fix, however, neglected to
3573 consider the case when local delivery batching is involved. The test for
3574 "previously delivered" was not happening when checking to see if an
3575 address could be batched with a previous (undelivered) one; under
3576 certain circumstances this could lead to multiple deliveries to the same
3579 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3580 in its include files, and this causes problems building Exim.
3582 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3583 header_syntax) but Exim was just ignoring anything given after a slash.
3584 In particular, this caused confusion with an attempt to use "verify =
3585 reverse_host_lookup/defer_ok". An error is now given when options are
3586 supplied for verify items that do not have them. (Maybe reverse_host_
3587 lookup should have a defer_ok option, but that's a different point.)
3589 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3590 defined by RFC 821) to 2048, because there were problems with some AUTH
3591 commands, and RFC 1869 says the size should be increased for extended
3592 SMTP commands that take arguments.
3594 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3597 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3598 "unknown" error; now it says that the functionality isn't in the binary.
3600 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3601 an address' error message when a string expansion fails (syntax or
3602 whatever). Otherwise the password may appear in the log. Following change
3603 PH/42 below, there is no longer a chance of it appearing in a bounce
3606 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3608 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3609 of its addresses was ever tried. (Bugzilla bug #2.)
3611 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3612 the result incorrectly in the debug output. (It correctly added a newline
3613 to what was transported.)
3615 TF/01 Added $received_time.
3617 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3618 commented out examples of how to interface to a virus scanner and to
3619 SpamAssassin. Also added commented examples of av_scanner and
3620 spamd_address settings.
3622 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3623 and controls are allowed in which ACLs. There were a couple of minor
3624 errors. Some of the entries in the conditions table (which is a table of
3625 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3626 negation of where the condition IS allowed.
3628 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3630 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3631 header file does not have a version number, so I've had to invent a new
3632 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3633 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3634 radiusclient), but it was contributed by a Radius user.
3636 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3637 files or directories, for OpenSSL.
3639 PH/19 When an Exim process that is running as root has to create an Exim log
3640 file, it does so in a subprocess that runs as exim:exim so as to get the
3641 ownership right at creation (otherwise, other Exim processes might see
3642 the file with the wrong ownership). There was no test for failure of this
3643 fork() call, which would lead to the process getting stuck as it waited
3644 for a non-existent subprocess. Forks do occasionally fail when resources
3645 run out. I reviewed all the other calls to fork(); they all seem to check
3648 PH/20 When checking for unexpected SMTP input at connect time (before writing
3649 the banner), Exim was not dealing correctly with a non-positive return
3650 from the read() function. If the client had disconnected by this time,
3651 the result was a log entry for a synchronization error with an empty
3652 string after "input=" when read() returned zero. If read() returned -1
3653 (an event I could not check), uninitialized data bytes were printed.
3654 There were reports of junk text (parts of files, etc) appearing after
3657 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3659 PH/22 Added support for macro redefinition, and (re)definition in between
3660 driver and ACL definitions.
3662 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3663 forgetting to use the resulting value; it was using the unexpanded value.
3665 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3666 hadn't been configured. The fix is from Juergen Kreileder, who
3667 understands it better than I do:
3669 "Here's what I see happening with three configured cyrus_sasl
3670 authenticators configured (plain, login, cram-md5):
3672 On startup auth_cyrus_sasl_init() gets called for each of these.
3673 This means three calls to sasl_listmech() without a specified mech_list.
3674 => SASL tests which mechs of all available mechs actually work
3675 => three warnings about OTP not working
3676 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3678 With the patch, sasl_listmech() also gets called three times. But now
3679 SASL's mech_list option is set to the server_mech specified in the the
3680 authenticator. Or in other words, the answer from sasl_listmech()
3681 gets limited to just the mech you're testing for (which is different
3683 => the return list contains just 'plain' or 'login', 'cram-md5' or
3684 nothing depending on the value of ob->server_mech.
3686 I've just tested the patch: Authentication still works fine,
3687 unavailable mechs specified in the exim configuration are still
3688 caught, and the auth.log warnings about OTP are gone."
3690 PH/25 When debugging is enabled, the contents of the command line are added
3691 to the debugging output, even when log_selector=+arguments is not
3694 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3695 answer is "GNU", and only if the return is "GNU/something" is the answer
3698 PH/27 $acl_verify_message is now set immediately after the failure of a
3699 verification in an ACL, and so is available in subsequent modifiers. In
3700 particular, the message can be preserved by coding like this:
3702 warn !verify = sender
3703 set acl_m0 = $acl_verify_message
3705 Previously, $acl_verify_message was set only while expanding "message"
3706 and "log_message" when a very denied access.
3708 PH/28 Modified OS/os.c-Linux with
3710 -#ifndef OS_LOAD_AVERAGE
3711 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3713 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3714 nomenclature these days.)
3716 PH/29 Installed patch from the Sieve maintainer that adds the options
3717 sieve_useraddress and sieve_subaddress to the redirect router.
3719 PH/30 In these circumstances:
3720 . Two addresses routed to the same list of hosts;
3721 . First host does not offer TLS;
3722 . First host accepts first address;
3723 . First host gives temporary error to second address;
3724 . Second host offers TLS and a TLS session is established;
3725 . Second host accepts second address.
3726 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3727 and peerdn, if requested) that were in fact used only for the second
3730 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3731 attention to any local part prefix or suffix that was matched by the
3732 router that accepted the address. It now behaves in the same way as it
3733 does for delivery: the affixes are removed from the local part unless
3734 rcpt_include_affixes is set on the transport.
3736 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3737 timeout during the DATA phase of an incoming message.
3739 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3740 applied a patch sanctioned by the Sieve maintainer.
3742 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3743 the uid or gid is negative. A case of a negative gid caused this to be
3744 noticed. The fix allows for either to be negative.
3746 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3747 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3750 PH/36 The change PH/12 above was broken. Fixed it.
3752 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3753 the grounds that routing the same address twice would always produce the
3754 same answer. This might have been true once, but it is certainly no
3755 longer true now. Routing a child address may depend on the previous
3756 routing that produced that child. Some complicated redirection strategies
3757 went wrong when messages had multiple recipients, and made Exim's
3758 behaviour dependent on the order in which the addresses were given.
3760 I have moved the duplicate checking until after the routing is complete.
3761 Exim scans the addresses that are assigned to local and remote
3762 transports, and removes any duplicates. This means that more work will be
3763 done, as duplicates will always all be routed, but duplicates are
3764 presumably rare, so I don't expect this is of any significance.
3766 For deliveries to pipes, files, and autoreplies, the duplicate checking
3767 still happens during the routing process, since they are not going to be
3770 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3771 It corrects a timeout issue with spamd. This is Ian's comment: "The
3772 background is that sometimes spamd either never reads data from a
3773 connection it has accepted, or it never writes response data. The exiscan
3774 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3775 blindly assumes that writes won't block so it may never time out."
3777 PH/39 Allow G after quota size as well as K and M.
3779 PH/40 The value set for $authenticated_id in an authenticator may not contain
3780 binary zeroes or newlines because the value is written to log lines and
3781 to spool files. There was no check on this. Now the value is run through
3782 the string_printing() function so that such characters are converted to
3783 printable escape sequences.
3785 PH/41 $message_linecount is a new variable that contains the total number of
3786 lines in the message. Compare $body_linecount, which is the count for the
3789 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3790 bounce and delay warning messages, except in certain special cases, which
3793 (a) An SMTP error message from a remote host;
3794 (b) A message specified in a :fail: redirection;
3795 (c) A message specified in a "fail" command in a system filter;
3796 (d) A message specified in a FAIL return from the queryprogram router;
3797 (e) A message specified by the cannot_route_message router option.
3799 In these cases only, Exim does include the error details in bounce and
3800 warning messages. There are also a few cases where bland messages such
3801 as "unrouteable address" or "local delivery error" are given.
3803 PH/43 $value is now also set for the "else" part of a ${run expansion.
3805 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3806 being worked on, but at least Exim now implements the latest version to
3809 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3810 process to complete was treated as a delivery failure, a timeout while
3811 writing the message to the pipe was logged, but erroneously treated as a
3812 successful delivery. Such timeouts include transport filter timeouts. For
3813 consistency with the overall process timeout, these timeouts are now
3814 treated as errors, giving rise to delivery failures by default. However,
3815 there is now a new Boolean option for the pipe transport called
3816 timeout_defer, which, if set TRUE, converts the failures into defers for
3817 both kinds of timeout. A transport filter timeout is now identified in
3820 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3821 systems where "make" and "gmake" are different, calling "gmake" at top
3822 level broke things. I've arranged for the value of $(MAKE) to be passed
3823 from the Makefile to this script so that it can call the same version of
3827 A note about Exim versions 4.44 and 4.50
3828 ----------------------------------------
3830 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3831 changes of various kinds. As a consequence, a big documentation update was
3832 needed. This delayed the release for rather longer than seemed good, especially
3833 in the light of a couple of (minor) security issues. Therefore, the changes
3834 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3835 release. So 4.44 and 4.50 are in effect two different branches that both start
3838 I have left the 4.50 change log unchanged; it contains all the changes since
3839 4.43. The change log for 4.44 is below; many of its items are identical to
3840 those for 4.50. This seems to be the most sensible way to preserve the
3841 historical information.
3847 1. Minor wording change to the doc/README.SIEVE file.
3849 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3850 computation of the current number of files was incorrect.
3852 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3853 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3856 4. Give more explanation in the error message when the command for a transport
3857 filter fails to execute.
3859 5. There are several places where Exim runs a non-Exim command in a
3860 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3861 was being done only for the command run by the queryprogram router. It is
3862 now done for all such subprocesses. The other cases are: ${run, transport
3863 filters, and the commands run by the lmtp and pipe transports.
3865 6. Added CONFIGURE_GROUP build-time option.
3867 7. Some older OS have a limit of 256 on the maximum number of file
3868 descriptors. Exim was using setrlimit() to set 1000 as a large value
3869 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3870 systems. I've change it so that if it can't get 1000, it tries for 256.
3872 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3873 was an oversight, and furthermore, ever since the addition of extra
3874 controls (e.g. 4.43/32), the checks on when to allow different forms of
3875 "control" were broken. There should now be diagnostics for all cases when a
3876 control that does not make sense is encountered.
3878 9. Added the /retain_sender option to "control=submission".
3880 10. $recipients is now available in the predata ACL (oversight).
3882 11. Tidy the search cache before the fork to do a delivery from a message
3883 received from the command line. Otherwise the child will trigger a lookup
3884 failure and thereby defer the delivery if it tries to use (for example) a
3885 cached ldap connection that the parent has called unbind on.
3887 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3888 of $address_data from the recipient verification was clobbered by the
3889 sender verification.
3891 13. The value of address_data from a sender verification is now available in
3892 $sender_address_data in subsequent conditions in the ACL statement.
3894 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3896 15. Added a new option "connect=<time>" to callout options, to set a different
3899 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3900 was its contents. (It was OK if the option was not defined at all.)
3902 17. A "Completed" log line is now written for messages that are removed from
3903 the spool by the -Mrm option.
3905 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3906 information about exactly what failed.
3908 19. Added -dd to debug only the daemon process.
3910 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3911 handles timeouts, both on the server side and network timeouts. Renamed the
3912 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3914 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3915 It is now set to "smtps".
3917 22. $host_address is now set to the target address during the checking of
3918 ignore_target_hosts.
3920 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3921 being passed; this would have caused $sender_host_name to have been used if
3922 matching the list had actually called for a host name (not very likely,
3923 since this list is usually IP addresses). A host name is now passed as
3926 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3927 code that specifies a non-restarting handler (typically sigaction() in
3928 modern systems) in an attempt to fix a rare and obscure crash bug.
3930 25. Narrowed the window for a race in the daemon that could cause it to ignore
3931 SIGCHLD signals. This is not a major problem, because they are used only to
3932 wake it up if nothing else does.
3934 26. A malformed maildirsize file could cause Exim to calculate negative values
3935 for the mailbox size or file count. Odd effects could occur as a result.
3936 The maildirsize information is now recalculated if the size or filecount
3939 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3940 support for a long time. Removed HAVE_SYS_VFS_H.
3942 28. Installed the latest version of exipick from John Jetmore.
3944 29. In an address list, if the pattern was not a regular expression, an empty
3945 subject address (from a bounce message) matched only if the pattern was an
3946 empty string. Non-empty patterns were not even tested. This was the wrong
3947 because it is perfectly reasonable to use an empty address as part of a
3948 database query. An empty address is now tested by patterns that are
3949 lookups. However, all the other forms of pattern expect the subject to
3950 contain a local part and a domain, and therefore, for them, an empty
3951 address still always fails if the pattern is not itself empty.
3953 30. Exim went into a mad DNS loop when attempting to do a callout where the
3954 host was specified on an smtp transport, and looking it up yielded more
3955 than one IP address.
3957 31. Re-factored the code for checking spool and log partition space into a
3958 function that finds that data and another that does the check. The former
3959 is then used to implement four new variables: $spool_space, $log_space,
3960 $spool_inodes, and $log_inodes.
3962 32. The RFC2047 encoding function was originally intended for short strings
3963 such as real names; it was not keeping to the 75-character limit for
3964 encoded words that the RFC imposes. It now respects the limit, and
3965 generates multiple encoded words if necessary. To be on the safe side, I
3966 have increased the buffer size for the ${rfc2047: expansion operator from
3969 33. It is now permitted to omit both strings after an "if" condition; if the
3970 condition is true, the result is "true". As before, when the second string
3971 is omitted, a false condition yields an empty string. This makes it less
3972 cumbersome to write custom ACL and router conditions.
3974 34. Failure to deliver a bounce message always caused it to be frozen, even if
3975 there was an errors_to setting on the router. The errors_to setting is now
3978 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3979 canonical form (fully expanded) before being placed in
3980 $sender_host_address.
3982 36. The table in the code that translates DNS record types into text (T_A to
3983 "A" for instance) was missing entries for NS and CNAME. It is just possible
3984 that this could have caused confusion if both these types were looked up
3985 for the same domain, because the text type is used as part of Exim's
3986 per-process caching. But the chance of anyone hitting this buglet seems
3989 37. The dnsdb lookup has been extended in a number of ways.
3991 (1) There is a new type, "zns", which walks up the domain tree until it
3992 finds some nameserver records. It should be used with care.
3994 (2) There is a new type, "mxh", which is like "mx" except that it returns
3995 just the host names, not the priorities.
3997 (3) It is now possible to give a list of domains (or IP addresses) to be
3998 looked up. The behaviour when one of the lookups defers can be
3999 controlled by a keyword.
4001 (4) It is now possible to specify the separator character for use when
4002 multiple records are returned.
4004 38. The dnslists ACL condition has been extended: it is now possible to supply
4005 a list of IP addresses and/or domains to be looked up in a particular DNS
4008 39. Added log_selector=+queue_time_overall.
4010 40. When running the queue in the test harness, wait just a tad after forking a
4011 delivery process, to get repeatability of debugging output.
4013 41. Include certificate and key file names in error message when GnuTLS fails
4014 to set them up, because the GnuTLS error message doesn't include the name
4015 of the failing file when there is a problem reading it.
4017 42. Allow both -bf and -bF in the same test run.
4019 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4021 44. The "Exiscan patch" is now merged into the mainline Exim source.
4023 45. Sometimes the final signoff response after QUIT could fail to get
4024 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4025 < 0 before doing a fflush(). This bug looks as though it goes back to the
4026 introduction of TLS in release 3.20, but "sometimes" must have been rare
4027 because the tests only now provoked it.
4029 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4030 (this can affect the format of dates).
4032 47. exim_tidydb, when checking for the continued existence of a message for
4033 which it has found a message-specific retry record, was not finding
4034 messages that were in split spool directories. Consequently, it was
4035 deleting retry records that should have stayed in existence.
4037 48. Steve fixed some bugs in eximstats.
4039 49. The SPA authentication driver was not abandoning authentication and moving
4040 on to the next authenticator when an expansion was forced to fail,
4041 contradicting the general specification for all authenticators. Instead it
4042 was generating a temporary error. It now behaves as specified.
4044 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4045 (the order specifies the preference for clients). The order is now AES256,
4046 AES128, 3DES, ARCFOUR128.
4048 51. Small patch to Sieve code - explicitly set From: when generating an
4051 52. Exim crashed if a remote delivery caused a very long error message to be
4052 recorded - for instance if somebody sent an entire SpamAssassin report back
4053 as a large number of 550 error lines. This bug was coincidentally fixed by
4054 increasing the size of one of Exim's internal buffers (big_buffer) that
4055 happened as part of the Exiscan merge. However, to be on the safe side, I
4056 have made the code more robust (and fixed the comments that describe what
4059 53. Now that there can be additional text after "Completed" in log lines (if
4060 the queue_time_overall log selector is set), a one-byte patch to exigrep
4061 was needed to allow it to recognize "Completed" as not the last thing in
4064 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4065 patch that reportedly fixes this has been added. I am not expert enough to
4066 create a test for it. This is what the patch creator wrote:
4068 "I found a little strange behaviour of ldap code when working with
4069 Windows 2003 AD Domain, where users was placed in more than one
4070 Organization Units. When I tried to give exim partial DN, the exit code
4071 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4072 But simultaneously result of request was absolutely normal ldap result,
4073 so I produce this patch..."
4075 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4076 REFERENCE, so I have modified the code to exclude the patch when that macro
4079 55. Some experimental protocols are using DNS PTR records for new purposes. The
4080 keys for these records are domain names, not reversed IP addresses. The
4081 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4082 leaves it alone. Component reversal etc. now happens only for IP addresses.
4085 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4087 57. Double the size of the debug message buffer (to 2048) so that more of very
4088 long debug lines gets shown.
4090 58. The exicyclog utility now does better if the number of log files to keep
4091 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4093 59. Two changes related to the smtp_active_hostname option:
4095 (1) $smtp_active_hostname is now available as a variable.
4096 (2) The default for smtp_banner uses $smtp_active_hostname instead
4097 of $primary_hostname.
4099 60. The host_aton() function is supposed to be passed a string that is known
4100 to be a valid IP address. However, in the case of IPv6 addresses, it was
4101 not checking this. This is a hostage to fortune. Exim now panics and dies
4102 if the condition is not met. A case was found where this could be provoked
4103 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4104 components; fortuitously, this particular loophole had already been fixed
4105 by change 4.50/55 above.
4107 If there are any other similar loopholes, the new check in host_aton()
4108 itself should stop them being exploited. The report I received stated that
4109 data on the command line could provoke the exploit when Exim was running as
4110 exim, but did not say which command line option was involved. All I could
4111 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4112 running as the user.
4115 61. There was a buffer overflow vulnerability in the SPA authentication code
4116 (which came originally from the Samba project). I have added a test to the
4117 spa_base64_to_bits() function which I hope fixes it.
4120 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4121 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4123 63. The daemon start-up calls getloadavg() while still root for those OS that
4124 need the first call to be done as root, but it missed one case: when
4125 deliver_queue_load_max is set with deliver_drop_privilege. This is
4126 necessary for the benefit of the queue runner, because there is no re-exec
4127 when deliver_drop_privilege is set.
4129 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4130 This has been fixed.
4132 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4133 was in use, was not putting the data itself into the right store pool;
4134 consequently, it could be overwritten for a subsequent message in the same
4135 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4138 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4140 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4141 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4142 the function were treating the return as a boolean value, which happened to
4143 work because 0=false and not-0=true, but is not correct code.
4145 68. The host_aton() function was not handling scoped IPv6 addresses (those
4146 with, for example, "%eth0" on the end) correctly.
4148 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4149 negated items (that is, ~something) in unsigned ints. Some compilers
4150 apparently mutter when there is no cast.
4152 70. If an address verification called from an ACL failed, and did not produce a
4153 user-specific message (i.e. there was only a "system" message), nothing was
4154 put in $acl_verify_message. In this situation, it now puts the system
4157 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4158 session; change 4.31/43 added the unwanted input to the log line - except
4159 that it did not do this in the start of session case. It now does.
4161 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4162 This is wrong and can cause the other end to generate a synchronization
4163 error if it is another Exim or anything else that does the synchronization
4164 check. A QUIT command is no longer sent after a timeout.
4166 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4167 during host lookups.
4169 74. The defer_ok option of callout verification was not working if it was used
4170 when verifying addresses in header lines, that is, for this case:
4172 verify = header_sender/callout=defer_ok
4174 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4175 those file descriptors could be used for SMTP connections. If anything
4176 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4177 could be sent to the SMTP client, causing chaos. The daemon now opens
4178 stdin, stdout, and stderr to /dev/null when it puts itself into the
4181 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4182 log by default. The user can override this with suitable Perl magic.
4184 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4185 the log message when discard triggers, was not working for the DATA ACL or
4186 for the non-SMTP ACL.
4188 78. Error message wording change in sieve.c.
4190 79. If smtp_accept_max_per_host was set, the number of connections could be
4191 restricted to fewer than expected, because the daemon was trying to set up
4192 a new connection before checking whether the processes handling previous
4193 connections had finished. The check for completed processes is now done
4194 earlier. On busy systems, this bug wouldn't be noticed because something
4195 else would have woken the daemon, and it would have reaped the completed
4198 80. If a message was submitted locally by a user whose login name contained one
4199 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4200 It caused a spool format error. I have fixed the spool reading code. A
4201 related problem was that the "from" clause in the Received: line became
4202 illegal because of the space(s). It is now covered by ${quote_local_part.
4204 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4207 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4209 83. Patch from Sieve maintainer to fix unterminated string problem in
4210 "vacation" handling.
4212 84. Some minor changes to the Linux configuration files to help with other
4213 OS variants using glibc.
4215 85. One more patch for Sieve to update vacation handling to latest spec.
4218 ----------------------------------------------------
4219 See the note above about the 4.44 and 4.50 releases.
4220 ----------------------------------------------------
4226 1. Change 4.43/35 introduced a bug that caused file counts to be
4227 incorrectly computed when quota_filecount was set in an appendfile
4230 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4231 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4234 3. Give more explanation in the error message when the command for a transport
4235 filter fails to execute.
4237 4. There are several places where Exim runs a non-Exim command in a
4238 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4239 was being done only for the command run by the queryprogram router. It is
4240 now done for all such subprocesses. The other cases are: ${run, transport
4241 filters, and the commands run by the lmtp and pipe transports.
4243 5. Some older OS have a limit of 256 on the maximum number of file
4244 descriptors. Exim was using setrlimit() to set 1000 as a large value
4245 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4246 systems. I've change it so that if it can't get 1000, it tries for 256.
4248 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4249 was an oversight, and furthermore, ever since the addition of extra
4250 controls (e.g. 4.43/32), the checks on when to allow different forms of
4251 "control" were broken. There should now be diagnostics for all cases when a
4252 control that does not make sense is encountered.
4254 7. $recipients is now available in the predata ACL (oversight).
4256 8. Tidy the search cache before the fork to do a delivery from a message
4257 received from the command line. Otherwise the child will trigger a lookup
4258 failure and thereby defer the delivery if it tries to use (for example) a
4259 cached ldap connection that the parent has called unbind on.
4261 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4262 of $address_data from the recipient verification was clobbered by the
4263 sender verification.
4265 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4266 was its contents. (It was OK if the option was not defined at all.)
4268 11. A "Completed" log line is now written for messages that are removed from
4269 the spool by the -Mrm option.
4271 12. $host_address is now set to the target address during the checking of
4272 ignore_target_hosts.
4274 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4275 being passed; this would have caused $sender_host_name to have been used if
4276 matching the list had actually called for a host name (not very likely,
4277 since this list is usually IP addresses). A host name is now passed as
4280 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4281 code that specifies a non-restarting handler (typically sigaction() in
4282 modern systems) in an attempt to fix a rare and obscure crash bug.
4284 15. Narrowed the window for a race in the daemon that could cause it to ignore
4285 SIGCHLD signals. This is not a major problem, because they are used only to
4286 wake it up if nothing else does.
4288 16. A malformed maildirsize file could cause Exim to calculate negative values
4289 for the mailbox size or file count. Odd effects could occur as a result.
4290 The maildirsize information is now recalculated if the size or filecount
4293 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4294 support for a long time. Removed HAVE_SYS_VFS_H.
4296 18. Updated exipick to current release from John Jetmore.
4298 19. Allow an empty sender to be matched against a lookup in an address list.
4299 Previously the only cases considered were a regular expression, or an
4302 20. Exim went into a mad DNS lookup loop when doing a callout where the
4303 host was specified on the transport, if the DNS lookup yielded more than
4306 21. The RFC2047 encoding function was originally intended for short strings
4307 such as real names; it was not keeping to the 75-character limit for
4308 encoded words that the RFC imposes. It now respects the limit, and
4309 generates multiple encoded words if necessary. To be on the safe side, I
4310 have increased the buffer size for the ${rfc2047: expansion operator from
4313 22. Failure to deliver a bounce message always caused it to be frozen, even if
4314 there was an errors_to setting on the router. The errors_to setting is now
4317 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4318 canonical form (fully expanded) before being placed in
4319 $sender_host_address.
4321 24. Updated eximstats to version 1.33
4323 25. Include certificate and key file names in error message when GnuTLS fails
4324 to set them up, because the GnuTLS error message doesn't include the name
4325 of the failing file when there is a problem reading it.
4327 26. Expand error message when OpenSSL has problems setting up cert/key files.
4330 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4331 (this can affect the format of dates).
4333 28. exim_tidydb, when checking for the continued existence of a message for
4334 which it has found a message-specific retry record, was not finding
4335 messages that were in split spool directories. Consequently, it was
4336 deleting retry records that should have stayed in existence.
4338 29. eximstats updated to version 1.35
4339 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4340 1.35 - bugfix such that pie charts by volume are generated correctly
4342 30. The SPA authentication driver was not abandoning authentication and moving
4343 on to the next authenticator when an expansion was forced to fail,
4344 contradicting the general specification for all authenticators. Instead it
4345 was generating a temporary error. It now behaves as specified.
4347 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4348 (the order specifies the preference for clients). The order is now AES256,
4349 AES128, 3DES, ARCFOUR128.
4351 31. Small patch to Sieve code - explicitly set From: when generating an
4354 32. Exim crashed if a remote delivery caused a very long error message to be
4355 recorded - for instance if somebody sent an entire SpamAssassin report back
4356 as a large number of 550 error lines. This bug was coincidentally fixed by
4357 increasing the size of one of Exim's internal buffers (big_buffer) that
4358 happened as part of the Exiscan merge. However, to be on the safe side, I
4359 have made the code more robust (and fixed the comments that describe what
4362 33. Some experimental protocols are using DNS PTR records for new purposes. The
4363 keys for these records are domain names, not reversed IP addresses. The
4364 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4365 leaves it alone. Component reversal etc. now happens only for IP addresses.
4368 34. The host_aton() function is supposed to be passed a string that is known
4369 to be a valid IP address. However, in the case of IPv6 addresses, it was
4370 not checking this. This is a hostage to fortune. Exim now panics and dies
4371 if the condition is not met. A case was found where this could be provoked
4372 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4373 components; fortuitously, this particular loophole had already been fixed
4374 by change 4.50/55 or 4.44/33 above.
4376 If there are any other similar loopholes, the new check in host_aton()
4377 itself should stop them being exploited. The report I received stated that
4378 data on the command line could provoke the exploit when Exim was running as
4379 exim, but did not say which command line option was involved. All I could
4380 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4381 running as the user.
4384 35. There was a buffer overflow vulnerability in the SPA authentication code
4385 (which came originally from the Samba project). I have added a test to the
4386 spa_base64_to_bits() function which I hope fixes it.
4389 36. The daemon start-up calls getloadavg() while still root for those OS that
4390 need the first call to be done as root, but it missed one case: when
4391 deliver_queue_load_max is set with deliver_drop_privilege. This is
4392 necessary for the benefit of the queue runner, because there is no re-exec
4393 when deliver_drop_privilege is set.
4395 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4396 was in use, was not putting the data itself into the right store pool;
4397 consequently, it could be overwritten for a subsequent message in the same
4398 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4401 38. Sometimes the final signoff response after QUIT could fail to get
4402 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4403 < 0 before doing a fflush(). This bug looks as though it goes back to the
4404 introduction of TLS in release 3.20, but "sometimes" must have been rare
4405 because the tests only now provoked it.
4411 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4412 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4413 Now it returns an integer. A number of calls were still expecting a T/F
4414 return. Fortuitously, in all cases, the tests worked in OK situations,
4415 which is the norm. However, things would have gone wrong on any write
4416 failures on the smtp file descriptor. This function is used when sending
4417 messages over SMTP and also when doing verify callouts.
4419 2. When Exim is called to do synchronous delivery of a locally submitted
4420 message (the -odf or -odi options), it no longer closes stderr before doing
4423 3. Implemented the mua_wrapper option.
4425 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4427 5. Implemented the functions header_remove(), header_testname(),
4428 header_add_at_position(), and receive_remove_recipient(), and exported them
4431 6. If an ACL "warn" statement specified the addition of headers, Exim already
4432 inserted X-ACL-Warn: at the start if there was no header name. However, it
4433 was not making this test for the second and subsequent header lines if
4434 there were newlines in the string. This meant that an invalid header could
4435 be inserted if Exim was badly configured.
4437 7. Allow an ACL "warn" statement to add header lines at the start or after all
4438 the Received: headers, as well as at the end.
4440 8. Added the rcpt_4xx retry error code.
4442 9. Added postmaster_mailfrom=xxx to callout verification option.
4444 10. Added mailfrom=xxxx to the callout verification option, for verify=
4447 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4448 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4450 12. Inserted some casts to stop certain compilers warning when using pointer
4451 differences as field lengths or precisions in printf-type calls (mostly
4452 affecting debugging statements).
4454 13. Added optional readline() support for -be (dynamically loaded).
4456 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4457 same clock tick as a message's arrival, so that its received time was the
4458 same as the "first fail" time on the retry record, and that message
4459 remained on the queue past the ultimate address timeout, every queue runner
4460 would try a delivery (because it was past the ultimate address timeout) but
4461 after another failure, the ultimate address timeout, which should have then
4462 bounced the address, did not kick in. This was a "< instead of <=" error;
4463 in most cases the first failure would have been in the next clock tick
4464 after the received time, and all would be well.
4466 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4467 being recognized when the domain list was tested by the match_domain
4468 condition in an expansion string.
4470 16. Added the ${str2b64: operator.
4472 17. Exim was always calling setrlimit() to set a large limit for the number of
4473 processes, without checking whether the existing limit was already
4474 adequate. (It did check for the limit on file descriptors.) Furthermore,
4475 errors from getrlimit() and setrlimit() were being ignored. Now they are
4476 logged to the main and panic logs, but Exim does carry on, to try to do its
4477 job under whatever limits there are.
4479 18. Imported PCRE 5.0.
4481 19. Trivial typo in log message " temporarily refused connection" (the leading
4484 20. If the log selector return_path_on_delivery was set and an address was
4485 redirected to /dev/null, the delivery process crashed because it assumed
4486 that a return path would always be set for a "successful" delivery. In this
4487 case, the whole delivery is bypassed as an optimization, and therefore no
4490 21. Internal re-arrangement: the function for sending a challenge and reading
4491 a response while authentication was assuming a zero-terminated challenge
4492 string. It's now changed to take a pointer and a length, to allow for
4493 binary data in such strings.
4495 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4497 23. Exim was not respecting finduser_retries when seeking the login of the
4498 uid under which it was called; it was always trying 10 times. (The default
4499 setting of finduser_retries is zero.) Also, it was sleeping after the final
4500 failure, which is pointless.
4502 24. Implemented tls_on_connect_ports.
4504 25. Implemented acl_smtp_predata.
4506 26. If the domain in control=submission is set empty, Exim assumes that the
4507 authenticated id is a complete email address when it generates From: or
4508 Sender: header lines.
4510 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4511 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4512 chown and chgrp in /bin and hostname in /usr/bin.
4514 28. Exim was keeping the "process log" file open after each use, just as it
4515 does for the main log. This opens the possibility of it remaining open for
4516 long periods when the USR1 signal hits a daemon. Occasional processlog
4517 errors were reported, that could have been caused by this. Anyway, it seems
4518 much more sensible not to leave this file open at all, so that is what now
4521 29. The long-running daemon process does not normally write to the log once it
4522 has entered its main loop, and it closes the log before doing so. This is
4523 so that log files can straightforwardly be renamed and moved. However,
4524 there are a couple of unusual error situations where the daemon does write
4525 log entries, and I had neglected to close the log afterwards.
4527 30. The text of an SMTP error response that was received during a remote
4528 delivery was being truncated at 512 bytes. This is too short for some of
4529 the long messages that one sometimes sees. I've increased the limit to
4532 31. It is now possible to make retry rules that apply only when a message has a
4533 specific sender, in particular, an empty sender.
4535 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4536 it possible to be selective about when SMTP synchronization is enforced.
4538 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4540 32. Implemented hosts_connection_nolog.
4542 33. Added an ACL for QUIT.
4544 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4547 35. Added mailbox_size and mailbox_filecount to appendfile.
4549 36. Added control = no_multiline_responses to ACLs.
4551 37. There was a bug in the logic of the code that waits for the clock to tick
4552 in the case where the clock went backwards by a substantial amount such
4553 that the microsecond fraction of "now" was more than the microsecond
4554 fraction of "then" (but the whole seconds number was less).
4556 38. Added support for the libradius Radius client library this is found on
4557 FreeBSD (previously only the radiusclient library was supported).
4563 1. When certain lookups returned multiple values in the form name=value, the
4564 quoting of the values was not always being done properly. Specifically:
4565 (a) If the value started with a double quote, but contained no whitespace,
4567 (b) If the value contained whitespace other than a space character (i.e.
4568 tabs or newlines or carriage returns) it was not quoted.
4569 This fix has been applied to the mysql and pgsql lookups by writing a
4570 separate quoting function and calling it from the lookup code. The fix
4571 should probably also be applied to nisplus, ibase and oracle lookups, but
4572 since I cannot test any of those, I have not disturbed their existing code.
4574 2. A hit in the callout cache for a specific address caused a log line with no
4575 reason for rejecting RCPT. Now it says "Previous (cached) callout
4576 verification failure".
4578 3. There was an off-by-one bug in the queryprogram router. An over-long
4579 return line was truncated at 256 instead of 255 characters, thereby
4580 overflowing its buffer with the terminating zero. As well as fixing this, I
4581 have increased the buffer size to 1024 (and made a note to document this).
4583 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4584 when Exim is waiting for an SMTP response from a remote server, Exim
4585 restarts its select() call on the socket, thereby resetting its timeout.
4586 This is not a problem when such interrupts are rare. Somebody set up a cron
4587 job to run exiwhat every 2 minutes, which is less than the normal select()
4588 timeout (5 or 10 minutes). This meant that the select() timeout never
4589 kicked in because it was always reset. I have fixed this by comparing the
4590 time when an interrupt arrives with the time at the start of the first call
4591 to select(). If more time than the timeout has elapsed, the interrupt is
4592 treated as a timeout.
4594 5. Some internal re-factoring in preparation for the addition of Sieve
4595 extensions (by MH). In particular, the "personal" test is moved to a
4596 separate function, and given an option for scanning Cc: and Bcc: (which is
4597 not set for Exim filters).
4599 6. When Exim created an email address using the login of the caller as the
4600 local part (e.g. when creating a From: or Sender: header line), it was not
4601 quoting the local part when it contained special characters such as @.
4603 7. Installed new OpenBSD configuration files.
4605 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4606 try to make them clearer.
4608 9. Callout options, other than the timeout value, were being ignored when
4609 verifying sender addresses in header lines. For example, when using
4611 verify = header_sender/callout=no_cache
4613 the cache was (incorrectly) being used.
4615 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4616 only the Cygwin environment.
4618 11. When return_path_on_delivery was set as a log selector, if different remote
4619 addresses in the same message used different return paths and parallel
4620 remote delivery occurred, the wrong values would sometimes be logged.
4621 (Whenever a remote delivery process finished, the return path value from
4622 the most recently started remote delivery process was logged.)
4624 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4625 lines when AUTH and/or TLS are in use. This is the "received protocol"
4626 field. Exim used to use "asmtp" for authenticated SMTP, without any
4627 indication (in the protocol name) for TLS use. Now it follows the RFC and
4628 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4629 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4630 names appear in log lines as well as in Received: header lines.
4632 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4633 extensions, and comparison tests, and to fix some bugs.
4635 14. Changes to the "personal" filter test:
4637 (1) The test was buggy in that it was just doing the equivalent of
4638 "contains" tests on header lines. For example, if a user's address was
4639 anne@some.where, the "personal" test would incorrectly be true for
4641 To: susanne@some.where
4643 This test is now done by extracting each address from the header in turn,
4644 and checking the entire address. Other tests that are part of "personal"
4645 are now done using regular expressions (for example, to check local parts
4646 of addresses in From: header lines).
4648 (2) The list of non-personal local parts in From: addresses has been
4649 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4650 taken from the Sieve specification recommendations.
4652 (3) If the message contains any header line starting with "List-" it is
4653 treated as non-personal.
4655 (4) The test for "circular" in the Subject: header line has been removed
4656 because it now seems ill-conceived.
4658 15. Minor typos in src/EDITME comments corrected.
4660 16. Installed latest exipick from John Jetmore.
4662 17. If headers_add on a router specified a text string that was too long for
4663 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4664 of string_sprintf() is now avoided.
4666 18. $message_body_size was not set (it was always zero) when running the DATA
4667 ACL and the local_scan() function.
4669 19. For the "mail" command in an Exim filter, no default was being set for
4670 the once_repeat time, causing a random time value to be used if "once" was
4671 specified. (If the value happened to be <= 0, no repeat happened.) The
4672 default is now 0s, meaning "never repeat". The "vacation" command was OK
4673 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4674 (I found it when inspecting the code).
4676 20. There is now an overall timeout for performing a callout verification. It
4677 defaults to 4 times the callout timeout, which applies to individual SMTP
4678 commands during the callout. The overall timeout applies when there is more
4679 than one host that can be tried. The timeout is checked before trying the
4680 next host. This prevents very long delays if there are a large number of
4681 hosts and all are timing out (e.g. when the network connections are timing
4682 out). The value of the overall timeout can be changed by specifying an
4683 additional sub-option for "callout", called "maxwait". For example:
4685 verify = sender/callout=5s,maxwait=20s
4687 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4688 to the end before writing, but this should make it even safer).
4690 22. Exim was forgetting that it had advertised PIPELINING for the second and
4691 subsequent messages on an SMTP connection. It was also not resetting its
4692 memory on STARTTLS and an internal HELO.
4694 23. When Exim logs an SMTP synchronization error within a session, it now
4695 records whether PIPELINING has been advertised or not.
4697 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4698 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4699 rather than long int.
4701 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4703 26. Added the never_mail option to autoreply.
4709 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4710 crash if the getsockname() call failed; this can happen if a connection is
4711 closed very soon after it is established. The problem was simply in the
4712 order in which certain operations were done, causing Exim to try to write
4713 to the SMTP stream before it had set up the file descriptor. The bug has
4714 been fixed by making things happen in the correct order.
4720 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4721 before the connection was closed, thus losing the rejection response.
4723 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4724 some early Solaris releases, but causes trouble in current releases where
4725 socklen_t is defined.
4727 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4730 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4733 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4734 that releasing the top of it at the end releases what was used for sub-
4735 expansions (unless the block got too big). However, discard this block if
4736 the first thing is a variable or header, so that we can use its block when
4737 it is dynamic (useful for very large $message_headers, for example).
4739 6. Lookups now cache *every* query, not just the most recent. A new, separate
4740 store pool is used for this. It can be recovered when all lookup caches are
4741 flushed. Lookups now release memory at the end of their result strings.
4742 This has involved some general refactoring of the lookup sources.
4744 7. Some code has been added to the store_xxx() functions to reduce the amount
4745 of flapping under certain conditions.
4747 8. log_incoming_interface used to affect only the <= reception log lines. Now
4748 it causes the local interface and port to be added to several more SMTP log
4749 lines, for example "SMTP connection from", and rejection lines.
4751 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4753 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4755 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4756 could be overwritten at the end of the current message (or the start of a
4757 new message if it was set in a HELO ACL). The value is now preserved for
4758 the duration of the SMTP connection.
4760 12. If a transport had a headers_rewrite setting, and a matching header line
4761 contained an unqualified address, that address was qualified, even if it
4762 did not match any rewriting rules. The underlying bug was that the values
4763 of the flags that permit the existence of unqualified sender and recipient
4764 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4765 non-local messages, and by -bnq for local messages) were not being
4766 preserved with the message after it was received.
4768 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4769 "next input=" as part of the text comprising the host identity instead of
4770 the correct text. The code was using the same buffer for two different
4771 strings. However, depending on which order the printing function evaluated
4772 its arguments, the bug did not always show up. Under Linux, for example, my
4773 test suite worked just fine.
4775 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4776 doesn't work with some older versions of Perl. It has been changed to "my",
4777 which in any case is probably the better facility to use.
4779 15. A really picky compiler found some instances of statements for creating
4780 error messages that either had too many or two few arguments for the format
4783 16. The size of the buffer for calls to the DNS resolver has been increased
4784 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4785 for addresses that have a lot of PTR records. This alleviates a problem; it
4786 does not fully solve it.
4788 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4789 buffer now truncates the list and logs the incident, which is the same
4790 action as happens when Exim is looking up a host name and its aliases.
4791 Previously in this situation something unpredictable would happen;
4792 sometimes it was "internal error: store_reset failed".
4794 18. If a server dropped the connection unexpectedly when an Exim client was
4795 using GnuTLS and trying to read a response, the client delivery process
4796 crashed while trying to generate an error log message.
4798 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4799 string, for example:
4801 warn message = H1: something\nH2: something
4803 the text was added as a single header line from Exim's point of view
4804 though it ended up OK in the delivered message. However, searching for the
4805 second and subsequent header lines using $h_h2: did not work. This has been
4806 fixed. Similarly, if a system filter added multiple headers in this way,
4807 the routers could not see them.
4809 20. Expanded the error message when iplsearch is called with an invalid key to
4810 suggest using net-iplsearch in a host list.
4812 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4813 ACLs are no longer actually imposed (and a message to that effect is
4816 22. If a "gecos" field in a passwd entry contained escaped characters, in
4817 particular, if it contained a \" sequence, Exim got it wrong when building
4818 a From: or a Sender: header from that name. A second bug also caused
4819 incorrect handling when an unquoted " was present following a character
4820 that needed quoting.
4822 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4823 was not being matched caselessly.
4825 24. Arranged for all hyphens in the exim.8 source to be escaped with
4828 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4829 properly. Recipient callout cache records were still being keyed to include
4830 the sender, even when use_sender was set false. This led to far more
4831 callouts that were necessary. The sender is no longer included in the key
4832 when use_sender is false.
4834 26. Added "control = submission" modifier to ACLs.
4836 27. Added the ${base62d: operator to decode base 62 numbers.
4838 28. dnsdb lookups can now access SRV records.
4840 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4841 the configuration file.
4843 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4844 (-v) mode. This makes the output for a verbose queue run more intelligible.
4846 31. Added a use_postmaster feature to recipient callouts.
4848 32. Added the $body_zerocount variable, containing the number of binary zero
4849 bytes in the message body.
4851 33. The time of last modification of the "new" subdirectory is now used as the
4852 "mailbox time last read" when there is a quota error for a maildir
4855 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4857 35. Added +ignore_unknown as a special item in host lists.
4859 36. Code for decoding IPv6 addresses in host lists is now included, even if
4860 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4861 address was recognized as an IP address, but was then not correctly decoded
4862 into binary, causing unexpected and incorrect effects when compared with
4869 1. Very minor rewording of debugging text in manualroute to say "list of
4870 hosts" instead of "hostlist".
4872 2. If verify=header_syntax was set, and a header line with an unqualified
4873 address (no domain) and a large number of spaces between the end of the
4874 name and the colon was received, the reception process suffered a buffer
4875 overflow, and (when I tested it) crashed. This was caused by some obsolete
4876 code that should have been removed. The fix is to remove it!
4878 3. When running in the test harness, delay a bit after writing a bounce
4879 message to get a bit more predictability in the log output.
4881 4. Added a call to search_tidyup() just before forking a reception process. In
4882 theory, someone could use a lookup in the expansion of smtp_accept_max_
4883 per_host which, without the tidyup, could leave open a database connection.
4885 5. Added the variables $recipient_data and $sender_data which get set from a
4886 lookup success in an ACL "recipients" or "senders" condition, or a router
4887 "senders" option, similar to $domain_data and $local_part_data.
4889 6. Moved the writing of debug_print from before to after the "senders" test
4892 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4893 problems for message scanning, either using a data ACL, or using
4894 local_scan() because the Received: header was not generated till after they
4895 were called (in order to set the time as the time of reception completion).
4896 I have revised the way this works. The header is now generated after the
4897 body is received, but before the ACL or local_scan() are called. After they
4898 are run, the timestamp in the header is updated.
4904 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4905 before starting a queue runner without re-exec. This happened only when
4906 deliver_drop_privilege was set or when the Exim user was set to root. The
4907 effect of the bug was that timeouts during subsequent deliveries caused
4908 crashes instead of being properly handled. The handler is now left at its
4909 default (and expected) setting.
4911 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4912 message, again when deliver_drop_privilege is set or Exim is run as root.
4913 The bug described in (1) was not present in this case, but the tidying up
4914 of the other signals was missing. I have made the two cases consistent.
4916 3. The ignore_target_hosts setting on a manualroute router was being ignored
4917 for hosts that were looked up using the /MX notation.
4919 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4922 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4923 operated on the sender address. After changing the $sender_address to <>
4924 for the sender address verify, Exim was re-instated it as the original
4925 (before rewriting) address, but remembering that it had rewritten it, so it
4926 wasn't rewriting it again. This bug also had the effect of breaking the
4927 sender address verification caching when the sender address was rewritten.
4929 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4930 This has been changed so that if the ip literal address matches
4931 ignore_target_hosts, the router declines.
4933 7. Added expansion conditions match_domain, match_address, and match_local_
4934 part (NOT match_host).
4936 8. The placeholder for the Received: header didn't have a length field set.
4938 9. Added code to Exim itself and to exim_lock to test for a specific race
4939 condition that could lead to file corruption when using MBX delivery. The
4940 issue is with the lockfile that is created in /tmp. If this file is removed
4941 after a process has opened it but before that process has acquired a lock,
4942 there is the potential for a second process to recreate the file and also
4943 acquire a lock. This could lead to two Exim processes writing to the file
4944 at the same time. The added code performs the same test as UW imapd; it
4945 checks after acquiring the lock that its file descriptor still refers to
4946 the same named file.
4948 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4949 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4952 11. Added the smtp_active_hostname option. If used, this will typically be made
4953 to depend on the incoming interface address. Because $interface_address is
4954 not set up until the daemon has forked a reception process, error responses
4955 that can happen earlier (such as "too many connections") no longer contain
4958 12. If an expansion in a condition on a "warn" statement fails because a lookup
4959 defers, the "warn" statement is abandoned, and the next ACL statement is
4960 processed. Previously this caused the whole ACL to be aborted.
4962 13. Added the iplsearch lookup type.
4964 14. Added ident_timeout as a log selector.
4966 15. Added tls_certificate_verified as a log selector.
4968 16. Added a global option tls_require_ciphers (compare the smtp transport
4969 option of the same name). This controls incoming TLS connections.
4971 17. I finally figured out how to make tls_require_ciphers do a similar thing
4972 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4973 before starting the TLS session.
4975 18. Tabs are now shown as \t in -bP output.
4977 19. If the log selector return_path_on_delivery was set, Exim crashed when
4978 bouncing a message because it had too many Received: header lines.
4980 20. If two routers both had headers_remove settings, and the first one included
4981 a superfluous trailing colon, the final name in the first list and the
4982 first name in the second list were incorrectly joined into one item (with a
4983 colon in the middle).
4989 1. Added -C and -D options to the exinext utility, mainly to make it easier
4990 to include in the automated testing, but these could be helpful when
4991 multiple configurations are in use.
4993 2. The exinext utility was not formatting the output nicely when there was
4994 an alternate port involved in the retry record key, nor when there was a
4995 message id as well (for retries that were specific to a specific message
4996 and a specific host). It was also confused by IPv6 addresses, because of
4997 the additional colons they contain. I have fixed the IPv4 problem, and
4998 patched it up to do a reasonable job for IPv6.
5000 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5001 delivery, the log line now contains "pipelined" if PIPELINING was used.
5003 4. An SMTP transport process used to panic and die if the bind() call to set
5004 an explicit outgoing interface failed. This has been changed; it is now
5005 treated in the same way as a connect() failure.
5007 5. A reference to $sender_host_name in the part of a conditional expansion
5008 that was being skipped was still causing a DNS lookup. This no longer
5011 6. The def: expansion condition was not recognizing references to header lines
5012 that used bh_ and bheader_.
5014 7. Added the _cache feature to named lists.
5016 8. The code for checking quota_filecount in the appendfile transport was
5017 allowing one more file than it should have been.
5019 9. For compatibility with Sendmail, the command line option
5027 and sets the incoming protocol and host name (for trusted callers). The
5028 host name and its colon can be omitted when only the protocol is to be set.
5029 Note the Exim already has two private options, -pd and -ps, that refer to
5030 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5031 "s", but I don't think that's a major issue.
5033 10. A number of refactoring changes to the code, none of which should affect
5036 (a) The number of logging options was getting close to filling up the
5037 32-bit word that was used as a bit map. I have split them into two classes:
5038 those that are passed in the argument to log_write(), and those that are
5039 only ever tested independently outside of that function. These are now in
5040 separate 32-bit words, so there is plenty of room for expansion again.
5041 There is no change in the user interface or the logging behaviour.
5043 (b) When building, for example, log lines, the code previously used a
5044 macro that called string_cat() twice, in order to add two strings. This is
5045 not really sufficiently general. Furthermore, there was one instance where
5046 it was actually wrong because one of the argument was used twice, and in
5047 one call a function was used. (As it happened, calling the function twice
5048 did not affect the overall behaviour.) The macro has been replaced by a
5049 function that can join an arbitrary number of extra strings onto a growing
5052 (c) The code for expansion conditions now uses a table and a binary chop
5053 instead of a serial search (which was left over from when there were very
5054 few conditions). Also, it now recognizes conditions like "pam" even when
5055 the relevant support is not compiled in: a suitably worded error message is
5056 given if an attempt is made to use such a condition.
5058 11. Added ${time_interval:xxxxx}.
5060 12. A bug was causing one of the ddress fields not to be passed back correctly
5061 from remote delivery subprocesses. The field in question was not being
5062 subsequently used, so this caused to problems in practice.
5064 13. Added new log selectors queue_time and deliver_time.
5066 14. Might have fixed a bug in maildirsizefile handling that threw up
5067 "unexpected character" debug warnings, and recalculated the data
5068 unnecessarily. In any case, I expanded the warning message to give more
5071 15. Added the message "Restricted characters in address" to the statements in
5072 the default ACL that block characters like @ and % in local parts.
5074 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5075 Three changes have been made:
5077 (a) There was a serious bug; a negative response to MAIL caused the whole
5078 recipient domain to be cached as invalid, thereby blocking all messages
5079 to all local parts at the same domain, from all senders. This bug has
5080 been fixed. The domain is no longer cached after a negative response to
5081 MAIL if the sender used is not empty.
5083 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5086 (c) A new callout option, "use_sender" has been added for people who want
5087 the modified behaviour.
5093 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5096 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5097 indeed breaks things for older releases.
5099 3. Added additional logging to the case where there is a problem reading data
5100 from a filter that is running in a subprocess using a pipe, in order to
5101 try to track down a specific problem.
5103 4. Testing facility fudge: when running in the test harness and attempting
5104 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5105 getting "No route to host". Convert this to a timeout.
5107 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5110 6. Some OS don't have socklen_t but use size_t instead. This affects the
5111 fifth argument of getsockopt() amongst other things. This is now
5112 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5113 can be set for individual OS. I have set it for SunOS5, OSF1, and
5114 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5115 some earlier ones do not.
5117 7. Change 4.30/15 was not doing the test caselessly.
5119 8. The standard form for an IPv6 address literal was being rejected by address
5120 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5121 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5122 this, as well as the form without the "IPv6" on the front (but only when
5123 address literals are enabled, of course).
5125 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5127 10. Exim crashed if a message with an empty sender address specified by -f
5128 encountered a router with an errors_to setting. This could be provoked only
5129 by a command such as
5133 where an empty string was supplied; "<>" did not hit this bug.
5135 11. Installed PCRE release 4.5.
5137 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5138 remained set. It is now erased.
5140 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5141 times from message ids (which are base 36 rather than the normal 62).
5143 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5144 were being counted as actual protocol errors, and logged if the log
5145 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5146 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5147 and DATA following a set of rejected RCPTs do not count as protocol errors.
5148 In other words, Exim assumes they were pipelined, though this may not
5149 actually be the case. Of course, in all cases the client gets an
5150 appropriate error code.
5152 15. If a lookup fails in an ACL condition, a message about the failure may
5153 be available; it is used if testing the ACL cannot continue, because most
5154 such messages specify what the cause of the deferral is. However, some
5155 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5156 that caused an old message to be retained and used if a later statement
5157 caused a defer, replacing the real cause of the deferral.
5159 16. If an IP address had so many PTR records that the DNS lookup buffer
5160 was not large enough to hold them, Exim could crash while trying to process
5161 the truncated data. It now detects and logs this case.
5163 17. Further to 4.21/58, another change has been made: if (and only if) the
5164 first line of a message (the first header line) ends with CRLF, a bare LF
5165 in a subsequent header line has a space inserted after it, so as not to
5166 terminate the header.
5168 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5169 unnecessarily, used atoi() instead of strtol(), and didn't check the
5170 termination when getting file sizes from file names by regex.
5172 19. Completely re-implemented the support for maildirsize files, in the light
5173 of a number of problems with the previous contributed implementation
5174 (4.30/29). In particular:
5176 . If the quota is zero, the maildirsize file is maintained, but no quota is
5179 . If the maildir directory does not exist, it is created before any attempt
5180 to write a maildirsize file.
5182 . The quota value in the file is just a cache; if the quota is changed in
5183 the transport, the new value overrides.
5185 . A regular expression is available for excluding directories from the
5188 20. The autoreply transport checks the characters in options that define the
5189 message's headers; it allows continued headers, but it was checking with
5190 isspace() after an embedded newline instead of explicitly looking for a
5193 21. If all the "regular" hosts to which an address was routed had passed their
5194 expiry times, and had not reached their retry times, the address was
5195 bounced, even if fallback hosts were defined. Now Exim should go on to try
5198 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5199 equivalent code in the SMTP transport. Some hosts send humungous responses
5200 to HELO/EHLO, more than 1024 it seems.
5202 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5203 gives compiler warnings in some environments. I've now done it "properly",
5206 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5207 (because of problems with the built-in one) was declared to return uschar *
5208 instead of char *, causing compiler failure.
5210 25. Fixed a file descriptor leak when processing alias/forward files.
5212 26. Fixed a minor format string issue in dbfn.c.
5214 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5216 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5217 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5220 29. When a sender address is verified, it is cached, to save repeating the test
5221 when there is more than one recipient in a message. However, when the
5222 verification involves a callout, it is possible for different callout
5223 options to be set for different recipients. It is too complicated to keep
5224 track of this in the cache, so now Exim always runs a verification when a
5225 callout is required, relying on the callout cache for the optimization.
5226 The overhead is duplication of the address routing, but this should not be
5229 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5230 to be verified with callout=postmaster, and the main callout worked but the
5231 postmaster check failed, the verification correctly failed. However, if a
5232 subsequent RCPT command asked for sender verification *without* the
5233 postmaster check, incorrect caching caused this verification also to fail,
5236 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5237 it was not caching the DNS options (qualify_single, search_parents) that
5238 were used when the lookup failed. A subsequent lookup with different
5239 options therefore always gave the same answer, though there were cases
5240 where it should not have. (Example: a "domains = !$mx_any" option on a
5241 dnslookup router: the "domains" option is always processed without any
5242 widening, but the router might have qualify_single set.) Now Exim uses the
5243 cached value only when the same options are set.
5245 32. Added John Jetmore's "exipick" utility to the distribution.
5247 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5248 than a timeout (e.g. a certificate is required, and is not provided), an
5249 Exim server now closes the connection immediately. Previously it waited for
5250 the client to close - but if the client is SSL, it seems that they each
5251 wait for each other, leading to a delay before one of them times out.
5253 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5254 maintained 0.8.x compatibility because I don't think many are using it, and
5255 it is clearly obsolete.
5257 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5260 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5261 client certificate was expired. A simple patch fixes this, though I don't
5262 understand the full logic of why the verify callback is called multiple
5265 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5266 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5267 which causes problems with some clients (such as the Certicom SSL Plus
5268 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5269 disables the coutermeasure allowing Eudora to connect."
5271 38. Exim was not checking that a write() to a log file succeeded. This could
5272 lead to Bad Things if a log got too big, in particular if it hit a file
5273 size limit. Exim now panics and dies if it cannot write to a log file, just
5274 as it does if it cannot open a log file.
5276 39. Modified OS/Makefile-Linux so that it now contains
5278 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5280 The two -D definitions ensure that Exim is compiled with large file
5281 support, which makes it possible to handle log files that are bigger than
5284 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5285 instance) a domain was checked against a named list that involved a lookup,
5286 causing $domain_data to be set, then another domain was checked against the
5287 same list, then the first domain was re-checked, the value of $domain_data
5288 after the final check could be wrong. In particular, if the second check
5289 failed, it could be set empty. This bug probably also applied to
5292 41. The strip_trailing_dot option was not being applied to the address given
5293 with the -f command-line option.
5295 42. The code for reading a message's header from the spool was incrementing
5296 $received_count, but never initializing it. This meant that the value was
5297 incorrect (doubled) while delivering a message in the same process in which
5298 it was received. In the most common configuration of Exim, this never
5299 happens - a fresh exec is done - but it can happen when
5300 deliver_drop_privilege is set.
5302 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5303 it now includes up to 150 characters of the unexpected data in the log
5306 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5307 and building data strings. The size of both of these buffers was 10 000
5308 bytes - far larger than anybody would *ever* want, thought I. Needless to
5309 say, somebody hit the limit. I have increased the maximum line length to
5310 20 000 and the maximum data length of concatenated lines to 100 000. I have
5311 also fixed two bugs, because there was no checking on these buffers. Tsk,
5312 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5313 buffer is too small.
5315 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5316 lsearch lookups. Now it does.
5318 46. When parsing a route_list item in a manualroute router, a fixed-length
5319 buffer was used for the list of hosts. I made this 1024 bytes long,
5320 thinking that nobody would ever have a list of hosts that long. Wrong.
5321 Somebody had a whole pile of complicated expansion conditions, and the
5322 string was silently truncated, leading to an expansion error. It turns out
5323 that it is easier to change to an unlimited length (owing to other changes
5324 that have happened since this code was originally written) than to build
5325 structure for giving a limitation error. The length of the item that
5326 expands into the list of hosts is now unlimited.
5328 47. The lsearch lookup could not handle data where the length of text line was
5329 more than 4095 characters. Such lines were truncated, leading to shortened
5330 data being returned. It should now handle lines of any length.
5332 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5333 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5336 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5337 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5338 no longer happens in comment lines. A list of replacements is now placed
5339 at the head of all of the source files, except those whose only change is
5340 to replace PERL_COMMAND in the very first #! line.
5342 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5343 messages on the queue, with a bottom-up merge sort, using code contributed
5344 by Michael Haardt. This should make operations like -bp somewhat faster on
5345 large queues. It won't affect queue runners, except when queue_run_in_order
5348 51. Installed eximstats 1.31 in the distribution.
5350 52. Added support for SRV lookups to the dnslookup router.
5352 53. If an ACL referred to $message_body or $message_body_end, the value was not
5353 reset for any messages that followed in the same SMTP session.
5355 54. The store-handling optimization for building very long strings was not
5356 differentiating between the different store pools. I don't think this
5357 actually made any difference in practice, but I've tidied it.
5359 55. While running the routers to verify a sender address, $sender_address
5360 was still set to the sender address. This is wrong, because when routing to
5361 send a bounce to the sender, it would be empty. Therefore, I have changed
5362 it so that, while verifying a sender address, $sender_address is set to <>.
5363 (There is no change to what happens when verifying a recipient address.)
5365 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5366 target A or AAAA records (if not already returned) without resetting the
5367 qualify_single or search_parents options of the DNS resolver. These are
5368 inappropriate in this case because the targets of MX and SRV records must
5369 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5370 target that, when qualified, matched something in the local domain. These
5371 two options are now turned off when doing these lookups.
5373 57. It seems that at least some releases of Reiserfs (which does not have the
5374 concept of a fixed number of inodes) returns zero and not -1 for the
5375 number of available inodes. This interacted badly with check_spool_inodes,
5376 which assumed that -1 was the "no such thing" setting. What I have done is
5377 to check that the total number of inodes is greater than zero before doing
5378 the test of how many are available.
5380 58. When a "warn" ACL statement has a log_message modifier, the message is
5381 remembered, and not repeated. This is to avoid a lot of repetition when a
5382 message has many recipients that cause the same warning to be written.
5383 Howewer, Exim was preserving the list of already written lines for an
5384 entire SMTP session, which doesn't seem right. The memory is now reset if a
5385 new message is started.
5387 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5388 debugging output unless log_rewrite was also set.
5390 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5391 of (int)(handle) when we know that handle contains (void *)(-1).
5393 61. The Exim daemon panic-logs an error return when it closes the incoming
5394 connection. However "connection reset by peer" seems to be common, and
5395 isn't really an error worthy of noting specially, so that particular error
5398 62. When Exim is trying to find all the local interfaces, it used to panic and
5399 die if the ioctl to get the interface flags failed. However, it seems that
5400 on at least one OS (Solaris 9) it is possible to have an interface that is
5401 included in the list of interfaces, but for which you get a failure error
5402 for this call. This happens when the interface is not "plumbed" into a
5403 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5404 failure of the "get flags" call assumes that the interface is down.
5406 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5407 makes life easier for people who are doing arithmetic on fields extracted
5408 from dates, where you often get leading zeros that should not be
5409 interpreted as octal.
5411 64. Added qualify_domain to the redirect router, to override the global
5414 65. If a pathologically long header line contained very many addresses (the
5415 report of this problem mentioned 10 000) and each of them was rewritten,
5416 Exim could use up a very large amount of memory. (It kept on making new
5417 copies of the header line as it rewrote, and never released the old ones.)
5418 At the expense of a bit more processing, the header rewriting function has
5419 been changed so that it no longer eats memory in this way.
5421 66. The generation of the Received: header has been moved from the time that a
5422 message starts to be received, to the time that it finishes. The timestamp
5423 in the Received: header should now be very close to that of the <= log
5424 line. There are two side-effects of this change:
5426 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5427 logged header lines no longer include the local Received: line, because
5428 it has not yet been created. The same applies to a copy of the message
5429 that is returned to a non-SMTP sender when a message is rejected.
5431 (b) When a filter file is tested using -bf, no additional Received: header
5432 is added to the test message. After some thought, I decided that this
5435 This change does not affect the value of $received_for. It is still set
5436 after address rewriting, but before local_scan() is called.
5438 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5440 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5441 gave an unhelpful panic error message, and a defer error. I have managed to
5442 change this behaviour so that it now rejects any supplied certificate,
5443 which seems right, as the list of acceptable certificates is empty.
5445 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5446 gave an unhelpful defer error. I have not managed to make this reject any
5447 supplied certificates, but the error message it gives is "no certificate
5448 supplied", which is not helpful.
5450 70. exigrep's output now also includes lines that are not associated with any
5451 message, but which match the given pattern. Implemented by a patch from
5452 Martin Sluka, which also tidied up the Perl a bit.
5454 71. Recipient callout verification, like sender verification, was using <> in
5455 the MAIL FROM command. This isn't really the right thing, since the actual
5456 sender may affect whether the remote host accepts the recipient or not. I
5457 have changed it to use the actual sender in the callout; this means that
5458 the cache record is now keyed on a recipient/sender pair, not just the
5459 recipient address. There doesn't seem to be a real danger of callout loops,
5460 since a callout by the remote host to check the sender would use <>.
5461 [SEE ABOVE: changed after hitting problems.]
5463 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5464 temporary errors. However, in the case of such a code being given after
5465 the end of a data transmission (i.e. after ".") Exim was failing to write
5466 a retry record for the message. (Yes, there was some broken host that was
5467 actually sending 8xx at this point.)
5469 73. An unknown lookup type in a host list could cause Exim to panic-die when
5470 the list was checked. (An example that provoked this was putting <; in the
5471 middle of a list instead of at the start.) If this happened during a DATA
5472 ACL check, a -D file could be left lying around. This kind of configuration
5473 error no longer causes Exim to die; instead it causes a defer errror. The
5474 incident is still logged to the main and panic logs.
5476 74. Buglet left over from Exim 3 conversion. The message "too many messages
5477 in one connection" was written to the rejectlog but not the mainlog, except
5478 when address rewriting (yes!) was being logged.
5480 75. Added write_rejectlog option.
5482 76. When a system filter was run not as root (that is, when system_filter_user
5483 was set), the values of the $n variables were not being returned to the
5484 main process; thus, they were not subsequently available in the $sn
5487 77. Added +return_path_on_delivery log selector.
5489 78. A connection timeout was being treated differently from recipients deferred
5490 when testing hosts_max_try with a message that was older than the host's
5491 retry timeout. (The host should not be counted, thus allowing all hosts to
5492 be tried at least once before bouncing.) This may have been the cause of an
5493 occasionally reported bug whereby a message would remain on the queue
5494 longer than the retry timeout, but would be bounced if a delivery was
5495 forced. I say "may" because I never totally pinned down the problem;
5496 setting up timeout/retry tests is difficult. See also the next item.
5498 79. The ultimate address timeout was not being applied to errors that involved
5499 a combination of host plus message (for example, a timeout on a MAIL
5500 command). When an address resolved to a number of possible hosts, and they
5501 were not all tried for each delivery (e.g. because of hosts_max_try), a
5502 message could remain on the queue longer than the retry timeout.
5504 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5507 81. Fixed an obscure SMTP outgoing bug which required at least the following
5508 conditions: (a) there was another message waiting for the same server;
5509 (b) the server returned 5xx to all RCPT commands in the first message so
5510 that the message was not completed; (c) the server dropped the connection
5511 or gave a negative response to the RSET that Exim sends to abort the
5512 transaction. The observed case was a dropped connection after DATA that had
5513 been sent in pipelining mode. That is, the server had advertised PIPELINING
5514 but was not implementing it correctly. The effect of the bug was incorrect
5515 behaviour, such as trying another host, and this could lead to a crash.
5521 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5522 and daemon.c were passed as pointers to ints; they should have been
5523 pointers to socklen_t variables (which are typically unsigned ints).
5525 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5528 3. Fixed a really odd bug that affected only the testing scheme; patching a
5529 certain fixed string in the binary changed the value of another string that
5530 happened to be identical to the end of the original first string.
5532 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5533 name", it returns that address as the IP address. On some operating
5534 systems (e.g. Solaris), it also passes back the IP address string as the
5535 "host name". However, on others (e.g. Linux), it passes back an empty
5536 string. Exim wasn't checking for this, and was changing the host name to an
5537 empty string, assuming it had been canonicized.
5539 5. Although rare, it is permitted to have more than one PTR record for a given
5540 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5541 all the names associated with an address, because they do in Solaris.
5542 However, it seems that they do not in Linux for data that comes from the
5543 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5544 I found this out when I moved to a new Linux workstation and tried to run
5545 the Exim test suite.
5547 To get round this problem I have changed the code so that it now does its
5548 own call to the DNS to look up PTR records when searching for a host name.
5549 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5550 addresses that are only in /etc/hosts are still found.
5552 This behaviour is, however, controlled by an option called host_lookup_
5553 order, which defaults to "bydns:byaddr". If people want to use the other
5554 order, or indeed, just use one or the other means of lookup, they can
5555 specify it in this variable.
5557 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5558 some operating systems, this comes back from gethostbyaddr() as an empty
5559 string, and this is what Exim used to test for. However, it seems that in
5560 other systems, "." is yielded. Exim now tests for this case too.
5562 7. The values of check_spool_space and check_log_space are now held internally
5563 as a number of kilobytes instead of an absolute number of bytes. If a
5564 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5565 kilobyte. This means that much larger values can be stored.
5567 8. Exim monitor: an attempt to get the action menu when not actually pointing
5568 at a message produces an empty menu entitled "No message selected". This
5569 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5570 no entries in it ("Shell widget menu has zero width and/or height"). So I
5571 have added a single, blank menu entry in this case.
5573 9. Added ${quote_local_part.
5575 10. MIME decoding is now applied to the contents of Subject: header lines when
5578 11. Now that a reference to $sender_host_address automatically causes a reverse
5579 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5580 host lookup before query-style lookups in lists that might use this
5581 variable. This has therefore been abolished, and the "net-" prefix is no
5582 longer necessary for query-style lookups.
5584 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5585 have been a typo for LFLAGS, so it has been changed.
5587 13. The install script calls Exim with "-C /dev/null" in order to find the
5588 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5589 to be output. Howeve, since Exim outputs its version number before the
5590 error, it didn't break the script. It just looked ugly. I fixed this by
5591 always allowing "-C /dev/null" if the caller is root.
5593 14. Ignore overlarge ACL variable number when reading spool file - insurance
5594 against a later release with more variables having written the file.
5596 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5597 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5598 this, as well as the form without the "IPv6" on the front.
5600 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5601 OS/Makefile-Darwin file.
5603 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5604 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5606 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5607 variable. However, some OS use ints for inodes. I've added cast to long int
5608 to get rid of the compiler warning.
5610 19. I had forgotten to lock out "/../" in configuration file names when
5611 ALT_CONFIG_PREFIX was set.
5613 20. Routers used for verification do not need to specify transports. However,
5614 if such a router generated a host list, and callout was configured, Exim
5615 crashed, because it could not find a port number from the (non-existent)
5616 transport. It now assumes port 25 in this circumstance.
5618 21. Added the -t option to exigrep.
5620 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5621 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5622 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5623 them is set, LOOKUP_LSEARCH is forced.
5625 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5627 24. Added sender and host information to the "rejected by local_scan()" log
5628 line; previously there was no indication of these.
5630 25. Added .include_if_exists.
5632 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5633 Linux. It turns out that not all file systems support this. Apparently some
5634 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5635 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5636 not supported on the file descriptor, is now ignored when Exim is trying to
5637 sync a directory. This applies only to Linux.
5639 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5641 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5642 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5645 29. Added support for maildirsize files from supplied patch (modified a bit).
5647 30. The use of :fail: followed by an empty string could lead Exim to respond to
5648 sender verification failures with (e.g.):
5650 550 Verification failed for <xxx>
5651 550 Sender verify failed
5653 where the first response line was missing the '-' that indicates it is not
5654 the final line of the response.
5656 31. The loop for finding the name of the user that called Exim had a hardwired
5657 limit of 10; it now uses the value of finduser_retries, which is used for
5658 all other user lookups.
5660 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5663 33. Exim was neglecting to zero errno before one call of strtol() when
5664 expanding a string and expecting an integer value. On some systems this
5665 resulted in spurious "integer overflow" errors. Also, it was casting the
5666 result into an int without checking.
5668 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5669 did not work. The code looks as if it has *never* worked, though it appears
5670 to have been documented since at least releast 1.62. I have made it work.
5672 35. The "timeout_DNS" error in retry rules, also documented since at least
5673 1.62, also never worked. As it isn't clear exactly what this means, and
5674 clearly it isn't a major issue, I have abolished the feature by treating it
5675 as "timeout", and writing a warning to the main and panic logs.
5677 36. The display of retry rules for -brt wasn't always showing the error code
5680 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5681 timeout_connect_A, timeout_connect_MX.
5683 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5684 to the empty sender.
5686 39. The daemon was not analysing the content of -oX till after it had closed
5687 stderr and disconnected from the controlling terminal. This meant that any
5688 syntax errors were only noted on the panic log, and the return code from
5689 the command was 0. By re-arranging the code a little, I've made the
5690 decoding happen first, so such errors now appear on stderr, and the return
5691 code is 1. However, the actual setting up of the sockets still happens in
5692 the disconnected process, so errors there are still only recorded on the
5695 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5696 connections (as happens on some IP stacks) was logged at start up time as
5697 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5698 it from "IPv6 and IPv4", which means that two separate sockets are being
5701 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5702 says whether AF_INET or AF_INET6 was passed as an argument.
5704 42. Exiwhat output was messed up when time zones were included in log
5707 43. Exiwhat now gives more information about the daemon's listening ports,
5708 and whether -tls-on-connect was used.
5710 44. The "port" option of the smtp transport is now expanded.
5712 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5713 silently ignored. Now an error message is written to the main and panic
5716 46. There's a new ACL modifier called "logwrite" which writes to a log file
5717 as soon as it is encountered.
5719 47. Added $local_user_uid and $local_user_gid at routing time.
5721 48. Exim crashed when trying to verify a sender address that was being
5724 49. Exim was recognizing only a space character after ".include". It now also
5725 recognizes a tab character.
5727 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5728 extracting the relevant information from the specification. The man page no
5729 longer contains scrambled data for the -d option, and I've added a section
5730 at the front about calling Exim under different names.
5732 51. Added "extra_headers" argument to the "mail" command in filter files.
5734 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5737 53. Installed eximstats 1.29.
5739 54. Added transport_filter_timeout as a generic transport option.
5741 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5742 Cc: header lines. This was required by RFC 822, but it not required by RFC
5745 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5746 incoming messages that did not have them. Now it does so only if the
5747 message originates locally, that is, if there is no associated remote host
5748 address. When Resent- header lines are present, this applies to the Resent-
5749 lines rather than the non-Resent- lines.
5751 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5752 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5754 58. Messages for configuration errors now include the name of the main
5755 configuration file - useful now that there may be more than one file in a
5756 list (.included file names were always shown).
5758 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5759 for those rare installations that do not start the daemon as root or run it
5760 setuid root. I've cut out the call to initgroups() if the daemon is not
5763 60. The Exim user and group can now be bound into the binary as text strings
5764 that are looked up at the start of Exim's processing.
5766 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5768 62. Added $mailstore_basename variable.
5770 63. Installed patch to sieve.c from Michael Haardt.
5772 64. When Exim failed to open the panic log after failing to open the main log,
5773 the original message it was trying to log was written to stderr and debug
5774 output, but if they were not available (the usual case in production), it
5775 was lost. Now it is written to syslog before the two lines that record the
5776 failures to open the logs.
5778 65. Users' Exim filters run in subprocesses under the user's uid. It is
5779 possible for a "deliver" command or an alias in a "personal" command to
5780 provoke an address rewrite. If logging of address rewriting is configured,
5781 this fails because the process is not running as root or exim. There may be
5782 a better way of dealing with this, but for the moment (because 4.30 needs
5783 to be released), I have disabled address rewrite logging when running a
5784 filter in a non-root, non-exim process.
5790 1. The buildconfig auxiliary program wasn't quoting the value set for
5791 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5792 not defined. This bug was masked in 4.22 by the effect that was fixed in
5795 2. Some messages that were rejected after a message id was allocated were
5796 shown as "incomplete" by exigrep. It no longer does this for messages that
5797 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5799 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5800 have allow_domain_literals set, the ID did not get logged in the <= line.
5801 Domain literals are now always recognized in Message-ID: header lines.
5803 4. The first argument for a ${extract expansion item is the key name or field
5804 number. Leading and trailing spaces in this item were not being ignored,
5805 causing some misleading effects.
5807 5. When deliver_drop_privilege was set, single queue runner processes started
5808 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5809 same command in the process it spins off) were not dropping privilege.
5811 6. When the daemon running as "exim" started a queue runner, it always
5812 re-executed Exim in the spun-off process. This is a waste of effort when
5813 deliver_drop_privilege is set. The new process now just calls the
5814 queue-runner function directly.
5820 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5823 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5824 ignored. Though the use of -oP was forcing the writing of a pid file, it
5825 was always written to the default place.
5827 3. If the message "no IP address found for host xxxx" is generated during
5828 incoming verification, it is now followed by identification of the incoming
5829 connection (so you can more easily find what provoked it).
5831 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5833 5. Added some features to "harden" Exim a bit more against certain attacks:
5835 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5836 be put in Local/Makefile. This is like the never_users runtime option,
5837 but it cannot be overridden. The default setting is "root".
5839 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5840 prefix string with which any file named in a -C command line option
5843 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5844 is retained for -C and -D only if the caller of Exim is root. Without
5845 it, the exim user may also use -C and -D and retain privilege.
5847 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5848 command line option is disabled.
5850 6. Macro names set by the -D option must start with an upper case letter, just
5851 like macro names defined in the configuration file.
5853 7. Added "dereference=" facility to LDAP.
5855 8. Two instances of the typo "uknown" in the source files are fixed.
5857 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5858 the Configure-Makefile script screwed up while processing it.
5860 10. Incorporated PCRE 4.4.
5862 11. The SMTP synchronization check was not operating right at the start of an
5863 SMTP session. For example, it could not catch a HELO sent before the client
5864 waited for the greeting. There is now a check for outstanding input at the
5865 point when the greeting is written. Because of the duplex, asynchronous
5866 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5867 way, but not yet received, when the check is performed.
5869 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5870 on TCP/IP sockets, because this apparently causes some broken clients to
5873 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5874 unchanged) from the Cygwin maintainer.
5876 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5877 was supported instead of testing for mbx. Effectively a typo.
5879 15. The spa authenticator server code was not checking that the input it
5880 received was valid base64.
5882 16. The debug output line for the "set" modifier in ACLs was not showing the
5883 name of the variable that was being set.
5885 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5887 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5888 lookup on its own. Something else was needed to trigger the lookup. For
5889 example, a match in host_lookup or the need for a host name in a host list.
5890 Now, if $sender_host_name is referenced and the host name has not yet been
5891 looked up, a lookup is performed. If the lookup fails, the variable remains
5892 empty, and $host_lookup_failed is set to "1".
5894 19. Added "eqi" as a case-independent comparison operator.
5896 20. The saslauthd authentication condition could segfault if neither service
5897 nor realm was specified.
5899 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5900 error message that was logged was misleading, and incoming SMTP
5901 connections were dropped. The message is now more accurate, and temporary
5902 errors are given to SMTP connections.
5904 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5905 (except RSET) with a 421 error, until QUIT is received. However, it was
5906 failing to send a response to QUIT.
5908 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5909 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5910 ACL is now run after the helo_try_verify_hosts code.
5912 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5913 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5914 case-independent, so other case variants are also recognized). Apparently
5915 some systems use these upper case variants.
5917 25. If more than two messages were waiting for the same host, and a transport
5918 filter was specified for the transport, Exim sent two messages over the
5919 same TCP/IP connection, and then failed with "socket operation on non-
5920 socket" when it tried to send the third.
5922 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5924 27. The extern definition of crypt16() in expand.c was not being excluded when
5925 the OS had its own crypt16() function.
5927 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5928 as a preferred synonym for return_size_limit, both as an option and as an
5931 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5933 30. Changed the default configuration ACL to relax the local part checking rule
5934 for addresses that are not in any local domains. For these addresses,
5935 slashes and pipe symbols are allowed within local parts, but the sequence
5936 /../ is explicitly forbidden.
5938 31. SPA server authentication was not clearing the challenge buffer before
5941 32. log_message in a "warn" ACL statement was writing to the reject log as
5942 well as to the main log, which contradicts the documentation and doesn't
5943 seem right (because no rejection is happening). So I have stopped it.
5945 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5946 However, I am unable to do any testing of this.
5948 34. Fixed an infelicity in the appendfile transport. When checking directories
5949 for a mailbox, to see if any needed to be created, it was accidentally
5950 using path names with one or more superfluous leading slashes; tracing
5951 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5953 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5954 contents are added to the log line that is written for every discarded
5955 recipient. (Previously a log_message setting was ignored.)
5957 36. The ${quote: operator now quotes the string if it is empty.
5959 37. The install script runs exim in order to find its version number. If for
5960 some reason other than non-existence or emptiness, which it checks, it
5961 could not run './exim', it was installing it with an empty version number,
5962 i.e. as "exim-". This error state is now caught, and the installation is
5965 38. An argument was missing from the function that creates an error message
5966 when Exim fails to connect to the socket for saslauthd authentication.
5967 This could cause Exim to crash, or give a corrupted message.
5969 39. Added isip, isip4, and isip6 to ${if conditions.
5971 40. The ACL variables $acl_xx are now saved with the message, and can be
5972 accessed later in routers, transports, and filters.
5974 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5975 strings in the file are not string-expanded.
5977 42. If a MAIL command specified a SIZE value that was too large to fit into an
5978 int variable, the check against message_size_limit failed. Such values are
5979 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5980 day this will have to be increased, but I don't think I want to be around
5981 when emails are that large.
5988 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5989 iconv() is not standard in FreeBSD.
5991 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5992 IPv6 enabled. The observed symptom was a segmentation fault on return from
5993 the function os_common_find_running_interfaces() in src/os.c.
5995 3. In the check_special_case() function in daemon.c I had used "errno" as an
5996 argument name, which causes warnings on some systems. This was basically a
5997 typo, since it was named "eno" in the comments!
5999 4. The code that waits for the clock to tick (at a resolution of some fraction
6000 of a second) so as to ensure message-id uniqueness was always waiting for
6001 at least one whole tick, when it could have waited for less. [This is
6002 almost certainly not relevant at current processor speeds, where it is
6003 unlikely to ever wait at all. But we try to future-proof.]
6005 5. The function that sleeps for a time interval that includes fractions of a
6006 second contained a race. It did not block SIGALRM between setting the
6007 timer, and suspending (a couple of lines later). If the interval was short
6008 and the sigsuspend() was delayed until after it had expired, the suspension
6009 never ended. On busy systems this could lead to processes getting stuck for
6012 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6013 process, before it forks any delivery processes. The open lookup caching
6014 mechanism meant that the open file or database connection was passed into
6015 the delivery process. The problem was that delivery processes always tidy
6016 up cached lookup data. This could cause a problem for the next delivery
6017 process started by the queue runner, because the external queue runner
6018 process does not know about the closure. So the next delivery process
6019 still has data in the lookup cache. In the case of a file lookup, there was
6020 no problem because closing a file descriptor in a subprocess doesn't affect
6021 the parent. However, if the lookup was caching a connection to a database,
6022 the connection was closed, and the second delivery process was likely to
6023 see errors such as "PGSQL: query failed: server closed the connection
6024 unexpectedly". The problem has been fixed by closing all cached lookups
6025 in a queue runner before running a delivery process.
6027 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6028 seem to have the "const" qualifier which it has on other OS. I've
6031 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6032 *of the same type* (client or server) with the same public name that an
6033 error should be diagnosed.
6035 9. When Exim looked up a host name for an IP address, but failed to find the
6036 original IP address when looking up the host name (a safety check), it
6037 output the message "<ip address> does not match any IP for NULL", which was
6038 confusing, to say the least. The bug was that the host name should have
6039 appeared instead of "NULL".
6041 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6042 user that is built into the binary, and the -C or -D options is used, root
6043 privilege is dropped before the configuration file is read. In addition,
6044 logging is switched to stderr instead of the normal log files. If the
6045 configuration then re-defines the Exim user, the unprivileged environment
6046 is probably not what is expected, so Exim logs a panic warning message (but
6049 However, if deliver_drop_privilege is set, the unprivileged state may well
6050 be exactly what is intended, so the warning has been cut out in that case,
6051 and Exim is allowed to try to write to its normal log files.
6057 1. smtp_return_error_details was not giving details for temporary sender
6058 or receiver verification errors.
6060 2. Diagnose a configuration error if two authenticators have the same public
6063 3. Exim used not to create the message log file for a message until the first
6064 delivery attempt. This could be confusing when incoming messages were held
6065 for policy or load reasons. The message log file is now created at the time
6066 the message is received, and an initial "Received" line is written to it.
6068 4. The automatically generated man page for command line options had a minor
6069 bug that caused no ill effects; however, a more serious problem was that
6070 the procedure for building the man page automatically didn't always
6071 operate. Consequently, release 4.20 contains an out-of-date version. This
6072 shouldn't happen again.
6074 5. When building Exim with embedded Perl support, the script that builds the
6075 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6076 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6078 6. The freeze_tell option was not being used for messages that were frozen on
6079 arrival, either by an ACL or by local_scan().
6081 7. Added the smtp_incomplete_transaction log selector.
6083 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6084 was accepting AUTH without a new EHLO.
6086 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6087 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6090 10. Logging of TCP/IP connections (when configured) now happens in the main
6091 daemon process instead of the child process, so that the TCP/IP connection
6092 count is more accurate (but it can never be perfect).
6094 11. The use of "drop" in a nested ACL was not being handled correctly in the
6095 outer ACL. Now, if condition failure induced by the nested "drop" causes
6096 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6097 or "require"), the connection is dropped.
6099 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6100 that yield "discard" can now be used with an "accept" or a "discard" verb,
6101 but an error is generated for any others (because I can't see a useful way
6102 to define what should happen).
6104 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6105 are now processed in the same way as lines in the Exim configuration file.
6106 In particular, continuation lines are supported.
6108 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6110 15. Added -ti meaning -t -i.
6112 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6113 domains, and warn by logging if others are found.
6115 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6116 returned from GIFCONF when Exim is trying to find the list of interfaces on
6117 a host. The code in os.c has been modified to copy each ifreq to an aligned
6118 structure in all cases.
6120 Also, in some cases, the returned ifreq's were being copied to a 'struct
6121 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6122 means the last couple of bytes of an IPv6 address could be chopped if the
6123 ifreq contained only a normal sockaddr (14 bytes storage).
6125 18. Named domain lists were not supported in the hosts_treat_as_local option.
6126 An entry such as +xxxx was not recognized, and was treated as a literal
6129 19. Ensure that header lines added by a DATA ACL are included in the reject log
6130 if the ACL subsequently rejects the message.
6132 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6133 MD5 (which is deprecated).
6135 21. When testing a filter file using -bf, Exim was writing a message when it
6136 took the sender from a "From " line in the message, but it was not doing so
6137 when it took $return_path from a Return-Path: header line. It now does.
6139 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6140 with a valid header line field name (a series of printing characters
6141 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6143 23. Changed "disc" in the source to "disk" to conform to the documentation and
6144 the book and for uniformity.
6146 24. Ignore Sendmail's -Ooption=value command line item.
6148 25. When execve() failed while trying to run a command in a pipe transport,
6149 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6150 could be confused with a return value of 69 from the command itself. This
6151 has been changed to 127, the value the shell returns if it is asked to run
6152 a non-existent command. The wording for the related log line suggests a
6153 non-existent command as the problem.
6155 26. If received_header_text expands to an empty string, do not add a Received:
6156 header line to the message. (Well, it adds a token one on the spool, but
6157 marks it "old" so that it doesn't get used or transmitted.)
6159 27. Installed eximstats 1.28 (addition of -nt option).
6161 28. There was no check for failure on the call to getsockname() in the daemon
6162 code. This can fail if there is a shortage of resources on the system, with
6163 ENOMEM, for example. A temporary error is now given on failure.
6165 29. Contrary to the C standard, it seems that in some environments, the
6166 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6167 program. Exim now does this explicitly; it affects the formatting of
6168 timestamps using strftime().
6170 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6171 complaints. I've now initialized all the variables, to avoid this.
6173 32. Header lines added by a system filter were not being "seen" during
6174 transport-time rewrites.
6176 33. The info_callback() function passed to OpenSSL is set up with type void
6177 (*)(SSL *, int, int), as described somewhere. However, when calling the
6178 function (actually a macro) that sets it up, the type void(*)() is
6179 expected. I've put in a cast to prevent warnings from picky compilers.
6181 34. If a DNS black list lookup found a CNAME record, but there were no A
6182 records associated with the domain it pointed at, Exim crashed.
6184 35. If a DNS black list lookup returned more than one A record, Exim ignored
6185 all but the first. It now scans all returned addresses if a particular IP
6186 value is being sought. In this situation, the contents of the
6187 $dnslist_value variable are a list of all the addresses, separated by a
6190 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6191 to accept a host name and all its aliases if the forward lookup for any of
6192 them yielded the IP address of the incoming connection. Now it accepts only
6193 those names whose forward lookup yields the correct IP address. Any other
6194 names are discarded. This closes a loophole whereby a rogue DNS
6195 administrator could create reverse DNS records to break through a
6196 wildcarded host restriction in an ACL.
6198 37. If a user filter or a system filter that ran in a subprocess used any of
6199 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6200 the wrong values were passed to the pipe command ($thisaddress had the
6201 value of $0, $0 had the value of $1, etc). This bug was introduced by
6202 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6204 38. Improved the line breaking for long SMTP error messages from ACLs.
6205 Previously, if there was no break point between 40 and 75 characters, Exim
6206 left the rest of the message alone. Two changes have been made: (a) I've
6207 reduced the minimum length to 35 characters; (b) if it can't find a break
6208 point between 35 and 75 characters, it looks ahead and uses the first one
6209 that it finds. This may give the occasional overlong line, but at least the
6210 remaining text gets split now.
6212 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6213 file descriptors might be low, and that setting 1000 would always raise it.
6214 It turns out that in some environments, the limit is already over 1000 and
6215 that lowering it causes trouble. So now Exim takes care not to decrease it.
6217 40. When delivering a message, the value of $return_path is set to $sender_
6218 address at the start of routing (routers may change the value). By an
6219 oversight, this default was not being set up when an address was tested by
6220 -bt or -bv, which affected the outcome if any router or filter referred to
6223 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6224 log only when "message" or "log_message" are set. However, if one of the
6225 conditions was an address verification, or a call to a nested ACL, the
6226 messages generated by the underlying test were being passed through. This
6227 no longer happens. The underlying message is available in $acl_verify_
6228 message for both "message" and "log_message" expansions, so it can be
6229 passed through if needed.
6231 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6232 new expansion $bh_ to give the encoded byte string without charset
6233 translation. Translation happens only if iconv() is available; HAVE_ICONV
6234 indicates this at build time. HEADERS_CHARSET gives the charset to
6235 translate to; headers_charset can change it in the configuration, and
6236 "headers charset" can change it in an individual filter file.
6238 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6239 that creates RFC 2047 encoded "words" labels them as that charset instead
6240 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6241 expansion operator; (ii) when Exim creates a From: line for a local
6242 message; (iii) when a header line is rewritten to include a "phrase" part.
6244 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6245 buggy, causing it to skip the first lines of messages whose message ID
6246 ended in 'D'. This would not have bitten before Exim release 4.14, because
6247 message IDs were unlikely to end in 'D' before then. The effect was to have
6248 incorrect size information for certain domains.
6250 45. #include "config.h" was missing at the start of the crypt16.c module. This
6251 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6254 46. If there was a timeout during a "random" callout check, Exim treated it as
6255 a failure of the random address, and carried on sending RSET and the real
6256 address. If the delay was just some slowness somewhere, the response to the
6257 original RCPT would be taken as a response to RSET and so on, causing
6258 mayhem of various kinds.
6260 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6261 when I implemented it. It didn't allow for the fact that some option values
6262 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6263 the right test for positive values.
6265 48. Domain names in DNS records are case-independent. Exim always looks them up
6266 in lower case. Some resolvers return domain names in exactly the case they
6267 appear in the zone file, that is, they may contain uppercase letters. Not
6268 all resolvers do this - some return always lower case. Exim was treating a
6269 change of case by a resolver as a change of domain, similar to a widening
6270 of a domain abbreviation. This triggered its re-routing code and so it was
6271 trying to route what was effectively the same domain again. This normally
6272 caused routing to fail (because the router wouldn't handle the domain
6273 twice). Now Exim checks for this case specially, and just changes the
6274 casing of the domain that it ultimately uses when it transmits the message
6277 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6280 50. If a filter generated a file delivery with a non-absolute name (possible if
6281 no home directory exists for the router), the forbid_file option was not
6284 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6285 the existing equality matching.
6287 52. Exim was using ints instead of ino_t variables in some places where it was
6288 dealing with inode numbers.
6290 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6291 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6292 called TMPDIR, and if it finds it is different, it changes its value.
6294 54. The smtp_printf() function is now made available to local_scan() so
6295 additional output lines can be written before returning. There is also an
6296 smtp_fflush() function to enable the detection of a dropped connection.
6297 The variables smtp_input and smtp_batched_input are exported to
6300 55. Changed the default runtime configuration: the message "Unknown user"
6301 has been removed from the ACL, and instead placed on the localuser router,
6302 using the cannot_route_message feature. This means that any verification
6303 failures that generate their own messages won't get overridden. Similarly,
6304 the "Unrouteable address" message that was in the ACL for unverifiable
6305 relay addresses has also been removed.
6307 56. Added hosts_avoid_esmtp to the smtp transport.
6309 57. The exicyclog script was not checking for the esoteric option
6310 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6311 will work only if exicyclog is run under the appropriate euid.
6313 58. Following a discussion on the list, the rules by which Exim recognises line
6314 endings on incoming messages have been changed. The -dropcr and drop_cr
6315 options are now no-ops, retained only for backwards compatibility. The
6316 following line terminators are recognized: LF CRLF CR. However, special
6317 processing applies to CR:
6319 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6320 nor a local message in the state where . is a terminator.
6322 (ii) If a bare CR is encountered in a header line, an extra space is added
6323 after the line terminator so as not to end the header. The reasoning
6324 behind this is that bare CRs in header lines are most likely either
6325 to be mistakes, or people trying to play silly games.
6327 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6328 was being incorrectly given as 18 bytes larger than it should have been.
6329 This is a VOB (very old bug).
6331 60. This may never have affected anything current, but just in case it has:
6332 When the local host is found other than at the start of a list of hosts,
6333 the local host, those with the same MX, and any that follow, are discarded.
6334 When the list in question was part of a longer list of hosts, the following
6335 hosts (not currently being processed) were also being discarded. This no
6336 longer happens. I'm not sure if this situation could ever has previously
6339 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6342 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6343 header. This is something that is recommended in a new Internet Draft, and
6344 is something that is documented as being done by Sendmail. There are two
6345 possible values. For messages generated by the autoreply transport, Exim
6348 Auto-Submitted: auto-replied
6350 whereas for all other generated messages (e.g. bounces) it adds
6352 Auto-Submitted: auto-generated
6354 63. The "personal" condition in filters now includes a test for the
6355 Auto-Submitted: header. If it contains the string "auto-" the message it
6356 not considered personal.
6358 64. Added rcpt_include_affixes as a generic transport option.
6360 65. Added queue_only_override (default true).
6362 66. Added the syslog_duplication option.
6364 67. If what should have been the first header line of a message consisted of
6365 a space followed by a colon, Exim was mis-interpreting it as a header line.
6366 It isn't of course - it is syntactically invalid and should therefore be
6367 treated as the start of the message body. The misbehaviour could have
6368 caused a number of strange effects, including loss of data in subsequent
6369 header lines, and spool format errors.
6371 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6372 client host had authenticated. This control can now be exercised by an ACL
6373 for more flexibility.
6375 69. By default, callouts do not happen when testing with -bh. There is now a
6376 variant, -bhc, which does actually run the callout code, including
6377 consulting and updating the callout cache.
6379 70. Added support for saslauthd authentication, courtesy of Alexander
6382 71. If statvfs() failed on the spool or log directories while checking their
6383 size for availability, Exim confusingly gave the error "space shortage".
6384 Furthermore, in debugging mode it crashed with a floating point exception.
6385 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6386 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6387 serious problem, Exim now writes to the main and panic logs when this
6388 happens, with details of the failure. It then refuses to accept the
6389 incoming message, giving the message "spool directory problem" or "log
6390 directory problem" with a 421 code for SMTP messages.
6392 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6393 0, 1, and 2 exist, because some OS complain for execs without them (see
6394 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6395 descriptors. However, the code omitted to check that the open succeeded,
6396 causing mysterious errors if for some reason the permissions on /dev/null
6397 got screwed. Now Exim writes a message to the main and panic logs, and
6398 bombs out if it can't open /dev/null.
6400 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6401 interact so that it is all more flexible. It is supposed to remain
6402 backwards compatible. Also added extra_local_interfaces.
6404 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6405 to bomb out with an assertion failure - to the client this appears as a
6406 connection drop. This problem occurs in the part of the code that was taken
6407 from the Samba project. Fortunately, the assertion is in a very simple
6408 function, so I have fixed this by reproducing the function inline in the
6409 one place where it is called, and arranging for authentication to fail
6410 instead of killing the process with assert().
6412 75. The SPA client code was not working when the server requested OEM rather
6413 than Unicode encoding.
6415 76. Added code to make require_files with a specific uid setting more usable in
6416 the case where statting the file as root fails - usually a non-root-mounted
6417 NFS file system. When this happens and the failure is EACCES, Exim now
6418 forks a subprocess and does the per-uid checking as the relevant uid.
6420 77. Added process_log_path.
6422 78. If log_file_path was not explicitly set, a setting of check_log_space or
6423 check_log_inodes was ignored.
6425 79. If a space check for the spool or log partitions fails, the incident is now
6426 logged. Of course, in the latter case the data may get lost...
6428 80. Added the %p formatting code to string_format() so that it can be used to
6429 print addresses in debug_print(). Adjusted all the address printing in the
6430 debugging in store.c to use %p rather than %d.
6432 81. There was a concern that a line of code in smtp_in.c could overflow a
6433 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6434 initially expressed, the concern was not well-founded, because trailing
6435 spaces are removed early. However, if the trailing spaces were followed by
6436 a NULL, they did not get removed, so the overflow was possible. Two fixes
6439 (a) I re-wrote the offending code in a cleaner fashion.
6440 (b) If an incoming SMTP command contains a NULL character, it is rejected
6443 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6444 runs initgroups(), so that if the Exim user is in any additional groups,
6445 they will be used during message reception.
6451 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob