1 Change log file for Exim from version 4.21
2 -------------------------------------------
7 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
8 and 3.4.4 - once the server is enabled to respond to an OCSP request
9 it does even when not requested, resulting in a stapling non-aware
10 client dropping the TLS connection.
12 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
13 support variable-length bit vectors. No functional change.
15 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
16 The I= interface field on outgoing lines is now after the H= remote
17 host field, same as incoming lines. There is a separate
18 outgoing_interface log selector which allows you to disable the
21 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
22 If not running log_selector +smtp_connection the mainlog would be held
23 open indefinitely after a "too many connections" event, including to a
24 deleted file after a log rotate. Leave the per net connection logging
25 leaving it open for efficiency as that will be quickly detected by the
26 check on the next write.
28 HS/01 Bug 1671: Fix post transport crash.
29 Processing the wait-<transport> messages could crash the delivery
30 process if the message IDs didn't exist for some reason. When
31 using 'split_spool_directory=yes' the construction of the spool
32 file name failed already, exposing the same netto behaviour.
34 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
35 mime_regex ACL conditions.
37 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
38 to DSN fail messages (bounces): remote IP, remote greeting, remote response
39 to HELO, local diagnostic string.
41 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
42 log line to debug. Even when configured with a tls authenticator many
43 client connections are expected to not authenticate in this way, so
44 an authenticate fail is not an error.
46 HS/02 Add the Exim version string to the process info. This way exiwhat
47 gives some more detail about the running daemon.
49 JH/06 Bug 1395: time-limit cacheing of DNS lookups, to the TTL value. This may
50 matter for fast-change records such as DNSBLs.
52 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
53 retry record, even if constant. There may be multiple transports with
54 different interface settings and the retry behaviour needs to be kept
57 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
59 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
61 JH/10 Bug 840: fix log_defer_output option of pipe transport
63 JH/11 Bug 830: use same host for all RCPTS of a message, even under
64 hosts_randomize. This matters a lot when combined with mua_wrapper.
66 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
67 ${quote_pgsql:<string>} operator.
69 JH/13 Bug 1708: avoid misaligned access in cached lookup.
71 JH/14 Change header file name for freeradius-client. Relevant if compiling
72 with Radius support; from the Gentoo tree and checked under Fedora.
74 JH/15 Bug 1712: Introduce $prdr_requested flag variable
76 JH/16 Bug 1714: Permit an empty string as expansion result for transport
77 option transport_filter, meaning no filtering.
79 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
81 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
82 defaults to "*" (all hosts). The variable is now available when not built
83 with TLS, default unset, mainly to enable keeping the testuite sane.
84 If a server certificate is not supplied (via tls_certificate) an error is
85 logged, and clients will find TLS connections fail on startup. Presumably
86 they will retry in-clear.
87 Packagers of Exim are strongly encouraged to create a server certificate
90 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
91 with the $config_file variable.
93 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
94 in transport context, after the attempt, and per-recipient. The latter type
95 is per host attempted. The event data is the error message, and the errno
96 information encodes the lookup type (A vs. MX) used for the (first) host,
97 and the trailing two digits of the smtp 4xx reponse.
99 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
100 to write to mainlog (or rejectlog, paniclog) in the window between file
101 creation and permissions/ownership being changed. Particularly affects
102 installations where exicyclog is run as root, rather than exim user;
103 result is that the running daemon panics and dies.
105 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
107 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
108 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
109 "pri" and "weight". Note that the previous implicit priority given by the
110 list order is no longer honoured.
112 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalisation
115 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
116 by defining SUPPORT_SOCKS.
118 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
119 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
120 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
121 variables are renamed to proxy_{local,external}_{address,port}.
123 JH/27 Move Internationalisation support from Experimental to mainline, enabled
124 for a build by defining SUPPORT_I18N
126 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
127 of the query string, and make ${quote_redis:} do that quoting.
129 JH/29 Move Events support from Experimental to mainline, enabled by default
130 and removable for a build by defining DISABLE_EVENT.
132 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
134 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
135 cached by the daemon.
137 JH/32 Move Redis support from Experimental to mainline, enabled for a build
138 by defining LOOKUP_REDIS. The libhiredis library is required.
140 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
141 keys are given for lookup.
143 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
144 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
145 only supported when built with TLS support. The PolarSSL SHA routines
146 are still used when the TLS library is too old for convenient support.
148 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
149 openssl_options), for security. OpenSSL forces this from version 1.1.0
150 server-side so match that on older versions.
152 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
153 allocation for $value could be released as the expansion processing
154 concluded, but leaving the global pointer active for it.
156 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
157 and to use the domains and local_parts ACL conditions.
159 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
160 incorrectly not doubled on cutthrough transmission, hence seen as a
161 body-termination at the receiving system - resulting in truncated mails.
162 Commonly the sender saw a TCP-level error, and retransmitted the message
163 via the normal store-and-forward channel. This could result in duplicates
164 received - but deduplicating mailstores were liable to retain only the
165 initial truncated version.
167 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
169 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
171 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
172 we're in there, support oversigning also; bug 1309.
174 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
176 HS/04 Add support for keep_environment and add_environment options.
178 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
179 either intentional arithmetic overflow during PRNG, or testing config-
186 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
189 JH/02 The smtp transport option "multi_domain" is now expanded.
191 JH/03 The smtp transport now requests PRDR by default, if the server offers
194 JH/04 Certificate name checking on server certificates, when exim is a client,
195 is now done by default. The transport option tls_verify_cert_hostnames
196 can be used to disable this per-host. The build option
197 EXPERIMENTAL_CERTNAMES is withdrawn.
199 JH/05 The value of the tls_verify_certificates smtp transport and main options
200 default to the word "system" to access the system default CA bundle.
201 For GnuTLS, only version 3.0.20 or later.
203 JH/06 Verification of the server certificate for a TLS connection is now tried
204 (but not required) by default. The verification status is now logged by
205 default, for both outbound TLS and client-certificate supplying inbound
208 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
211 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
212 Status Notification (bounce) messages are now MIME format per RFC 3464.
213 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
214 under the control of the dsn_advertise_hosts option, and routers may
215 have a dsn_lasthop option.
217 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
218 default, modifiable by a malware= option. The list separator for
219 the options can now be changed in the usual way. Bug 68.
221 JH/10 The smtp_receive_timeout main option is now expanded before use.
223 JH/11 The incoming_interface log option now also enables logging of the
224 local interface on delivery outgoing connections.
226 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
227 if the interface and destination host and port all match.
229 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
232 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
233 Patch from Andrew Lewis.
235 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
236 now supports optional time-restrictions, weighting, and priority
237 modifiers per server. Patch originally by <rommer@active.by>.
239 JH/16 The spamd_address main option now supports a mixed list of local
240 and remote servers. Remote servers can be IPv6 addresses, and
241 specify a port-range.
243 JH/17 Bug 68: The spamd_address main option now supports an optional
244 timeout value per server.
246 JH/18 Bug 1581: Router and transport options headers_add/remove can
247 now have the list separator specified.
249 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
252 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
255 JH/21 Support for the A6 type of dns record is withdrawn.
257 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
258 rather than the verbs used.
260 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
261 from 255 to 1024 chars.
263 JH/24 Verification callouts now attempt to use TLS by default.
265 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
266 are generic router options now. The defaults didn't change.
268 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
269 Original patch from Alexander Shikoff, worked over by JH.
271 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
272 files if ZCAT_COMMAND is not executable.
274 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
276 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
278 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
279 Normally benign, it bites when the pair was led to by a CNAME;
280 modern usage is to not canoicalize the domain to a CNAME target
281 (and we were inconsistent anyway for A-only vs AAAA+A).
283 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
285 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
286 when evaluating $sender_host_dnssec.
288 JH/31 Check the HELO verification lookup for DNSSEC, adding new
289 $sender_helo_dnssec variable.
291 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
293 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
295 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
297 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
298 documented as working, but never had. Support all but $spam_report.
300 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
301 added for tls authenticator.
306 TL/01 When running the test suite, the README says that variables such as
307 no_msglog_check are global and can be placed anywhere in a specific
308 test's script, however it was observed that placement needed to be near
309 the beginning for it to behave that way. Changed the runtest perl
310 script to read through the entire script once to detect and set these
311 variables, reset to the beginning of the script, and then run through
312 the script parsing/test process like normal.
314 TL/02 The BSD's have an arc4random API. One of the functions to induce
315 adding randomness was arc4random_stir(), but it has been removed in
316 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
317 function when detected.
319 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
320 cause callback expansion.
322 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
323 syntax errors in an expansion can be treated as a string instead of
324 logging or causing an error, due to the internal use of bool_lax
325 instead of bool when processing it.
327 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
328 server certificates when making smtp deliveries.
330 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
332 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
334 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
336 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
337 Merged patch from Sebastian Wiedenroth.
339 JH/05 Fix results-pipe from transport process. Several recipients, combined
340 with certificate use, exposed issues where response data items split
341 over buffer boundaries were not parsed properly. This eventually
342 resulted in duplicates being sent. This issue only became common enough
343 to notice due to the introduction of conection certificate information,
344 the item size being so much larger. Found and fixed by Wolfgang Breyha.
346 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
347 size buffer was used, resulting in syntax errors when an expansion
350 JH/07 Add support for directories of certificates when compiled with a GnuTLS
351 version 3.3.6 or later.
353 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
354 is EXPERIMENTAL_EVENT, the main-configuration and transport options
355 both become "event_action", the variables become $event_name, $event_data
356 and $event_defer_errno. There is a new variable $verify_mode, usable in
357 routers, transports and related events. The tls:cert event is now also
358 raised for inbound connections, if the main configuration event_action
361 TL/06 In test suite, disable OCSP for old versions of openssl which contained
362 early OCSP support, but no stapling (appears to be less than 1.0.0).
364 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
365 server certificate names available under the smtp transport option
366 "tls_verify_cert_hostname" now do not permit multi-component wildcard
369 JH/10 Time-related extraction expansions from certificates now use the main
370 option "timezone" setting for output formatting, and are consistent
371 between OpenSSL and GnuTLS compilations. Bug 1541.
373 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
374 encoded parameter in the incoming message. Bug 1558.
376 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
377 include certificate info, eximon was claiming there were spoolfile
380 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
382 JH/14 Log delivery-related information more consistently, using the sequence
383 "H=<name> [<ip>]" wherever possible.
385 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
386 are problematic for Debian distribution, omit them from the release
389 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
391 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
394 JH/17 Fix a null-indirection in certextract expansions when a nondefault
395 output list separator was used.
400 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
401 checkers that were complaining about end of non-void function with no
404 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
405 This was a regression intruduced in 4.83 by another bugfix.
407 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
409 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
410 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
416 TF/01 Correctly close the server side of TLS when forking for delivery.
418 When a message was received over SMTP with TLS, Exim failed to clear up
419 the incoming connection properly after forking off the child process to
420 deliver the message. In some situations the subsequent outgoing
421 delivery connection happened to have the same fd number as the incoming
422 connection previously had. Exim would try to use TLS and fail, logging
423 a "Bad file descriptor" error.
425 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
426 utilities have not been installed.
428 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
429 temporary space as the ACL may create new global variables.
431 TL/01 LDAP support uses per connection or global context settings, depending
432 upon the detected version of the libraries at build time.
434 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
435 to extract and use the src ip:port in logging and expansions as if it
436 were a direct connection from the outside internet. PPv2 support was
437 updated based on HAProxy spec change in May 2014.
439 JH/02 Add ${listextract {number}{list}{success}{fail}}.
441 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
442 Properly escape header and check for NULL return.
444 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
447 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
449 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
450 characters in header names, implemented as a verify condition.
451 Contributed by Michael Fischer v. Mollard.
453 TL/05 Rename SPF condition results err_perm and err_temp to standardized
454 results permerror and temperror. Previous values are deprecated but
455 still accepted. In a future release, err_perm and err_temp will be
456 completely removed, which will be a backward incompatibility if the
457 ACL tests for either of these two old results. Patch contributed by
458 user bes-internal on the mailing list.
460 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
462 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
463 selectors, in both main and reject logs.
465 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
468 JH/07 Add malware type "sock" for talking to simple daemon.
470 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
472 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
473 routers/transports under cutthrough routing.
475 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
476 numbers. Touch up "bool" conditional to keep the same definition.
478 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
480 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
482 JH/12 Expand items in router/transport headers_add or headers_remove lists
483 individually rather than the list as a whole. Bug 1452.
485 Required for reasonable handling of multiple headers_ options when
486 they may be empty; requires that headers_remove items with embedded
487 colons must have them doubled (or the list-separator changed).
489 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
490 view the policy declared in the DMARC record. Currently, $dmarc_status
491 is a combined value of both the record presence and the result of the
494 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
496 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
497 dnslookup router and the smtp transport (applying to the forward
500 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
501 of ldap servers used for a specific lookup. Patch provided by Heiko
504 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
505 New variable $lookup_dnssec_authenticated for observability.
507 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
508 Patch submitted by Lars Timman.
510 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
512 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
513 Requires trusted mode and valid format message id, aborts otherwise.
514 Patch contributed by Heiko Schlichting.
516 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
517 certextract with support for various fields. Bug 1358.
519 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
520 is requested by default, modifiable by smtp transport option
523 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
524 operate on certificate variables to give certificate fingerprints
525 Also new ${sha256:cert_variable}.
527 JH/23 The PRDR feature is moved from being Experimental into the mainline.
529 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
530 Christian Aistleitner.
532 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
534 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
535 file. Patch from Wolfgang Breyha.
537 JH/25 Expand the coverage of the delivery $host and $host_address to
538 client authenticators run in verify callout. Bug 1476.
540 JH/26 Port service names are now accepted for tls_on_connect_ports, to
541 align with daemon_smtp_ports. Bug 72.
543 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
544 support and error reporting did not work properly.
546 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
547 and is readable. Patch from Andrew Colin Kissa.
549 TL/14 Enhance documentation of ${run expansion and how it parses the
550 commandline after expansion, particularly in the case when an
551 unquoted variable expansion results in an empty value.
553 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
555 PP/02 Fix internal collision of T_APL on systems which support RFC3123
556 by renaming away from it. Addresses GH issue 15, reported by
559 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
561 TL/15 SECURITY: prevent double expansion in math comparison functions
562 (can expand unsanitized data). Not remotely exploitable.
569 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
571 PP/02 Make -n do something, by making it not do something.
572 When combined with -bP, the name of an option is not output.
574 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
577 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
578 $sender_host_name and config options to manage this, and basic check
581 PP/05 DSCP support for outbound connections and control modifier for inbound.
583 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
584 (Only plugin which currently uses this is kerberos4, which nobody should
585 be using, but we should make it available and other future plugins might
586 conceivably use it, even though it would break NAT; stuff *should* be
587 using channel bindings instead).
589 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
590 name; added for Sendmail compatibility; requires admin caller.
591 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
592 just ignore it); requires trusted caller.
593 Also parse but ignore: -Ac -Am -X<logfile>
596 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
598 TL/02 Add +smtp_confirmation as a default logging option.
600 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
601 Patch by Magnus Holmgren from 2007-02-20.
603 TL/04 Bugzilla 1281 - Spec typo.
604 Bugzilla 1283 - Spec typo.
605 Bugzilla 1290 - Spec grammar fixes.
607 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
609 TL/06 Add Experimental DMARC support using libopendmarc libraries.
611 TL/07 Fix an out of order global option causing a segfault. Reported to dev
612 mailing list by by Dmitry Isaikin.
614 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
616 JH/02 Support "G" suffix to numbers in ${if comparisons.
618 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
620 NM/01 Bugzilla 1197 - Spec typo
621 Bugzilla 1196 - Spec examples corrections
623 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
625 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
626 gnutls_enable_pkcs11, but renamed to more accurately indicate its
629 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
630 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
632 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
633 "acl {{name}{arg}...}", and optional args on acl condition
636 JH/05 Permit multiple router/transport headers_add/remove lines.
638 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
640 JH/07 Avoid using a waiting database for a single-message-only transport.
641 Performance patch from Paul Fisher. Bugzilla 1262.
643 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
646 JH/09 Add $headers_added variable, with content from use of ACL modifier
647 add_header (but not yet added to the message). Bugzilla 199.
649 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
650 Pulled from Bugzilla 817 by Wolfgang Breyha.
652 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
654 (nb: this is the same fix as in Exim 4.80.1)
656 JH/11 Add A= logging on delivery lines, and a client_set_id option on
659 JH/12 Add optional authenticated_sender logging to A= and a log_selector
662 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
664 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
665 advertise SMTP AUTH mechanism to us, instead of a generic
666 protocol violation error. Also, make Exim more robust to bad
667 data from the Dovecot auth socket.
669 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
671 When a queue runner is handling a message, Exim first routes the
672 recipient addresses, during which it prunes them based on the retry
673 hints database. After that it attempts to deliver the message to
674 any remaining recipients. It then updates the hints database using
677 So if a recipient address works intermittently, it can get repeatedly
678 deferred at routing time. The retry hints record remains fresh so the
679 address never reaches the final cutoff time.
681 This is a fairly common occurrence when a user is bumping up against
682 their storage quota. Exim had some logic in its local delivery code
683 to deal with this. However it did not apply to per-recipient defers
684 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
686 This change adds a proper retry rule check during routing so that the
687 final cutoff time is checked against the message's age. We only do
688 this check if there is an address retry record and there is not a
689 domain retry record; this implies that previous attempts to handle
690 the address had the retry_use_local_parts option turned on. We use
691 this as an approximation for the destination being like a local
692 delivery, as in LMTP.
694 I suspect this new check makes the old local delivery cutoff check
695 redundant, but I have not verified this so I left the code in place.
697 TF/02 Correct gecos expansion when From: is a prefix of the username.
699 Test 0254 submits a message to Exim with the header
703 When I ran the test suite under the user fanf2, Exim expanded
704 the header to contain my full name, whereas it should have added
705 a Resent-Sender: header. It erroneously treats any prefix of the
706 username as equal to the username.
708 This change corrects that bug.
710 GF/01 DCC debug and logging tidyup
711 Error conditions log to paniclog rather than rejectlog.
712 Debug lines prefixed by "DCC: " to remove any ambiguity.
714 TF/03 Avoid unnecessary rebuilds of lookup-related code.
716 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
717 Bug spotted by Jeremy Harris; was flawed since initial commit.
718 Would have resulted in OCSP responses post-SNI triggering an Exim
719 NULL dereference and crash.
721 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
723 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
724 Bug detection, analysis and fix by Samuel Thibault.
725 Bugzilla 1331, Debian bug #698092.
727 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
729 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
730 Server implementation by Todd Lyons, client by JH.
731 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
732 config variable "prdr_enable" controls whether the server
733 advertises the facility. If the client requests PRDR a new
734 acl_data_smtp_prdr ACL is called once for each recipient, after
735 the body content is received and before the acl_smtp_data ACL.
736 The client is controlled by bolth of: a hosts_try_prdr option
737 on the smtp transport, and the server advertisement.
738 Default client logging of deliveries and rejections involving
739 PRDR are flagged with the string "PRDR".
741 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
742 fclose(). Diagnosis by Todd Lyons.
744 PP/17 Update configure.default to handle IPv6 localhost better.
745 Patch by Alain Williams (plus minor tweaks).
748 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
749 This is now consistent with GnuTLS, and is now documented: the
750 previous undocumented portable approach to treating the option as
751 unset was to force an expansion failure. That still works, and
752 an empty string is now equivalent.
754 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
755 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
756 not performing validation itself.
758 PP/20 Added force_command boolean option to pipe transport.
759 Patch from Nick Koston, of cPanel Inc.
761 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
764 TF/04 Added udpsend ACL modifer and hexquote expansion operator
766 PP/21 Fix eximon continuous updating with timestamped log-files.
767 Broken in a format-string cleanup in 4.80, missed when I repaired the
768 other false fix of the same issue.
769 Report and fix from Heiko Schlichting.
772 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
773 Report from Prashanth Katuri.
775 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
776 It's SecureTransport, so affects any MacOS clients which use the
777 system-integrated TLS libraries, including email clients.
779 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
780 using a MIME ACL for non-SMTP local injection.
781 Report and assistance in diagnosis by Warren Baker.
783 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
785 JH/16 Fix comparisons for 64b. Bugzilla 1385.
787 TL/09 Add expansion variable $authenticated_fail_id to keep track of
788 last id that failed so it may be referenced in subsequent ACL's.
790 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
793 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
794 ldap library initialization, allowing self-signed CA's to be
795 used. Also properly sets require_cert option later in code by
796 using NULL (global ldap config) instead of ldap handle (per
797 session). Bug diagnosis and testing by alxgomz.
799 TL/12 Enhanced documentation in the ratelimit.pl script provided in
800 the src/util/ subdirectory.
802 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
803 renamed to Transport Post Delivery Action by Jeremy Harris, as
806 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
807 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
808 redis_servers = needs to be configured which will be used by the redis
809 lookup. Patch from Warren Baker, of The Packet Hub.
811 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
813 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
814 hostname or reverse DNS when processing a host list. Used suggestions
815 from multiple comments on this bug.
817 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
819 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
820 Missed a few lines, added it to make the runtest require no keyboard
823 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
824 contains upper case chars. Make router use caseful_local_part.
826 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
827 support when GnuTLS has been built with p11-kit.
833 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
835 This, or similar/improved, will also be change PP/11 of 4.82.
841 PP/01 Handle short writes when writing local log-files.
842 In practice, only affects FreeBSD (8 onwards).
843 Bugzilla 1053, with thanks to Dmitry Isaikin.
845 NM/01 Bugzilla 949 - Documentation tweak
847 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
850 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
852 PP/02 Implemented gsasl authenticator.
854 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
856 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
857 `pkg-config foo` for cflags/libs.
859 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
860 with rest of GSASL and with heimdal_gssapi.
862 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
863 `pkg-config foo` for cflags/libs for the TLS implementation.
865 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
866 properties get this fed in as external SSF. A number of robustness
867 and debugging improvements to the cyrus_sasl authenticator.
869 PP/08 cyrus_sasl server now expands the server_realm option.
871 PP/09 Bugzilla 1214 - Log authentication information in reject log.
872 Patch by Jeremy Harris.
874 PP/10 Added dbmjz lookup type.
876 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
878 PP/12 MAIL args handles TAB as well as SP, for better interop with
879 non-compliant senders.
880 Analysis and variant patch by Todd Lyons.
882 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
883 Bug report from Lars Müller <lars@samba.org> (via SUSE),
884 Patch from Dirk Mueller <dmueller@suse.com>
886 PP/13 tls_peerdn now print-escaped for spool files.
887 Observed some $tls_peerdn in wild which contained \n, which resulted
888 in spool file corruption.
890 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
891 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
892 or write after TLS renegotiation, which otherwise led to messages
895 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
896 as a tracking header (ie: a signed header comes before the signature).
897 Patch from Wolfgang Breyha.
899 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
900 comma-sep list; embedded commas doubled.
902 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
904 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
906 Report and patch from Dmitry Banschikov.
908 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
909 Removed SSL_clear() after SSL_new() which led to protocol negotiation
910 failures. We appear to now support TLS1.1+ with Exim.
912 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
913 lets Exim select keys and certificates based upon TLS SNI from client.
914 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
915 before an outbound SMTP session. New log_selector, +tls_sni.
917 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
918 NULL dereference. Report and patch from Alun Jones.
920 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
921 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
922 Not seeing resolver debug output on NetBSD, but suspect this is a
923 resolver implementation change.
925 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
926 Left warnings. Added "eximon gdb" invocation mode.
928 PP/21 Defaulting "accept_8bitmime" to true, not false.
930 PP/22 Added -bw for inetd wait mode support.
932 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
933 locate the relevant includes and libraries. Made this the default.
935 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
936 Bugzilla 1246, report and most of solution from Tomasz Kusy.
938 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
939 This may cause build issues on older platforms.
941 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
942 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
943 gnutls_require_mac & gnutls_require_protocols (no longer supported).
944 Added SNI support via GnuTLS too.
945 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
947 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
949 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
950 Applied second patch from Janne, implementing suggestion to default
951 multiple-strings-in-record handling to match SPF spec.
953 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
955 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
956 read-only, out of scope).
957 Patch from Wolfgang Breyha, report from Stuart Northfield.
959 PP/29 Fix three issues highlighted by clang analyser static analysis.
960 Only crash-plausible issue would require the Cambridge-specific
961 iplookup router and a misconfiguration.
962 Report from Marcin Mirosław.
964 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
966 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
967 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
968 As part of this, removing so much warning spew let me fix some minor
969 real issues in debug logging.
971 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
972 assignment on my part. Fixed.
974 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
975 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
976 Janne Snabb (who went above and beyond: thank you).
978 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
979 string otherwise requires a connection and a bunch more work and it's
980 relatively easy to get wrong. Should also expose TLS library linkage
983 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
984 64-bit ${eval} (JH/03).
986 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
987 GNU libc to support some of the 64-bit stuff, should not lead to
988 conflicts. Defined before os.h is pulled in, so if a given platform
989 needs to override this, it can.
991 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
992 protection layer was required, which is not implemented.
993 Bugzilla 1254, patch from Wolfgang Breyha.
995 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
996 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
997 tls_dhparam take prime identifiers. Also unbreak combination of
998 OpenSSL+DH_params+TLSSNI.
1000 PP/39 Disable SSLv2 by default in OpenSSL support.
1006 PP/01 Solaris build fix for Oracle's LDAP libraries.
1007 Bugzilla 1109, patch from Stephen Usher.
1009 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1011 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1014 TF/02 Fix a couple more cases where we did not log the error message
1015 when unlink() failed. See also change 4.74-TF/03.
1017 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1018 lock up or crash if it happened to be inside a call to libc when it
1019 got a SIGUSR1 from exiwhat.
1021 The SIGUSR1 handler appends the current process status to the process
1022 log which is later printed by exiwhat. It used to use the general
1023 purpose logging code to do this, but several functions it calls are
1024 not safe for signals.
1026 The new output code in the SIGUSR1 handler is specific to the process
1027 log, and simple enough that it's easy to inspect for signal safety.
1028 Removing some special cases also simplifies the general logging code.
1029 Removing the spurious timestamps from the process log simplifies
1032 TF/04 Improved ratelimit ACL condition.
1034 The /noupdate option has been deprecated in favour of /readonly which
1035 has clearer semantics. The /leaky, /strict, and /readonly update modes
1036 are mutually exclusive. The update mode is no longer included in the
1037 database key; it just determines when the database is updated. (This
1038 means that when you upgrde Exim will forget old rate measurements.)
1040 Exim now checks that the per_* options are used with an update mode that
1041 makes sense for the current ACL. For example, when Exim is processing a
1042 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1043 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1044 must specify per_mail/readonly. If you omit the update mode it defaults to
1045 /leaky where that makes sense (as before) or /readonly where required.
1047 The /noupdate option is now undocumented but still supported for
1048 backwards compatibility. It is equivalent to /readonly except that in
1049 ACLs where /readonly is required you may specify /leaky/noupdate or
1050 /strict/noupdate which are treated the same as /readonly.
1052 A useful new feature is the /count= option. This is a generalization
1053 of the per_byte option, so that you can measure the throughput of other
1054 aggregate values. For example, the per_byte option is now equivalent
1055 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1057 The per_rcpt option has been generalized using the /count= mechanism
1058 (though it's more complicated than the per_byte equivalence). When it is
1059 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1060 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1061 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1062 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1063 Note that using per_rcpt with a non-readonly update mode in more than
1064 one ACL will cause the recipients to be double-counted. (The per_mail
1065 and per_byte options don't have this problem.)
1067 The handling of very low rates has changed slightly. If the computed rate
1068 is less than the event's count (usually one) then this event is the first
1069 after a long gap. In this case the rate is set to the same as this event's
1070 count, so that the first message of a spam run is counted properly.
1072 The major new feature is a mechanism for counting the rate of unique
1073 events. The new per_addr option counts the number of different
1074 recipients that someone has sent messages to in the last time period. It
1075 behaves like per_rcpt if all the recipient addresses are different, but
1076 duplicate recipient addresses do not increase the measured rate. Like
1077 the /count= option this is a general mechanism, so the per_addr option
1078 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1079 example, measure the rate that a client uses different sender addresses
1080 with the options per_mail/unique=$sender_address. There are further
1081 details in the main documentation.
1083 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1085 TF/06 Removed a few PCRE remnants.
1087 TF/07 Automatically extract Exim's version number from tags in the git
1088 repository when doing development or release builds.
1090 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1091 Bugzilla 879. Patch from Paul Fisher.
1093 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1094 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1097 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1099 PP/05 Variable $av_failed, true if the AV scanner deferred.
1100 Bugzilla 1078. Patch from John Horne.
1102 PP/06 Stop make process more reliably on build failure.
1103 Bugzilla 1087. Patch from Heiko Schlittermann.
1105 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1106 Bugzilla 1089. Patch from Heiko Schlittermann.
1108 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1109 Bugzilla 1131. Patch from Holger Weiß.
1111 PP/09 Handle IPv6 addresses with SPF.
1112 Bugzilla 860. Patch from Wolfgang Breyha.
1114 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1116 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1119 PP/11 match_* no longer expand right-hand-side by default.
1120 New compile-time build option, EXPAND_LISTMATCH_RHS.
1121 New expansion conditions, "inlist", "inlisti".
1123 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1125 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1127 PP/14 fix log_write() format string regression from TF/03.
1128 Bugzilla 1152. Patch from Dmitry Isaikin.
1134 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1136 PP/02 Harmonised TLS library version reporting; only show if debugging.
1137 Layout now matches that introduced for other libraries in 4.74 PP/03.
1139 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1141 PP/04 New "dns_use_edns0" global option.
1143 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1146 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1147 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1149 TK/01 Updated PolarSSL code to 0.14.2.
1150 Bugzilla 1097. Patch from Andreas Metzler.
1152 PP/07 Catch divide-by-zero in ${eval:...}.
1153 Fixes bugzilla 1102.
1155 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1158 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1159 format-string attack -- SECURITY: remote arbitrary code execution.
1161 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1162 time unintentionally subject to list matching rules, letting the header
1163 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1164 arbitrary string expansion). This allowed for information disclosure.
1166 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1167 INT_MIN/-1 -- value coerced to INT_MAX.
1173 NM/01 Workround for PCRE version dependency in version reporting
1176 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1177 This fixes portability to compilers other than gcc, notably
1178 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1180 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1181 makefiles for portability to HP-UX and POSIX correctness.
1183 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1184 Also via indented variable definition in the Makefile.
1185 (Debugging by Oliver Heesakkers).
1187 PP/02 Restore caching of spamd results with expanded spamd_address.
1188 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1190 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1191 Improves build reliability. Fix from: Frank Elsner
1193 NM/02 Fix wide character breakage in the rfc2047 coding
1194 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1196 NM/03 Allow underscore in dnslist lookups
1197 Fixes bug 1026. Patch from Graeme Fowler
1199 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1200 Code patches from Adam Ciarcinski of NetBSD.
1202 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1205 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1206 is logged, to avoid truncation. Patch from John Horne.
1208 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1209 Patch from Jakob Hirsch.
1211 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1212 SQL string expansion failure details.
1213 Patch from Andrey Oktyabrski.
1215 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1216 Patch from Simon Arlott.
1218 PP/09 New lookups functionality failed to compile on old gcc which rejects
1219 extern declarations in function scope.
1220 Patch from Oliver Fleischmann
1222 PP/10 Use sig_atomic_t for flags set from signal handlers.
1223 Check getgroups() return and improve debugging.
1224 Fixed developed for diagnosis in bug 927 (which turned out to be
1227 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1228 Patch from Mark Zealey.
1230 PP/12 Bugzilla 1056: Improved spamd server selection.
1231 Patch from Mark Zealey.
1233 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1234 Based on patch from Heiko Schlittermann.
1236 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1237 Patch from Uwe Doering, sign-off by Michael Haardt.
1239 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1240 variable declaration deep within a block. Bug and patch from
1243 PP/15 lookups-Makefile IRIX compatibilty coercion.
1245 PP/16 Make DISABLE_DKIM build knob functional.
1247 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1248 Patch by Simon Arlott
1250 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1251 variable argument macros. Our copy now differs from upstream.
1257 TF/01 Failure to get a lock on a hints database can have serious
1258 consequences so log it to the panic log.
1260 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1261 controlled using the smtp_confirmation log selector.
1263 TF/03 Include the error message when we fail to unlink a spool file.
1265 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1266 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1267 for maintaining out-of-tree patches for some time.
1269 PP/01 Bugzilla 139: Documentation and portability issues.
1270 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1271 Handle per-OS dynamic-module compilation flags.
1273 PP/02 Let /dev/null have normal permissions.
1274 The 4.73 fixes were a little too stringent and complained about the
1275 permissions on /dev/null. Exempt it from some checks.
1276 Reported by Andreas M. Kirchwitz.
1278 PP/03 Report version information for many libraries, including
1279 Exim version information for dynamically loaded libraries. Created
1280 version.h, now support a version extension string for distributors
1281 who patch heavily. Dynamic module ABI change.
1283 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1284 privilege escalation vulnerability whereby the Exim run-time user
1285 can cause root to append content of the attacker's choosing to
1288 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1291 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1292 If dropping privileges for untrusted macros, we disabled normal logging
1293 on the basis that it would fail; for the Exim run-time user, this is not
1294 the case, and it resulted in successful deliveries going unlogged.
1295 Fixed. Reported by Andreas Metzler.
1301 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1302 only prepend for the Resent-* case. Fixes regression introduced in
1303 Exim 4.70 by NM/22 for Bugzilla 607.
1305 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1306 increasing numbers of administrators be bitten by this.
1308 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1310 PP/03 Bugzilla 994: added openssl_options main configuration option.
1312 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1314 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1316 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1318 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1319 without a peer certificate, leading to a segfault because of an
1320 assumption that peers always have certificates. Be a little more
1321 paranoid. Problem reported by Martin Tscholak.
1323 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1324 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1325 NB: ClamAV planning to remove STREAM in "middle of 2010".
1326 CL also introduces -bmalware, various -d+acl logging additions and
1327 more caution in buffer sizes.
1329 PP/09 Implemented reverse_ip expansion operator.
1331 PP/10 Bugzilla 937: provide a "debug" ACL control.
1333 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1335 PP/12 Bugzilla 973: Implement --version.
1337 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1339 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1341 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1343 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1344 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1345 ignore trailing whitespace.
1347 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1349 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1352 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1353 Notification from Dr Andrew Aitchison.
1355 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1356 ExtendedDetectionInfo response format.
1357 Notification from John Horne.
1359 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1362 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1363 XSL and documented dependency on system catalogs, with examples of how
1366 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1369 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1370 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1371 configuration file which is writeable by the Exim user or group.
1373 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1374 of configuration files to cover files specified with the -C option if
1375 they are going to be used with root privileges, not just the default
1378 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1379 option (effectively making it always true).
1381 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1382 files to be used while preserving root privileges.
1384 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1385 that rogue child processes cannot use them.
1387 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1388 run-time user, instead of root.
1390 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1391 Exim run-time user without dropping privileges.
1393 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1394 result string, instead of calling string_vformat() twice with the same
1397 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1398 for other users. Others should always drop root privileges if they use
1399 -C on the command line, even for a whitelisted configure file.
1401 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1403 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1409 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1410 $data_path, and $header_path variables; fixed documentation bugs and
1413 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1414 exipick to access non-standard spools, including the "frozen" queue
1417 NM/01 Bugzilla 965: Support mysql stored procedures.
1418 Patch from Alain Williams
1420 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1422 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1423 Patch from Andreas Metzler
1425 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1426 Patch from Kirill Miazine
1428 NM/05 Bugzilla 671: Added umask to procmail example.
1430 JJ/03 installed exipick 20100323.0, fixing doc bug
1432 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1433 directory. Notification and patch from Dan Rosenberg.
1435 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1437 TK/02 Improve log output when DKIM signing operation fails.
1439 MH/01 Treat the transport option dkim_domain as a colon separated
1440 list, not as a single string, and sign the message with each element,
1441 omitting multiple occurences of the same signer.
1443 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1444 Bugzilla 985, 986. Patch by Simon Arlott
1446 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1447 Patch by Simon Arlott
1449 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1450 MBX locking. Notification from Dan Rosenberg.
1456 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1458 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1460 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1462 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1464 NM/04 Bugzilla 915: Flag broken perl installation during build.
1470 TK/01 Added patch by Johannes Berg that expands the main option
1471 "spamd_address" if it starts with a dollar sign.
1473 TK/02 Write list of recipients to X-Envelope-Sender header when building
1474 the mbox-format spool file for content scanning (suggested by Jakob
1477 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1478 (http://www.dcc-servers.net/) support via dccifd. Activated by
1479 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1481 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1482 by Mark Daniel Reidel <mr@df.eu>.
1484 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1485 When building exim an external PCRE library is now needed -
1486 PCRE is a system library on the majority of modern systems.
1487 See entry on PCRE_LIBS in EDITME file.
1489 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1490 conversation. Added nologin parameter to request.
1491 Patch contributed by Kirill Miazine.
1493 TF/01 Do not log submission mode rewrites if they do not change the address.
1495 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1497 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1498 log files in place. Contributed by Roberto Lima.
1500 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1502 TF/03 Bugzilla 615: When checking the local_parts router precondition
1503 after a local_part_suffix or local_part_prefix option, Exim now
1504 does not use the address's named list lookup cache, since this
1505 contains cached lookups for the whole local part.
1507 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1508 Robert Millan. Documentation is in experimental-spec.txt.
1510 TF/04 Bugzilla 668: Fix parallel build (make -j).
1512 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1514 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1515 Patch provided by Jan Srzednicki.
1517 TF/05 Leading white space used to be stripped from $spam_report which
1518 wrecked the formatting. Now it is preserved.
1520 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1521 that they are available at delivery time.
1523 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1525 TF/08 TLS error reporting now respects the incoming_interface and
1526 incoming_port log selectors.
1528 TF/09 Produce a more useful error message if an SMTP transport's hosts
1529 setting expands to an empty string.
1531 NM/06 Bugzilla 744: EXPN did not work under TLS.
1532 Patch provided by Phil Pennock.
1534 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1535 Patch provided by Richard Godbee.
1537 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1538 acl_smtp_notquit, added index entry.
1540 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1541 Patch provided by Eugene Bujak.
1543 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1544 accept(). Patch provided by Maxim Dounin.
1546 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1547 Patch provided by Phil Pennock.
1549 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1551 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1552 Patch provided by Brad "anomie" Jorsch.
1554 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1555 Patch provided by Dean Brooks.
1557 TK/05 Add native DKIM support (does not depend on external libraries).
1559 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1560 Patch provided by Graeme Fowler.
1562 NM/16 Bugzilla 851: Documentation example syntax fix.
1564 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1566 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1569 NM/19 Bugzilla 745: TLS version reporting.
1570 Patch provided by Phil Pennock.
1572 NM/20 Bugzilla 167: bool: condition support.
1573 Patch provided by Phil Pennock.
1575 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1576 clients. Patch provided by Phil Pennock.
1578 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1579 Patch provided by Brad "anomie" Jorsch.
1581 NM/23 Bugzilla 687: Fix misparses in eximstats.
1582 Patch provided by Heiko Schlittermann.
1584 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1585 Patch provided by Heiko Schlittermann.
1587 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1588 plus update to original patch.
1590 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1592 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1593 Patch provided by David Brownlee.
1595 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1597 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1599 NM/30 Bugzilla 888: TLS documentation bugfixes.
1601 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1603 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1604 Unlike the original bugzilla I have changed all shell scripts in src tree.
1606 NM/33 Bugzilla 898: Transport filter timeout fix.
1607 Patch by Todd Rinaldo.
1609 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1610 Patch by Serge Demonchaux.
1612 NM/35 Bugzilla 39: Base64 decode bug fixes.
1613 Patch by Jakob Hirsch.
1615 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1617 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1619 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1621 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1627 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1628 ALT-N's libdkim that I have put here:
1629 http://duncanthrax.net/exim-experimental/
1631 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1632 were called 'true' and it seems that C99 defines that as a reserved
1633 keyword to be used with 'bool' variable types. That means you could
1634 not include C99-style headers which use bools without triggering
1635 build errors in sieve.c.
1637 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1638 as mailq or other aliases. Changed the --help handling significantly
1639 to do whats expected. exim_usage() emits usage/help information.
1641 SC/01 Added the -bylocaldomain option to eximstats.
1643 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1645 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1647 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1653 PH/01 Another patch from the Sieve maintainer.
1655 PH/02 When an IPv6 address is converted to a string for single-key lookup
1656 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1657 dots are used instead of colons so that keys in lsearch files need not
1658 contain colons. This was done some time before quoting was made available
1659 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1660 (notated using the quote facility) so as to distinguish them from IPv4
1661 keys. This meant that lookups for IP addresses in host lists did not work
1662 for iplsearch lookups.
1664 This has been fixed by arranging for IPv6 addresses to be expressed with
1665 colons if the lookup type is iplsearch. This is not incompatible, because
1666 previously such lookups could never work.
1668 The situation is now rather anomolous, since one *can* have colons in
1669 ordinary lsearch keys. However, making the change in all cases is
1670 incompatible and would probably break a number of configurations.
1672 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1675 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1676 conversion specification without a maximum field width, thereby enabling
1677 a rogue spamd server to cause a buffer overflow. While nobody in their
1678 right mind would setup Exim to query an untrusted spamd server, an
1679 attacker that gains access to a server running spamd could potentially
1680 exploit this vulnerability to run arbitrary code as the Exim user.
1682 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1683 $primary_hostname instead of what libspf2 thinks the hosts name is.
1685 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1686 a directory entry by the name of the lookup key. Previously, if a
1687 symlink pointed to a non-existing file or a file in a directory that
1688 Exim lacked permissions to read, a lookup for a key matching that
1689 symlink would fail. Now it is enough that a matching directory entry
1690 exists, symlink or not. (Bugzilla 503.)
1692 PH/03 The body_linecount and body_zerocount variables are now exported in the
1695 PH/04 Added the $dnslist_matched variable.
1697 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1698 This means they are set thereafter only if the connection becomes
1701 PH/06 Added the client_condition to authenticators so that some can be skipped
1702 by clients under certain conditions.
1704 PH/07 The error message for a badly-placed control=no_multiline_responses left
1705 "_responses" off the end of the name.
1707 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1709 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1710 (without spaces) instead of just copying the configuration text.
1712 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1714 PH/11 Added $max_received_linelength.
1716 PH/12 Added +ignore_defer and +include_defer to host lists.
1718 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1719 way in which PCRE > 7.0 is built.
1721 PH/14 Implemented queue_only_load_latch.
1723 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1724 MAIL command. The effect was to mangle the value on 64-bit systems.
1726 PH/16 Another patch from the Sieve maintainer.
1728 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1730 PH/18 If a system quota error occurred while trying to create the file for
1731 a maildir delivery, the message "Mailbox is full" was not appended to the
1732 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1733 only to a quota excession during the actual writing of the file.
1735 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1736 characters?) which causes problems in log lines. The DN values are now
1737 passed through string_printing() before being added to log lines.
1739 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1740 and InterBase are left for another time.)
1742 PH/21 Added message_body_newlines option.
1744 PH/22 Guard against possible overflow in moan_check_errorcopy().
1746 PH/23 POSIX allows open() to be a macro; guard against that.
1748 PH/24 If the recipient of an error message contained an @ in the local part
1749 (suitably quoted, of course), incorrect values were put in $domain and
1750 $local_part during the evaluation of errors_copy.
1756 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1757 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1760 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1761 issue a MAIL command.
1763 PH/02 In an ACL statement such as
1765 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1767 if a client was not listed at all, or was listed with a value other than
1768 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1769 the condition was not true (as it should be), so access was not denied.
1770 The bug was that the ! inversion was incorrectly passed on to the second
1771 item. This has been fixed.
1773 PH/03 Added additional dnslists conditions == and =& which are different from
1774 = and & when the dns lookup returns more than one IP address.
1776 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1777 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1779 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1780 FSYNC, which compiles an option called disable_fsync that allows for
1781 bypassing fsync(). The documentation is heavily laced with warnings.
1783 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1785 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1786 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1787 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1788 including adding "make clean"; (3) Added -fPIC when compiling the test
1789 dynamically loaded module, to get rid of a warning.
1791 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1792 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1793 = 0s. The bug is otherwise harmless.
1795 PH/07 There was a bug in the dovecot authenticator such that the value of
1796 $auth1 could be overwritten, and so not correctly preserved, after a
1797 successful authentication. This usually meant that the value preserved by
1798 the server_setid option was incorrect.
1800 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1802 PH/09 Installed PCRE release 7.0.
1804 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1805 run for batched SMTP input. It is now run at the start of every message
1806 in the batch. While fixing this I discovered that the process information
1807 (output by running exiwhat) was not always getting set for -bs and -bS
1808 input. This is fixed, and it now also says "batched" for BSMTP.
1810 PH/11 Added control=no_pipelining.
1812 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1813 patch, slightly modified), and move the expansion of helo_data till after
1814 the connection is made in the smtp transport (so it can use these
1817 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1819 PH/14 Added log_selector = +pid.
1821 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1823 PH/16 Add ${if forany and ${if forall.
1825 PH/17 Added dsn_from option to vary the From: line in DSNs.
1827 PH/18 Flush SMTP output before performing a callout, unless control =
1828 no_callout_flush is set.
1830 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1831 was true (the default) a successful delivery failed to delete the retry
1832 item, thus causing premature timeout of the address. The bug is now
1835 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1837 PH/21 Long custom messages for fakedefer and fakereject are now split up
1838 into multiline reponses in the same way that messages for "deny" and
1839 other ACL rejections are.
1841 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1842 with slight modification.
1844 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1845 draft, changing the syntax and factoring some duplicate code".
1847 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1848 for deliveries of the second and subsequent messages over the same SMTP
1851 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1852 ${reduce, with only minor "tidies".
1854 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1856 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1857 expansion side effects.
1859 PH/27 When a message times out after an over-quota error from an Exim-imposed
1860 quota, the bounce message says "mailbox is full". This message was not
1861 being given when it was a system quota that was exceeded. It now should
1864 MH/03 Made $recipients available in local_scan(). local_scan() already has
1865 better access to the recipient list through recipients_list[], but
1866 $recipients can be useful in postmaster-provided expansion strings.
1868 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1869 in the case of a MAIL command with additional options following the
1870 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1871 were accidentally chopped off.
1873 PH/29 SMTP synchronization checks are implemented when a command is read -
1874 there is a check that no more input is waiting when there shouldn't be
1875 any. However, for some commands, a delay in an ACL can mean that it is
1876 some time before the response is written. In this time, more input might
1877 arrive, invalidly. So now there are extra checks after an ACL has run for
1878 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1879 pipelining has not been advertised.
1881 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1883 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1884 returned in the SMTP response when smtp_return_error_details was false.
1885 This has been fixed.
1887 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1888 instead of the C I/O that was originally supplied, because problems were
1889 reported on Solaris.
1891 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1892 Exim which did not show up earlier: it was assuming that a call to
1893 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1894 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1895 was a macro that became an assignment, so it seemed to work. This has
1896 changed to a proper function call with a void return, hence the compile
1897 error. Exim's code has been fixed.
1899 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1902 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1904 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1906 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1907 because it turns out to be too big or there is a timeout) there is no
1908 "Completed" line in the log. When some messages of this type were
1909 selected by exigrep, they were listed as "not completed". Others were
1910 picked up by some special patterns. I have improved the selection
1911 criteria to be more general.
1913 PH/38 The host_find_failed option in the manualroute router can now be set
1914 to "ignore", to completely ignore a host whose IP address cannot be
1915 found. If all hosts are ignored, the behaviour is controlled by the new
1916 host_all_ignored option.
1918 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1919 homing or because of multiple MX records with /mx) generated more than
1920 one IP address, and the following item turned out to be the local host,
1921 all the secondary addresses of the first item were incorrectly removed
1922 from the list, along with the local host and any following hosts (which
1923 is what is supposed to happen).
1925 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1926 whoever called Exim into the -H file. In the case of the daemon it was
1927 behaving confusingly. When first started, it used values for whoever
1928 started the daemon, but after a SIGHUP it used the Exim user (because it
1929 calls itself on a restart). I have changed the code so that it now always
1932 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1933 message are rejected with the same error (e.g. no authentication or bad
1934 sender address), and a DATA command is nevertheless sent (as can happen
1935 with PIPELINING or a stupid MUA), the error message that was given to the
1936 RCPT commands is included in the rejection of the DATA command. This is
1937 intended to be helpful for MUAs that show only the final error to their
1940 PH/42 Another patch from the Sieve maintainer.
1942 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1943 Eximstats - Fixed some broken HTML links and added missing column headers
1945 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1946 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1948 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1950 PH/43 Yet another patch from the Sieve maintainer.
1952 PH/44 I found a way to check for a TCP/IP connection going away before sending
1953 the response to the final '.' that terminates a message, but only in the
1954 case where the client has not sent further data following the '.'
1955 (unfortunately, this is allowed). However, in many cases there won't be
1956 any further data because there won't be any more messages to send. A call
1957 to select() can be used: if it shows that the input is "ready", there is
1958 either input waiting, or the socket has been closed. An attempt to read
1959 the next input character can distinguish the two cases. Previously, Exim
1960 would have sent an OK response which the client would never have see.
1961 This could lead to message repetition. This fix should cure that, at
1962 least in a lot of common cases.
1964 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1965 advertised in response to EHLO.
1971 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1972 fixed by 4.65/MH/01 (is this a record?) are fixed:
1974 (i) An empty string was always treated as zero by the numeric comparison
1975 operators. This behaviour has been restored.
1977 (ii) It is documented that the numeric comparison operators always treat
1978 their arguments as decimal numbers. This was broken in that numbers
1979 starting with 0 were being interpreted as octal.
1981 While fixing these problems I realized that there was another issue that
1982 hadn't been noticed. Values of message_size_limit (both the global option
1983 and the transport option) were treated as octal if they started with 0.
1984 The documentation was vague. These values are now always treated as
1985 decimal, and I will make that clear in the documentation.
1991 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1992 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1995 MH/01 Don't check that the operands of numeric comparison operators are
1996 integers when their expansion is in "skipping" mode (fixes bug
1997 introduced by 4.64-PH/07).
1999 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2000 child addresses, Exim now panics and dies. Previously, because the count
2001 is held in a short int, deliveries were likely to be lost. As such a
2002 large number of recipients for a single message is ridiculous
2003 (performance will be very, very poor), I have chosen to impose a limit
2004 rather than extend the field.
2010 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2011 leftover -K file (the existence of which was triggered by #402).
2012 While we were at it, introduced process PID as part of the -K
2013 filename. This should rule out race conditions when creating
2016 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2017 processing considerably. Previous code took too long for large mails,
2018 triggering a timeout which in turn triggers #401.
2020 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2021 in the DK code in transports.c. sendfile() is not really portable,
2022 hence the _LINUX specificness.
2024 TF/01 In the add_headers option to the mail command in an Exim filter,
2025 there was a bug that Exim would claim a syntax error in any
2026 header after the first one which had an odd number of characters
2029 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2030 callout verification, Exim cached a "reject" for the entire domain. This
2031 is correct for most verifications, but it is not correct for a recipient
2032 verification with use_sender or use_postmaster set, because in that case
2033 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2034 case of MAIL FROM:<> rejection from other early rejections (e.g.
2035 rejection of HELO). When verifying a recipient using a non-null MAIL
2036 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2037 Whatever the result of the callout, the value of the domain cache is
2038 left unchanged (for any other kind of callout, getting as far as trying
2039 RCPT means that the domain itself is ok).
2041 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2044 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2045 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2046 session, the daemon ignores ECONNECT errors and logs others; it now
2047 ignores EPIPE as well.
2049 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2050 (quoted-printable decoding).
2052 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2053 later the small subsequent patch to fix an introduced bug.
2055 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2057 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2059 PH/08 An error is now given if message_size_limit is specified negative.
2061 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2062 to be given (somewhat) arbitrary names.
2064 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2067 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2068 miscellaneous code fixes
2070 PH/10 Added the log_reject_target ACL modifier to specify where to log
2073 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2074 hostname. This is wrong, because it relates to the incoming message (and
2075 probably the interface on which it is arriving) and not to the outgoing
2076 callout (which could be using a different interface). This has been
2077 changed to use the value of the helo_data option from the smtp transport
2078 instead - this is what is used when a message is actually being sent. If
2079 there is no remote transport (possible with a router that sets up host
2080 addresses), $smtp_active_hostname is used.
2082 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2083 tweaks were necessary in order to get it to work (see also 21 below):
2084 (a) The code assumed that strncpy() returns a negative number on buffer
2085 overflow, which isn't the case. Replaced with Exim's string_format()
2087 (b) There were several signed/unsigned issues. I just did the minimum
2088 hacking in of casts. There is scope for a larger refactoring.
2089 (c) The code used strcasecmp() which is not a standard C function.
2090 Replaced with Exim's strcmpic() function.
2091 (d) The code set only $1; it now sets $auth1 as well.
2092 (e) A simple test gave the error "authentication client didn't specify
2093 service in request". It would seem that Dovecot has changed its
2094 interface. Fortunately there's a specification; I followed it and
2095 changed what the client sends and it appears to be working now.
2097 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2100 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2101 address A is aliased to B and C, where B exists and C does not. Without
2102 -v the output is "A verified" because verification stops after a
2103 successful redirection if more than one address is generated. However,
2104 with -v the child addresses are also verified. Exim was outputting "A
2105 failed to verify" and then showing the successful verification for C,
2106 with its parentage. It now outputs "B failed to verify", showing B's
2107 parentage before showing the successful verification of C.
2109 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2110 look up a TXT record in a specific list after matching in a combined
2113 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2114 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2115 they consult the DNS. I had assumed they would set it the way they
2116 wanted; and indeed my experiments on Linux seem to show that in some
2117 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2118 To be on the safe side, however, I have now made the interface to
2119 host_find_byname() similar to host_find_bydns(), with an argument
2120 containing the DNS resolver options. The host_find_byname() function now
2121 sets these options at its start, just as host_find_bydns() does. The smtp
2122 transport options dns_qualify_single and dns_search_parents are passed to
2123 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2124 of host_find_byname() use the default settings of RES_DEFNAMES
2125 (qualify_single) but not RES_DNSRCH (search_parents).
2127 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2128 spool_read_header() do less string testing, by means of a preliminary
2129 switch on the second character of optional "-foo" lines. (This is
2130 overdue, caused by the large number of possibilities that now exist.
2131 Originally there were few.) While I was there, I also converted the
2132 str(n)cmp tests so they don't re-test the leading "-" and the first
2133 character, in the hope this might squeeze out yet more improvement.
2135 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2136 flag allowing group syntax was set by the header_syntax check but not
2137 turned off, possible causing trouble later; (2) The flag was not being
2138 set at all for the header_verify test, causing "group"-style headers to
2139 be rejected. I have now set it in this case, and also caused header_
2140 verify to ignore an empty address taken from a group. While doing this, I
2141 came across some other cases where the code for allowing group syntax
2142 while scanning a header line wasn't quite right (mostly, not resetting
2143 the flag correctly in the right place). These bugs could have caused
2144 trouble for malformed header lines. I hope it is now all correct.
2146 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2147 with the "reply" argument non-NULL. The code, however (which originally
2148 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2149 but it didn't always do it. This confused somebody who was copying the
2150 code for some other use. I have removed all the tests.
2152 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2153 feature that was used to support insecure browsers during the U.S. crypto
2154 embargo. It requires special client support, and Exim is probably the
2155 only MTA that supported it -- and would never use it because real RSA is
2156 always available. This code has been removed, because it had the bad
2157 effect of slowing Exim down by computing (never used) parameters for the
2158 RSA_EXPORT functionality.
2160 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2161 authenticator to fail if there's a tab character in the incoming data
2162 (there should never be unless someone is messing about, as it's supposed
2163 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2164 if the connection is using TLS or if the remote IP is the same as the
2165 local IP, and the "valid-client-cert option" if a client certificate has
2168 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2169 authenticators. This can be used for authorization after authentication
2170 succeeds. (In the case of plaintext, it servers for both authentication
2173 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2174 if any retry times were supplied.
2176 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2177 connection, where there is no client IP address to check. In this
2178 situation, the verify now always succeeds.
2180 PH/25 Applied John Jetmore's -Mset patch.
2182 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2184 PH/27 In a string expansion for a processed (not raw) header when multiple
2185 headers of the same name were present, leading whitespace was being
2186 removed from all of them, but trailing whitespace was being removed only
2187 from the last one. Now trailing whitespace is removed from each header
2188 before concatenation. Completely empty headers in a concatenation (as
2189 before) are ignored.
2191 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2192 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2194 PH/29 [Removed. This was a change that I later backed out, and forgot to
2195 correct the ChangeLog entry (that I had efficiently created) before
2196 committing the later change.]
2198 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2199 address errors (4xx response to RCPT) over the same connection as other
2200 messages routed to the same hosts. Such deliveries are always "forced",
2201 so retry times are not inspected. This resulted in far too many retries
2202 for the affected addresses. The effect occurred only when there were more
2203 hosts than the hosts_max_try setting in the smtp transport when it had
2204 the 4xx errors. Those hosts that it had tried were not added to the list
2205 of hosts for which the message was waiting, so if all were tried, there
2206 was no problem. Two fixes have been applied:
2208 (i) If there are any address or message errors in an SMTP delivery, none
2209 of the hosts (tried or untried) are now added to the list of hosts
2210 for which the message is waiting, so the message should not be a
2211 candidate for sending over the same connection that was used for a
2212 successful delivery of some other message. This seems entirely
2213 reasonable: after all the message is NOT "waiting for some host".
2214 This is so "obvious" that I'm not sure why it wasn't done
2215 previously. Hope I haven't missed anything, but it can't do any
2216 harm, as the worst effect is to miss an optimization.
2218 (ii) If, despite (i), such a delivery is accidentally attempted, the
2219 routing retry time is respected, so at least it doesn't keep
2220 hammering the server.
2222 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2223 in ${readsocket because some servers need this prod.
2225 PH/32 Added some extra debug output when updating a wait-xxx database.
2227 PH/33 The hint "could be header name not terminated by colon", which has been
2228 given for certain expansion errors for a long time, was not being given
2229 for the ${if def:h_colon_omitted{... case.
2231 PH/34 The spec says: "With one important exception, whenever a domain list is
2232 being scanned, $domain contains the subject domain." There was at least
2233 one case where this was not true.
2235 PH/35 The error "getsockname() failed: connection reset by peer" was being
2236 written to the panic log as well as the main log, but it isn't really
2237 panic-worthy as it just means the connection died rather early on. I have
2238 removed the panic log writing for the ECONNRESET error when getsockname()
2241 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2242 runs only) independently of the message's sender address. This meant
2243 that, if the 4xx error was in fact related to the sender, a different
2244 message to the same recipient with a different sender could confuse
2245 things. In particualar, this can happen when sending to a greylisting
2246 server, but other circumstances could also provoke similar problems.
2247 I have changed the default so that the retry time for these errors is now
2248 based a combination of the sender and recipient addresses. This change
2249 can be overridden by setting address_retry_include_sender=false in the
2252 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2253 remote server are returned as part of bounce messages. This was not
2254 happening for LMTP over a pipe (the lmtp transport), but now it is the
2255 same for both kinds of LMTP.
2257 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2258 in header lines that were in fact CNAMEs. This is no longer the case.
2260 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2261 and queue runs started by the daemon processed all messages. This has
2262 been fixed so that -R and -S can now usefully be given with -q<time>.
2264 PH/40 Import PCRE release 6.7 (fixes some bugs).
2266 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2268 PH/42 Give an error if -q is specified more than once.
2270 PH/43 Renamed the variables $interface_address and $interface_port as
2271 $received_ip_address and $received_port, to make it clear that these
2272 values apply to message reception, and not to the outgoing interface when
2273 a message is delivered. (The old names remain recognized, of course.)
2275 PH/44 There was no timeout on the connect() call when using a Unix domain
2276 socket in the ${readsocket expansion. There now is.
2278 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2279 be meaningful with "accept".
2281 SC/01 Eximstats V1.43
2282 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2284 SC/02 Eximstats V1.44
2285 Use a glob alias rather than an array ref in the generated
2286 parser. This improves both readability and performance.
2288 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2289 Collect SpamAssassin and rejection statistics.
2290 Don't display local sender or destination tables unless
2291 there is data to show.
2292 Added average volumes into the top table text output.
2294 SC/04 Eximstats V1.46
2295 Collect data on the number of addresses (recipients)
2296 as well as the number of messages.
2298 SC/05 Eximstats V1.47
2299 Added 'Message too big' to the list of mail rejection
2300 reasons (thanks to Marco Gaiarin).
2302 SC/06 Eximstats V1.48
2303 Mainlog lines which have GMT offsets and are too short to
2304 have a flag are now skipped.
2306 SC/07 Eximstats V1.49 (Alain Williams)
2307 Added the -emptyok flag.
2309 SC/08 Eximstats V1.50
2310 Fixes for obtaining the IP address from reject messages.
2312 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2313 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2314 whitesspace changes from 4.64-PH/27
2316 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2319 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2320 are found, allow negative numbers in numeric criteria)
2322 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2324 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2325 to match changes made in 4.64-PH/43
2327 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2329 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2330 "domain" is actually an IP address, and force a failure. This locks out
2331 those revolvers/nameservers that support "A-for-A" lookups, in
2332 contravention of the specifications.
2334 PH/48 When a host name was looked up from an IP address, and the subsequent
2335 forward lookup of the name timed out, the host name was left in
2336 $sender_host_name, contrary to the specification.
2338 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2339 restricted to single-key lookups, Exim was not diagnosing an error if
2340 * or *@ was used with a query-style lookup.
2342 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2344 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2345 long ago, but noone interested enough thought of it. Let's just say that
2346 the "1.1" means that there are some new functions that weren't there at
2347 some point in the past.
2349 PH/51 Error processing for expansion failure of helo_data from an smtp
2350 transport during callout processing was broken.
2352 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2353 tested/used via the -bh/-bhc/-bs options.
2355 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2356 bug, fixed in subsequent PCRE releases).
2358 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2359 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2361 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2367 SC/01 Use a glob alias rather than an array ref in eximstats generated
2368 parser. This improves both readability and performance.
2370 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2371 Don't display local sender or destination tables in eximstats unless
2372 there is data to show.
2373 Added average volumes into the eximstats top table text output.
2375 SC/03 Collect data on the number of addresses (recipients) as well
2376 as the number of messages in eximstats.
2378 TF/01 Correct an error in the documentation for the redirect router. Exim
2379 does (usually) call initgroups() when daemonizing.
2381 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2382 with consistent privilege compared to when running as a daemon.
2384 TF/03 Note in the spec that $authenticated_id is not set for local
2385 submissions from trusted users.
2387 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2388 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2390 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2391 by adding some example configuration directives to the default
2392 configuration file. A little bit of work is required to uncomment the
2393 directives and define how usernames and passwords are checked, but
2394 there is now a framework to start from.
2396 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2397 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2398 without this. I don't know how relevant this is to other LDAP libraries.
2400 PH/02 Add the verb name to the "unknown ACL verb" error.
2402 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2404 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2406 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2407 directory not expanded when it should be if an expanded home directory
2408 was set for the address (which is overridden by the transport).
2410 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2413 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2414 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2415 because it is too late at that time, and has no effect.
2417 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2418 security issue with \' (bugzilla #107). I could not use the
2419 PQescapeStringConn() function, because it needs a PGconn value as one of
2422 PH/08 When testing addresses using -bt, indicate those final addresses that
2423 are duplicates that would not cause an additional delivery. At least one
2424 person was confused, thinking that -bt output corresponded to deliveries.
2425 (Suppressing duplicates isn't a good idea as you lose the information
2426 about possibly different redirections that led to the duplicates.)
2428 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2429 systems where poll() doesn't work, in particular OS X.
2431 PH/10 Added more information to debugging output for retry time not reached.
2433 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2434 operations in malware.c.
2436 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2439 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2440 syslog_duplication set false, log lines that would normally be written
2441 both the the main log and to the reject log were not written to syslog at
2444 PH/14 In the default configuration, change the use of "message" in ACL warn
2445 statements to "add_header".
2447 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2448 not followed by a command (e.g. "seen endif").
2450 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2451 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2454 PH/17 Added extra conditions to the default value of delay_warning_condition
2458 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2459 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2460 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2463 The Auto-Submitted: and various List- headers are standardised, whereas I
2464 don't think Precedence: ever was.
2466 PH/18 Refactored debugging code in route_finduser() to show more information,
2467 in particular, the error code if getpwnam() issues one.
2469 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2470 This is apparently needed in addition to the PH/07 change above to avoid
2471 any possible encoding problems.
2473 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2474 but not after initializing Perl.
2476 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2477 output them only if debugging. By default they are written stderr,
2478 apparently, which is not desirable.
2480 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2483 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2486 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2488 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2489 authenticated or an ident call has been made. Suppress the default
2490 values for $authenticated_id and $authenticated_sender (but permit -oMai
2491 and -oMas) when testing with -bh.
2493 PH/24 Re-jigged the order of the tests in the default configuration so that the
2494 tests for valid domains and recipients precede the DNS black list and CSA
2495 tests, on the grounds that those ones are more expensive.
2497 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2498 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2499 This bug exists in every version of Exim that I still have, right back to
2502 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2503 However, an attempt to turn on case-sensitivity in a regex key by
2504 including (?-i) didn't work because the subject string was already
2505 lowercased, and the effects were non-intuitive. It turns out that a
2506 one-line patch can be used to allow (?-i) to work as expected.
2512 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2513 other effects) broke the use of negated acl sub-conditions.
2515 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2518 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2519 "Deny" causes Exim to reject the incoming connection with a 554 error.
2520 Unfortunately, if there is a major crisis, such as a disk failure,
2521 tcp-wrappers gives "deny", whereas what one would like would be some
2522 kind of temporary error. A kludge has been added to help with this.
2523 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2524 554 error is used if errno is still zero or contains ENOENT (which occurs
2525 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2528 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2530 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2531 errors. Otherwise a message that provokes a temporary error (when other
2532 messages do not) can cause a whole host to time out.
2534 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2535 addresses were routed directly to files or pipes from a redirect router.
2536 File deliveries just didn't batch; pipe deliveries might have suffered
2539 PH/06 A failure to get a lock for a hints database would erroneously always say
2540 "Failed to get write lock", even when it was really a read lock.
2542 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2543 of 0600. This has been changed to use the value of the lockfile_mode
2544 option (which defaults to 0600).
2546 PH/08 Applied small patch from the Sieve maintainer.
2548 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2549 folder from quota calculations, a direct delivery into this folder messed
2550 up the contents of the maildirsize file. This was because the regex was
2551 used only to exclude .Trash (or whatever) when the size of the mailbox
2552 was calculated. There was no check that a delivery was happening into an
2553 excluded directory. This bug has been fixed by ignoring all quota
2554 processing for deliveries into excluded directories.
2556 PH/10 Added the maildirfolder_create_regex option to appendfile.
2562 PH/01 The code for finding all the local interface addresses on a FreeBSD
2563 system running IPv6 was broken. This may well have applied to all BSD
2564 systems, as well as to others that have similar system calls. The broken
2565 code found IPv4 interfaces correctly, but gave incorrect values for the
2566 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2567 that it would not match correctly against @[] and not recognize the IPv6
2570 PH/02 The ipliteral router was not recognizing addresses of the form user@
2571 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2573 PH/03 Added disable_ipv6.
2575 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2576 decoded form, because it is most often used to construct To: headers
2577 lines in autoreplies, and the decoded form may well be syntactically
2578 invalid. However, $reply_address has leading white space removed, and all
2579 newlines turned into spaces so that the autoreply transport does not
2582 PH/05 If group was specified without a user on a router, and no group or user
2583 was specified on a transport, the group from the router was ignored.
2585 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2586 for visible compile-time settings that can be used to change these
2587 numbers, for those that want even more. Backwards compatibility with old
2588 spool files has been maintained. However, going back to a previous Exim
2589 release will lost any variables that are in spool files.
2591 PH/07 Two small changes when running in the test harness: increase delay when
2592 passing a TCP/IP connection to a new process, in case the original
2593 process has to generate a bounce, and remove special handling of
2594 127.0.0.2 (sic), which is no longer necessary.
2596 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2597 be the same on different OS.
2599 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2602 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2603 whether --show-vars was specified or not
2605 JJ/02 exipick: Added support for new ACL variable spool format introduced
2608 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2609 syntactically invalid From: or Reply-to: line, and a filter used this to
2610 generate an autoreply, and therefore failed to obtain an address for the
2611 autoreply, Exim could try to deliver to a non-existent relative file
2612 name, causing unrelated and misleading errors. What now happens is that
2613 it logs this as a hard delivery error, but does not attempt to create a
2616 PH/11 The exinext utility has a -C option for testing purposes, but although
2617 the given file was scanned by exinext itself; it wasn't being passed on
2618 when Exim was called.
2620 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2621 an end-of-file indication when reading a command response.
2623 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2624 compiled. In many other places in Exim, IPv6 addresses are always
2625 recognized, so I have changed this. It also means that IPv4 domain
2626 literals of the form [IPV4:n.n.n.n] are now always recognized.
2628 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2629 used if the router is not running as root, for example, when verifying at
2630 ACL time, or when using -bh. The debugging output from this situation was
2631 non-existent - all you got was a failure to exec. I have made two
2634 (a) Failures to set uid/gid, the current directory, or a process leader
2635 in a subprocess such as that created by queryprogram now generate
2636 suitable debugging ouput when -d is set.
2638 (b) The queryprogram router detects when it is not running as root,
2639 outputs suitable debugging information if -d is set, and then runs
2640 the subprocess without attempting to change uid/gid.
2642 PH/15 Minor change to Makefile for building test_host (undocumented testing
2645 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2646 additional section of a DNS packet that returns MX or SRV records.
2647 Instead, it always explicitly searches for A/AAAA records. This avoids
2648 major problems that occur when a DNS server includes only records of one
2649 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2650 fixed another bug: if SRV records were looked up and the corresponding
2651 address records were *not* found in the additional section, the port
2652 values from the SRV records were lost.
2654 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2655 using the correct key (the original address) when searching the retry
2656 rules in order to find which one to use for generating the retry hint.
2658 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2659 adding the default one. Similarly, if it contains a Reply-To: header, the
2660 errors_reply_to option, if set, is not used.
2662 PH/19 When calculating a retry time, Exim used to measure the "time since
2663 failure" by looking at the "first failed" field in the retry record. Now
2664 it does not use this if it is later than than the arrival time of the
2665 message. Instead it uses the arrival time. This makes for better
2666 behaviour in cases where some deliveries succeed, thus re-setting the
2667 "first failed" field. An example is a quota failure for a huge message
2668 when small messages continue to be delivered. Without this change, the
2669 "time since failure" will always be short, possible causing more frequent
2670 delivery attempts for the huge message than are intended.
2671 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2673 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2674 $1, $2, $3) because the numerical variables can be reset during some
2675 expansion items (e.g. "match"), thereby losing the authentication data.
2677 PH/21 Make -bV show the size of off_t variables so that the test suite can
2678 decide whether to run tests for quotas > 2G.
2680 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2681 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2682 filecount value is greater than 2G or if a quota value is greater than 2G
2683 on a system where the size of off_t is not greater than 4, a panic error
2686 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2687 never match. The debug and -bh output now contains an explicit error
2688 message indicating a malformed IPv4 address or mask.
2690 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2691 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2692 PH/23 above applies.
2694 PH/25 Do not write to syslog when running in the test harness. The only
2695 occasion when this arises is a failure to open the main or panic logs
2696 (for which there is an explicit test).
2698 PH/26 Added the /no_tell option to "control=freeze".
2700 PH/27 If a host name lookup failed very early in a connection, for example, if
2701 the IP address matched host_lookup and the reverse lookup yielded a name
2702 that did not have a forward lookup, an error message of the form "no IP
2703 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2704 could be logged. Now it outputs the IP address instead of "NULL".
2706 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2707 allows the sender and the authenticated sender to be set when
2708 submitting a message from within Exim. Since child_open_exim() is
2709 documented for local_scan(), the new function should be too.
2711 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2712 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2713 results in an empty string is now treated as unset.
2715 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2717 PH/31 Added sender_verify_fail logging option.
2719 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2720 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2721 tidied the source and removed it altogether.
2723 PH/33 When a queue run was abandoned because the load average was too high, a
2724 log line was always written; now it is written only if the queue_run log
2725 selector is set. In addition, the log line for abandonment now contains
2726 information about the queue run such as the pid. This is always present
2727 in "start" and "stop" lines but was omitted from the "abandon" line.
2729 PH/34 Omit spaces between a header name and the colon in the error message that
2730 is given when verify = headers_syntax fails (if there are lots of them,
2731 the message gets confusing).
2733 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2734 names, as there are now some PTR records that contain slashes. This check
2735 is only to protect against broken name servers that fall over on strange
2736 characters, so the fact that it applies to all lookups doesn't matter.
2738 PH/36 Now that the new test suite is complete, we can remove some of the
2739 special code in Exim that was needed for the old test suite. For example,
2740 sorting DNS records because real resolvers return them in an arbitrary
2741 order. The new test suite's fake resolver always returns records in the
2744 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2745 bounces) except when queue_only is set, to avoid logging races between
2746 the different processes.
2748 PH/38 Panic-die if .include specifies a non-absolute path.
2750 PH/39 A tweak to the "H" retry rule from its user.
2752 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2753 a label. They prevented compilation on older perls.
2755 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2756 a warning to be raised on newish perls.
2758 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2759 on queue. Changes to match documented behaviour of showing count of
2760 messages matching specified criteria.
2762 PH/40 Changed the default ident timeout from 30s to 5s.
2764 PH/41 Added support for the use of login_cap features, on those BSD systems
2765 that have them, for controlling the resources used by pipe deliveries.
2767 PH/42 The content-scanning code uses fopen() to create files in which to put
2768 message data. Previously it was not paying any attention to the mode of
2769 the files. Exim runs with umask(0) because the rest of the code creates
2770 files with open(), and sets the required mode explicitly. Thus, these
2771 files were ending up world-writeable. This was not a big issue, because,
2772 being within the spool directory, they were not world-accessible. I have
2773 created a function called modefopen, which takes an additional mode
2774 argument. It sets umask(777), creates the file, chmods it to the required
2775 mode, then resets the umask. All the relevant calls to fopen() in the
2776 content scanning code have been changed to use this function.
2778 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2779 to 24 hours. This avoids potential overflow problems when processing G
2780 and H retry rules. I suspect nobody ever tinkers with this value.
2782 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2784 PH/45 When the plaintext authenticator is running as a client, the server's
2785 challenges are checked to ensure they are valid base64 strings. By
2786 default, the authentication attempt is cancelled if an invalid string is
2787 received. Setting client_ignore_invalid_base64 true ignores these errors.
2788 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2789 they are received. Thus, the responses can be made to depend on the
2790 challenges. If an invalid string is ignored, an empty string is placed in
2793 PH/46 Messages that are created by the autoreply transport now contains a
2794 References: header, in accordance with RFCs 2822 and 3834.
2796 PH/47 Added authenticated_sender_force to the smtp transport.
2798 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2800 PH/49 Installed latest patch from the Sieve maintainer.
2802 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2803 was also set, the appendfile transport was unnecessarily scanning a
2804 directory of message files (e.g. for maildir delivery) to find the count
2805 of files (along with the size), even though it did not need this
2806 information. It now does the scan only if it needs to find either the
2807 size of the count of files.
2809 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2811 PH/52 Two bugs concerned with error handling when the smtp transport is
2814 (i) Exim was not creating retry information for temporary errors given
2815 for individual recipients after the DATA command when the smtp transport
2816 was used in LMTP mode. This meant that they could be retried too
2817 frequently, and not timed out correctly.
2819 (ii) Exim was setting the flag that allows error details to be returned
2820 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2821 recipients that were returned after the DATA command.
2823 PH/53 This is related to PH/52, but is more general: for any failing address,
2824 when detailed error information was permitted to be returned to the
2825 sender, but the error was temporary, then after the final timeout, only
2826 "retry timeout exceeded" was returned. Now it returns the full error as
2827 well as "retry timeout exceeded".
2829 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2830 do this, and (what is worse) MTAs that accept it.
2832 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2833 will now be deprecated.
2835 PH/56 New os.c-cygwin from the Cygwin maintainer.
2837 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2838 formats (previously only available in exim formats via -bpr, -bpru,
2839 and -bpra. Now also available in native and exiqgrep formats)
2841 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2842 with very large, slow to parse queues
2844 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2846 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2848 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2849 responses to authentication challenges, though it was showing the
2850 challenges; (ii) I've removed the CR characters from the debug output for
2853 PH/58 Allow for the insertion of a newline as well as a space when a string
2854 is turned into more than one encoded-word during RFC 2047 encoding. The
2855 Sieve code now uses this.
2857 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2858 data_4xx, lost_connection, tls_required.
2860 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2861 message was being sent as an SMTP response.
2863 PH/61 Add -l and -k options to exicyclog.
2865 PH/62 When verifying, if an address was redirected to one new address, so that
2866 verification continued, and the new address failed or deferred after
2867 having set something in $address_data, the value of $address_data was not
2868 passed back to the ACL. This was different to the case when no
2869 redirection occurred. The value is now passed back in both cases.
2871 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2872 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2873 use login_cap.h, so on its own it isn't the distinguishing feature. The
2874 new name refers directly to the setclassresources() function.
2876 PH/65 Added configuration files for NetBSD3.
2878 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2880 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2881 is preferred over IPv4.
2883 PH/68 The bounce_return_message and bounce_return_body options were not being
2884 honoured for bounces generated during the reception of non-SMTP messages.
2885 In particular, this applied to messages rejected by the ACL. This bug has
2886 been fixed. However, if bounce_return_message is true and bounce_return_
2887 body is false, the headers that are returned for a non-SMTP message
2888 include only those that have been read before the error was detected.
2889 (In the case of an ACL rejection, they have all been read.)
2891 PH/69 The HTML version of the specification is now built in a directory called
2892 spec_html instead of spec.html, because the latter looks like a path with
2893 a MIME-type, and this confuses some software.
2895 PH/70 Catch two compiler warnings in sieve.c.
2897 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2898 function verify_get_ident() calls ip_connect() to connect a socket, but
2899 if the "connect()" function timed out, ip_connect() used to close the
2900 socket. However, verify_get_ident() also closes the socket later, and in
2901 between Exim writes to the log, which may get opened at this point. When
2902 the socket was closed in ip_connect(), the log could get the same file
2903 descriptor number as the socket. This naturally causes chaos. The fix is
2904 not to close the socket in ip_connect(); the socket should be closed by
2905 the function that creates it. There was only one place in the code where
2906 this was missing, in the iplookup router, which I don't think anybody now
2907 uses, but I've fixed it anyway.
2909 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2910 well as to direct DNS lookups. Otherwise the handling of names in host
2911 lists is inconsistent and therefore confusing.
2917 PH/01 Two changes to the default runtime configuration:
2919 (1) Move the checks for relay_from_hosts and authenticated clients from
2920 after to before the (commented out) DNS black list checks.
2922 (2) Add control=submission to the relay_from_hosts and authenticated
2923 clients checks, on the grounds that messages accepted by these
2924 statements are most likely to be submissions.
2926 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2928 (1) Generate an error if the third argument for the ${prvs expansion is
2931 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2934 (3) Reset the variables that are obtained from the first argument of
2935 ${prvscheck and used in the second argument before leaving the code,
2936 because their memory is reclaimed, so using them afterwards may do
2939 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2940 one (it's much easier than Tom thought :-).
2942 (5) Because of (4), we can now allow for the use of $prvscheck_result
2943 inside the third argument.
2945 PH/03 For some reason, the default setting of PATH when running a command from
2946 a pipe transport was just "/usr/bin". I have changed it to
2949 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2950 anything to be listed in the output from -bV.
2952 PH/05 When a filter generated an autoreply, the entire To: header line was
2953 quoted in the delivery log line, like this:
2955 => >A.N.Other <ano@some.domain> <original@ddress> ...
2957 This has been changed so that it extracts the operative address. There
2958 may be more than one such address. If so, they are comma-separated, like
2961 => >ano@some.domain,ona@other.domain <original@ddress> ...
2963 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2964 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2965 not being looked up in the rDNS to get a host name, Exim was showing the
2966 IP address twice in Received: lines, even though the IP addresses were
2967 identical. For example:
2969 Received: from [1.2.3.4] (helo=[1.2.3.4])
2971 However, if the real host name was known, it was omitting the HELO data
2972 if it matched the actual IP address. This has been tidied up so that it
2973 doesn't show the same IP address twice.
2975 PH/07 When both +timestamp and +memory debugging was on, the value given by
2976 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2977 called by the debug printing, thereby overwriting the timestamp buffer.
2978 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2980 PH/08 When the original message was included in an autoreply transport, it
2981 always said "this is a copy of the message, including all the headers",
2982 even if body_only or headers_only was set. It now gives an appropriate
2985 PH/09 Applied a patch from the Sieve maintainer which:
2987 o fixes some comments
2988 o adds the (disabled) notify extension core
2989 o adds some debug output for the result of if/elsif tests
2990 o points to the current vacation draft in the documentation
2991 and documents the missing references header update
2995 o fixes a bug in processing the envelope test (when testing
2996 multiple envelope elements, the last element determinted the
2999 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3000 Electronic Mail") by including:
3002 Auto-submitted: auto-generated
3004 in the messages that it generates (bounce messages and others, such as
3005 warnings). In the case of bounce messages for non-SMTP mesages, there was
3006 also a typo: it was using "Auto_submitted" (underscore instead of
3007 hyphen). Since every message generated by Exim is necessarily in response
3008 to another message, thes have all been changed to:
3010 Auto-Submitted: auto-replied
3012 in accordance with these statements in the RFC:
3014 The auto-replied keyword:
3016 - SHOULD be used on messages sent in direct response to another
3017 message by an automatic process,
3019 - MUST NOT be used on manually-generated messages,
3021 - MAY be used on Delivery Status Notifications (DSNs) and Message
3022 Disposition Notifications (MDNs),
3024 - MUST NOT be used on messages generated by automatic or periodic
3025 processes, except for messages which are automatic responses to
3028 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3029 to the default Received: header definition.
3031 PH/12 Added log selector acl_warn_skipped (default on).
3033 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3034 variables because (a) they are in the wrong storage pool and (b) even if
3035 they were copied, it wouldn't work properly because of the caching.
3037 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3038 checking when decoding. Apparently there are clients that generate
3039 overlong encoded strings. Why am I not surprised?
3041 PH/15 If the first argument of "${if match_address" was not empty, but did not
3042 contain an "@" character, Exim crashed. Now it writes a panic log message
3043 and treats the condition as false.
3045 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3047 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3048 extension "envelope-auth". The code is finished and in agreement with
3049 other implementations, but there is no documentation so far and in fact,
3050 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3051 not changing the active code.
3053 Print executed "if" and "elsif" statements when debugging is used. This
3054 helps a great deal to understand what a filter does.
3056 Document more things not specified clearly in RFC3028. I had all this
3057 sorted out, when out of a sudden new issues came to my mind. Oops."
3059 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3062 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3063 canonical form (as documented). However, after a host name lookup from
3064 the IP address, check_host() was doing a simple string comparison with
3065 addresses acquired from the DNS when checking that the found name did
3066 have the original IP as one of its addresses. Since any found IPv6
3067 addresses are likely to be in abbreviated form, the comparison could
3068 fail. Luckily, there already exists a function for doing the comparison
3069 by converting both addresses to binary, so now that is used instead of
3070 the text comparison.
3072 PH/20 There was another similar case to PH/19, when a complete host name was
3073 given in a host list; looking up its IP address could give an abbreviated
3074 form, whereas the current host's name might or might not be abbreviated.
3075 The same fix has been applied.
3081 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3082 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3085 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3086 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3088 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3090 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3091 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3092 Also, for "Auto-Submitted", treat anything other than "no" as
3093 non-personal, in accordance with RFC 3834. (Previously it treated
3094 anything starting "auto-" as non-personal.)
3096 TF/01 The control=submission/name=... option had a problem with syntax
3097 errors if the name included a slash character. The /name= option
3098 now slurps the rest of the string, so it can include any characters
3099 but it must come last in the list of options (after /sender_retain
3102 PH/05 Some modifications to the interface to the fake nameserver for the new
3110 TK/01 Added the "success_on_redirect" address verification option. See
3111 NewStuff for rationale and an example.
3113 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3115 PH/02 Patch to exigrep to allow it to work on syslog lines.
3117 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3118 fread() to skip over the body file's header line, because in Cygwin the
3119 header line is locked and is inaccessible.
3121 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3122 co-exist for some time) to make it clear that it is the Exim ID that is
3123 referenced, not the Message-ID: header line.
3125 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3126 string_format() function, because snprintf() does not exist on all
3129 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3130 ${stat: expansion item.
3132 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3133 protocol synchronization error", to keep the pedants happy.
3135 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3136 well as for IRIX systems, when gcc is being used. See the host.c source
3139 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3141 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3144 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3145 transport and to the smtp transport in LMTP mode.
3147 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3149 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3150 run a filter in a subprocess. This could lead to confusion in subsequent
3151 lookups in the parent process. There should also be a search_tidyup() at
3152 the end of the subprocess.
3154 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3155 only if the host matched helo_try_verify_hosts, which caused the
3156 verification to occur when the EHLO/HELO command was issued. The ACL just
3157 tested the remembered result. Now, if a previous verification attempt has
3158 not happened, "verify = helo" does it there and then.
3160 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3162 TK/03 Fix log output including CR from clamd.
3164 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3165 exist provoked a memory error which could cause a segfault.
3167 PH/15 Installed PCRE 6.2
3169 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3171 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3172 of the problem. Specifically, suggested +O2 rather than +O1 for the
3175 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3177 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3178 record, the port was not correctly logged when the outgoing_port log
3179 selector was set (it logged the transort's default port).
3181 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3182 fallback_hosts, and "hosts" in the smtp transport.
3184 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3185 host errors such as "Connection refused".
3187 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3188 authentication with radiusclient 0.4.9:
3190 - Error returned from rc_read_config was caught wrongly
3191 - Username/password not passed on to radius server due to wrong length.
3193 The presumption is that some radiusclient API changes for 4.51/PH/17
3194 were not taken care of correctly. The code is still untested by me (my
3195 Linux distribution still has 0.3.2 of radiusclient), but it was
3196 contributed by a Radius user.
3198 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3199 expanding the "port" option of the smtp transport.
3201 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3202 while reading a MIME header. Thanks to Tom Hughes for a patch.
3204 PH/24 Include config.h inside local_scan.h so that configuration settings are
3207 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3208 that in an ACL for RCPT (for example), you can examine exactly what was
3211 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3212 commands, but it was not correctly comparing the address with the actual
3213 client host address. Thus, it would show the EHLO address in Received:
3214 header lines when this was not necessary.
3216 PH/27 Added the % operator to ${eval:}.
3218 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3219 it should be ignoring failures (because with -C, for example, it has lost
3220 privilege). It wasn't ignoring creation failures other than "already
3223 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3226 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3227 because an input line was too long, either on its own, or by virtue of
3228 too many continuations, the temporary file was not being removed, and the
3229 return code was incorrect.
3231 PH/31 Missing "BOOL" in function definition in filtertest.c.
3233 PH/32 Applied Sieve patches from the maintainer.
3235 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3237 PH/33 Added "verify = not_blind".
3239 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3240 Local/Makefile (with some defaults set). These are used in built scripts
3241 such as exicyclog, but they have never been used in the exim_install
3242 script (though there are many overriding facilities there). I have
3243 arranged that the exim_install script now takes note of these two
3246 PH/35 Installed configuration files for Dragonfly.
3248 PH/36 When a locally submitted message by a trusted user did not contain a
3249 From: header, and the sender address was obtained from -f or from an SMTP
3250 MAIL command, and the trusted user did not use -F to supply a sender
3251 name, $originator_name was incorrectly used when constructing a From:
3252 header. Furthermore, $originator_name was used for submission mode
3253 messages from external hosts without From: headers in a similar way,
3254 which is clearly wrong.
3256 PH/37 Added control=suppress_local_fixups.
3258 PH/38 When log_selector = +received_sender was set, and the addition of the
3259 sender made the log line's construction buffer exactly full, or one byte
3260 less than full, an overflow happened when the terminating "\n" was
3263 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3264 when the result of a list match is failure because a DNS lookup failed.
3266 PH/40 RM_COMMAND is now used in the building process.
3268 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3269 the "build-* directories that it finds.
3271 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3272 address in a domain literal was a prefix of an interface address.
3274 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3275 when verifying a sender address, unless rewrite_headers is false.
3277 PH/44 Wrote a long comment about why errors_to addresses are verified as
3278 recipients, not senders.
3280 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3281 the ratelimit ACL was added.
3283 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3285 PH/46 Added extra information about PostgreSQL errors to the error string.
3287 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3288 avoiding the need to install special zones in a real server. This is
3289 backwards compatible; if it can't find the fake resolver, it drops back.
3290 Thus, both old and new test suites can be run.
3292 TF/02 Added util/ratelimit.pl
3294 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3295 clock is set back in time.
3297 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3298 Candler <B.Candler@pobox.com>.
3300 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3301 OK for addresses that are the result of redirections.
3303 PH/48 A number of further additions for the benefit of the new test suite,
3304 including a fake gethostbyname() that interfaces to the fake DNS resolver
3307 TF/06 The fix for widen_domains has also been applied to qualify_single and
3308 search_parents which are the other dnslookup options that can cause
3311 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3314 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3316 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3317 service" if it encountered an MX record with an empty target hostname.
3318 The message is now "an MX or SRV record indicated no SMTP service".
3320 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3321 if the DNS of the sending site is misconfigured. This is quite a
3322 common situation. This change restores the behaviour of treating a
3323 helo verification defer as a failure.
3325 PH/51 If self=fail was set on a router, the bounce message did not include the
3326 actual error message.
3332 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3334 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3335 command itself ended in error, the underlying message about the transport
3336 filter timeout was being overwritten with the pipe command error. Now the
3337 underlying error message should be appended to the second error message.
3339 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3341 PH/02 Reduce the amount of output that "make" produces by default. Full output
3342 can still be requested.
3344 PH/03 The warning log line about a condition test deferring for a "warn" verb
3345 was being output only once per connection, rather than after each
3346 occurrence (because it was using the same function as for successful
3347 "warn" verbs). This seems wrong, so I have changed it.
3349 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3350 it should not have, which might have caused a crash in the right
3351 circumstances, but probably never did.
3353 PH/04 Installed a modified version of Tony Finch's patch to make submission
3354 mode fix the return path as well as the Sender: header line, and to
3355 add a /name= option so that you can make the user's friendly name appear
3358 TF/03 Added the control = fakedefer ACL modifier.
3360 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3361 Mark Lowes for thorough testing.
3363 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3365 TK/03 Merged latest SRS patch from Miles Wilton.
3367 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3368 with the definition in sysexits.h (which is #included earlier).
3369 Fortunately, Exim does not actually use EX_OK. The code used to try to
3370 preserve the sysexits.h value, by assumimg that macro definitions were
3371 scanned for macro replacements. I have been disabused of this notion,
3372 so now the code just undefines EX_OK before #including unistd.h.
3374 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3375 in the smtp transport. When a block could not be written in a single
3376 write() function, the timeout was being re-applied to each part-write.
3377 This seems wrong - if the receiver was accepting one byte at a time it
3378 would take for ever. The timeout is now adjusted when this happens. It
3379 doesn't have to be particularly precise.
3381 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3382 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3384 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3385 without a domain if the check to <postmaster@domain> fails.
3387 SC/01 Eximstats: added -xls and the ability to specify output files
3388 (patch written by Frank Heydlauf).
3390 SC/02 Eximstats: use FileHandles for outputing results.
3392 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3394 SC/04 Eximstats: fixed display of large numbers with -nvr option
3396 SC/05 Eximstats: fixed merging of reports with empty tables.
3398 SC/06 Eximstats: added the -include_original_destination flag
3400 SC/07 Eximstats: removed tabs and trailing whitespace.
3402 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3404 TK/06 MBOX spool code: Add real "From " MBOX separator line
3405 so the .eml file is really in mbox format (even though
3406 most programs do not really care). Patch from Alex Miller.
3408 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3409 The latter is generated from $received_to and is only set if the
3410 message has one envelope recipient. SA can use these headers,
3411 obviously out-of-the-box. Patch from Alex Miller.
3413 PH/08 The ${def test on a variable was returning false if the variable's
3414 value was "0", contrary to what the specification has always said!
3415 The result should be true unless the variable is empty.
3417 PH/09 The syntax error of a character other than { following "${if
3418 def:variable_name" (after optional whitespace) was not being diagnosed.
3419 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3420 accidental colon was present, for example, could give incorrect results.
3422 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3423 result is used (not including appendfile, where other changes are about
3426 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3427 This involved changing a lot of size variables from int to off_t. It
3428 should work with maildirs and everything.
3430 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3431 spamd dying while we are connected to it.
3433 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3436 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3437 function for BDB 4.3.
3439 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3441 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3442 the sources for PCRE have been re-arranged and more files are now
3445 PH/15 The code I had for printing potentially long long variables in PH/11
3446 above was not the best (it lost precision). The length of off_t variables
3447 is now inspected at build time, and an appropriate printing format (%ld
3448 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3449 to be "long long int" or "long int". This is needed for the internal
3450 formatting function string_vformat().
3452 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3453 the configuration file to be ":syslog", then the script "guesses" where
3454 the logs files are, rather than using the compiled in default. In our
3455 case the guess is not the same as the compiled default, so the script
3456 suddenly stopped working when I started to use syslog. The patch checks
3457 to see if log_file_path is "". If so, it attempts to read it from exim
3458 with no configuration file to get the compiled in version, before it
3459 falls back to the previous guessing code."
3461 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3462 implementing BATV in an Exim configuration. See NewStuff for the gory
3465 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3466 Makefile that are specific to HP-UX.
3468 PH/18 If the "use_postmaster" option was set for a recipient callout together
3469 with the "random" option, the postmaster address was used as the MAIL
3470 FROM address for the random test, but not for the subsequent recipient
3471 test. It is now used for both.
3473 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3474 patch removes a few documentation additions to RFC 3028, because the
3475 latest draft now contains them. It adds the new en;ascii-case comparator
3476 and a new error check for 8bit text in MIME parts. Comparator and
3477 require names are now matched exactly. I enabled the subaddress
3478 extension, but it is not well tested yet (read: it works for me)."
3480 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3481 rework some of the code of TK/09 above to avoid the hardwired use of
3482 "%lld" and "long long". Replaced the call to snprintf() with a call to
3485 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3486 records point to non-existent hosts", "retry timeout exceeded", and
3487 "retry time not reached for any host after a long failure period".
3489 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3490 experimental DomainKeys support:
3492 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3493 (2) On an error such as an illegally used "control", the wrong name for
3494 the control was given.
3496 These problems did NOT occur unless DomainKeys support was compiled.
3498 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3500 PH/24 Added ${if match_ip condition.
3502 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3503 databases so that it will be absolutely obvious if a crash occurs in the
3504 DB library. This is a regular occurrence (often caused by mis-matched
3507 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3508 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3509 picked up on a user's system that detects such things. There doesn't seem
3510 to be a gcc warning option for this - only an attribute that has to be
3511 put on the function's prototype. It seems that in Fedora Core 4 they have
3512 set this on a number of new functions. No doubt there will be more in due
3515 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3516 specify a transport. However, if an address that was verified by such a
3517 router was the subject of a callout, Exim crashed because it tried to
3518 read the rcpt_include_affixes from the non-existent transport. Now it
3519 just assumes that the setting of that option is false. This bug was
3520 introduced by 4.51/PH/31.
3522 PH/28 Changed -d+all to exclude +memory, because that information is very
3523 rarely of interest, but it makes the output a lot bigger. People tend to
3524 do -d+all out of habit.
3526 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3527 code in os-type was giving problems when libc.so lives in lib64, like on
3530 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3531 aren't the modern standard, and it seems that some systems' include files
3532 don't always have them. Exim was already checking for some of the newer
3533 ones like T_AAAA, and defining it itself. I've added checks for all the
3534 record types that Exim uses.
3536 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3537 not automatically generating a new one, as it is supposed to. This
3538 prevented TLS from working. If the file did exist, but contained invalid
3539 data, a new version was generated, as expected. It was only the case of a
3540 non-existent file that was broken.
3542 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3543 with a change in libdomainkeys > 0.64.
3545 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3546 from DNS. If the selector record carries the flag, it now has
3547 precedence over the domain-wide flag.
3549 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3551 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3552 the use of a port name that isn't defined in /etc/services) occurred, the
3553 message was deferred as in a normal delivery, and thus remained on the
3554 spool, instead of being failed because of the mua_wrapper setting. This
3555 is now fixed, and I tidied up some of the mua_wrapper messages at the
3558 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3559 the messages in a hash of arrays rather than using individual hashes.
3560 This is a bit cleaner and results in dramatic memory savings, albeit
3561 at a slight CPU cost.
3563 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3564 as requested by Marc Sherman.
3566 SC/10 Eximstats: added histograms for user specified patterns as requested
3569 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3571 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3572 fopen() in the content-scanning modules that did not already have it.
3578 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3579 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3581 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3583 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3585 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3586 to test Sieve filters that use "vacation".
3588 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3589 that changes the way the GnuTLS parameters are stored in the cache file.
3590 The new format can be generated externally. For backward compatibility,
3591 if the data in the cache doesn't make sense, Exim assumes it has read an
3592 old-format file, and it generates new data and writes a new file. This
3593 means that you can't go back to an older release without removing the
3596 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3597 work if there are any delivery delays because "one_time" forces the
3598 parent to be marked "delivered", so its unseen clone is never tried
3599 again. For this reason, Exim now forbids the simultaneous setting of
3602 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3603 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3604 entry if you want to know the details. The fix, however, neglected to
3605 consider the case when local delivery batching is involved. The test for
3606 "previously delivered" was not happening when checking to see if an
3607 address could be batched with a previous (undelivered) one; under
3608 certain circumstances this could lead to multiple deliveries to the same
3611 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3612 in its include files, and this causes problems building Exim.
3614 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3615 header_syntax) but Exim was just ignoring anything given after a slash.
3616 In particular, this caused confusion with an attempt to use "verify =
3617 reverse_host_lookup/defer_ok". An error is now given when options are
3618 supplied for verify items that do not have them. (Maybe reverse_host_
3619 lookup should have a defer_ok option, but that's a different point.)
3621 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3622 defined by RFC 821) to 2048, because there were problems with some AUTH
3623 commands, and RFC 1869 says the size should be increased for extended
3624 SMTP commands that take arguments.
3626 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3629 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3630 "unknown" error; now it says that the functionality isn't in the binary.
3632 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3633 an address' error message when a string expansion fails (syntax or
3634 whatever). Otherwise the password may appear in the log. Following change
3635 PH/42 below, there is no longer a chance of it appearing in a bounce
3638 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3640 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3641 of its addresses was ever tried. (Bugzilla bug #2.)
3643 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3644 the result incorrectly in the debug output. (It correctly added a newline
3645 to what was transported.)
3647 TF/01 Added $received_time.
3649 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3650 commented out examples of how to interface to a virus scanner and to
3651 SpamAssassin. Also added commented examples of av_scanner and
3652 spamd_address settings.
3654 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3655 and controls are allowed in which ACLs. There were a couple of minor
3656 errors. Some of the entries in the conditions table (which is a table of
3657 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3658 negation of where the condition IS allowed.
3660 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3662 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3663 header file does not have a version number, so I've had to invent a new
3664 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3665 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3666 radiusclient), but it was contributed by a Radius user.
3668 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3669 files or directories, for OpenSSL.
3671 PH/19 When an Exim process that is running as root has to create an Exim log
3672 file, it does so in a subprocess that runs as exim:exim so as to get the
3673 ownership right at creation (otherwise, other Exim processes might see
3674 the file with the wrong ownership). There was no test for failure of this
3675 fork() call, which would lead to the process getting stuck as it waited
3676 for a non-existent subprocess. Forks do occasionally fail when resources
3677 run out. I reviewed all the other calls to fork(); they all seem to check
3680 PH/20 When checking for unexpected SMTP input at connect time (before writing
3681 the banner), Exim was not dealing correctly with a non-positive return
3682 from the read() function. If the client had disconnected by this time,
3683 the result was a log entry for a synchronization error with an empty
3684 string after "input=" when read() returned zero. If read() returned -1
3685 (an event I could not check), uninitialized data bytes were printed.
3686 There were reports of junk text (parts of files, etc) appearing after
3689 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3691 PH/22 Added support for macro redefinition, and (re)definition in between
3692 driver and ACL definitions.
3694 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3695 forgetting to use the resulting value; it was using the unexpanded value.
3697 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3698 hadn't been configured. The fix is from Juergen Kreileder, who
3699 understands it better than I do:
3701 "Here's what I see happening with three configured cyrus_sasl
3702 authenticators configured (plain, login, cram-md5):
3704 On startup auth_cyrus_sasl_init() gets called for each of these.
3705 This means three calls to sasl_listmech() without a specified mech_list.
3706 => SASL tests which mechs of all available mechs actually work
3707 => three warnings about OTP not working
3708 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3710 With the patch, sasl_listmech() also gets called three times. But now
3711 SASL's mech_list option is set to the server_mech specified in the the
3712 authenticator. Or in other words, the answer from sasl_listmech()
3713 gets limited to just the mech you're testing for (which is different
3715 => the return list contains just 'plain' or 'login', 'cram-md5' or
3716 nothing depending on the value of ob->server_mech.
3718 I've just tested the patch: Authentication still works fine,
3719 unavailable mechs specified in the exim configuration are still
3720 caught, and the auth.log warnings about OTP are gone."
3722 PH/25 When debugging is enabled, the contents of the command line are added
3723 to the debugging output, even when log_selector=+arguments is not
3726 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3727 answer is "GNU", and only if the return is "GNU/something" is the answer
3730 PH/27 $acl_verify_message is now set immediately after the failure of a
3731 verification in an ACL, and so is available in subsequent modifiers. In
3732 particular, the message can be preserved by coding like this:
3734 warn !verify = sender
3735 set acl_m0 = $acl_verify_message
3737 Previously, $acl_verify_message was set only while expanding "message"
3738 and "log_message" when a very denied access.
3740 PH/28 Modified OS/os.c-Linux with
3742 -#ifndef OS_LOAD_AVERAGE
3743 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3745 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3746 nomenclature these days.)
3748 PH/29 Installed patch from the Sieve maintainer that adds the options
3749 sieve_useraddress and sieve_subaddress to the redirect router.
3751 PH/30 In these circumstances:
3752 . Two addresses routed to the same list of hosts;
3753 . First host does not offer TLS;
3754 . First host accepts first address;
3755 . First host gives temporary error to second address;
3756 . Second host offers TLS and a TLS session is established;
3757 . Second host accepts second address.
3758 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3759 and peerdn, if requested) that were in fact used only for the second
3762 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3763 attention to any local part prefix or suffix that was matched by the
3764 router that accepted the address. It now behaves in the same way as it
3765 does for delivery: the affixes are removed from the local part unless
3766 rcpt_include_affixes is set on the transport.
3768 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3769 timeout during the DATA phase of an incoming message.
3771 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3772 applied a patch sanctioned by the Sieve maintainer.
3774 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3775 the uid or gid is negative. A case of a negative gid caused this to be
3776 noticed. The fix allows for either to be negative.
3778 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3779 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3782 PH/36 The change PH/12 above was broken. Fixed it.
3784 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3785 the grounds that routing the same address twice would always produce the
3786 same answer. This might have been true once, but it is certainly no
3787 longer true now. Routing a child address may depend on the previous
3788 routing that produced that child. Some complicated redirection strategies
3789 went wrong when messages had multiple recipients, and made Exim's
3790 behaviour dependent on the order in which the addresses were given.
3792 I have moved the duplicate checking until after the routing is complete.
3793 Exim scans the addresses that are assigned to local and remote
3794 transports, and removes any duplicates. This means that more work will be
3795 done, as duplicates will always all be routed, but duplicates are
3796 presumably rare, so I don't expect this is of any significance.
3798 For deliveries to pipes, files, and autoreplies, the duplicate checking
3799 still happens during the routing process, since they are not going to be
3802 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3803 It corrects a timeout issue with spamd. This is Ian's comment: "The
3804 background is that sometimes spamd either never reads data from a
3805 connection it has accepted, or it never writes response data. The exiscan
3806 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3807 blindly assumes that writes won't block so it may never time out."
3809 PH/39 Allow G after quota size as well as K and M.
3811 PH/40 The value set for $authenticated_id in an authenticator may not contain
3812 binary zeroes or newlines because the value is written to log lines and
3813 to spool files. There was no check on this. Now the value is run through
3814 the string_printing() function so that such characters are converted to
3815 printable escape sequences.
3817 PH/41 $message_linecount is a new variable that contains the total number of
3818 lines in the message. Compare $body_linecount, which is the count for the
3821 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3822 bounce and delay warning messages, except in certain special cases, which
3825 (a) An SMTP error message from a remote host;
3826 (b) A message specified in a :fail: redirection;
3827 (c) A message specified in a "fail" command in a system filter;
3828 (d) A message specified in a FAIL return from the queryprogram router;
3829 (e) A message specified by the cannot_route_message router option.
3831 In these cases only, Exim does include the error details in bounce and
3832 warning messages. There are also a few cases where bland messages such
3833 as "unrouteable address" or "local delivery error" are given.
3835 PH/43 $value is now also set for the "else" part of a ${run expansion.
3837 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3838 being worked on, but at least Exim now implements the latest version to
3841 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3842 process to complete was treated as a delivery failure, a timeout while
3843 writing the message to the pipe was logged, but erroneously treated as a
3844 successful delivery. Such timeouts include transport filter timeouts. For
3845 consistency with the overall process timeout, these timeouts are now
3846 treated as errors, giving rise to delivery failures by default. However,
3847 there is now a new Boolean option for the pipe transport called
3848 timeout_defer, which, if set TRUE, converts the failures into defers for
3849 both kinds of timeout. A transport filter timeout is now identified in
3852 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3853 systems where "make" and "gmake" are different, calling "gmake" at top
3854 level broke things. I've arranged for the value of $(MAKE) to be passed
3855 from the Makefile to this script so that it can call the same version of
3859 A note about Exim versions 4.44 and 4.50
3860 ----------------------------------------
3862 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3863 changes of various kinds. As a consequence, a big documentation update was
3864 needed. This delayed the release for rather longer than seemed good, especially
3865 in the light of a couple of (minor) security issues. Therefore, the changes
3866 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3867 release. So 4.44 and 4.50 are in effect two different branches that both start
3870 I have left the 4.50 change log unchanged; it contains all the changes since
3871 4.43. The change log for 4.44 is below; many of its items are identical to
3872 those for 4.50. This seems to be the most sensible way to preserve the
3873 historical information.
3879 1. Minor wording change to the doc/README.SIEVE file.
3881 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3882 computation of the current number of files was incorrect.
3884 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3885 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3888 4. Give more explanation in the error message when the command for a transport
3889 filter fails to execute.
3891 5. There are several places where Exim runs a non-Exim command in a
3892 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3893 was being done only for the command run by the queryprogram router. It is
3894 now done for all such subprocesses. The other cases are: ${run, transport
3895 filters, and the commands run by the lmtp and pipe transports.
3897 6. Added CONFIGURE_GROUP build-time option.
3899 7. Some older OS have a limit of 256 on the maximum number of file
3900 descriptors. Exim was using setrlimit() to set 1000 as a large value
3901 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3902 systems. I've change it so that if it can't get 1000, it tries for 256.
3904 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3905 was an oversight, and furthermore, ever since the addition of extra
3906 controls (e.g. 4.43/32), the checks on when to allow different forms of
3907 "control" were broken. There should now be diagnostics for all cases when a
3908 control that does not make sense is encountered.
3910 9. Added the /retain_sender option to "control=submission".
3912 10. $recipients is now available in the predata ACL (oversight).
3914 11. Tidy the search cache before the fork to do a delivery from a message
3915 received from the command line. Otherwise the child will trigger a lookup
3916 failure and thereby defer the delivery if it tries to use (for example) a
3917 cached ldap connection that the parent has called unbind on.
3919 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3920 of $address_data from the recipient verification was clobbered by the
3921 sender verification.
3923 13. The value of address_data from a sender verification is now available in
3924 $sender_address_data in subsequent conditions in the ACL statement.
3926 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3928 15. Added a new option "connect=<time>" to callout options, to set a different
3931 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3932 was its contents. (It was OK if the option was not defined at all.)
3934 17. A "Completed" log line is now written for messages that are removed from
3935 the spool by the -Mrm option.
3937 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3938 information about exactly what failed.
3940 19. Added -dd to debug only the daemon process.
3942 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3943 handles timeouts, both on the server side and network timeouts. Renamed the
3944 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3946 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3947 It is now set to "smtps".
3949 22. $host_address is now set to the target address during the checking of
3950 ignore_target_hosts.
3952 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3953 being passed; this would have caused $sender_host_name to have been used if
3954 matching the list had actually called for a host name (not very likely,
3955 since this list is usually IP addresses). A host name is now passed as
3958 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3959 code that specifies a non-restarting handler (typically sigaction() in
3960 modern systems) in an attempt to fix a rare and obscure crash bug.
3962 25. Narrowed the window for a race in the daemon that could cause it to ignore
3963 SIGCHLD signals. This is not a major problem, because they are used only to
3964 wake it up if nothing else does.
3966 26. A malformed maildirsize file could cause Exim to calculate negative values
3967 for the mailbox size or file count. Odd effects could occur as a result.
3968 The maildirsize information is now recalculated if the size or filecount
3971 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3972 support for a long time. Removed HAVE_SYS_VFS_H.
3974 28. Installed the latest version of exipick from John Jetmore.
3976 29. In an address list, if the pattern was not a regular expression, an empty
3977 subject address (from a bounce message) matched only if the pattern was an
3978 empty string. Non-empty patterns were not even tested. This was the wrong
3979 because it is perfectly reasonable to use an empty address as part of a
3980 database query. An empty address is now tested by patterns that are
3981 lookups. However, all the other forms of pattern expect the subject to
3982 contain a local part and a domain, and therefore, for them, an empty
3983 address still always fails if the pattern is not itself empty.
3985 30. Exim went into a mad DNS loop when attempting to do a callout where the
3986 host was specified on an smtp transport, and looking it up yielded more
3987 than one IP address.
3989 31. Re-factored the code for checking spool and log partition space into a
3990 function that finds that data and another that does the check. The former
3991 is then used to implement four new variables: $spool_space, $log_space,
3992 $spool_inodes, and $log_inodes.
3994 32. The RFC2047 encoding function was originally intended for short strings
3995 such as real names; it was not keeping to the 75-character limit for
3996 encoded words that the RFC imposes. It now respects the limit, and
3997 generates multiple encoded words if necessary. To be on the safe side, I
3998 have increased the buffer size for the ${rfc2047: expansion operator from
4001 33. It is now permitted to omit both strings after an "if" condition; if the
4002 condition is true, the result is "true". As before, when the second string
4003 is omitted, a false condition yields an empty string. This makes it less
4004 cumbersome to write custom ACL and router conditions.
4006 34. Failure to deliver a bounce message always caused it to be frozen, even if
4007 there was an errors_to setting on the router. The errors_to setting is now
4010 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4011 canonical form (fully expanded) before being placed in
4012 $sender_host_address.
4014 36. The table in the code that translates DNS record types into text (T_A to
4015 "A" for instance) was missing entries for NS and CNAME. It is just possible
4016 that this could have caused confusion if both these types were looked up
4017 for the same domain, because the text type is used as part of Exim's
4018 per-process caching. But the chance of anyone hitting this buglet seems
4021 37. The dnsdb lookup has been extended in a number of ways.
4023 (1) There is a new type, "zns", which walks up the domain tree until it
4024 finds some nameserver records. It should be used with care.
4026 (2) There is a new type, "mxh", which is like "mx" except that it returns
4027 just the host names, not the priorities.
4029 (3) It is now possible to give a list of domains (or IP addresses) to be
4030 looked up. The behaviour when one of the lookups defers can be
4031 controlled by a keyword.
4033 (4) It is now possible to specify the separator character for use when
4034 multiple records are returned.
4036 38. The dnslists ACL condition has been extended: it is now possible to supply
4037 a list of IP addresses and/or domains to be looked up in a particular DNS
4040 39. Added log_selector=+queue_time_overall.
4042 40. When running the queue in the test harness, wait just a tad after forking a
4043 delivery process, to get repeatability of debugging output.
4045 41. Include certificate and key file names in error message when GnuTLS fails
4046 to set them up, because the GnuTLS error message doesn't include the name
4047 of the failing file when there is a problem reading it.
4049 42. Allow both -bf and -bF in the same test run.
4051 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4053 44. The "Exiscan patch" is now merged into the mainline Exim source.
4055 45. Sometimes the final signoff response after QUIT could fail to get
4056 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4057 < 0 before doing a fflush(). This bug looks as though it goes back to the
4058 introduction of TLS in release 3.20, but "sometimes" must have been rare
4059 because the tests only now provoked it.
4061 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4062 (this can affect the format of dates).
4064 47. exim_tidydb, when checking for the continued existence of a message for
4065 which it has found a message-specific retry record, was not finding
4066 messages that were in split spool directories. Consequently, it was
4067 deleting retry records that should have stayed in existence.
4069 48. Steve fixed some bugs in eximstats.
4071 49. The SPA authentication driver was not abandoning authentication and moving
4072 on to the next authenticator when an expansion was forced to fail,
4073 contradicting the general specification for all authenticators. Instead it
4074 was generating a temporary error. It now behaves as specified.
4076 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4077 (the order specifies the preference for clients). The order is now AES256,
4078 AES128, 3DES, ARCFOUR128.
4080 51. Small patch to Sieve code - explicitly set From: when generating an
4083 52. Exim crashed if a remote delivery caused a very long error message to be
4084 recorded - for instance if somebody sent an entire SpamAssassin report back
4085 as a large number of 550 error lines. This bug was coincidentally fixed by
4086 increasing the size of one of Exim's internal buffers (big_buffer) that
4087 happened as part of the Exiscan merge. However, to be on the safe side, I
4088 have made the code more robust (and fixed the comments that describe what
4091 53. Now that there can be additional text after "Completed" in log lines (if
4092 the queue_time_overall log selector is set), a one-byte patch to exigrep
4093 was needed to allow it to recognize "Completed" as not the last thing in
4096 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4097 patch that reportedly fixes this has been added. I am not expert enough to
4098 create a test for it. This is what the patch creator wrote:
4100 "I found a little strange behaviour of ldap code when working with
4101 Windows 2003 AD Domain, where users was placed in more than one
4102 Organization Units. When I tried to give exim partial DN, the exit code
4103 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4104 But simultaneously result of request was absolutely normal ldap result,
4105 so I produce this patch..."
4107 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4108 REFERENCE, so I have modified the code to exclude the patch when that macro
4111 55. Some experimental protocols are using DNS PTR records for new purposes. The
4112 keys for these records are domain names, not reversed IP addresses. The
4113 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4114 leaves it alone. Component reversal etc. now happens only for IP addresses.
4117 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4119 57. Double the size of the debug message buffer (to 2048) so that more of very
4120 long debug lines gets shown.
4122 58. The exicyclog utility now does better if the number of log files to keep
4123 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4125 59. Two changes related to the smtp_active_hostname option:
4127 (1) $smtp_active_hostname is now available as a variable.
4128 (2) The default for smtp_banner uses $smtp_active_hostname instead
4129 of $primary_hostname.
4131 60. The host_aton() function is supposed to be passed a string that is known
4132 to be a valid IP address. However, in the case of IPv6 addresses, it was
4133 not checking this. This is a hostage to fortune. Exim now panics and dies
4134 if the condition is not met. A case was found where this could be provoked
4135 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4136 components; fortuitously, this particular loophole had already been fixed
4137 by change 4.50/55 above.
4139 If there are any other similar loopholes, the new check in host_aton()
4140 itself should stop them being exploited. The report I received stated that
4141 data on the command line could provoke the exploit when Exim was running as
4142 exim, but did not say which command line option was involved. All I could
4143 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4144 running as the user.
4147 61. There was a buffer overflow vulnerability in the SPA authentication code
4148 (which came originally from the Samba project). I have added a test to the
4149 spa_base64_to_bits() function which I hope fixes it.
4152 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4153 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4155 63. The daemon start-up calls getloadavg() while still root for those OS that
4156 need the first call to be done as root, but it missed one case: when
4157 deliver_queue_load_max is set with deliver_drop_privilege. This is
4158 necessary for the benefit of the queue runner, because there is no re-exec
4159 when deliver_drop_privilege is set.
4161 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4162 This has been fixed.
4164 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4165 was in use, was not putting the data itself into the right store pool;
4166 consequently, it could be overwritten for a subsequent message in the same
4167 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4170 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4172 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4173 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4174 the function were treating the return as a boolean value, which happened to
4175 work because 0=false and not-0=true, but is not correct code.
4177 68. The host_aton() function was not handling scoped IPv6 addresses (those
4178 with, for example, "%eth0" on the end) correctly.
4180 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4181 negated items (that is, ~something) in unsigned ints. Some compilers
4182 apparently mutter when there is no cast.
4184 70. If an address verification called from an ACL failed, and did not produce a
4185 user-specific message (i.e. there was only a "system" message), nothing was
4186 put in $acl_verify_message. In this situation, it now puts the system
4189 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4190 session; change 4.31/43 added the unwanted input to the log line - except
4191 that it did not do this in the start of session case. It now does.
4193 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4194 This is wrong and can cause the other end to generate a synchronization
4195 error if it is another Exim or anything else that does the synchronization
4196 check. A QUIT command is no longer sent after a timeout.
4198 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4199 during host lookups.
4201 74. The defer_ok option of callout verification was not working if it was used
4202 when verifying addresses in header lines, that is, for this case:
4204 verify = header_sender/callout=defer_ok
4206 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4207 those file descriptors could be used for SMTP connections. If anything
4208 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4209 could be sent to the SMTP client, causing chaos. The daemon now opens
4210 stdin, stdout, and stderr to /dev/null when it puts itself into the
4213 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4214 log by default. The user can override this with suitable Perl magic.
4216 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4217 the log message when discard triggers, was not working for the DATA ACL or
4218 for the non-SMTP ACL.
4220 78. Error message wording change in sieve.c.
4222 79. If smtp_accept_max_per_host was set, the number of connections could be
4223 restricted to fewer than expected, because the daemon was trying to set up
4224 a new connection before checking whether the processes handling previous
4225 connections had finished. The check for completed processes is now done
4226 earlier. On busy systems, this bug wouldn't be noticed because something
4227 else would have woken the daemon, and it would have reaped the completed
4230 80. If a message was submitted locally by a user whose login name contained one
4231 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4232 It caused a spool format error. I have fixed the spool reading code. A
4233 related problem was that the "from" clause in the Received: line became
4234 illegal because of the space(s). It is now covered by ${quote_local_part.
4236 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4239 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4241 83. Patch from Sieve maintainer to fix unterminated string problem in
4242 "vacation" handling.
4244 84. Some minor changes to the Linux configuration files to help with other
4245 OS variants using glibc.
4247 85. One more patch for Sieve to update vacation handling to latest spec.
4250 ----------------------------------------------------
4251 See the note above about the 4.44 and 4.50 releases.
4252 ----------------------------------------------------
4258 1. Change 4.43/35 introduced a bug that caused file counts to be
4259 incorrectly computed when quota_filecount was set in an appendfile
4262 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4263 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4266 3. Give more explanation in the error message when the command for a transport
4267 filter fails to execute.
4269 4. There are several places where Exim runs a non-Exim command in a
4270 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4271 was being done only for the command run by the queryprogram router. It is
4272 now done for all such subprocesses. The other cases are: ${run, transport
4273 filters, and the commands run by the lmtp and pipe transports.
4275 5. Some older OS have a limit of 256 on the maximum number of file
4276 descriptors. Exim was using setrlimit() to set 1000 as a large value
4277 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4278 systems. I've change it so that if it can't get 1000, it tries for 256.
4280 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4281 was an oversight, and furthermore, ever since the addition of extra
4282 controls (e.g. 4.43/32), the checks on when to allow different forms of
4283 "control" were broken. There should now be diagnostics for all cases when a
4284 control that does not make sense is encountered.
4286 7. $recipients is now available in the predata ACL (oversight).
4288 8. Tidy the search cache before the fork to do a delivery from a message
4289 received from the command line. Otherwise the child will trigger a lookup
4290 failure and thereby defer the delivery if it tries to use (for example) a
4291 cached ldap connection that the parent has called unbind on.
4293 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4294 of $address_data from the recipient verification was clobbered by the
4295 sender verification.
4297 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4298 was its contents. (It was OK if the option was not defined at all.)
4300 11. A "Completed" log line is now written for messages that are removed from
4301 the spool by the -Mrm option.
4303 12. $host_address is now set to the target address during the checking of
4304 ignore_target_hosts.
4306 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4307 being passed; this would have caused $sender_host_name to have been used if
4308 matching the list had actually called for a host name (not very likely,
4309 since this list is usually IP addresses). A host name is now passed as
4312 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4313 code that specifies a non-restarting handler (typically sigaction() in
4314 modern systems) in an attempt to fix a rare and obscure crash bug.
4316 15. Narrowed the window for a race in the daemon that could cause it to ignore
4317 SIGCHLD signals. This is not a major problem, because they are used only to
4318 wake it up if nothing else does.
4320 16. A malformed maildirsize file could cause Exim to calculate negative values
4321 for the mailbox size or file count. Odd effects could occur as a result.
4322 The maildirsize information is now recalculated if the size or filecount
4325 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4326 support for a long time. Removed HAVE_SYS_VFS_H.
4328 18. Updated exipick to current release from John Jetmore.
4330 19. Allow an empty sender to be matched against a lookup in an address list.
4331 Previously the only cases considered were a regular expression, or an
4334 20. Exim went into a mad DNS lookup loop when doing a callout where the
4335 host was specified on the transport, if the DNS lookup yielded more than
4338 21. The RFC2047 encoding function was originally intended for short strings
4339 such as real names; it was not keeping to the 75-character limit for
4340 encoded words that the RFC imposes. It now respects the limit, and
4341 generates multiple encoded words if necessary. To be on the safe side, I
4342 have increased the buffer size for the ${rfc2047: expansion operator from
4345 22. Failure to deliver a bounce message always caused it to be frozen, even if
4346 there was an errors_to setting on the router. The errors_to setting is now
4349 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4350 canonical form (fully expanded) before being placed in
4351 $sender_host_address.
4353 24. Updated eximstats to version 1.33
4355 25. Include certificate and key file names in error message when GnuTLS fails
4356 to set them up, because the GnuTLS error message doesn't include the name
4357 of the failing file when there is a problem reading it.
4359 26. Expand error message when OpenSSL has problems setting up cert/key files.
4362 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4363 (this can affect the format of dates).
4365 28. exim_tidydb, when checking for the continued existence of a message for
4366 which it has found a message-specific retry record, was not finding
4367 messages that were in split spool directories. Consequently, it was
4368 deleting retry records that should have stayed in existence.
4370 29. eximstats updated to version 1.35
4371 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4372 1.35 - bugfix such that pie charts by volume are generated correctly
4374 30. The SPA authentication driver was not abandoning authentication and moving
4375 on to the next authenticator when an expansion was forced to fail,
4376 contradicting the general specification for all authenticators. Instead it
4377 was generating a temporary error. It now behaves as specified.
4379 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4380 (the order specifies the preference for clients). The order is now AES256,
4381 AES128, 3DES, ARCFOUR128.
4383 31. Small patch to Sieve code - explicitly set From: when generating an
4386 32. Exim crashed if a remote delivery caused a very long error message to be
4387 recorded - for instance if somebody sent an entire SpamAssassin report back
4388 as a large number of 550 error lines. This bug was coincidentally fixed by
4389 increasing the size of one of Exim's internal buffers (big_buffer) that
4390 happened as part of the Exiscan merge. However, to be on the safe side, I
4391 have made the code more robust (and fixed the comments that describe what
4394 33. Some experimental protocols are using DNS PTR records for new purposes. The
4395 keys for these records are domain names, not reversed IP addresses. The
4396 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4397 leaves it alone. Component reversal etc. now happens only for IP addresses.
4400 34. The host_aton() function is supposed to be passed a string that is known
4401 to be a valid IP address. However, in the case of IPv6 addresses, it was
4402 not checking this. This is a hostage to fortune. Exim now panics and dies
4403 if the condition is not met. A case was found where this could be provoked
4404 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4405 components; fortuitously, this particular loophole had already been fixed
4406 by change 4.50/55 or 4.44/33 above.
4408 If there are any other similar loopholes, the new check in host_aton()
4409 itself should stop them being exploited. The report I received stated that
4410 data on the command line could provoke the exploit when Exim was running as
4411 exim, but did not say which command line option was involved. All I could
4412 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4413 running as the user.
4416 35. There was a buffer overflow vulnerability in the SPA authentication code
4417 (which came originally from the Samba project). I have added a test to the
4418 spa_base64_to_bits() function which I hope fixes it.
4421 36. The daemon start-up calls getloadavg() while still root for those OS that
4422 need the first call to be done as root, but it missed one case: when
4423 deliver_queue_load_max is set with deliver_drop_privilege. This is
4424 necessary for the benefit of the queue runner, because there is no re-exec
4425 when deliver_drop_privilege is set.
4427 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4428 was in use, was not putting the data itself into the right store pool;
4429 consequently, it could be overwritten for a subsequent message in the same
4430 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4433 38. Sometimes the final signoff response after QUIT could fail to get
4434 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4435 < 0 before doing a fflush(). This bug looks as though it goes back to the
4436 introduction of TLS in release 3.20, but "sometimes" must have been rare
4437 because the tests only now provoked it.
4443 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4444 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4445 Now it returns an integer. A number of calls were still expecting a T/F
4446 return. Fortuitously, in all cases, the tests worked in OK situations,
4447 which is the norm. However, things would have gone wrong on any write
4448 failures on the smtp file descriptor. This function is used when sending
4449 messages over SMTP and also when doing verify callouts.
4451 2. When Exim is called to do synchronous delivery of a locally submitted
4452 message (the -odf or -odi options), it no longer closes stderr before doing
4455 3. Implemented the mua_wrapper option.
4457 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4459 5. Implemented the functions header_remove(), header_testname(),
4460 header_add_at_position(), and receive_remove_recipient(), and exported them
4463 6. If an ACL "warn" statement specified the addition of headers, Exim already
4464 inserted X-ACL-Warn: at the start if there was no header name. However, it
4465 was not making this test for the second and subsequent header lines if
4466 there were newlines in the string. This meant that an invalid header could
4467 be inserted if Exim was badly configured.
4469 7. Allow an ACL "warn" statement to add header lines at the start or after all
4470 the Received: headers, as well as at the end.
4472 8. Added the rcpt_4xx retry error code.
4474 9. Added postmaster_mailfrom=xxx to callout verification option.
4476 10. Added mailfrom=xxxx to the callout verification option, for verify=
4479 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4480 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4482 12. Inserted some casts to stop certain compilers warning when using pointer
4483 differences as field lengths or precisions in printf-type calls (mostly
4484 affecting debugging statements).
4486 13. Added optional readline() support for -be (dynamically loaded).
4488 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4489 same clock tick as a message's arrival, so that its received time was the
4490 same as the "first fail" time on the retry record, and that message
4491 remained on the queue past the ultimate address timeout, every queue runner
4492 would try a delivery (because it was past the ultimate address timeout) but
4493 after another failure, the ultimate address timeout, which should have then
4494 bounced the address, did not kick in. This was a "< instead of <=" error;
4495 in most cases the first failure would have been in the next clock tick
4496 after the received time, and all would be well.
4498 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4499 being recognized when the domain list was tested by the match_domain
4500 condition in an expansion string.
4502 16. Added the ${str2b64: operator.
4504 17. Exim was always calling setrlimit() to set a large limit for the number of
4505 processes, without checking whether the existing limit was already
4506 adequate. (It did check for the limit on file descriptors.) Furthermore,
4507 errors from getrlimit() and setrlimit() were being ignored. Now they are
4508 logged to the main and panic logs, but Exim does carry on, to try to do its
4509 job under whatever limits there are.
4511 18. Imported PCRE 5.0.
4513 19. Trivial typo in log message " temporarily refused connection" (the leading
4516 20. If the log selector return_path_on_delivery was set and an address was
4517 redirected to /dev/null, the delivery process crashed because it assumed
4518 that a return path would always be set for a "successful" delivery. In this
4519 case, the whole delivery is bypassed as an optimization, and therefore no
4522 21. Internal re-arrangement: the function for sending a challenge and reading
4523 a response while authentication was assuming a zero-terminated challenge
4524 string. It's now changed to take a pointer and a length, to allow for
4525 binary data in such strings.
4527 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4529 23. Exim was not respecting finduser_retries when seeking the login of the
4530 uid under which it was called; it was always trying 10 times. (The default
4531 setting of finduser_retries is zero.) Also, it was sleeping after the final
4532 failure, which is pointless.
4534 24. Implemented tls_on_connect_ports.
4536 25. Implemented acl_smtp_predata.
4538 26. If the domain in control=submission is set empty, Exim assumes that the
4539 authenticated id is a complete email address when it generates From: or
4540 Sender: header lines.
4542 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4543 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4544 chown and chgrp in /bin and hostname in /usr/bin.
4546 28. Exim was keeping the "process log" file open after each use, just as it
4547 does for the main log. This opens the possibility of it remaining open for
4548 long periods when the USR1 signal hits a daemon. Occasional processlog
4549 errors were reported, that could have been caused by this. Anyway, it seems
4550 much more sensible not to leave this file open at all, so that is what now
4553 29. The long-running daemon process does not normally write to the log once it
4554 has entered its main loop, and it closes the log before doing so. This is
4555 so that log files can straightforwardly be renamed and moved. However,
4556 there are a couple of unusual error situations where the daemon does write
4557 log entries, and I had neglected to close the log afterwards.
4559 30. The text of an SMTP error response that was received during a remote
4560 delivery was being truncated at 512 bytes. This is too short for some of
4561 the long messages that one sometimes sees. I've increased the limit to
4564 31. It is now possible to make retry rules that apply only when a message has a
4565 specific sender, in particular, an empty sender.
4567 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4568 it possible to be selective about when SMTP synchronization is enforced.
4570 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4572 32. Implemented hosts_connection_nolog.
4574 33. Added an ACL for QUIT.
4576 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4579 35. Added mailbox_size and mailbox_filecount to appendfile.
4581 36. Added control = no_multiline_responses to ACLs.
4583 37. There was a bug in the logic of the code that waits for the clock to tick
4584 in the case where the clock went backwards by a substantial amount such
4585 that the microsecond fraction of "now" was more than the microsecond
4586 fraction of "then" (but the whole seconds number was less).
4588 38. Added support for the libradius Radius client library this is found on
4589 FreeBSD (previously only the radiusclient library was supported).
4595 1. When certain lookups returned multiple values in the form name=value, the
4596 quoting of the values was not always being done properly. Specifically:
4597 (a) If the value started with a double quote, but contained no whitespace,
4599 (b) If the value contained whitespace other than a space character (i.e.
4600 tabs or newlines or carriage returns) it was not quoted.
4601 This fix has been applied to the mysql and pgsql lookups by writing a
4602 separate quoting function and calling it from the lookup code. The fix
4603 should probably also be applied to nisplus, ibase and oracle lookups, but
4604 since I cannot test any of those, I have not disturbed their existing code.
4606 2. A hit in the callout cache for a specific address caused a log line with no
4607 reason for rejecting RCPT. Now it says "Previous (cached) callout
4608 verification failure".
4610 3. There was an off-by-one bug in the queryprogram router. An over-long
4611 return line was truncated at 256 instead of 255 characters, thereby
4612 overflowing its buffer with the terminating zero. As well as fixing this, I
4613 have increased the buffer size to 1024 (and made a note to document this).
4615 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4616 when Exim is waiting for an SMTP response from a remote server, Exim
4617 restarts its select() call on the socket, thereby resetting its timeout.
4618 This is not a problem when such interrupts are rare. Somebody set up a cron
4619 job to run exiwhat every 2 minutes, which is less than the normal select()
4620 timeout (5 or 10 minutes). This meant that the select() timeout never
4621 kicked in because it was always reset. I have fixed this by comparing the
4622 time when an interrupt arrives with the time at the start of the first call
4623 to select(). If more time than the timeout has elapsed, the interrupt is
4624 treated as a timeout.
4626 5. Some internal re-factoring in preparation for the addition of Sieve
4627 extensions (by MH). In particular, the "personal" test is moved to a
4628 separate function, and given an option for scanning Cc: and Bcc: (which is
4629 not set for Exim filters).
4631 6. When Exim created an email address using the login of the caller as the
4632 local part (e.g. when creating a From: or Sender: header line), it was not
4633 quoting the local part when it contained special characters such as @.
4635 7. Installed new OpenBSD configuration files.
4637 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4638 try to make them clearer.
4640 9. Callout options, other than the timeout value, were being ignored when
4641 verifying sender addresses in header lines. For example, when using
4643 verify = header_sender/callout=no_cache
4645 the cache was (incorrectly) being used.
4647 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4648 only the Cygwin environment.
4650 11. When return_path_on_delivery was set as a log selector, if different remote
4651 addresses in the same message used different return paths and parallel
4652 remote delivery occurred, the wrong values would sometimes be logged.
4653 (Whenever a remote delivery process finished, the return path value from
4654 the most recently started remote delivery process was logged.)
4656 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4657 lines when AUTH and/or TLS are in use. This is the "received protocol"
4658 field. Exim used to use "asmtp" for authenticated SMTP, without any
4659 indication (in the protocol name) for TLS use. Now it follows the RFC and
4660 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4661 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4662 names appear in log lines as well as in Received: header lines.
4664 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4665 extensions, and comparison tests, and to fix some bugs.
4667 14. Changes to the "personal" filter test:
4669 (1) The test was buggy in that it was just doing the equivalent of
4670 "contains" tests on header lines. For example, if a user's address was
4671 anne@some.where, the "personal" test would incorrectly be true for
4673 To: susanne@some.where
4675 This test is now done by extracting each address from the header in turn,
4676 and checking the entire address. Other tests that are part of "personal"
4677 are now done using regular expressions (for example, to check local parts
4678 of addresses in From: header lines).
4680 (2) The list of non-personal local parts in From: addresses has been
4681 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4682 taken from the Sieve specification recommendations.
4684 (3) If the message contains any header line starting with "List-" it is
4685 treated as non-personal.
4687 (4) The test for "circular" in the Subject: header line has been removed
4688 because it now seems ill-conceived.
4690 15. Minor typos in src/EDITME comments corrected.
4692 16. Installed latest exipick from John Jetmore.
4694 17. If headers_add on a router specified a text string that was too long for
4695 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4696 of string_sprintf() is now avoided.
4698 18. $message_body_size was not set (it was always zero) when running the DATA
4699 ACL and the local_scan() function.
4701 19. For the "mail" command in an Exim filter, no default was being set for
4702 the once_repeat time, causing a random time value to be used if "once" was
4703 specified. (If the value happened to be <= 0, no repeat happened.) The
4704 default is now 0s, meaning "never repeat". The "vacation" command was OK
4705 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4706 (I found it when inspecting the code).
4708 20. There is now an overall timeout for performing a callout verification. It
4709 defaults to 4 times the callout timeout, which applies to individual SMTP
4710 commands during the callout. The overall timeout applies when there is more
4711 than one host that can be tried. The timeout is checked before trying the
4712 next host. This prevents very long delays if there are a large number of
4713 hosts and all are timing out (e.g. when the network connections are timing
4714 out). The value of the overall timeout can be changed by specifying an
4715 additional sub-option for "callout", called "maxwait". For example:
4717 verify = sender/callout=5s,maxwait=20s
4719 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4720 to the end before writing, but this should make it even safer).
4722 22. Exim was forgetting that it had advertised PIPELINING for the second and
4723 subsequent messages on an SMTP connection. It was also not resetting its
4724 memory on STARTTLS and an internal HELO.
4726 23. When Exim logs an SMTP synchronization error within a session, it now
4727 records whether PIPELINING has been advertised or not.
4729 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4730 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4731 rather than long int.
4733 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4735 26. Added the never_mail option to autoreply.
4741 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4742 crash if the getsockname() call failed; this can happen if a connection is
4743 closed very soon after it is established. The problem was simply in the
4744 order in which certain operations were done, causing Exim to try to write
4745 to the SMTP stream before it had set up the file descriptor. The bug has
4746 been fixed by making things happen in the correct order.
4752 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4753 before the connection was closed, thus losing the rejection response.
4755 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4756 some early Solaris releases, but causes trouble in current releases where
4757 socklen_t is defined.
4759 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4762 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4765 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4766 that releasing the top of it at the end releases what was used for sub-
4767 expansions (unless the block got too big). However, discard this block if
4768 the first thing is a variable or header, so that we can use its block when
4769 it is dynamic (useful for very large $message_headers, for example).
4771 6. Lookups now cache *every* query, not just the most recent. A new, separate
4772 store pool is used for this. It can be recovered when all lookup caches are
4773 flushed. Lookups now release memory at the end of their result strings.
4774 This has involved some general refactoring of the lookup sources.
4776 7. Some code has been added to the store_xxx() functions to reduce the amount
4777 of flapping under certain conditions.
4779 8. log_incoming_interface used to affect only the <= reception log lines. Now
4780 it causes the local interface and port to be added to several more SMTP log
4781 lines, for example "SMTP connection from", and rejection lines.
4783 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4785 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4787 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4788 could be overwritten at the end of the current message (or the start of a
4789 new message if it was set in a HELO ACL). The value is now preserved for
4790 the duration of the SMTP connection.
4792 12. If a transport had a headers_rewrite setting, and a matching header line
4793 contained an unqualified address, that address was qualified, even if it
4794 did not match any rewriting rules. The underlying bug was that the values
4795 of the flags that permit the existence of unqualified sender and recipient
4796 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4797 non-local messages, and by -bnq for local messages) were not being
4798 preserved with the message after it was received.
4800 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4801 "next input=" as part of the text comprising the host identity instead of
4802 the correct text. The code was using the same buffer for two different
4803 strings. However, depending on which order the printing function evaluated
4804 its arguments, the bug did not always show up. Under Linux, for example, my
4805 test suite worked just fine.
4807 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4808 doesn't work with some older versions of Perl. It has been changed to "my",
4809 which in any case is probably the better facility to use.
4811 15. A really picky compiler found some instances of statements for creating
4812 error messages that either had too many or two few arguments for the format
4815 16. The size of the buffer for calls to the DNS resolver has been increased
4816 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4817 for addresses that have a lot of PTR records. This alleviates a problem; it
4818 does not fully solve it.
4820 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4821 buffer now truncates the list and logs the incident, which is the same
4822 action as happens when Exim is looking up a host name and its aliases.
4823 Previously in this situation something unpredictable would happen;
4824 sometimes it was "internal error: store_reset failed".
4826 18. If a server dropped the connection unexpectedly when an Exim client was
4827 using GnuTLS and trying to read a response, the client delivery process
4828 crashed while trying to generate an error log message.
4830 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4831 string, for example:
4833 warn message = H1: something\nH2: something
4835 the text was added as a single header line from Exim's point of view
4836 though it ended up OK in the delivered message. However, searching for the
4837 second and subsequent header lines using $h_h2: did not work. This has been
4838 fixed. Similarly, if a system filter added multiple headers in this way,
4839 the routers could not see them.
4841 20. Expanded the error message when iplsearch is called with an invalid key to
4842 suggest using net-iplsearch in a host list.
4844 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4845 ACLs are no longer actually imposed (and a message to that effect is
4848 22. If a "gecos" field in a passwd entry contained escaped characters, in
4849 particular, if it contained a \" sequence, Exim got it wrong when building
4850 a From: or a Sender: header from that name. A second bug also caused
4851 incorrect handling when an unquoted " was present following a character
4852 that needed quoting.
4854 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4855 was not being matched caselessly.
4857 24. Arranged for all hyphens in the exim.8 source to be escaped with
4860 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4861 properly. Recipient callout cache records were still being keyed to include
4862 the sender, even when use_sender was set false. This led to far more
4863 callouts that were necessary. The sender is no longer included in the key
4864 when use_sender is false.
4866 26. Added "control = submission" modifier to ACLs.
4868 27. Added the ${base62d: operator to decode base 62 numbers.
4870 28. dnsdb lookups can now access SRV records.
4872 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4873 the configuration file.
4875 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4876 (-v) mode. This makes the output for a verbose queue run more intelligible.
4878 31. Added a use_postmaster feature to recipient callouts.
4880 32. Added the $body_zerocount variable, containing the number of binary zero
4881 bytes in the message body.
4883 33. The time of last modification of the "new" subdirectory is now used as the
4884 "mailbox time last read" when there is a quota error for a maildir
4887 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4889 35. Added +ignore_unknown as a special item in host lists.
4891 36. Code for decoding IPv6 addresses in host lists is now included, even if
4892 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4893 address was recognized as an IP address, but was then not correctly decoded
4894 into binary, causing unexpected and incorrect effects when compared with
4901 1. Very minor rewording of debugging text in manualroute to say "list of
4902 hosts" instead of "hostlist".
4904 2. If verify=header_syntax was set, and a header line with an unqualified
4905 address (no domain) and a large number of spaces between the end of the
4906 name and the colon was received, the reception process suffered a buffer
4907 overflow, and (when I tested it) crashed. This was caused by some obsolete
4908 code that should have been removed. The fix is to remove it!
4910 3. When running in the test harness, delay a bit after writing a bounce
4911 message to get a bit more predictability in the log output.
4913 4. Added a call to search_tidyup() just before forking a reception process. In
4914 theory, someone could use a lookup in the expansion of smtp_accept_max_
4915 per_host which, without the tidyup, could leave open a database connection.
4917 5. Added the variables $recipient_data and $sender_data which get set from a
4918 lookup success in an ACL "recipients" or "senders" condition, or a router
4919 "senders" option, similar to $domain_data and $local_part_data.
4921 6. Moved the writing of debug_print from before to after the "senders" test
4924 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4925 problems for message scanning, either using a data ACL, or using
4926 local_scan() because the Received: header was not generated till after they
4927 were called (in order to set the time as the time of reception completion).
4928 I have revised the way this works. The header is now generated after the
4929 body is received, but before the ACL or local_scan() are called. After they
4930 are run, the timestamp in the header is updated.
4936 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4937 before starting a queue runner without re-exec. This happened only when
4938 deliver_drop_privilege was set or when the Exim user was set to root. The
4939 effect of the bug was that timeouts during subsequent deliveries caused
4940 crashes instead of being properly handled. The handler is now left at its
4941 default (and expected) setting.
4943 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4944 message, again when deliver_drop_privilege is set or Exim is run as root.
4945 The bug described in (1) was not present in this case, but the tidying up
4946 of the other signals was missing. I have made the two cases consistent.
4948 3. The ignore_target_hosts setting on a manualroute router was being ignored
4949 for hosts that were looked up using the /MX notation.
4951 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4954 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4955 operated on the sender address. After changing the $sender_address to <>
4956 for the sender address verify, Exim was re-instated it as the original
4957 (before rewriting) address, but remembering that it had rewritten it, so it
4958 wasn't rewriting it again. This bug also had the effect of breaking the
4959 sender address verification caching when the sender address was rewritten.
4961 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4962 This has been changed so that if the ip literal address matches
4963 ignore_target_hosts, the router declines.
4965 7. Added expansion conditions match_domain, match_address, and match_local_
4966 part (NOT match_host).
4968 8. The placeholder for the Received: header didn't have a length field set.
4970 9. Added code to Exim itself and to exim_lock to test for a specific race
4971 condition that could lead to file corruption when using MBX delivery. The
4972 issue is with the lockfile that is created in /tmp. If this file is removed
4973 after a process has opened it but before that process has acquired a lock,
4974 there is the potential for a second process to recreate the file and also
4975 acquire a lock. This could lead to two Exim processes writing to the file
4976 at the same time. The added code performs the same test as UW imapd; it
4977 checks after acquiring the lock that its file descriptor still refers to
4978 the same named file.
4980 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4981 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4984 11. Added the smtp_active_hostname option. If used, this will typically be made
4985 to depend on the incoming interface address. Because $interface_address is
4986 not set up until the daemon has forked a reception process, error responses
4987 that can happen earlier (such as "too many connections") no longer contain
4990 12. If an expansion in a condition on a "warn" statement fails because a lookup
4991 defers, the "warn" statement is abandoned, and the next ACL statement is
4992 processed. Previously this caused the whole ACL to be aborted.
4994 13. Added the iplsearch lookup type.
4996 14. Added ident_timeout as a log selector.
4998 15. Added tls_certificate_verified as a log selector.
5000 16. Added a global option tls_require_ciphers (compare the smtp transport
5001 option of the same name). This controls incoming TLS connections.
5003 17. I finally figured out how to make tls_require_ciphers do a similar thing
5004 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5005 before starting the TLS session.
5007 18. Tabs are now shown as \t in -bP output.
5009 19. If the log selector return_path_on_delivery was set, Exim crashed when
5010 bouncing a message because it had too many Received: header lines.
5012 20. If two routers both had headers_remove settings, and the first one included
5013 a superfluous trailing colon, the final name in the first list and the
5014 first name in the second list were incorrectly joined into one item (with a
5015 colon in the middle).
5021 1. Added -C and -D options to the exinext utility, mainly to make it easier
5022 to include in the automated testing, but these could be helpful when
5023 multiple configurations are in use.
5025 2. The exinext utility was not formatting the output nicely when there was
5026 an alternate port involved in the retry record key, nor when there was a
5027 message id as well (for retries that were specific to a specific message
5028 and a specific host). It was also confused by IPv6 addresses, because of
5029 the additional colons they contain. I have fixed the IPv4 problem, and
5030 patched it up to do a reasonable job for IPv6.
5032 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5033 delivery, the log line now contains "pipelined" if PIPELINING was used.
5035 4. An SMTP transport process used to panic and die if the bind() call to set
5036 an explicit outgoing interface failed. This has been changed; it is now
5037 treated in the same way as a connect() failure.
5039 5. A reference to $sender_host_name in the part of a conditional expansion
5040 that was being skipped was still causing a DNS lookup. This no longer
5043 6. The def: expansion condition was not recognizing references to header lines
5044 that used bh_ and bheader_.
5046 7. Added the _cache feature to named lists.
5048 8. The code for checking quota_filecount in the appendfile transport was
5049 allowing one more file than it should have been.
5051 9. For compatibility with Sendmail, the command line option
5059 and sets the incoming protocol and host name (for trusted callers). The
5060 host name and its colon can be omitted when only the protocol is to be set.
5061 Note the Exim already has two private options, -pd and -ps, that refer to
5062 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5063 "s", but I don't think that's a major issue.
5065 10. A number of refactoring changes to the code, none of which should affect
5068 (a) The number of logging options was getting close to filling up the
5069 32-bit word that was used as a bit map. I have split them into two classes:
5070 those that are passed in the argument to log_write(), and those that are
5071 only ever tested independently outside of that function. These are now in
5072 separate 32-bit words, so there is plenty of room for expansion again.
5073 There is no change in the user interface or the logging behaviour.
5075 (b) When building, for example, log lines, the code previously used a
5076 macro that called string_cat() twice, in order to add two strings. This is
5077 not really sufficiently general. Furthermore, there was one instance where
5078 it was actually wrong because one of the argument was used twice, and in
5079 one call a function was used. (As it happened, calling the function twice
5080 did not affect the overall behaviour.) The macro has been replaced by a
5081 function that can join an arbitrary number of extra strings onto a growing
5084 (c) The code for expansion conditions now uses a table and a binary chop
5085 instead of a serial search (which was left over from when there were very
5086 few conditions). Also, it now recognizes conditions like "pam" even when
5087 the relevant support is not compiled in: a suitably worded error message is
5088 given if an attempt is made to use such a condition.
5090 11. Added ${time_interval:xxxxx}.
5092 12. A bug was causing one of the ddress fields not to be passed back correctly
5093 from remote delivery subprocesses. The field in question was not being
5094 subsequently used, so this caused to problems in practice.
5096 13. Added new log selectors queue_time and deliver_time.
5098 14. Might have fixed a bug in maildirsizefile handling that threw up
5099 "unexpected character" debug warnings, and recalculated the data
5100 unnecessarily. In any case, I expanded the warning message to give more
5103 15. Added the message "Restricted characters in address" to the statements in
5104 the default ACL that block characters like @ and % in local parts.
5106 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5107 Three changes have been made:
5109 (a) There was a serious bug; a negative response to MAIL caused the whole
5110 recipient domain to be cached as invalid, thereby blocking all messages
5111 to all local parts at the same domain, from all senders. This bug has
5112 been fixed. The domain is no longer cached after a negative response to
5113 MAIL if the sender used is not empty.
5115 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5118 (c) A new callout option, "use_sender" has been added for people who want
5119 the modified behaviour.
5125 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5128 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5129 indeed breaks things for older releases.
5131 3. Added additional logging to the case where there is a problem reading data
5132 from a filter that is running in a subprocess using a pipe, in order to
5133 try to track down a specific problem.
5135 4. Testing facility fudge: when running in the test harness and attempting
5136 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5137 getting "No route to host". Convert this to a timeout.
5139 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5142 6. Some OS don't have socklen_t but use size_t instead. This affects the
5143 fifth argument of getsockopt() amongst other things. This is now
5144 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5145 can be set for individual OS. I have set it for SunOS5, OSF1, and
5146 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5147 some earlier ones do not.
5149 7. Change 4.30/15 was not doing the test caselessly.
5151 8. The standard form for an IPv6 address literal was being rejected by address
5152 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5153 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5154 this, as well as the form without the "IPv6" on the front (but only when
5155 address literals are enabled, of course).
5157 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5159 10. Exim crashed if a message with an empty sender address specified by -f
5160 encountered a router with an errors_to setting. This could be provoked only
5161 by a command such as
5165 where an empty string was supplied; "<>" did not hit this bug.
5167 11. Installed PCRE release 4.5.
5169 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5170 remained set. It is now erased.
5172 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5173 times from message ids (which are base 36 rather than the normal 62).
5175 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5176 were being counted as actual protocol errors, and logged if the log
5177 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5178 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5179 and DATA following a set of rejected RCPTs do not count as protocol errors.
5180 In other words, Exim assumes they were pipelined, though this may not
5181 actually be the case. Of course, in all cases the client gets an
5182 appropriate error code.
5184 15. If a lookup fails in an ACL condition, a message about the failure may
5185 be available; it is used if testing the ACL cannot continue, because most
5186 such messages specify what the cause of the deferral is. However, some
5187 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5188 that caused an old message to be retained and used if a later statement
5189 caused a defer, replacing the real cause of the deferral.
5191 16. If an IP address had so many PTR records that the DNS lookup buffer
5192 was not large enough to hold them, Exim could crash while trying to process
5193 the truncated data. It now detects and logs this case.
5195 17. Further to 4.21/58, another change has been made: if (and only if) the
5196 first line of a message (the first header line) ends with CRLF, a bare LF
5197 in a subsequent header line has a space inserted after it, so as not to
5198 terminate the header.
5200 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5201 unnecessarily, used atoi() instead of strtol(), and didn't check the
5202 termination when getting file sizes from file names by regex.
5204 19. Completely re-implemented the support for maildirsize files, in the light
5205 of a number of problems with the previous contributed implementation
5206 (4.30/29). In particular:
5208 . If the quota is zero, the maildirsize file is maintained, but no quota is
5211 . If the maildir directory does not exist, it is created before any attempt
5212 to write a maildirsize file.
5214 . The quota value in the file is just a cache; if the quota is changed in
5215 the transport, the new value overrides.
5217 . A regular expression is available for excluding directories from the
5220 20. The autoreply transport checks the characters in options that define the
5221 message's headers; it allows continued headers, but it was checking with
5222 isspace() after an embedded newline instead of explicitly looking for a
5225 21. If all the "regular" hosts to which an address was routed had passed their
5226 expiry times, and had not reached their retry times, the address was
5227 bounced, even if fallback hosts were defined. Now Exim should go on to try
5230 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5231 equivalent code in the SMTP transport. Some hosts send humungous responses
5232 to HELO/EHLO, more than 1024 it seems.
5234 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5235 gives compiler warnings in some environments. I've now done it "properly",
5238 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5239 (because of problems with the built-in one) was declared to return uschar *
5240 instead of char *, causing compiler failure.
5242 25. Fixed a file descriptor leak when processing alias/forward files.
5244 26. Fixed a minor format string issue in dbfn.c.
5246 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5248 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5249 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5252 29. When a sender address is verified, it is cached, to save repeating the test
5253 when there is more than one recipient in a message. However, when the
5254 verification involves a callout, it is possible for different callout
5255 options to be set for different recipients. It is too complicated to keep
5256 track of this in the cache, so now Exim always runs a verification when a
5257 callout is required, relying on the callout cache for the optimization.
5258 The overhead is duplication of the address routing, but this should not be
5261 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5262 to be verified with callout=postmaster, and the main callout worked but the
5263 postmaster check failed, the verification correctly failed. However, if a
5264 subsequent RCPT command asked for sender verification *without* the
5265 postmaster check, incorrect caching caused this verification also to fail,
5268 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5269 it was not caching the DNS options (qualify_single, search_parents) that
5270 were used when the lookup failed. A subsequent lookup with different
5271 options therefore always gave the same answer, though there were cases
5272 where it should not have. (Example: a "domains = !$mx_any" option on a
5273 dnslookup router: the "domains" option is always processed without any
5274 widening, but the router might have qualify_single set.) Now Exim uses the
5275 cached value only when the same options are set.
5277 32. Added John Jetmore's "exipick" utility to the distribution.
5279 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5280 than a timeout (e.g. a certificate is required, and is not provided), an
5281 Exim server now closes the connection immediately. Previously it waited for
5282 the client to close - but if the client is SSL, it seems that they each
5283 wait for each other, leading to a delay before one of them times out.
5285 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5286 maintained 0.8.x compatibility because I don't think many are using it, and
5287 it is clearly obsolete.
5289 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5292 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5293 client certificate was expired. A simple patch fixes this, though I don't
5294 understand the full logic of why the verify callback is called multiple
5297 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5298 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5299 which causes problems with some clients (such as the Certicom SSL Plus
5300 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5301 disables the coutermeasure allowing Eudora to connect."
5303 38. Exim was not checking that a write() to a log file succeeded. This could
5304 lead to Bad Things if a log got too big, in particular if it hit a file
5305 size limit. Exim now panics and dies if it cannot write to a log file, just
5306 as it does if it cannot open a log file.
5308 39. Modified OS/Makefile-Linux so that it now contains
5310 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5312 The two -D definitions ensure that Exim is compiled with large file
5313 support, which makes it possible to handle log files that are bigger than
5316 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5317 instance) a domain was checked against a named list that involved a lookup,
5318 causing $domain_data to be set, then another domain was checked against the
5319 same list, then the first domain was re-checked, the value of $domain_data
5320 after the final check could be wrong. In particular, if the second check
5321 failed, it could be set empty. This bug probably also applied to
5324 41. The strip_trailing_dot option was not being applied to the address given
5325 with the -f command-line option.
5327 42. The code for reading a message's header from the spool was incrementing
5328 $received_count, but never initializing it. This meant that the value was
5329 incorrect (doubled) while delivering a message in the same process in which
5330 it was received. In the most common configuration of Exim, this never
5331 happens - a fresh exec is done - but it can happen when
5332 deliver_drop_privilege is set.
5334 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5335 it now includes up to 150 characters of the unexpected data in the log
5338 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5339 and building data strings. The size of both of these buffers was 10 000
5340 bytes - far larger than anybody would *ever* want, thought I. Needless to
5341 say, somebody hit the limit. I have increased the maximum line length to
5342 20 000 and the maximum data length of concatenated lines to 100 000. I have
5343 also fixed two bugs, because there was no checking on these buffers. Tsk,
5344 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5345 buffer is too small.
5347 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5348 lsearch lookups. Now it does.
5350 46. When parsing a route_list item in a manualroute router, a fixed-length
5351 buffer was used for the list of hosts. I made this 1024 bytes long,
5352 thinking that nobody would ever have a list of hosts that long. Wrong.
5353 Somebody had a whole pile of complicated expansion conditions, and the
5354 string was silently truncated, leading to an expansion error. It turns out
5355 that it is easier to change to an unlimited length (owing to other changes
5356 that have happened since this code was originally written) than to build
5357 structure for giving a limitation error. The length of the item that
5358 expands into the list of hosts is now unlimited.
5360 47. The lsearch lookup could not handle data where the length of text line was
5361 more than 4095 characters. Such lines were truncated, leading to shortened
5362 data being returned. It should now handle lines of any length.
5364 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5365 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5368 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5369 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5370 no longer happens in comment lines. A list of replacements is now placed
5371 at the head of all of the source files, except those whose only change is
5372 to replace PERL_COMMAND in the very first #! line.
5374 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5375 messages on the queue, with a bottom-up merge sort, using code contributed
5376 by Michael Haardt. This should make operations like -bp somewhat faster on
5377 large queues. It won't affect queue runners, except when queue_run_in_order
5380 51. Installed eximstats 1.31 in the distribution.
5382 52. Added support for SRV lookups to the dnslookup router.
5384 53. If an ACL referred to $message_body or $message_body_end, the value was not
5385 reset for any messages that followed in the same SMTP session.
5387 54. The store-handling optimization for building very long strings was not
5388 differentiating between the different store pools. I don't think this
5389 actually made any difference in practice, but I've tidied it.
5391 55. While running the routers to verify a sender address, $sender_address
5392 was still set to the sender address. This is wrong, because when routing to
5393 send a bounce to the sender, it would be empty. Therefore, I have changed
5394 it so that, while verifying a sender address, $sender_address is set to <>.
5395 (There is no change to what happens when verifying a recipient address.)
5397 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5398 target A or AAAA records (if not already returned) without resetting the
5399 qualify_single or search_parents options of the DNS resolver. These are
5400 inappropriate in this case because the targets of MX and SRV records must
5401 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5402 target that, when qualified, matched something in the local domain. These
5403 two options are now turned off when doing these lookups.
5405 57. It seems that at least some releases of Reiserfs (which does not have the
5406 concept of a fixed number of inodes) returns zero and not -1 for the
5407 number of available inodes. This interacted badly with check_spool_inodes,
5408 which assumed that -1 was the "no such thing" setting. What I have done is
5409 to check that the total number of inodes is greater than zero before doing
5410 the test of how many are available.
5412 58. When a "warn" ACL statement has a log_message modifier, the message is
5413 remembered, and not repeated. This is to avoid a lot of repetition when a
5414 message has many recipients that cause the same warning to be written.
5415 Howewer, Exim was preserving the list of already written lines for an
5416 entire SMTP session, which doesn't seem right. The memory is now reset if a
5417 new message is started.
5419 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5420 debugging output unless log_rewrite was also set.
5422 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5423 of (int)(handle) when we know that handle contains (void *)(-1).
5425 61. The Exim daemon panic-logs an error return when it closes the incoming
5426 connection. However "connection reset by peer" seems to be common, and
5427 isn't really an error worthy of noting specially, so that particular error
5430 62. When Exim is trying to find all the local interfaces, it used to panic and
5431 die if the ioctl to get the interface flags failed. However, it seems that
5432 on at least one OS (Solaris 9) it is possible to have an interface that is
5433 included in the list of interfaces, but for which you get a failure error
5434 for this call. This happens when the interface is not "plumbed" into a
5435 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5436 failure of the "get flags" call assumes that the interface is down.
5438 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5439 makes life easier for people who are doing arithmetic on fields extracted
5440 from dates, where you often get leading zeros that should not be
5441 interpreted as octal.
5443 64. Added qualify_domain to the redirect router, to override the global
5446 65. If a pathologically long header line contained very many addresses (the
5447 report of this problem mentioned 10 000) and each of them was rewritten,
5448 Exim could use up a very large amount of memory. (It kept on making new
5449 copies of the header line as it rewrote, and never released the old ones.)
5450 At the expense of a bit more processing, the header rewriting function has
5451 been changed so that it no longer eats memory in this way.
5453 66. The generation of the Received: header has been moved from the time that a
5454 message starts to be received, to the time that it finishes. The timestamp
5455 in the Received: header should now be very close to that of the <= log
5456 line. There are two side-effects of this change:
5458 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5459 logged header lines no longer include the local Received: line, because
5460 it has not yet been created. The same applies to a copy of the message
5461 that is returned to a non-SMTP sender when a message is rejected.
5463 (b) When a filter file is tested using -bf, no additional Received: header
5464 is added to the test message. After some thought, I decided that this
5467 This change does not affect the value of $received_for. It is still set
5468 after address rewriting, but before local_scan() is called.
5470 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5472 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5473 gave an unhelpful panic error message, and a defer error. I have managed to
5474 change this behaviour so that it now rejects any supplied certificate,
5475 which seems right, as the list of acceptable certificates is empty.
5477 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5478 gave an unhelpful defer error. I have not managed to make this reject any
5479 supplied certificates, but the error message it gives is "no certificate
5480 supplied", which is not helpful.
5482 70. exigrep's output now also includes lines that are not associated with any
5483 message, but which match the given pattern. Implemented by a patch from
5484 Martin Sluka, which also tidied up the Perl a bit.
5486 71. Recipient callout verification, like sender verification, was using <> in
5487 the MAIL FROM command. This isn't really the right thing, since the actual
5488 sender may affect whether the remote host accepts the recipient or not. I
5489 have changed it to use the actual sender in the callout; this means that
5490 the cache record is now keyed on a recipient/sender pair, not just the
5491 recipient address. There doesn't seem to be a real danger of callout loops,
5492 since a callout by the remote host to check the sender would use <>.
5493 [SEE ABOVE: changed after hitting problems.]
5495 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5496 temporary errors. However, in the case of such a code being given after
5497 the end of a data transmission (i.e. after ".") Exim was failing to write
5498 a retry record for the message. (Yes, there was some broken host that was
5499 actually sending 8xx at this point.)
5501 73. An unknown lookup type in a host list could cause Exim to panic-die when
5502 the list was checked. (An example that provoked this was putting <; in the
5503 middle of a list instead of at the start.) If this happened during a DATA
5504 ACL check, a -D file could be left lying around. This kind of configuration
5505 error no longer causes Exim to die; instead it causes a defer errror. The
5506 incident is still logged to the main and panic logs.
5508 74. Buglet left over from Exim 3 conversion. The message "too many messages
5509 in one connection" was written to the rejectlog but not the mainlog, except
5510 when address rewriting (yes!) was being logged.
5512 75. Added write_rejectlog option.
5514 76. When a system filter was run not as root (that is, when system_filter_user
5515 was set), the values of the $n variables were not being returned to the
5516 main process; thus, they were not subsequently available in the $sn
5519 77. Added +return_path_on_delivery log selector.
5521 78. A connection timeout was being treated differently from recipients deferred
5522 when testing hosts_max_try with a message that was older than the host's
5523 retry timeout. (The host should not be counted, thus allowing all hosts to
5524 be tried at least once before bouncing.) This may have been the cause of an
5525 occasionally reported bug whereby a message would remain on the queue
5526 longer than the retry timeout, but would be bounced if a delivery was
5527 forced. I say "may" because I never totally pinned down the problem;
5528 setting up timeout/retry tests is difficult. See also the next item.
5530 79. The ultimate address timeout was not being applied to errors that involved
5531 a combination of host plus message (for example, a timeout on a MAIL
5532 command). When an address resolved to a number of possible hosts, and they
5533 were not all tried for each delivery (e.g. because of hosts_max_try), a
5534 message could remain on the queue longer than the retry timeout.
5536 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5539 81. Fixed an obscure SMTP outgoing bug which required at least the following
5540 conditions: (a) there was another message waiting for the same server;
5541 (b) the server returned 5xx to all RCPT commands in the first message so
5542 that the message was not completed; (c) the server dropped the connection
5543 or gave a negative response to the RSET that Exim sends to abort the
5544 transaction. The observed case was a dropped connection after DATA that had
5545 been sent in pipelining mode. That is, the server had advertised PIPELINING
5546 but was not implementing it correctly. The effect of the bug was incorrect
5547 behaviour, such as trying another host, and this could lead to a crash.
5553 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5554 and daemon.c were passed as pointers to ints; they should have been
5555 pointers to socklen_t variables (which are typically unsigned ints).
5557 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5560 3. Fixed a really odd bug that affected only the testing scheme; patching a
5561 certain fixed string in the binary changed the value of another string that
5562 happened to be identical to the end of the original first string.
5564 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5565 name", it returns that address as the IP address. On some operating
5566 systems (e.g. Solaris), it also passes back the IP address string as the
5567 "host name". However, on others (e.g. Linux), it passes back an empty
5568 string. Exim wasn't checking for this, and was changing the host name to an
5569 empty string, assuming it had been canonicized.
5571 5. Although rare, it is permitted to have more than one PTR record for a given
5572 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5573 all the names associated with an address, because they do in Solaris.
5574 However, it seems that they do not in Linux for data that comes from the
5575 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5576 I found this out when I moved to a new Linux workstation and tried to run
5577 the Exim test suite.
5579 To get round this problem I have changed the code so that it now does its
5580 own call to the DNS to look up PTR records when searching for a host name.
5581 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5582 addresses that are only in /etc/hosts are still found.
5584 This behaviour is, however, controlled by an option called host_lookup_
5585 order, which defaults to "bydns:byaddr". If people want to use the other
5586 order, or indeed, just use one or the other means of lookup, they can
5587 specify it in this variable.
5589 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5590 some operating systems, this comes back from gethostbyaddr() as an empty
5591 string, and this is what Exim used to test for. However, it seems that in
5592 other systems, "." is yielded. Exim now tests for this case too.
5594 7. The values of check_spool_space and check_log_space are now held internally
5595 as a number of kilobytes instead of an absolute number of bytes. If a
5596 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5597 kilobyte. This means that much larger values can be stored.
5599 8. Exim monitor: an attempt to get the action menu when not actually pointing
5600 at a message produces an empty menu entitled "No message selected". This
5601 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5602 no entries in it ("Shell widget menu has zero width and/or height"). So I
5603 have added a single, blank menu entry in this case.
5605 9. Added ${quote_local_part.
5607 10. MIME decoding is now applied to the contents of Subject: header lines when
5610 11. Now that a reference to $sender_host_address automatically causes a reverse
5611 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5612 host lookup before query-style lookups in lists that might use this
5613 variable. This has therefore been abolished, and the "net-" prefix is no
5614 longer necessary for query-style lookups.
5616 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5617 have been a typo for LFLAGS, so it has been changed.
5619 13. The install script calls Exim with "-C /dev/null" in order to find the
5620 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5621 to be output. Howeve, since Exim outputs its version number before the
5622 error, it didn't break the script. It just looked ugly. I fixed this by
5623 always allowing "-C /dev/null" if the caller is root.
5625 14. Ignore overlarge ACL variable number when reading spool file - insurance
5626 against a later release with more variables having written the file.
5628 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5629 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5630 this, as well as the form without the "IPv6" on the front.
5632 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5633 OS/Makefile-Darwin file.
5635 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5636 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5638 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5639 variable. However, some OS use ints for inodes. I've added cast to long int
5640 to get rid of the compiler warning.
5642 19. I had forgotten to lock out "/../" in configuration file names when
5643 ALT_CONFIG_PREFIX was set.
5645 20. Routers used for verification do not need to specify transports. However,
5646 if such a router generated a host list, and callout was configured, Exim
5647 crashed, because it could not find a port number from the (non-existent)
5648 transport. It now assumes port 25 in this circumstance.
5650 21. Added the -t option to exigrep.
5652 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5653 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5654 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5655 them is set, LOOKUP_LSEARCH is forced.
5657 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5659 24. Added sender and host information to the "rejected by local_scan()" log
5660 line; previously there was no indication of these.
5662 25. Added .include_if_exists.
5664 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5665 Linux. It turns out that not all file systems support this. Apparently some
5666 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5667 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5668 not supported on the file descriptor, is now ignored when Exim is trying to
5669 sync a directory. This applies only to Linux.
5671 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5673 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5674 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5677 29. Added support for maildirsize files from supplied patch (modified a bit).
5679 30. The use of :fail: followed by an empty string could lead Exim to respond to
5680 sender verification failures with (e.g.):
5682 550 Verification failed for <xxx>
5683 550 Sender verify failed
5685 where the first response line was missing the '-' that indicates it is not
5686 the final line of the response.
5688 31. The loop for finding the name of the user that called Exim had a hardwired
5689 limit of 10; it now uses the value of finduser_retries, which is used for
5690 all other user lookups.
5692 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5695 33. Exim was neglecting to zero errno before one call of strtol() when
5696 expanding a string and expecting an integer value. On some systems this
5697 resulted in spurious "integer overflow" errors. Also, it was casting the
5698 result into an int without checking.
5700 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5701 did not work. The code looks as if it has *never* worked, though it appears
5702 to have been documented since at least releast 1.62. I have made it work.
5704 35. The "timeout_DNS" error in retry rules, also documented since at least
5705 1.62, also never worked. As it isn't clear exactly what this means, and
5706 clearly it isn't a major issue, I have abolished the feature by treating it
5707 as "timeout", and writing a warning to the main and panic logs.
5709 36. The display of retry rules for -brt wasn't always showing the error code
5712 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5713 timeout_connect_A, timeout_connect_MX.
5715 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5716 to the empty sender.
5718 39. The daemon was not analysing the content of -oX till after it had closed
5719 stderr and disconnected from the controlling terminal. This meant that any
5720 syntax errors were only noted on the panic log, and the return code from
5721 the command was 0. By re-arranging the code a little, I've made the
5722 decoding happen first, so such errors now appear on stderr, and the return
5723 code is 1. However, the actual setting up of the sockets still happens in
5724 the disconnected process, so errors there are still only recorded on the
5727 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5728 connections (as happens on some IP stacks) was logged at start up time as
5729 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5730 it from "IPv6 and IPv4", which means that two separate sockets are being
5733 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5734 says whether AF_INET or AF_INET6 was passed as an argument.
5736 42. Exiwhat output was messed up when time zones were included in log
5739 43. Exiwhat now gives more information about the daemon's listening ports,
5740 and whether -tls-on-connect was used.
5742 44. The "port" option of the smtp transport is now expanded.
5744 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5745 silently ignored. Now an error message is written to the main and panic
5748 46. There's a new ACL modifier called "logwrite" which writes to a log file
5749 as soon as it is encountered.
5751 47. Added $local_user_uid and $local_user_gid at routing time.
5753 48. Exim crashed when trying to verify a sender address that was being
5756 49. Exim was recognizing only a space character after ".include". It now also
5757 recognizes a tab character.
5759 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5760 extracting the relevant information from the specification. The man page no
5761 longer contains scrambled data for the -d option, and I've added a section
5762 at the front about calling Exim under different names.
5764 51. Added "extra_headers" argument to the "mail" command in filter files.
5766 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5769 53. Installed eximstats 1.29.
5771 54. Added transport_filter_timeout as a generic transport option.
5773 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5774 Cc: header lines. This was required by RFC 822, but it not required by RFC
5777 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5778 incoming messages that did not have them. Now it does so only if the
5779 message originates locally, that is, if there is no associated remote host
5780 address. When Resent- header lines are present, this applies to the Resent-
5781 lines rather than the non-Resent- lines.
5783 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5784 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5786 58. Messages for configuration errors now include the name of the main
5787 configuration file - useful now that there may be more than one file in a
5788 list (.included file names were always shown).
5790 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5791 for those rare installations that do not start the daemon as root or run it
5792 setuid root. I've cut out the call to initgroups() if the daemon is not
5795 60. The Exim user and group can now be bound into the binary as text strings
5796 that are looked up at the start of Exim's processing.
5798 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5800 62. Added $mailstore_basename variable.
5802 63. Installed patch to sieve.c from Michael Haardt.
5804 64. When Exim failed to open the panic log after failing to open the main log,
5805 the original message it was trying to log was written to stderr and debug
5806 output, but if they were not available (the usual case in production), it
5807 was lost. Now it is written to syslog before the two lines that record the
5808 failures to open the logs.
5810 65. Users' Exim filters run in subprocesses under the user's uid. It is
5811 possible for a "deliver" command or an alias in a "personal" command to
5812 provoke an address rewrite. If logging of address rewriting is configured,
5813 this fails because the process is not running as root or exim. There may be
5814 a better way of dealing with this, but for the moment (because 4.30 needs
5815 to be released), I have disabled address rewrite logging when running a
5816 filter in a non-root, non-exim process.
5822 1. The buildconfig auxiliary program wasn't quoting the value set for
5823 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5824 not defined. This bug was masked in 4.22 by the effect that was fixed in
5827 2. Some messages that were rejected after a message id was allocated were
5828 shown as "incomplete" by exigrep. It no longer does this for messages that
5829 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5831 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5832 have allow_domain_literals set, the ID did not get logged in the <= line.
5833 Domain literals are now always recognized in Message-ID: header lines.
5835 4. The first argument for a ${extract expansion item is the key name or field
5836 number. Leading and trailing spaces in this item were not being ignored,
5837 causing some misleading effects.
5839 5. When deliver_drop_privilege was set, single queue runner processes started
5840 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5841 same command in the process it spins off) were not dropping privilege.
5843 6. When the daemon running as "exim" started a queue runner, it always
5844 re-executed Exim in the spun-off process. This is a waste of effort when
5845 deliver_drop_privilege is set. The new process now just calls the
5846 queue-runner function directly.
5852 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5855 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5856 ignored. Though the use of -oP was forcing the writing of a pid file, it
5857 was always written to the default place.
5859 3. If the message "no IP address found for host xxxx" is generated during
5860 incoming verification, it is now followed by identification of the incoming
5861 connection (so you can more easily find what provoked it).
5863 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5865 5. Added some features to "harden" Exim a bit more against certain attacks:
5867 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5868 be put in Local/Makefile. This is like the never_users runtime option,
5869 but it cannot be overridden. The default setting is "root".
5871 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5872 prefix string with which any file named in a -C command line option
5875 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5876 is retained for -C and -D only if the caller of Exim is root. Without
5877 it, the exim user may also use -C and -D and retain privilege.
5879 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5880 command line option is disabled.
5882 6. Macro names set by the -D option must start with an upper case letter, just
5883 like macro names defined in the configuration file.
5885 7. Added "dereference=" facility to LDAP.
5887 8. Two instances of the typo "uknown" in the source files are fixed.
5889 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5890 the Configure-Makefile script screwed up while processing it.
5892 10. Incorporated PCRE 4.4.
5894 11. The SMTP synchronization check was not operating right at the start of an
5895 SMTP session. For example, it could not catch a HELO sent before the client
5896 waited for the greeting. There is now a check for outstanding input at the
5897 point when the greeting is written. Because of the duplex, asynchronous
5898 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5899 way, but not yet received, when the check is performed.
5901 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5902 on TCP/IP sockets, because this apparently causes some broken clients to
5905 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5906 unchanged) from the Cygwin maintainer.
5908 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5909 was supported instead of testing for mbx. Effectively a typo.
5911 15. The spa authenticator server code was not checking that the input it
5912 received was valid base64.
5914 16. The debug output line for the "set" modifier in ACLs was not showing the
5915 name of the variable that was being set.
5917 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5919 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5920 lookup on its own. Something else was needed to trigger the lookup. For
5921 example, a match in host_lookup or the need for a host name in a host list.
5922 Now, if $sender_host_name is referenced and the host name has not yet been
5923 looked up, a lookup is performed. If the lookup fails, the variable remains
5924 empty, and $host_lookup_failed is set to "1".
5926 19. Added "eqi" as a case-independent comparison operator.
5928 20. The saslauthd authentication condition could segfault if neither service
5929 nor realm was specified.
5931 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5932 error message that was logged was misleading, and incoming SMTP
5933 connections were dropped. The message is now more accurate, and temporary
5934 errors are given to SMTP connections.
5936 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5937 (except RSET) with a 421 error, until QUIT is received. However, it was
5938 failing to send a response to QUIT.
5940 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5941 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5942 ACL is now run after the helo_try_verify_hosts code.
5944 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5945 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5946 case-independent, so other case variants are also recognized). Apparently
5947 some systems use these upper case variants.
5949 25. If more than two messages were waiting for the same host, and a transport
5950 filter was specified for the transport, Exim sent two messages over the
5951 same TCP/IP connection, and then failed with "socket operation on non-
5952 socket" when it tried to send the third.
5954 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5956 27. The extern definition of crypt16() in expand.c was not being excluded when
5957 the OS had its own crypt16() function.
5959 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5960 as a preferred synonym for return_size_limit, both as an option and as an
5963 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5965 30. Changed the default configuration ACL to relax the local part checking rule
5966 for addresses that are not in any local domains. For these addresses,
5967 slashes and pipe symbols are allowed within local parts, but the sequence
5968 /../ is explicitly forbidden.
5970 31. SPA server authentication was not clearing the challenge buffer before
5973 32. log_message in a "warn" ACL statement was writing to the reject log as
5974 well as to the main log, which contradicts the documentation and doesn't
5975 seem right (because no rejection is happening). So I have stopped it.
5977 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5978 However, I am unable to do any testing of this.
5980 34. Fixed an infelicity in the appendfile transport. When checking directories
5981 for a mailbox, to see if any needed to be created, it was accidentally
5982 using path names with one or more superfluous leading slashes; tracing
5983 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5985 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5986 contents are added to the log line that is written for every discarded
5987 recipient. (Previously a log_message setting was ignored.)
5989 36. The ${quote: operator now quotes the string if it is empty.
5991 37. The install script runs exim in order to find its version number. If for
5992 some reason other than non-existence or emptiness, which it checks, it
5993 could not run './exim', it was installing it with an empty version number,
5994 i.e. as "exim-". This error state is now caught, and the installation is
5997 38. An argument was missing from the function that creates an error message
5998 when Exim fails to connect to the socket for saslauthd authentication.
5999 This could cause Exim to crash, or give a corrupted message.
6001 39. Added isip, isip4, and isip6 to ${if conditions.
6003 40. The ACL variables $acl_xx are now saved with the message, and can be
6004 accessed later in routers, transports, and filters.
6006 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6007 strings in the file are not string-expanded.
6009 42. If a MAIL command specified a SIZE value that was too large to fit into an
6010 int variable, the check against message_size_limit failed. Such values are
6011 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6012 day this will have to be increased, but I don't think I want to be around
6013 when emails are that large.
6020 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6021 iconv() is not standard in FreeBSD.
6023 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6024 IPv6 enabled. The observed symptom was a segmentation fault on return from
6025 the function os_common_find_running_interfaces() in src/os.c.
6027 3. In the check_special_case() function in daemon.c I had used "errno" as an
6028 argument name, which causes warnings on some systems. This was basically a
6029 typo, since it was named "eno" in the comments!
6031 4. The code that waits for the clock to tick (at a resolution of some fraction
6032 of a second) so as to ensure message-id uniqueness was always waiting for
6033 at least one whole tick, when it could have waited for less. [This is
6034 almost certainly not relevant at current processor speeds, where it is
6035 unlikely to ever wait at all. But we try to future-proof.]
6037 5. The function that sleeps for a time interval that includes fractions of a
6038 second contained a race. It did not block SIGALRM between setting the
6039 timer, and suspending (a couple of lines later). If the interval was short
6040 and the sigsuspend() was delayed until after it had expired, the suspension
6041 never ended. On busy systems this could lead to processes getting stuck for
6044 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6045 process, before it forks any delivery processes. The open lookup caching
6046 mechanism meant that the open file or database connection was passed into
6047 the delivery process. The problem was that delivery processes always tidy
6048 up cached lookup data. This could cause a problem for the next delivery
6049 process started by the queue runner, because the external queue runner
6050 process does not know about the closure. So the next delivery process
6051 still has data in the lookup cache. In the case of a file lookup, there was
6052 no problem because closing a file descriptor in a subprocess doesn't affect
6053 the parent. However, if the lookup was caching a connection to a database,
6054 the connection was closed, and the second delivery process was likely to
6055 see errors such as "PGSQL: query failed: server closed the connection
6056 unexpectedly". The problem has been fixed by closing all cached lookups
6057 in a queue runner before running a delivery process.
6059 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6060 seem to have the "const" qualifier which it has on other OS. I've
6063 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6064 *of the same type* (client or server) with the same public name that an
6065 error should be diagnosed.
6067 9. When Exim looked up a host name for an IP address, but failed to find the
6068 original IP address when looking up the host name (a safety check), it
6069 output the message "<ip address> does not match any IP for NULL", which was
6070 confusing, to say the least. The bug was that the host name should have
6071 appeared instead of "NULL".
6073 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6074 user that is built into the binary, and the -C or -D options is used, root
6075 privilege is dropped before the configuration file is read. In addition,
6076 logging is switched to stderr instead of the normal log files. If the
6077 configuration then re-defines the Exim user, the unprivileged environment
6078 is probably not what is expected, so Exim logs a panic warning message (but
6081 However, if deliver_drop_privilege is set, the unprivileged state may well
6082 be exactly what is intended, so the warning has been cut out in that case,
6083 and Exim is allowed to try to write to its normal log files.
6089 1. smtp_return_error_details was not giving details for temporary sender
6090 or receiver verification errors.
6092 2. Diagnose a configuration error if two authenticators have the same public
6095 3. Exim used not to create the message log file for a message until the first
6096 delivery attempt. This could be confusing when incoming messages were held
6097 for policy or load reasons. The message log file is now created at the time
6098 the message is received, and an initial "Received" line is written to it.
6100 4. The automatically generated man page for command line options had a minor
6101 bug that caused no ill effects; however, a more serious problem was that
6102 the procedure for building the man page automatically didn't always
6103 operate. Consequently, release 4.20 contains an out-of-date version. This
6104 shouldn't happen again.
6106 5. When building Exim with embedded Perl support, the script that builds the
6107 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6108 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6110 6. The freeze_tell option was not being used for messages that were frozen on
6111 arrival, either by an ACL or by local_scan().
6113 7. Added the smtp_incomplete_transaction log selector.
6115 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6116 was accepting AUTH without a new EHLO.
6118 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6119 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6122 10. Logging of TCP/IP connections (when configured) now happens in the main
6123 daemon process instead of the child process, so that the TCP/IP connection
6124 count is more accurate (but it can never be perfect).
6126 11. The use of "drop" in a nested ACL was not being handled correctly in the
6127 outer ACL. Now, if condition failure induced by the nested "drop" causes
6128 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6129 or "require"), the connection is dropped.
6131 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6132 that yield "discard" can now be used with an "accept" or a "discard" verb,
6133 but an error is generated for any others (because I can't see a useful way
6134 to define what should happen).
6136 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6137 are now processed in the same way as lines in the Exim configuration file.
6138 In particular, continuation lines are supported.
6140 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6142 15. Added -ti meaning -t -i.
6144 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6145 domains, and warn by logging if others are found.
6147 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6148 returned from GIFCONF when Exim is trying to find the list of interfaces on
6149 a host. The code in os.c has been modified to copy each ifreq to an aligned
6150 structure in all cases.
6152 Also, in some cases, the returned ifreq's were being copied to a 'struct
6153 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6154 means the last couple of bytes of an IPv6 address could be chopped if the
6155 ifreq contained only a normal sockaddr (14 bytes storage).
6157 18. Named domain lists were not supported in the hosts_treat_as_local option.
6158 An entry such as +xxxx was not recognized, and was treated as a literal
6161 19. Ensure that header lines added by a DATA ACL are included in the reject log
6162 if the ACL subsequently rejects the message.
6164 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6165 MD5 (which is deprecated).
6167 21. When testing a filter file using -bf, Exim was writing a message when it
6168 took the sender from a "From " line in the message, but it was not doing so
6169 when it took $return_path from a Return-Path: header line. It now does.
6171 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6172 with a valid header line field name (a series of printing characters
6173 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6175 23. Changed "disc" in the source to "disk" to conform to the documentation and
6176 the book and for uniformity.
6178 24. Ignore Sendmail's -Ooption=value command line item.
6180 25. When execve() failed while trying to run a command in a pipe transport,
6181 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6182 could be confused with a return value of 69 from the command itself. This
6183 has been changed to 127, the value the shell returns if it is asked to run
6184 a non-existent command. The wording for the related log line suggests a
6185 non-existent command as the problem.
6187 26. If received_header_text expands to an empty string, do not add a Received:
6188 header line to the message. (Well, it adds a token one on the spool, but
6189 marks it "old" so that it doesn't get used or transmitted.)
6191 27. Installed eximstats 1.28 (addition of -nt option).
6193 28. There was no check for failure on the call to getsockname() in the daemon
6194 code. This can fail if there is a shortage of resources on the system, with
6195 ENOMEM, for example. A temporary error is now given on failure.
6197 29. Contrary to the C standard, it seems that in some environments, the
6198 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6199 program. Exim now does this explicitly; it affects the formatting of
6200 timestamps using strftime().
6202 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6203 complaints. I've now initialized all the variables, to avoid this.
6205 32. Header lines added by a system filter were not being "seen" during
6206 transport-time rewrites.
6208 33. The info_callback() function passed to OpenSSL is set up with type void
6209 (*)(SSL *, int, int), as described somewhere. However, when calling the
6210 function (actually a macro) that sets it up, the type void(*)() is
6211 expected. I've put in a cast to prevent warnings from picky compilers.
6213 34. If a DNS black list lookup found a CNAME record, but there were no A
6214 records associated with the domain it pointed at, Exim crashed.
6216 35. If a DNS black list lookup returned more than one A record, Exim ignored
6217 all but the first. It now scans all returned addresses if a particular IP
6218 value is being sought. In this situation, the contents of the
6219 $dnslist_value variable are a list of all the addresses, separated by a
6222 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6223 to accept a host name and all its aliases if the forward lookup for any of
6224 them yielded the IP address of the incoming connection. Now it accepts only
6225 those names whose forward lookup yields the correct IP address. Any other
6226 names are discarded. This closes a loophole whereby a rogue DNS
6227 administrator could create reverse DNS records to break through a
6228 wildcarded host restriction in an ACL.
6230 37. If a user filter or a system filter that ran in a subprocess used any of
6231 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6232 the wrong values were passed to the pipe command ($thisaddress had the
6233 value of $0, $0 had the value of $1, etc). This bug was introduced by
6234 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6236 38. Improved the line breaking for long SMTP error messages from ACLs.
6237 Previously, if there was no break point between 40 and 75 characters, Exim
6238 left the rest of the message alone. Two changes have been made: (a) I've
6239 reduced the minimum length to 35 characters; (b) if it can't find a break
6240 point between 35 and 75 characters, it looks ahead and uses the first one
6241 that it finds. This may give the occasional overlong line, but at least the
6242 remaining text gets split now.
6244 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6245 file descriptors might be low, and that setting 1000 would always raise it.
6246 It turns out that in some environments, the limit is already over 1000 and
6247 that lowering it causes trouble. So now Exim takes care not to decrease it.
6249 40. When delivering a message, the value of $return_path is set to $sender_
6250 address at the start of routing (routers may change the value). By an
6251 oversight, this default was not being set up when an address was tested by
6252 -bt or -bv, which affected the outcome if any router or filter referred to
6255 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6256 log only when "message" or "log_message" are set. However, if one of the
6257 conditions was an address verification, or a call to a nested ACL, the
6258 messages generated by the underlying test were being passed through. This
6259 no longer happens. The underlying message is available in $acl_verify_
6260 message for both "message" and "log_message" expansions, so it can be
6261 passed through if needed.
6263 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6264 new expansion $bh_ to give the encoded byte string without charset
6265 translation. Translation happens only if iconv() is available; HAVE_ICONV
6266 indicates this at build time. HEADERS_CHARSET gives the charset to
6267 translate to; headers_charset can change it in the configuration, and
6268 "headers charset" can change it in an individual filter file.
6270 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6271 that creates RFC 2047 encoded "words" labels them as that charset instead
6272 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6273 expansion operator; (ii) when Exim creates a From: line for a local
6274 message; (iii) when a header line is rewritten to include a "phrase" part.
6276 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6277 buggy, causing it to skip the first lines of messages whose message ID
6278 ended in 'D'. This would not have bitten before Exim release 4.14, because
6279 message IDs were unlikely to end in 'D' before then. The effect was to have
6280 incorrect size information for certain domains.
6282 45. #include "config.h" was missing at the start of the crypt16.c module. This
6283 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6286 46. If there was a timeout during a "random" callout check, Exim treated it as
6287 a failure of the random address, and carried on sending RSET and the real
6288 address. If the delay was just some slowness somewhere, the response to the
6289 original RCPT would be taken as a response to RSET and so on, causing
6290 mayhem of various kinds.
6292 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6293 when I implemented it. It didn't allow for the fact that some option values
6294 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6295 the right test for positive values.
6297 48. Domain names in DNS records are case-independent. Exim always looks them up
6298 in lower case. Some resolvers return domain names in exactly the case they
6299 appear in the zone file, that is, they may contain uppercase letters. Not
6300 all resolvers do this - some return always lower case. Exim was treating a
6301 change of case by a resolver as a change of domain, similar to a widening
6302 of a domain abbreviation. This triggered its re-routing code and so it was
6303 trying to route what was effectively the same domain again. This normally
6304 caused routing to fail (because the router wouldn't handle the domain
6305 twice). Now Exim checks for this case specially, and just changes the
6306 casing of the domain that it ultimately uses when it transmits the message
6309 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6312 50. If a filter generated a file delivery with a non-absolute name (possible if
6313 no home directory exists for the router), the forbid_file option was not
6316 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6317 the existing equality matching.
6319 52. Exim was using ints instead of ino_t variables in some places where it was
6320 dealing with inode numbers.
6322 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6323 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6324 called TMPDIR, and if it finds it is different, it changes its value.
6326 54. The smtp_printf() function is now made available to local_scan() so
6327 additional output lines can be written before returning. There is also an
6328 smtp_fflush() function to enable the detection of a dropped connection.
6329 The variables smtp_input and smtp_batched_input are exported to
6332 55. Changed the default runtime configuration: the message "Unknown user"
6333 has been removed from the ACL, and instead placed on the localuser router,
6334 using the cannot_route_message feature. This means that any verification
6335 failures that generate their own messages won't get overridden. Similarly,
6336 the "Unrouteable address" message that was in the ACL for unverifiable
6337 relay addresses has also been removed.
6339 56. Added hosts_avoid_esmtp to the smtp transport.
6341 57. The exicyclog script was not checking for the esoteric option
6342 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6343 will work only if exicyclog is run under the appropriate euid.
6345 58. Following a discussion on the list, the rules by which Exim recognises line
6346 endings on incoming messages have been changed. The -dropcr and drop_cr
6347 options are now no-ops, retained only for backwards compatibility. The
6348 following line terminators are recognized: LF CRLF CR. However, special
6349 processing applies to CR:
6351 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6352 nor a local message in the state where . is a terminator.
6354 (ii) If a bare CR is encountered in a header line, an extra space is added
6355 after the line terminator so as not to end the header. The reasoning
6356 behind this is that bare CRs in header lines are most likely either
6357 to be mistakes, or people trying to play silly games.
6359 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6360 was being incorrectly given as 18 bytes larger than it should have been.
6361 This is a VOB (very old bug).
6363 60. This may never have affected anything current, but just in case it has:
6364 When the local host is found other than at the start of a list of hosts,
6365 the local host, those with the same MX, and any that follow, are discarded.
6366 When the list in question was part of a longer list of hosts, the following
6367 hosts (not currently being processed) were also being discarded. This no
6368 longer happens. I'm not sure if this situation could ever has previously
6371 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6374 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6375 header. This is something that is recommended in a new Internet Draft, and
6376 is something that is documented as being done by Sendmail. There are two
6377 possible values. For messages generated by the autoreply transport, Exim
6380 Auto-Submitted: auto-replied
6382 whereas for all other generated messages (e.g. bounces) it adds
6384 Auto-Submitted: auto-generated
6386 63. The "personal" condition in filters now includes a test for the
6387 Auto-Submitted: header. If it contains the string "auto-" the message it
6388 not considered personal.
6390 64. Added rcpt_include_affixes as a generic transport option.
6392 65. Added queue_only_override (default true).
6394 66. Added the syslog_duplication option.
6396 67. If what should have been the first header line of a message consisted of
6397 a space followed by a colon, Exim was mis-interpreting it as a header line.
6398 It isn't of course - it is syntactically invalid and should therefore be
6399 treated as the start of the message body. The misbehaviour could have
6400 caused a number of strange effects, including loss of data in subsequent
6401 header lines, and spool format errors.
6403 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6404 client host had authenticated. This control can now be exercised by an ACL
6405 for more flexibility.
6407 69. By default, callouts do not happen when testing with -bh. There is now a
6408 variant, -bhc, which does actually run the callout code, including
6409 consulting and updating the callout cache.
6411 70. Added support for saslauthd authentication, courtesy of Alexander
6414 71. If statvfs() failed on the spool or log directories while checking their
6415 size for availability, Exim confusingly gave the error "space shortage".
6416 Furthermore, in debugging mode it crashed with a floating point exception.
6417 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6418 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6419 serious problem, Exim now writes to the main and panic logs when this
6420 happens, with details of the failure. It then refuses to accept the
6421 incoming message, giving the message "spool directory problem" or "log
6422 directory problem" with a 421 code for SMTP messages.
6424 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6425 0, 1, and 2 exist, because some OS complain for execs without them (see
6426 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6427 descriptors. However, the code omitted to check that the open succeeded,
6428 causing mysterious errors if for some reason the permissions on /dev/null
6429 got screwed. Now Exim writes a message to the main and panic logs, and
6430 bombs out if it can't open /dev/null.
6432 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6433 interact so that it is all more flexible. It is supposed to remain
6434 backwards compatible. Also added extra_local_interfaces.
6436 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6437 to bomb out with an assertion failure - to the client this appears as a
6438 connection drop. This problem occurs in the part of the code that was taken
6439 from the Samba project. Fortunately, the assertion is in a very simple
6440 function, so I have fixed this by reproducing the function inline in the
6441 one place where it is called, and arranging for authentication to fail
6442 instead of killing the process with assert().
6444 75. The SPA client code was not working when the server requested OEM rather
6445 than Unicode encoding.
6447 76. Added code to make require_files with a specific uid setting more usable in
6448 the case where statting the file as root fails - usually a non-root-mounted
6449 NFS file system. When this happens and the failure is EACCES, Exim now
6450 forks a subprocess and does the per-uid checking as the relevant uid.
6452 77. Added process_log_path.
6454 78. If log_file_path was not explicitly set, a setting of check_log_space or
6455 check_log_inodes was ignored.
6457 79. If a space check for the spool or log partitions fails, the incident is now
6458 logged. Of course, in the latter case the data may get lost...
6460 80. Added the %p formatting code to string_format() so that it can be used to
6461 print addresses in debug_print(). Adjusted all the address printing in the
6462 debugging in store.c to use %p rather than %d.
6464 81. There was a concern that a line of code in smtp_in.c could overflow a
6465 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6466 initially expressed, the concern was not well-founded, because trailing
6467 spaces are removed early. However, if the trailing spaces were followed by
6468 a NULL, they did not get removed, so the overflow was possible. Two fixes
6471 (a) I re-wrote the offending code in a cleaner fashion.
6472 (b) If an incoming SMTP command contains a NULL character, it is rejected
6475 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6476 runs initgroups(), so that if the Exim user is in any additional groups,
6477 they will be used during message reception.
6483 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob