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-
182 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
183 delivery resulted in actual delivery. Cancel cutthrough before DATA
186 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
187 rejects a recipient: pass the reject to the originator.
189 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
190 Many were false-positives and ignorable, but it's worth fixing the
193 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
194 for the new environment-manipulation done at startup. Move the routines
195 from being local to tls.c to being global via the os.c file.
197 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
198 an extract embedded as result-arg for a map, the first arg for extract
199 is unavailable so we cannot tell if this is a numbered or keyed
200 extraction. Accept either.
206 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
209 JH/02 The smtp transport option "multi_domain" is now expanded.
211 JH/03 The smtp transport now requests PRDR by default, if the server offers
214 JH/04 Certificate name checking on server certificates, when exim is a client,
215 is now done by default. The transport option tls_verify_cert_hostnames
216 can be used to disable this per-host. The build option
217 EXPERIMENTAL_CERTNAMES is withdrawn.
219 JH/05 The value of the tls_verify_certificates smtp transport and main options
220 default to the word "system" to access the system default CA bundle.
221 For GnuTLS, only version 3.0.20 or later.
223 JH/06 Verification of the server certificate for a TLS connection is now tried
224 (but not required) by default. The verification status is now logged by
225 default, for both outbound TLS and client-certificate supplying inbound
228 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
231 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
232 Status Notification (bounce) messages are now MIME format per RFC 3464.
233 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
234 under the control of the dsn_advertise_hosts option, and routers may
235 have a dsn_lasthop option.
237 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
238 default, modifiable by a malware= option. The list separator for
239 the options can now be changed in the usual way. Bug 68.
241 JH/10 The smtp_receive_timeout main option is now expanded before use.
243 JH/11 The incoming_interface log option now also enables logging of the
244 local interface on delivery outgoing connections.
246 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
247 if the interface and destination host and port all match.
249 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
252 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
253 Patch from Andrew Lewis.
255 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
256 now supports optional time-restrictions, weighting, and priority
257 modifiers per server. Patch originally by <rommer@active.by>.
259 JH/16 The spamd_address main option now supports a mixed list of local
260 and remote servers. Remote servers can be IPv6 addresses, and
261 specify a port-range.
263 JH/17 Bug 68: The spamd_address main option now supports an optional
264 timeout value per server.
266 JH/18 Bug 1581: Router and transport options headers_add/remove can
267 now have the list separator specified.
269 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
272 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
275 JH/21 Support for the A6 type of dns record is withdrawn.
277 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
278 rather than the verbs used.
280 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
281 from 255 to 1024 chars.
283 JH/24 Verification callouts now attempt to use TLS by default.
285 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
286 are generic router options now. The defaults didn't change.
288 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
289 Original patch from Alexander Shikoff, worked over by JH.
291 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
292 files if ZCAT_COMMAND is not executable.
294 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
296 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
298 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
299 Normally benign, it bites when the pair was led to by a CNAME;
300 modern usage is to not canoicalize the domain to a CNAME target
301 (and we were inconsistent anyway for A-only vs AAAA+A).
303 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
305 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
306 when evaluating $sender_host_dnssec.
308 JH/31 Check the HELO verification lookup for DNSSEC, adding new
309 $sender_helo_dnssec variable.
311 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
313 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
315 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
317 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
318 documented as working, but never had. Support all but $spam_report.
320 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
321 added for tls authenticator.
326 TL/01 When running the test suite, the README says that variables such as
327 no_msglog_check are global and can be placed anywhere in a specific
328 test's script, however it was observed that placement needed to be near
329 the beginning for it to behave that way. Changed the runtest perl
330 script to read through the entire script once to detect and set these
331 variables, reset to the beginning of the script, and then run through
332 the script parsing/test process like normal.
334 TL/02 The BSD's have an arc4random API. One of the functions to induce
335 adding randomness was arc4random_stir(), but it has been removed in
336 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
337 function when detected.
339 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
340 cause callback expansion.
342 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
343 syntax errors in an expansion can be treated as a string instead of
344 logging or causing an error, due to the internal use of bool_lax
345 instead of bool when processing it.
347 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
348 server certificates when making smtp deliveries.
350 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
352 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
354 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
356 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
357 Merged patch from Sebastian Wiedenroth.
359 JH/05 Fix results-pipe from transport process. Several recipients, combined
360 with certificate use, exposed issues where response data items split
361 over buffer boundaries were not parsed properly. This eventually
362 resulted in duplicates being sent. This issue only became common enough
363 to notice due to the introduction of conection certificate information,
364 the item size being so much larger. Found and fixed by Wolfgang Breyha.
366 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
367 size buffer was used, resulting in syntax errors when an expansion
370 JH/07 Add support for directories of certificates when compiled with a GnuTLS
371 version 3.3.6 or later.
373 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
374 is EXPERIMENTAL_EVENT, the main-configuration and transport options
375 both become "event_action", the variables become $event_name, $event_data
376 and $event_defer_errno. There is a new variable $verify_mode, usable in
377 routers, transports and related events. The tls:cert event is now also
378 raised for inbound connections, if the main configuration event_action
381 TL/06 In test suite, disable OCSP for old versions of openssl which contained
382 early OCSP support, but no stapling (appears to be less than 1.0.0).
384 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
385 server certificate names available under the smtp transport option
386 "tls_verify_cert_hostname" now do not permit multi-component wildcard
389 JH/10 Time-related extraction expansions from certificates now use the main
390 option "timezone" setting for output formatting, and are consistent
391 between OpenSSL and GnuTLS compilations. Bug 1541.
393 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
394 encoded parameter in the incoming message. Bug 1558.
396 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
397 include certificate info, eximon was claiming there were spoolfile
400 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
402 JH/14 Log delivery-related information more consistently, using the sequence
403 "H=<name> [<ip>]" wherever possible.
405 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
406 are problematic for Debian distribution, omit them from the release
409 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
411 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
414 JH/17 Fix a null-indirection in certextract expansions when a nondefault
415 output list separator was used.
420 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
421 checkers that were complaining about end of non-void function with no
424 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
425 This was a regression intruduced in 4.83 by another bugfix.
427 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
429 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
430 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
436 TF/01 Correctly close the server side of TLS when forking for delivery.
438 When a message was received over SMTP with TLS, Exim failed to clear up
439 the incoming connection properly after forking off the child process to
440 deliver the message. In some situations the subsequent outgoing
441 delivery connection happened to have the same fd number as the incoming
442 connection previously had. Exim would try to use TLS and fail, logging
443 a "Bad file descriptor" error.
445 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
446 utilities have not been installed.
448 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
449 temporary space as the ACL may create new global variables.
451 TL/01 LDAP support uses per connection or global context settings, depending
452 upon the detected version of the libraries at build time.
454 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
455 to extract and use the src ip:port in logging and expansions as if it
456 were a direct connection from the outside internet. PPv2 support was
457 updated based on HAProxy spec change in May 2014.
459 JH/02 Add ${listextract {number}{list}{success}{fail}}.
461 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
462 Properly escape header and check for NULL return.
464 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
467 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
469 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
470 characters in header names, implemented as a verify condition.
471 Contributed by Michael Fischer v. Mollard.
473 TL/05 Rename SPF condition results err_perm and err_temp to standardized
474 results permerror and temperror. Previous values are deprecated but
475 still accepted. In a future release, err_perm and err_temp will be
476 completely removed, which will be a backward incompatibility if the
477 ACL tests for either of these two old results. Patch contributed by
478 user bes-internal on the mailing list.
480 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
482 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
483 selectors, in both main and reject logs.
485 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
488 JH/07 Add malware type "sock" for talking to simple daemon.
490 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
492 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
493 routers/transports under cutthrough routing.
495 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
496 numbers. Touch up "bool" conditional to keep the same definition.
498 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
500 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
502 JH/12 Expand items in router/transport headers_add or headers_remove lists
503 individually rather than the list as a whole. Bug 1452.
505 Required for reasonable handling of multiple headers_ options when
506 they may be empty; requires that headers_remove items with embedded
507 colons must have them doubled (or the list-separator changed).
509 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
510 view the policy declared in the DMARC record. Currently, $dmarc_status
511 is a combined value of both the record presence and the result of the
514 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
516 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
517 dnslookup router and the smtp transport (applying to the forward
520 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
521 of ldap servers used for a specific lookup. Patch provided by Heiko
524 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
525 New variable $lookup_dnssec_authenticated for observability.
527 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
528 Patch submitted by Lars Timman.
530 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
532 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
533 Requires trusted mode and valid format message id, aborts otherwise.
534 Patch contributed by Heiko Schlichting.
536 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
537 certextract with support for various fields. Bug 1358.
539 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
540 is requested by default, modifiable by smtp transport option
543 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
544 operate on certificate variables to give certificate fingerprints
545 Also new ${sha256:cert_variable}.
547 JH/23 The PRDR feature is moved from being Experimental into the mainline.
549 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
550 Christian Aistleitner.
552 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
554 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
555 file. Patch from Wolfgang Breyha.
557 JH/25 Expand the coverage of the delivery $host and $host_address to
558 client authenticators run in verify callout. Bug 1476.
560 JH/26 Port service names are now accepted for tls_on_connect_ports, to
561 align with daemon_smtp_ports. Bug 72.
563 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
564 support and error reporting did not work properly.
566 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
567 and is readable. Patch from Andrew Colin Kissa.
569 TL/14 Enhance documentation of ${run expansion and how it parses the
570 commandline after expansion, particularly in the case when an
571 unquoted variable expansion results in an empty value.
573 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
575 PP/02 Fix internal collision of T_APL on systems which support RFC3123
576 by renaming away from it. Addresses GH issue 15, reported by
579 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
581 TL/15 SECURITY: prevent double expansion in math comparison functions
582 (can expand unsanitized data). Not remotely exploitable.
589 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
591 PP/02 Make -n do something, by making it not do something.
592 When combined with -bP, the name of an option is not output.
594 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
597 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
598 $sender_host_name and config options to manage this, and basic check
601 PP/05 DSCP support for outbound connections and control modifier for inbound.
603 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
604 (Only plugin which currently uses this is kerberos4, which nobody should
605 be using, but we should make it available and other future plugins might
606 conceivably use it, even though it would break NAT; stuff *should* be
607 using channel bindings instead).
609 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
610 name; added for Sendmail compatibility; requires admin caller.
611 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
612 just ignore it); requires trusted caller.
613 Also parse but ignore: -Ac -Am -X<logfile>
616 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
618 TL/02 Add +smtp_confirmation as a default logging option.
620 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
621 Patch by Magnus Holmgren from 2007-02-20.
623 TL/04 Bugzilla 1281 - Spec typo.
624 Bugzilla 1283 - Spec typo.
625 Bugzilla 1290 - Spec grammar fixes.
627 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
629 TL/06 Add Experimental DMARC support using libopendmarc libraries.
631 TL/07 Fix an out of order global option causing a segfault. Reported to dev
632 mailing list by by Dmitry Isaikin.
634 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
636 JH/02 Support "G" suffix to numbers in ${if comparisons.
638 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
640 NM/01 Bugzilla 1197 - Spec typo
641 Bugzilla 1196 - Spec examples corrections
643 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
645 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
646 gnutls_enable_pkcs11, but renamed to more accurately indicate its
649 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
650 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
652 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
653 "acl {{name}{arg}...}", and optional args on acl condition
656 JH/05 Permit multiple router/transport headers_add/remove lines.
658 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
660 JH/07 Avoid using a waiting database for a single-message-only transport.
661 Performance patch from Paul Fisher. Bugzilla 1262.
663 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
666 JH/09 Add $headers_added variable, with content from use of ACL modifier
667 add_header (but not yet added to the message). Bugzilla 199.
669 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
670 Pulled from Bugzilla 817 by Wolfgang Breyha.
672 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
674 (nb: this is the same fix as in Exim 4.80.1)
676 JH/11 Add A= logging on delivery lines, and a client_set_id option on
679 JH/12 Add optional authenticated_sender logging to A= and a log_selector
682 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
684 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
685 advertise SMTP AUTH mechanism to us, instead of a generic
686 protocol violation error. Also, make Exim more robust to bad
687 data from the Dovecot auth socket.
689 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
691 When a queue runner is handling a message, Exim first routes the
692 recipient addresses, during which it prunes them based on the retry
693 hints database. After that it attempts to deliver the message to
694 any remaining recipients. It then updates the hints database using
697 So if a recipient address works intermittently, it can get repeatedly
698 deferred at routing time. The retry hints record remains fresh so the
699 address never reaches the final cutoff time.
701 This is a fairly common occurrence when a user is bumping up against
702 their storage quota. Exim had some logic in its local delivery code
703 to deal with this. However it did not apply to per-recipient defers
704 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
706 This change adds a proper retry rule check during routing so that the
707 final cutoff time is checked against the message's age. We only do
708 this check if there is an address retry record and there is not a
709 domain retry record; this implies that previous attempts to handle
710 the address had the retry_use_local_parts option turned on. We use
711 this as an approximation for the destination being like a local
712 delivery, as in LMTP.
714 I suspect this new check makes the old local delivery cutoff check
715 redundant, but I have not verified this so I left the code in place.
717 TF/02 Correct gecos expansion when From: is a prefix of the username.
719 Test 0254 submits a message to Exim with the header
723 When I ran the test suite under the user fanf2, Exim expanded
724 the header to contain my full name, whereas it should have added
725 a Resent-Sender: header. It erroneously treats any prefix of the
726 username as equal to the username.
728 This change corrects that bug.
730 GF/01 DCC debug and logging tidyup
731 Error conditions log to paniclog rather than rejectlog.
732 Debug lines prefixed by "DCC: " to remove any ambiguity.
734 TF/03 Avoid unnecessary rebuilds of lookup-related code.
736 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
737 Bug spotted by Jeremy Harris; was flawed since initial commit.
738 Would have resulted in OCSP responses post-SNI triggering an Exim
739 NULL dereference and crash.
741 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
743 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
744 Bug detection, analysis and fix by Samuel Thibault.
745 Bugzilla 1331, Debian bug #698092.
747 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
749 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
750 Server implementation by Todd Lyons, client by JH.
751 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
752 config variable "prdr_enable" controls whether the server
753 advertises the facility. If the client requests PRDR a new
754 acl_data_smtp_prdr ACL is called once for each recipient, after
755 the body content is received and before the acl_smtp_data ACL.
756 The client is controlled by bolth of: a hosts_try_prdr option
757 on the smtp transport, and the server advertisement.
758 Default client logging of deliveries and rejections involving
759 PRDR are flagged with the string "PRDR".
761 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
762 fclose(). Diagnosis by Todd Lyons.
764 PP/17 Update configure.default to handle IPv6 localhost better.
765 Patch by Alain Williams (plus minor tweaks).
768 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
769 This is now consistent with GnuTLS, and is now documented: the
770 previous undocumented portable approach to treating the option as
771 unset was to force an expansion failure. That still works, and
772 an empty string is now equivalent.
774 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
775 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
776 not performing validation itself.
778 PP/20 Added force_command boolean option to pipe transport.
779 Patch from Nick Koston, of cPanel Inc.
781 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
784 TF/04 Added udpsend ACL modifer and hexquote expansion operator
786 PP/21 Fix eximon continuous updating with timestamped log-files.
787 Broken in a format-string cleanup in 4.80, missed when I repaired the
788 other false fix of the same issue.
789 Report and fix from Heiko Schlichting.
792 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
793 Report from Prashanth Katuri.
795 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
796 It's SecureTransport, so affects any MacOS clients which use the
797 system-integrated TLS libraries, including email clients.
799 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
800 using a MIME ACL for non-SMTP local injection.
801 Report and assistance in diagnosis by Warren Baker.
803 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
805 JH/16 Fix comparisons for 64b. Bugzilla 1385.
807 TL/09 Add expansion variable $authenticated_fail_id to keep track of
808 last id that failed so it may be referenced in subsequent ACL's.
810 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
813 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
814 ldap library initialization, allowing self-signed CA's to be
815 used. Also properly sets require_cert option later in code by
816 using NULL (global ldap config) instead of ldap handle (per
817 session). Bug diagnosis and testing by alxgomz.
819 TL/12 Enhanced documentation in the ratelimit.pl script provided in
820 the src/util/ subdirectory.
822 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
823 renamed to Transport Post Delivery Action by Jeremy Harris, as
826 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
827 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
828 redis_servers = needs to be configured which will be used by the redis
829 lookup. Patch from Warren Baker, of The Packet Hub.
831 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
833 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
834 hostname or reverse DNS when processing a host list. Used suggestions
835 from multiple comments on this bug.
837 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
839 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
840 Missed a few lines, added it to make the runtest require no keyboard
843 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
844 contains upper case chars. Make router use caseful_local_part.
846 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
847 support when GnuTLS has been built with p11-kit.
853 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
855 This, or similar/improved, will also be change PP/11 of 4.82.
861 PP/01 Handle short writes when writing local log-files.
862 In practice, only affects FreeBSD (8 onwards).
863 Bugzilla 1053, with thanks to Dmitry Isaikin.
865 NM/01 Bugzilla 949 - Documentation tweak
867 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
870 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
872 PP/02 Implemented gsasl authenticator.
874 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
876 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
877 `pkg-config foo` for cflags/libs.
879 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
880 with rest of GSASL and with heimdal_gssapi.
882 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
883 `pkg-config foo` for cflags/libs for the TLS implementation.
885 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
886 properties get this fed in as external SSF. A number of robustness
887 and debugging improvements to the cyrus_sasl authenticator.
889 PP/08 cyrus_sasl server now expands the server_realm option.
891 PP/09 Bugzilla 1214 - Log authentication information in reject log.
892 Patch by Jeremy Harris.
894 PP/10 Added dbmjz lookup type.
896 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
898 PP/12 MAIL args handles TAB as well as SP, for better interop with
899 non-compliant senders.
900 Analysis and variant patch by Todd Lyons.
902 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
903 Bug report from Lars Müller <lars@samba.org> (via SUSE),
904 Patch from Dirk Mueller <dmueller@suse.com>
906 PP/13 tls_peerdn now print-escaped for spool files.
907 Observed some $tls_peerdn in wild which contained \n, which resulted
908 in spool file corruption.
910 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
911 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
912 or write after TLS renegotiation, which otherwise led to messages
915 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
916 as a tracking header (ie: a signed header comes before the signature).
917 Patch from Wolfgang Breyha.
919 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
920 comma-sep list; embedded commas doubled.
922 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
924 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
926 Report and patch from Dmitry Banschikov.
928 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
929 Removed SSL_clear() after SSL_new() which led to protocol negotiation
930 failures. We appear to now support TLS1.1+ with Exim.
932 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
933 lets Exim select keys and certificates based upon TLS SNI from client.
934 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
935 before an outbound SMTP session. New log_selector, +tls_sni.
937 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
938 NULL dereference. Report and patch from Alun Jones.
940 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
941 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
942 Not seeing resolver debug output on NetBSD, but suspect this is a
943 resolver implementation change.
945 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
946 Left warnings. Added "eximon gdb" invocation mode.
948 PP/21 Defaulting "accept_8bitmime" to true, not false.
950 PP/22 Added -bw for inetd wait mode support.
952 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
953 locate the relevant includes and libraries. Made this the default.
955 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
956 Bugzilla 1246, report and most of solution from Tomasz Kusy.
958 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
959 This may cause build issues on older platforms.
961 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
962 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
963 gnutls_require_mac & gnutls_require_protocols (no longer supported).
964 Added SNI support via GnuTLS too.
965 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
967 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
969 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
970 Applied second patch from Janne, implementing suggestion to default
971 multiple-strings-in-record handling to match SPF spec.
973 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
975 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
976 read-only, out of scope).
977 Patch from Wolfgang Breyha, report from Stuart Northfield.
979 PP/29 Fix three issues highlighted by clang analyser static analysis.
980 Only crash-plausible issue would require the Cambridge-specific
981 iplookup router and a misconfiguration.
982 Report from Marcin Mirosław.
984 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
986 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
987 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
988 As part of this, removing so much warning spew let me fix some minor
989 real issues in debug logging.
991 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
992 assignment on my part. Fixed.
994 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
995 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
996 Janne Snabb (who went above and beyond: thank you).
998 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
999 string otherwise requires a connection and a bunch more work and it's
1000 relatively easy to get wrong. Should also expose TLS library linkage
1003 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1004 64-bit ${eval} (JH/03).
1006 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1007 GNU libc to support some of the 64-bit stuff, should not lead to
1008 conflicts. Defined before os.h is pulled in, so if a given platform
1009 needs to override this, it can.
1011 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1012 protection layer was required, which is not implemented.
1013 Bugzilla 1254, patch from Wolfgang Breyha.
1015 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1016 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1017 tls_dhparam take prime identifiers. Also unbreak combination of
1018 OpenSSL+DH_params+TLSSNI.
1020 PP/39 Disable SSLv2 by default in OpenSSL support.
1026 PP/01 Solaris build fix for Oracle's LDAP libraries.
1027 Bugzilla 1109, patch from Stephen Usher.
1029 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1031 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1034 TF/02 Fix a couple more cases where we did not log the error message
1035 when unlink() failed. See also change 4.74-TF/03.
1037 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1038 lock up or crash if it happened to be inside a call to libc when it
1039 got a SIGUSR1 from exiwhat.
1041 The SIGUSR1 handler appends the current process status to the process
1042 log which is later printed by exiwhat. It used to use the general
1043 purpose logging code to do this, but several functions it calls are
1044 not safe for signals.
1046 The new output code in the SIGUSR1 handler is specific to the process
1047 log, and simple enough that it's easy to inspect for signal safety.
1048 Removing some special cases also simplifies the general logging code.
1049 Removing the spurious timestamps from the process log simplifies
1052 TF/04 Improved ratelimit ACL condition.
1054 The /noupdate option has been deprecated in favour of /readonly which
1055 has clearer semantics. The /leaky, /strict, and /readonly update modes
1056 are mutually exclusive. The update mode is no longer included in the
1057 database key; it just determines when the database is updated. (This
1058 means that when you upgrde Exim will forget old rate measurements.)
1060 Exim now checks that the per_* options are used with an update mode that
1061 makes sense for the current ACL. For example, when Exim is processing a
1062 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1063 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1064 must specify per_mail/readonly. If you omit the update mode it defaults to
1065 /leaky where that makes sense (as before) or /readonly where required.
1067 The /noupdate option is now undocumented but still supported for
1068 backwards compatibility. It is equivalent to /readonly except that in
1069 ACLs where /readonly is required you may specify /leaky/noupdate or
1070 /strict/noupdate which are treated the same as /readonly.
1072 A useful new feature is the /count= option. This is a generalization
1073 of the per_byte option, so that you can measure the throughput of other
1074 aggregate values. For example, the per_byte option is now equivalent
1075 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1077 The per_rcpt option has been generalized using the /count= mechanism
1078 (though it's more complicated than the per_byte equivalence). When it is
1079 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1080 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1081 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1082 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1083 Note that using per_rcpt with a non-readonly update mode in more than
1084 one ACL will cause the recipients to be double-counted. (The per_mail
1085 and per_byte options don't have this problem.)
1087 The handling of very low rates has changed slightly. If the computed rate
1088 is less than the event's count (usually one) then this event is the first
1089 after a long gap. In this case the rate is set to the same as this event's
1090 count, so that the first message of a spam run is counted properly.
1092 The major new feature is a mechanism for counting the rate of unique
1093 events. The new per_addr option counts the number of different
1094 recipients that someone has sent messages to in the last time period. It
1095 behaves like per_rcpt if all the recipient addresses are different, but
1096 duplicate recipient addresses do not increase the measured rate. Like
1097 the /count= option this is a general mechanism, so the per_addr option
1098 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1099 example, measure the rate that a client uses different sender addresses
1100 with the options per_mail/unique=$sender_address. There are further
1101 details in the main documentation.
1103 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1105 TF/06 Removed a few PCRE remnants.
1107 TF/07 Automatically extract Exim's version number from tags in the git
1108 repository when doing development or release builds.
1110 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1111 Bugzilla 879. Patch from Paul Fisher.
1113 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1114 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1117 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1119 PP/05 Variable $av_failed, true if the AV scanner deferred.
1120 Bugzilla 1078. Patch from John Horne.
1122 PP/06 Stop make process more reliably on build failure.
1123 Bugzilla 1087. Patch from Heiko Schlittermann.
1125 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1126 Bugzilla 1089. Patch from Heiko Schlittermann.
1128 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1129 Bugzilla 1131. Patch from Holger Weiß.
1131 PP/09 Handle IPv6 addresses with SPF.
1132 Bugzilla 860. Patch from Wolfgang Breyha.
1134 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1136 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1139 PP/11 match_* no longer expand right-hand-side by default.
1140 New compile-time build option, EXPAND_LISTMATCH_RHS.
1141 New expansion conditions, "inlist", "inlisti".
1143 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1145 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1147 PP/14 fix log_write() format string regression from TF/03.
1148 Bugzilla 1152. Patch from Dmitry Isaikin.
1154 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1156 PP/02 Harmonised TLS library version reporting; only show if debugging.
1157 Layout now matches that introduced for other libraries in 4.74 PP/03.
1159 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1161 PP/04 New "dns_use_edns0" global option.
1163 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1166 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1167 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1169 TK/01 Updated PolarSSL code to 0.14.2.
1170 Bugzilla 1097. Patch from Andreas Metzler.
1172 PP/07 Catch divide-by-zero in ${eval:...}.
1173 Fixes bugzilla 1102.
1175 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1178 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1179 format-string attack -- SECURITY: remote arbitrary code execution.
1181 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1182 time unintentionally subject to list matching rules, letting the header
1183 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1184 arbitrary string expansion). This allowed for information disclosure.
1186 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1187 INT_MIN/-1 -- value coerced to INT_MAX.
1193 NM/01 Workround for PCRE version dependency in version reporting
1196 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1197 This fixes portability to compilers other than gcc, notably
1198 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1200 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1201 makefiles for portability to HP-UX and POSIX correctness.
1203 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1204 Also via indented variable definition in the Makefile.
1205 (Debugging by Oliver Heesakkers).
1207 PP/02 Restore caching of spamd results with expanded spamd_address.
1208 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1210 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1211 Improves build reliability. Fix from: Frank Elsner
1213 NM/02 Fix wide character breakage in the rfc2047 coding
1214 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1216 NM/03 Allow underscore in dnslist lookups
1217 Fixes bug 1026. Patch from Graeme Fowler
1219 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1220 Code patches from Adam Ciarcinski of NetBSD.
1222 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1225 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1226 is logged, to avoid truncation. Patch from John Horne.
1228 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1229 Patch from Jakob Hirsch.
1231 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1232 SQL string expansion failure details.
1233 Patch from Andrey Oktyabrski.
1235 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1236 Patch from Simon Arlott.
1238 PP/09 New lookups functionality failed to compile on old gcc which rejects
1239 extern declarations in function scope.
1240 Patch from Oliver Fleischmann
1242 PP/10 Use sig_atomic_t for flags set from signal handlers.
1243 Check getgroups() return and improve debugging.
1244 Fixed developed for diagnosis in bug 927 (which turned out to be
1247 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1248 Patch from Mark Zealey.
1250 PP/12 Bugzilla 1056: Improved spamd server selection.
1251 Patch from Mark Zealey.
1253 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1254 Based on patch from Heiko Schlittermann.
1256 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1257 Patch from Uwe Doering, sign-off by Michael Haardt.
1259 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1260 variable declaration deep within a block. Bug and patch from
1263 PP/15 lookups-Makefile IRIX compatibilty coercion.
1265 PP/16 Make DISABLE_DKIM build knob functional.
1267 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1268 Patch by Simon Arlott
1270 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1271 variable argument macros. Our copy now differs from upstream.
1277 TF/01 Failure to get a lock on a hints database can have serious
1278 consequences so log it to the panic log.
1280 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1281 controlled using the smtp_confirmation log selector.
1283 TF/03 Include the error message when we fail to unlink a spool file.
1285 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1286 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1287 for maintaining out-of-tree patches for some time.
1289 PP/01 Bugzilla 139: Documentation and portability issues.
1290 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1291 Handle per-OS dynamic-module compilation flags.
1293 PP/02 Let /dev/null have normal permissions.
1294 The 4.73 fixes were a little too stringent and complained about the
1295 permissions on /dev/null. Exempt it from some checks.
1296 Reported by Andreas M. Kirchwitz.
1298 PP/03 Report version information for many libraries, including
1299 Exim version information for dynamically loaded libraries. Created
1300 version.h, now support a version extension string for distributors
1301 who patch heavily. Dynamic module ABI change.
1303 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1304 privilege escalation vulnerability whereby the Exim run-time user
1305 can cause root to append content of the attacker's choosing to
1308 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1311 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1312 If dropping privileges for untrusted macros, we disabled normal logging
1313 on the basis that it would fail; for the Exim run-time user, this is not
1314 the case, and it resulted in successful deliveries going unlogged.
1315 Fixed. Reported by Andreas Metzler.
1321 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1322 only prepend for the Resent-* case. Fixes regression introduced in
1323 Exim 4.70 by NM/22 for Bugzilla 607.
1325 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1326 increasing numbers of administrators be bitten by this.
1328 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1330 PP/03 Bugzilla 994: added openssl_options main configuration option.
1332 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1334 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1336 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1338 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1339 without a peer certificate, leading to a segfault because of an
1340 assumption that peers always have certificates. Be a little more
1341 paranoid. Problem reported by Martin Tscholak.
1343 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1344 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1345 NB: ClamAV planning to remove STREAM in "middle of 2010".
1346 CL also introduces -bmalware, various -d+acl logging additions and
1347 more caution in buffer sizes.
1349 PP/09 Implemented reverse_ip expansion operator.
1351 PP/10 Bugzilla 937: provide a "debug" ACL control.
1353 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1355 PP/12 Bugzilla 973: Implement --version.
1357 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1359 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1361 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1363 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1364 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1365 ignore trailing whitespace.
1367 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1369 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1372 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1373 Notification from Dr Andrew Aitchison.
1375 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1376 ExtendedDetectionInfo response format.
1377 Notification from John Horne.
1379 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1382 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1383 XSL and documented dependency on system catalogs, with examples of how
1386 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1389 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1390 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1391 configuration file which is writeable by the Exim user or group.
1393 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1394 of configuration files to cover files specified with the -C option if
1395 they are going to be used with root privileges, not just the default
1398 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1399 option (effectively making it always true).
1401 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1402 files to be used while preserving root privileges.
1404 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1405 that rogue child processes cannot use them.
1407 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1408 run-time user, instead of root.
1410 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1411 Exim run-time user without dropping privileges.
1413 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1414 result string, instead of calling string_vformat() twice with the same
1417 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1418 for other users. Others should always drop root privileges if they use
1419 -C on the command line, even for a whitelisted configure file.
1421 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1423 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1429 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1430 $data_path, and $header_path variables; fixed documentation bugs and
1433 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1434 exipick to access non-standard spools, including the "frozen" queue
1437 NM/01 Bugzilla 965: Support mysql stored procedures.
1438 Patch from Alain Williams
1440 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1442 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1443 Patch from Andreas Metzler
1445 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1446 Patch from Kirill Miazine
1448 NM/05 Bugzilla 671: Added umask to procmail example.
1450 JJ/03 installed exipick 20100323.0, fixing doc bug
1452 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1453 directory. Notification and patch from Dan Rosenberg.
1455 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1457 TK/02 Improve log output when DKIM signing operation fails.
1459 MH/01 Treat the transport option dkim_domain as a colon separated
1460 list, not as a single string, and sign the message with each element,
1461 omitting multiple occurences of the same signer.
1463 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1464 Bugzilla 985, 986. Patch by Simon Arlott
1466 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1467 Patch by Simon Arlott
1469 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1470 MBX locking. Notification from Dan Rosenberg.
1476 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1478 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1480 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1482 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1484 NM/04 Bugzilla 915: Flag broken perl installation during build.
1490 TK/01 Added patch by Johannes Berg that expands the main option
1491 "spamd_address" if it starts with a dollar sign.
1493 TK/02 Write list of recipients to X-Envelope-Sender header when building
1494 the mbox-format spool file for content scanning (suggested by Jakob
1497 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1498 (http://www.dcc-servers.net/) support via dccifd. Activated by
1499 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1501 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1502 by Mark Daniel Reidel <mr@df.eu>.
1504 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1505 When building exim an external PCRE library is now needed -
1506 PCRE is a system library on the majority of modern systems.
1507 See entry on PCRE_LIBS in EDITME file.
1509 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1510 conversation. Added nologin parameter to request.
1511 Patch contributed by Kirill Miazine.
1513 TF/01 Do not log submission mode rewrites if they do not change the address.
1515 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1517 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1518 log files in place. Contributed by Roberto Lima.
1520 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1522 TF/03 Bugzilla 615: When checking the local_parts router precondition
1523 after a local_part_suffix or local_part_prefix option, Exim now
1524 does not use the address's named list lookup cache, since this
1525 contains cached lookups for the whole local part.
1527 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1528 Robert Millan. Documentation is in experimental-spec.txt.
1530 TF/04 Bugzilla 668: Fix parallel build (make -j).
1532 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1534 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1535 Patch provided by Jan Srzednicki.
1537 TF/05 Leading white space used to be stripped from $spam_report which
1538 wrecked the formatting. Now it is preserved.
1540 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1541 that they are available at delivery time.
1543 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1545 TF/08 TLS error reporting now respects the incoming_interface and
1546 incoming_port log selectors.
1548 TF/09 Produce a more useful error message if an SMTP transport's hosts
1549 setting expands to an empty string.
1551 NM/06 Bugzilla 744: EXPN did not work under TLS.
1552 Patch provided by Phil Pennock.
1554 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1555 Patch provided by Richard Godbee.
1557 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1558 acl_smtp_notquit, added index entry.
1560 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1561 Patch provided by Eugene Bujak.
1563 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1564 accept(). Patch provided by Maxim Dounin.
1566 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1567 Patch provided by Phil Pennock.
1569 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1571 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1572 Patch provided by Brad "anomie" Jorsch.
1574 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1575 Patch provided by Dean Brooks.
1577 TK/05 Add native DKIM support (does not depend on external libraries).
1579 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1580 Patch provided by Graeme Fowler.
1582 NM/16 Bugzilla 851: Documentation example syntax fix.
1584 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1586 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1589 NM/19 Bugzilla 745: TLS version reporting.
1590 Patch provided by Phil Pennock.
1592 NM/20 Bugzilla 167: bool: condition support.
1593 Patch provided by Phil Pennock.
1595 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1596 clients. Patch provided by Phil Pennock.
1598 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1599 Patch provided by Brad "anomie" Jorsch.
1601 NM/23 Bugzilla 687: Fix misparses in eximstats.
1602 Patch provided by Heiko Schlittermann.
1604 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1605 Patch provided by Heiko Schlittermann.
1607 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1608 plus update to original patch.
1610 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1612 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1613 Patch provided by David Brownlee.
1615 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1617 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1619 NM/30 Bugzilla 888: TLS documentation bugfixes.
1621 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1623 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1624 Unlike the original bugzilla I have changed all shell scripts in src tree.
1626 NM/33 Bugzilla 898: Transport filter timeout fix.
1627 Patch by Todd Rinaldo.
1629 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1630 Patch by Serge Demonchaux.
1632 NM/35 Bugzilla 39: Base64 decode bug fixes.
1633 Patch by Jakob Hirsch.
1635 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1637 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1639 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1641 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1647 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1648 ALT-N's libdkim that I have put here:
1649 http://duncanthrax.net/exim-experimental/
1651 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1652 were called 'true' and it seems that C99 defines that as a reserved
1653 keyword to be used with 'bool' variable types. That means you could
1654 not include C99-style headers which use bools without triggering
1655 build errors in sieve.c.
1657 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1658 as mailq or other aliases. Changed the --help handling significantly
1659 to do whats expected. exim_usage() emits usage/help information.
1661 SC/01 Added the -bylocaldomain option to eximstats.
1663 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1665 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1667 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1673 PH/01 Another patch from the Sieve maintainer.
1675 PH/02 When an IPv6 address is converted to a string for single-key lookup
1676 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1677 dots are used instead of colons so that keys in lsearch files need not
1678 contain colons. This was done some time before quoting was made available
1679 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1680 (notated using the quote facility) so as to distinguish them from IPv4
1681 keys. This meant that lookups for IP addresses in host lists did not work
1682 for iplsearch lookups.
1684 This has been fixed by arranging for IPv6 addresses to be expressed with
1685 colons if the lookup type is iplsearch. This is not incompatible, because
1686 previously such lookups could never work.
1688 The situation is now rather anomolous, since one *can* have colons in
1689 ordinary lsearch keys. However, making the change in all cases is
1690 incompatible and would probably break a number of configurations.
1692 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1695 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1696 conversion specification without a maximum field width, thereby enabling
1697 a rogue spamd server to cause a buffer overflow. While nobody in their
1698 right mind would setup Exim to query an untrusted spamd server, an
1699 attacker that gains access to a server running spamd could potentially
1700 exploit this vulnerability to run arbitrary code as the Exim user.
1702 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1703 $primary_hostname instead of what libspf2 thinks the hosts name is.
1705 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1706 a directory entry by the name of the lookup key. Previously, if a
1707 symlink pointed to a non-existing file or a file in a directory that
1708 Exim lacked permissions to read, a lookup for a key matching that
1709 symlink would fail. Now it is enough that a matching directory entry
1710 exists, symlink or not. (Bugzilla 503.)
1712 PH/03 The body_linecount and body_zerocount variables are now exported in the
1715 PH/04 Added the $dnslist_matched variable.
1717 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1718 This means they are set thereafter only if the connection becomes
1721 PH/06 Added the client_condition to authenticators so that some can be skipped
1722 by clients under certain conditions.
1724 PH/07 The error message for a badly-placed control=no_multiline_responses left
1725 "_responses" off the end of the name.
1727 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1729 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1730 (without spaces) instead of just copying the configuration text.
1732 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1734 PH/11 Added $max_received_linelength.
1736 PH/12 Added +ignore_defer and +include_defer to host lists.
1738 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1739 way in which PCRE > 7.0 is built.
1741 PH/14 Implemented queue_only_load_latch.
1743 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1744 MAIL command. The effect was to mangle the value on 64-bit systems.
1746 PH/16 Another patch from the Sieve maintainer.
1748 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1750 PH/18 If a system quota error occurred while trying to create the file for
1751 a maildir delivery, the message "Mailbox is full" was not appended to the
1752 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1753 only to a quota excession during the actual writing of the file.
1755 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1756 characters?) which causes problems in log lines. The DN values are now
1757 passed through string_printing() before being added to log lines.
1759 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1760 and InterBase are left for another time.)
1762 PH/21 Added message_body_newlines option.
1764 PH/22 Guard against possible overflow in moan_check_errorcopy().
1766 PH/23 POSIX allows open() to be a macro; guard against that.
1768 PH/24 If the recipient of an error message contained an @ in the local part
1769 (suitably quoted, of course), incorrect values were put in $domain and
1770 $local_part during the evaluation of errors_copy.
1776 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1777 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1780 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1781 issue a MAIL command.
1783 PH/02 In an ACL statement such as
1785 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1787 if a client was not listed at all, or was listed with a value other than
1788 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1789 the condition was not true (as it should be), so access was not denied.
1790 The bug was that the ! inversion was incorrectly passed on to the second
1791 item. This has been fixed.
1793 PH/03 Added additional dnslists conditions == and =& which are different from
1794 = and & when the dns lookup returns more than one IP address.
1796 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1797 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1799 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1800 FSYNC, which compiles an option called disable_fsync that allows for
1801 bypassing fsync(). The documentation is heavily laced with warnings.
1803 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1805 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1806 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1807 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1808 including adding "make clean"; (3) Added -fPIC when compiling the test
1809 dynamically loaded module, to get rid of a warning.
1811 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1812 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1813 = 0s. The bug is otherwise harmless.
1815 PH/07 There was a bug in the dovecot authenticator such that the value of
1816 $auth1 could be overwritten, and so not correctly preserved, after a
1817 successful authentication. This usually meant that the value preserved by
1818 the server_setid option was incorrect.
1820 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1822 PH/09 Installed PCRE release 7.0.
1824 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1825 run for batched SMTP input. It is now run at the start of every message
1826 in the batch. While fixing this I discovered that the process information
1827 (output by running exiwhat) was not always getting set for -bs and -bS
1828 input. This is fixed, and it now also says "batched" for BSMTP.
1830 PH/11 Added control=no_pipelining.
1832 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1833 patch, slightly modified), and move the expansion of helo_data till after
1834 the connection is made in the smtp transport (so it can use these
1837 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1839 PH/14 Added log_selector = +pid.
1841 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1843 PH/16 Add ${if forany and ${if forall.
1845 PH/17 Added dsn_from option to vary the From: line in DSNs.
1847 PH/18 Flush SMTP output before performing a callout, unless control =
1848 no_callout_flush is set.
1850 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1851 was true (the default) a successful delivery failed to delete the retry
1852 item, thus causing premature timeout of the address. The bug is now
1855 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1857 PH/21 Long custom messages for fakedefer and fakereject are now split up
1858 into multiline reponses in the same way that messages for "deny" and
1859 other ACL rejections are.
1861 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1862 with slight modification.
1864 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1865 draft, changing the syntax and factoring some duplicate code".
1867 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1868 for deliveries of the second and subsequent messages over the same SMTP
1871 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1872 ${reduce, with only minor "tidies".
1874 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1876 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1877 expansion side effects.
1879 PH/27 When a message times out after an over-quota error from an Exim-imposed
1880 quota, the bounce message says "mailbox is full". This message was not
1881 being given when it was a system quota that was exceeded. It now should
1884 MH/03 Made $recipients available in local_scan(). local_scan() already has
1885 better access to the recipient list through recipients_list[], but
1886 $recipients can be useful in postmaster-provided expansion strings.
1888 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1889 in the case of a MAIL command with additional options following the
1890 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1891 were accidentally chopped off.
1893 PH/29 SMTP synchronization checks are implemented when a command is read -
1894 there is a check that no more input is waiting when there shouldn't be
1895 any. However, for some commands, a delay in an ACL can mean that it is
1896 some time before the response is written. In this time, more input might
1897 arrive, invalidly. So now there are extra checks after an ACL has run for
1898 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1899 pipelining has not been advertised.
1901 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1903 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1904 returned in the SMTP response when smtp_return_error_details was false.
1905 This has been fixed.
1907 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1908 instead of the C I/O that was originally supplied, because problems were
1909 reported on Solaris.
1911 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1912 Exim which did not show up earlier: it was assuming that a call to
1913 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1914 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1915 was a macro that became an assignment, so it seemed to work. This has
1916 changed to a proper function call with a void return, hence the compile
1917 error. Exim's code has been fixed.
1919 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1922 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1924 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1926 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1927 because it turns out to be too big or there is a timeout) there is no
1928 "Completed" line in the log. When some messages of this type were
1929 selected by exigrep, they were listed as "not completed". Others were
1930 picked up by some special patterns. I have improved the selection
1931 criteria to be more general.
1933 PH/38 The host_find_failed option in the manualroute router can now be set
1934 to "ignore", to completely ignore a host whose IP address cannot be
1935 found. If all hosts are ignored, the behaviour is controlled by the new
1936 host_all_ignored option.
1938 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1939 homing or because of multiple MX records with /mx) generated more than
1940 one IP address, and the following item turned out to be the local host,
1941 all the secondary addresses of the first item were incorrectly removed
1942 from the list, along with the local host and any following hosts (which
1943 is what is supposed to happen).
1945 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1946 whoever called Exim into the -H file. In the case of the daemon it was
1947 behaving confusingly. When first started, it used values for whoever
1948 started the daemon, but after a SIGHUP it used the Exim user (because it
1949 calls itself on a restart). I have changed the code so that it now always
1952 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1953 message are rejected with the same error (e.g. no authentication or bad
1954 sender address), and a DATA command is nevertheless sent (as can happen
1955 with PIPELINING or a stupid MUA), the error message that was given to the
1956 RCPT commands is included in the rejection of the DATA command. This is
1957 intended to be helpful for MUAs that show only the final error to their
1960 PH/42 Another patch from the Sieve maintainer.
1962 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1963 Eximstats - Fixed some broken HTML links and added missing column headers
1965 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1966 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1968 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1970 PH/43 Yet another patch from the Sieve maintainer.
1972 PH/44 I found a way to check for a TCP/IP connection going away before sending
1973 the response to the final '.' that terminates a message, but only in the
1974 case where the client has not sent further data following the '.'
1975 (unfortunately, this is allowed). However, in many cases there won't be
1976 any further data because there won't be any more messages to send. A call
1977 to select() can be used: if it shows that the input is "ready", there is
1978 either input waiting, or the socket has been closed. An attempt to read
1979 the next input character can distinguish the two cases. Previously, Exim
1980 would have sent an OK response which the client would never have see.
1981 This could lead to message repetition. This fix should cure that, at
1982 least in a lot of common cases.
1984 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1985 advertised in response to EHLO.
1991 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1992 fixed by 4.65/MH/01 (is this a record?) are fixed:
1994 (i) An empty string was always treated as zero by the numeric comparison
1995 operators. This behaviour has been restored.
1997 (ii) It is documented that the numeric comparison operators always treat
1998 their arguments as decimal numbers. This was broken in that numbers
1999 starting with 0 were being interpreted as octal.
2001 While fixing these problems I realized that there was another issue that
2002 hadn't been noticed. Values of message_size_limit (both the global option
2003 and the transport option) were treated as octal if they started with 0.
2004 The documentation was vague. These values are now always treated as
2005 decimal, and I will make that clear in the documentation.
2011 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2012 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2015 MH/01 Don't check that the operands of numeric comparison operators are
2016 integers when their expansion is in "skipping" mode (fixes bug
2017 introduced by 4.64-PH/07).
2019 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2020 child addresses, Exim now panics and dies. Previously, because the count
2021 is held in a short int, deliveries were likely to be lost. As such a
2022 large number of recipients for a single message is ridiculous
2023 (performance will be very, very poor), I have chosen to impose a limit
2024 rather than extend the field.
2030 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2031 leftover -K file (the existence of which was triggered by #402).
2032 While we were at it, introduced process PID as part of the -K
2033 filename. This should rule out race conditions when creating
2036 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2037 processing considerably. Previous code took too long for large mails,
2038 triggering a timeout which in turn triggers #401.
2040 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2041 in the DK code in transports.c. sendfile() is not really portable,
2042 hence the _LINUX specificness.
2044 TF/01 In the add_headers option to the mail command in an Exim filter,
2045 there was a bug that Exim would claim a syntax error in any
2046 header after the first one which had an odd number of characters
2049 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2050 callout verification, Exim cached a "reject" for the entire domain. This
2051 is correct for most verifications, but it is not correct for a recipient
2052 verification with use_sender or use_postmaster set, because in that case
2053 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2054 case of MAIL FROM:<> rejection from other early rejections (e.g.
2055 rejection of HELO). When verifying a recipient using a non-null MAIL
2056 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2057 Whatever the result of the callout, the value of the domain cache is
2058 left unchanged (for any other kind of callout, getting as far as trying
2059 RCPT means that the domain itself is ok).
2061 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2064 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2065 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2066 session, the daemon ignores ECONNECT errors and logs others; it now
2067 ignores EPIPE as well.
2069 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2070 (quoted-printable decoding).
2072 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2073 later the small subsequent patch to fix an introduced bug.
2075 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2077 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2079 PH/08 An error is now given if message_size_limit is specified negative.
2081 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2082 to be given (somewhat) arbitrary names.
2084 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2087 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2088 miscellaneous code fixes
2090 PH/10 Added the log_reject_target ACL modifier to specify where to log
2093 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2094 hostname. This is wrong, because it relates to the incoming message (and
2095 probably the interface on which it is arriving) and not to the outgoing
2096 callout (which could be using a different interface). This has been
2097 changed to use the value of the helo_data option from the smtp transport
2098 instead - this is what is used when a message is actually being sent. If
2099 there is no remote transport (possible with a router that sets up host
2100 addresses), $smtp_active_hostname is used.
2102 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2103 tweaks were necessary in order to get it to work (see also 21 below):
2104 (a) The code assumed that strncpy() returns a negative number on buffer
2105 overflow, which isn't the case. Replaced with Exim's string_format()
2107 (b) There were several signed/unsigned issues. I just did the minimum
2108 hacking in of casts. There is scope for a larger refactoring.
2109 (c) The code used strcasecmp() which is not a standard C function.
2110 Replaced with Exim's strcmpic() function.
2111 (d) The code set only $1; it now sets $auth1 as well.
2112 (e) A simple test gave the error "authentication client didn't specify
2113 service in request". It would seem that Dovecot has changed its
2114 interface. Fortunately there's a specification; I followed it and
2115 changed what the client sends and it appears to be working now.
2117 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2120 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2121 address A is aliased to B and C, where B exists and C does not. Without
2122 -v the output is "A verified" because verification stops after a
2123 successful redirection if more than one address is generated. However,
2124 with -v the child addresses are also verified. Exim was outputting "A
2125 failed to verify" and then showing the successful verification for C,
2126 with its parentage. It now outputs "B failed to verify", showing B's
2127 parentage before showing the successful verification of C.
2129 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2130 look up a TXT record in a specific list after matching in a combined
2133 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2134 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2135 they consult the DNS. I had assumed they would set it the way they
2136 wanted; and indeed my experiments on Linux seem to show that in some
2137 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2138 To be on the safe side, however, I have now made the interface to
2139 host_find_byname() similar to host_find_bydns(), with an argument
2140 containing the DNS resolver options. The host_find_byname() function now
2141 sets these options at its start, just as host_find_bydns() does. The smtp
2142 transport options dns_qualify_single and dns_search_parents are passed to
2143 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2144 of host_find_byname() use the default settings of RES_DEFNAMES
2145 (qualify_single) but not RES_DNSRCH (search_parents).
2147 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2148 spool_read_header() do less string testing, by means of a preliminary
2149 switch on the second character of optional "-foo" lines. (This is
2150 overdue, caused by the large number of possibilities that now exist.
2151 Originally there were few.) While I was there, I also converted the
2152 str(n)cmp tests so they don't re-test the leading "-" and the first
2153 character, in the hope this might squeeze out yet more improvement.
2155 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2156 flag allowing group syntax was set by the header_syntax check but not
2157 turned off, possible causing trouble later; (2) The flag was not being
2158 set at all for the header_verify test, causing "group"-style headers to
2159 be rejected. I have now set it in this case, and also caused header_
2160 verify to ignore an empty address taken from a group. While doing this, I
2161 came across some other cases where the code for allowing group syntax
2162 while scanning a header line wasn't quite right (mostly, not resetting
2163 the flag correctly in the right place). These bugs could have caused
2164 trouble for malformed header lines. I hope it is now all correct.
2166 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2167 with the "reply" argument non-NULL. The code, however (which originally
2168 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2169 but it didn't always do it. This confused somebody who was copying the
2170 code for some other use. I have removed all the tests.
2172 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2173 feature that was used to support insecure browsers during the U.S. crypto
2174 embargo. It requires special client support, and Exim is probably the
2175 only MTA that supported it -- and would never use it because real RSA is
2176 always available. This code has been removed, because it had the bad
2177 effect of slowing Exim down by computing (never used) parameters for the
2178 RSA_EXPORT functionality.
2180 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2181 authenticator to fail if there's a tab character in the incoming data
2182 (there should never be unless someone is messing about, as it's supposed
2183 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2184 if the connection is using TLS or if the remote IP is the same as the
2185 local IP, and the "valid-client-cert option" if a client certificate has
2188 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2189 authenticators. This can be used for authorization after authentication
2190 succeeds. (In the case of plaintext, it servers for both authentication
2193 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2194 if any retry times were supplied.
2196 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2197 connection, where there is no client IP address to check. In this
2198 situation, the verify now always succeeds.
2200 PH/25 Applied John Jetmore's -Mset patch.
2202 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2204 PH/27 In a string expansion for a processed (not raw) header when multiple
2205 headers of the same name were present, leading whitespace was being
2206 removed from all of them, but trailing whitespace was being removed only
2207 from the last one. Now trailing whitespace is removed from each header
2208 before concatenation. Completely empty headers in a concatenation (as
2209 before) are ignored.
2211 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2212 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2214 PH/29 [Removed. This was a change that I later backed out, and forgot to
2215 correct the ChangeLog entry (that I had efficiently created) before
2216 committing the later change.]
2218 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2219 address errors (4xx response to RCPT) over the same connection as other
2220 messages routed to the same hosts. Such deliveries are always "forced",
2221 so retry times are not inspected. This resulted in far too many retries
2222 for the affected addresses. The effect occurred only when there were more
2223 hosts than the hosts_max_try setting in the smtp transport when it had
2224 the 4xx errors. Those hosts that it had tried were not added to the list
2225 of hosts for which the message was waiting, so if all were tried, there
2226 was no problem. Two fixes have been applied:
2228 (i) If there are any address or message errors in an SMTP delivery, none
2229 of the hosts (tried or untried) are now added to the list of hosts
2230 for which the message is waiting, so the message should not be a
2231 candidate for sending over the same connection that was used for a
2232 successful delivery of some other message. This seems entirely
2233 reasonable: after all the message is NOT "waiting for some host".
2234 This is so "obvious" that I'm not sure why it wasn't done
2235 previously. Hope I haven't missed anything, but it can't do any
2236 harm, as the worst effect is to miss an optimization.
2238 (ii) If, despite (i), such a delivery is accidentally attempted, the
2239 routing retry time is respected, so at least it doesn't keep
2240 hammering the server.
2242 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2243 in ${readsocket because some servers need this prod.
2245 PH/32 Added some extra debug output when updating a wait-xxx database.
2247 PH/33 The hint "could be header name not terminated by colon", which has been
2248 given for certain expansion errors for a long time, was not being given
2249 for the ${if def:h_colon_omitted{... case.
2251 PH/34 The spec says: "With one important exception, whenever a domain list is
2252 being scanned, $domain contains the subject domain." There was at least
2253 one case where this was not true.
2255 PH/35 The error "getsockname() failed: connection reset by peer" was being
2256 written to the panic log as well as the main log, but it isn't really
2257 panic-worthy as it just means the connection died rather early on. I have
2258 removed the panic log writing for the ECONNRESET error when getsockname()
2261 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2262 runs only) independently of the message's sender address. This meant
2263 that, if the 4xx error was in fact related to the sender, a different
2264 message to the same recipient with a different sender could confuse
2265 things. In particualar, this can happen when sending to a greylisting
2266 server, but other circumstances could also provoke similar problems.
2267 I have changed the default so that the retry time for these errors is now
2268 based a combination of the sender and recipient addresses. This change
2269 can be overridden by setting address_retry_include_sender=false in the
2272 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2273 remote server are returned as part of bounce messages. This was not
2274 happening for LMTP over a pipe (the lmtp transport), but now it is the
2275 same for both kinds of LMTP.
2277 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2278 in header lines that were in fact CNAMEs. This is no longer the case.
2280 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2281 and queue runs started by the daemon processed all messages. This has
2282 been fixed so that -R and -S can now usefully be given with -q<time>.
2284 PH/40 Import PCRE release 6.7 (fixes some bugs).
2286 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2288 PH/42 Give an error if -q is specified more than once.
2290 PH/43 Renamed the variables $interface_address and $interface_port as
2291 $received_ip_address and $received_port, to make it clear that these
2292 values apply to message reception, and not to the outgoing interface when
2293 a message is delivered. (The old names remain recognized, of course.)
2295 PH/44 There was no timeout on the connect() call when using a Unix domain
2296 socket in the ${readsocket expansion. There now is.
2298 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2299 be meaningful with "accept".
2301 SC/01 Eximstats V1.43
2302 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2304 SC/02 Eximstats V1.44
2305 Use a glob alias rather than an array ref in the generated
2306 parser. This improves both readability and performance.
2308 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2309 Collect SpamAssassin and rejection statistics.
2310 Don't display local sender or destination tables unless
2311 there is data to show.
2312 Added average volumes into the top table text output.
2314 SC/04 Eximstats V1.46
2315 Collect data on the number of addresses (recipients)
2316 as well as the number of messages.
2318 SC/05 Eximstats V1.47
2319 Added 'Message too big' to the list of mail rejection
2320 reasons (thanks to Marco Gaiarin).
2322 SC/06 Eximstats V1.48
2323 Mainlog lines which have GMT offsets and are too short to
2324 have a flag are now skipped.
2326 SC/07 Eximstats V1.49 (Alain Williams)
2327 Added the -emptyok flag.
2329 SC/08 Eximstats V1.50
2330 Fixes for obtaining the IP address from reject messages.
2332 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2333 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2334 whitesspace changes from 4.64-PH/27
2336 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2339 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2340 are found, allow negative numbers in numeric criteria)
2342 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2344 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2345 to match changes made in 4.64-PH/43
2347 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2349 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2350 "domain" is actually an IP address, and force a failure. This locks out
2351 those revolvers/nameservers that support "A-for-A" lookups, in
2352 contravention of the specifications.
2354 PH/48 When a host name was looked up from an IP address, and the subsequent
2355 forward lookup of the name timed out, the host name was left in
2356 $sender_host_name, contrary to the specification.
2358 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2359 restricted to single-key lookups, Exim was not diagnosing an error if
2360 * or *@ was used with a query-style lookup.
2362 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2364 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2365 long ago, but noone interested enough thought of it. Let's just say that
2366 the "1.1" means that there are some new functions that weren't there at
2367 some point in the past.
2369 PH/51 Error processing for expansion failure of helo_data from an smtp
2370 transport during callout processing was broken.
2372 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2373 tested/used via the -bh/-bhc/-bs options.
2375 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2376 bug, fixed in subsequent PCRE releases).
2378 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2379 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2381 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2387 SC/01 Use a glob alias rather than an array ref in eximstats generated
2388 parser. This improves both readability and performance.
2390 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2391 Don't display local sender or destination tables in eximstats unless
2392 there is data to show.
2393 Added average volumes into the eximstats top table text output.
2395 SC/03 Collect data on the number of addresses (recipients) as well
2396 as the number of messages in eximstats.
2398 TF/01 Correct an error in the documentation for the redirect router. Exim
2399 does (usually) call initgroups() when daemonizing.
2401 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2402 with consistent privilege compared to when running as a daemon.
2404 TF/03 Note in the spec that $authenticated_id is not set for local
2405 submissions from trusted users.
2407 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2408 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2410 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2411 by adding some example configuration directives to the default
2412 configuration file. A little bit of work is required to uncomment the
2413 directives and define how usernames and passwords are checked, but
2414 there is now a framework to start from.
2416 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2417 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2418 without this. I don't know how relevant this is to other LDAP libraries.
2420 PH/02 Add the verb name to the "unknown ACL verb" error.
2422 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2424 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2426 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2427 directory not expanded when it should be if an expanded home directory
2428 was set for the address (which is overridden by the transport).
2430 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2433 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2434 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2435 because it is too late at that time, and has no effect.
2437 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2438 security issue with \' (bugzilla #107). I could not use the
2439 PQescapeStringConn() function, because it needs a PGconn value as one of
2442 PH/08 When testing addresses using -bt, indicate those final addresses that
2443 are duplicates that would not cause an additional delivery. At least one
2444 person was confused, thinking that -bt output corresponded to deliveries.
2445 (Suppressing duplicates isn't a good idea as you lose the information
2446 about possibly different redirections that led to the duplicates.)
2448 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2449 systems where poll() doesn't work, in particular OS X.
2451 PH/10 Added more information to debugging output for retry time not reached.
2453 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2454 operations in malware.c.
2456 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2459 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2460 syslog_duplication set false, log lines that would normally be written
2461 both the the main log and to the reject log were not written to syslog at
2464 PH/14 In the default configuration, change the use of "message" in ACL warn
2465 statements to "add_header".
2467 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2468 not followed by a command (e.g. "seen endif").
2470 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2471 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2474 PH/17 Added extra conditions to the default value of delay_warning_condition
2478 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2479 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2480 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2483 The Auto-Submitted: and various List- headers are standardised, whereas I
2484 don't think Precedence: ever was.
2486 PH/18 Refactored debugging code in route_finduser() to show more information,
2487 in particular, the error code if getpwnam() issues one.
2489 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2490 This is apparently needed in addition to the PH/07 change above to avoid
2491 any possible encoding problems.
2493 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2494 but not after initializing Perl.
2496 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2497 output them only if debugging. By default they are written stderr,
2498 apparently, which is not desirable.
2500 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2503 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2506 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2508 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2509 authenticated or an ident call has been made. Suppress the default
2510 values for $authenticated_id and $authenticated_sender (but permit -oMai
2511 and -oMas) when testing with -bh.
2513 PH/24 Re-jigged the order of the tests in the default configuration so that the
2514 tests for valid domains and recipients precede the DNS black list and CSA
2515 tests, on the grounds that those ones are more expensive.
2517 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2518 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2519 This bug exists in every version of Exim that I still have, right back to
2522 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2523 However, an attempt to turn on case-sensitivity in a regex key by
2524 including (?-i) didn't work because the subject string was already
2525 lowercased, and the effects were non-intuitive. It turns out that a
2526 one-line patch can be used to allow (?-i) to work as expected.
2532 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2533 other effects) broke the use of negated acl sub-conditions.
2535 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2538 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2539 "Deny" causes Exim to reject the incoming connection with a 554 error.
2540 Unfortunately, if there is a major crisis, such as a disk failure,
2541 tcp-wrappers gives "deny", whereas what one would like would be some
2542 kind of temporary error. A kludge has been added to help with this.
2543 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2544 554 error is used if errno is still zero or contains ENOENT (which occurs
2545 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2548 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2550 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2551 errors. Otherwise a message that provokes a temporary error (when other
2552 messages do not) can cause a whole host to time out.
2554 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2555 addresses were routed directly to files or pipes from a redirect router.
2556 File deliveries just didn't batch; pipe deliveries might have suffered
2559 PH/06 A failure to get a lock for a hints database would erroneously always say
2560 "Failed to get write lock", even when it was really a read lock.
2562 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2563 of 0600. This has been changed to use the value of the lockfile_mode
2564 option (which defaults to 0600).
2566 PH/08 Applied small patch from the Sieve maintainer.
2568 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2569 folder from quota calculations, a direct delivery into this folder messed
2570 up the contents of the maildirsize file. This was because the regex was
2571 used only to exclude .Trash (or whatever) when the size of the mailbox
2572 was calculated. There was no check that a delivery was happening into an
2573 excluded directory. This bug has been fixed by ignoring all quota
2574 processing for deliveries into excluded directories.
2576 PH/10 Added the maildirfolder_create_regex option to appendfile.
2582 PH/01 The code for finding all the local interface addresses on a FreeBSD
2583 system running IPv6 was broken. This may well have applied to all BSD
2584 systems, as well as to others that have similar system calls. The broken
2585 code found IPv4 interfaces correctly, but gave incorrect values for the
2586 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2587 that it would not match correctly against @[] and not recognize the IPv6
2590 PH/02 The ipliteral router was not recognizing addresses of the form user@
2591 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2593 PH/03 Added disable_ipv6.
2595 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2596 decoded form, because it is most often used to construct To: headers
2597 lines in autoreplies, and the decoded form may well be syntactically
2598 invalid. However, $reply_address has leading white space removed, and all
2599 newlines turned into spaces so that the autoreply transport does not
2602 PH/05 If group was specified without a user on a router, and no group or user
2603 was specified on a transport, the group from the router was ignored.
2605 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2606 for visible compile-time settings that can be used to change these
2607 numbers, for those that want even more. Backwards compatibility with old
2608 spool files has been maintained. However, going back to a previous Exim
2609 release will lost any variables that are in spool files.
2611 PH/07 Two small changes when running in the test harness: increase delay when
2612 passing a TCP/IP connection to a new process, in case the original
2613 process has to generate a bounce, and remove special handling of
2614 127.0.0.2 (sic), which is no longer necessary.
2616 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2617 be the same on different OS.
2619 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2622 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2623 whether --show-vars was specified or not
2625 JJ/02 exipick: Added support for new ACL variable spool format introduced
2628 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2629 syntactically invalid From: or Reply-to: line, and a filter used this to
2630 generate an autoreply, and therefore failed to obtain an address for the
2631 autoreply, Exim could try to deliver to a non-existent relative file
2632 name, causing unrelated and misleading errors. What now happens is that
2633 it logs this as a hard delivery error, but does not attempt to create a
2636 PH/11 The exinext utility has a -C option for testing purposes, but although
2637 the given file was scanned by exinext itself; it wasn't being passed on
2638 when Exim was called.
2640 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2641 an end-of-file indication when reading a command response.
2643 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2644 compiled. In many other places in Exim, IPv6 addresses are always
2645 recognized, so I have changed this. It also means that IPv4 domain
2646 literals of the form [IPV4:n.n.n.n] are now always recognized.
2648 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2649 used if the router is not running as root, for example, when verifying at
2650 ACL time, or when using -bh. The debugging output from this situation was
2651 non-existent - all you got was a failure to exec. I have made two
2654 (a) Failures to set uid/gid, the current directory, or a process leader
2655 in a subprocess such as that created by queryprogram now generate
2656 suitable debugging ouput when -d is set.
2658 (b) The queryprogram router detects when it is not running as root,
2659 outputs suitable debugging information if -d is set, and then runs
2660 the subprocess without attempting to change uid/gid.
2662 PH/15 Minor change to Makefile for building test_host (undocumented testing
2665 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2666 additional section of a DNS packet that returns MX or SRV records.
2667 Instead, it always explicitly searches for A/AAAA records. This avoids
2668 major problems that occur when a DNS server includes only records of one
2669 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2670 fixed another bug: if SRV records were looked up and the corresponding
2671 address records were *not* found in the additional section, the port
2672 values from the SRV records were lost.
2674 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2675 using the correct key (the original address) when searching the retry
2676 rules in order to find which one to use for generating the retry hint.
2678 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2679 adding the default one. Similarly, if it contains a Reply-To: header, the
2680 errors_reply_to option, if set, is not used.
2682 PH/19 When calculating a retry time, Exim used to measure the "time since
2683 failure" by looking at the "first failed" field in the retry record. Now
2684 it does not use this if it is later than than the arrival time of the
2685 message. Instead it uses the arrival time. This makes for better
2686 behaviour in cases where some deliveries succeed, thus re-setting the
2687 "first failed" field. An example is a quota failure for a huge message
2688 when small messages continue to be delivered. Without this change, the
2689 "time since failure" will always be short, possible causing more frequent
2690 delivery attempts for the huge message than are intended.
2691 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2693 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2694 $1, $2, $3) because the numerical variables can be reset during some
2695 expansion items (e.g. "match"), thereby losing the authentication data.
2697 PH/21 Make -bV show the size of off_t variables so that the test suite can
2698 decide whether to run tests for quotas > 2G.
2700 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2701 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2702 filecount value is greater than 2G or if a quota value is greater than 2G
2703 on a system where the size of off_t is not greater than 4, a panic error
2706 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2707 never match. The debug and -bh output now contains an explicit error
2708 message indicating a malformed IPv4 address or mask.
2710 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2711 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2712 PH/23 above applies.
2714 PH/25 Do not write to syslog when running in the test harness. The only
2715 occasion when this arises is a failure to open the main or panic logs
2716 (for which there is an explicit test).
2718 PH/26 Added the /no_tell option to "control=freeze".
2720 PH/27 If a host name lookup failed very early in a connection, for example, if
2721 the IP address matched host_lookup and the reverse lookup yielded a name
2722 that did not have a forward lookup, an error message of the form "no IP
2723 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2724 could be logged. Now it outputs the IP address instead of "NULL".
2726 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2727 allows the sender and the authenticated sender to be set when
2728 submitting a message from within Exim. Since child_open_exim() is
2729 documented for local_scan(), the new function should be too.
2731 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2732 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2733 results in an empty string is now treated as unset.
2735 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2737 PH/31 Added sender_verify_fail logging option.
2739 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2740 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2741 tidied the source and removed it altogether.
2743 PH/33 When a queue run was abandoned because the load average was too high, a
2744 log line was always written; now it is written only if the queue_run log
2745 selector is set. In addition, the log line for abandonment now contains
2746 information about the queue run such as the pid. This is always present
2747 in "start" and "stop" lines but was omitted from the "abandon" line.
2749 PH/34 Omit spaces between a header name and the colon in the error message that
2750 is given when verify = headers_syntax fails (if there are lots of them,
2751 the message gets confusing).
2753 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2754 names, as there are now some PTR records that contain slashes. This check
2755 is only to protect against broken name servers that fall over on strange
2756 characters, so the fact that it applies to all lookups doesn't matter.
2758 PH/36 Now that the new test suite is complete, we can remove some of the
2759 special code in Exim that was needed for the old test suite. For example,
2760 sorting DNS records because real resolvers return them in an arbitrary
2761 order. The new test suite's fake resolver always returns records in the
2764 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2765 bounces) except when queue_only is set, to avoid logging races between
2766 the different processes.
2768 PH/38 Panic-die if .include specifies a non-absolute path.
2770 PH/39 A tweak to the "H" retry rule from its user.
2772 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2773 a label. They prevented compilation on older perls.
2775 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2776 a warning to be raised on newish perls.
2778 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2779 on queue. Changes to match documented behaviour of showing count of
2780 messages matching specified criteria.
2782 PH/40 Changed the default ident timeout from 30s to 5s.
2784 PH/41 Added support for the use of login_cap features, on those BSD systems
2785 that have them, for controlling the resources used by pipe deliveries.
2787 PH/42 The content-scanning code uses fopen() to create files in which to put
2788 message data. Previously it was not paying any attention to the mode of
2789 the files. Exim runs with umask(0) because the rest of the code creates
2790 files with open(), and sets the required mode explicitly. Thus, these
2791 files were ending up world-writeable. This was not a big issue, because,
2792 being within the spool directory, they were not world-accessible. I have
2793 created a function called modefopen, which takes an additional mode
2794 argument. It sets umask(777), creates the file, chmods it to the required
2795 mode, then resets the umask. All the relevant calls to fopen() in the
2796 content scanning code have been changed to use this function.
2798 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2799 to 24 hours. This avoids potential overflow problems when processing G
2800 and H retry rules. I suspect nobody ever tinkers with this value.
2802 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2804 PH/45 When the plaintext authenticator is running as a client, the server's
2805 challenges are checked to ensure they are valid base64 strings. By
2806 default, the authentication attempt is cancelled if an invalid string is
2807 received. Setting client_ignore_invalid_base64 true ignores these errors.
2808 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2809 they are received. Thus, the responses can be made to depend on the
2810 challenges. If an invalid string is ignored, an empty string is placed in
2813 PH/46 Messages that are created by the autoreply transport now contains a
2814 References: header, in accordance with RFCs 2822 and 3834.
2816 PH/47 Added authenticated_sender_force to the smtp transport.
2818 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2820 PH/49 Installed latest patch from the Sieve maintainer.
2822 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2823 was also set, the appendfile transport was unnecessarily scanning a
2824 directory of message files (e.g. for maildir delivery) to find the count
2825 of files (along with the size), even though it did not need this
2826 information. It now does the scan only if it needs to find either the
2827 size of the count of files.
2829 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2831 PH/52 Two bugs concerned with error handling when the smtp transport is
2834 (i) Exim was not creating retry information for temporary errors given
2835 for individual recipients after the DATA command when the smtp transport
2836 was used in LMTP mode. This meant that they could be retried too
2837 frequently, and not timed out correctly.
2839 (ii) Exim was setting the flag that allows error details to be returned
2840 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2841 recipients that were returned after the DATA command.
2843 PH/53 This is related to PH/52, but is more general: for any failing address,
2844 when detailed error information was permitted to be returned to the
2845 sender, but the error was temporary, then after the final timeout, only
2846 "retry timeout exceeded" was returned. Now it returns the full error as
2847 well as "retry timeout exceeded".
2849 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2850 do this, and (what is worse) MTAs that accept it.
2852 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2853 will now be deprecated.
2855 PH/56 New os.c-cygwin from the Cygwin maintainer.
2857 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2858 formats (previously only available in exim formats via -bpr, -bpru,
2859 and -bpra. Now also available in native and exiqgrep formats)
2861 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2862 with very large, slow to parse queues
2864 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2866 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2868 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2869 responses to authentication challenges, though it was showing the
2870 challenges; (ii) I've removed the CR characters from the debug output for
2873 PH/58 Allow for the insertion of a newline as well as a space when a string
2874 is turned into more than one encoded-word during RFC 2047 encoding. The
2875 Sieve code now uses this.
2877 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2878 data_4xx, lost_connection, tls_required.
2880 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2881 message was being sent as an SMTP response.
2883 PH/61 Add -l and -k options to exicyclog.
2885 PH/62 When verifying, if an address was redirected to one new address, so that
2886 verification continued, and the new address failed or deferred after
2887 having set something in $address_data, the value of $address_data was not
2888 passed back to the ACL. This was different to the case when no
2889 redirection occurred. The value is now passed back in both cases.
2891 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2892 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2893 use login_cap.h, so on its own it isn't the distinguishing feature. The
2894 new name refers directly to the setclassresources() function.
2896 PH/65 Added configuration files for NetBSD3.
2898 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2900 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2901 is preferred over IPv4.
2903 PH/68 The bounce_return_message and bounce_return_body options were not being
2904 honoured for bounces generated during the reception of non-SMTP messages.
2905 In particular, this applied to messages rejected by the ACL. This bug has
2906 been fixed. However, if bounce_return_message is true and bounce_return_
2907 body is false, the headers that are returned for a non-SMTP message
2908 include only those that have been read before the error was detected.
2909 (In the case of an ACL rejection, they have all been read.)
2911 PH/69 The HTML version of the specification is now built in a directory called
2912 spec_html instead of spec.html, because the latter looks like a path with
2913 a MIME-type, and this confuses some software.
2915 PH/70 Catch two compiler warnings in sieve.c.
2917 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2918 function verify_get_ident() calls ip_connect() to connect a socket, but
2919 if the "connect()" function timed out, ip_connect() used to close the
2920 socket. However, verify_get_ident() also closes the socket later, and in
2921 between Exim writes to the log, which may get opened at this point. When
2922 the socket was closed in ip_connect(), the log could get the same file
2923 descriptor number as the socket. This naturally causes chaos. The fix is
2924 not to close the socket in ip_connect(); the socket should be closed by
2925 the function that creates it. There was only one place in the code where
2926 this was missing, in the iplookup router, which I don't think anybody now
2927 uses, but I've fixed it anyway.
2929 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2930 well as to direct DNS lookups. Otherwise the handling of names in host
2931 lists is inconsistent and therefore confusing.
2937 PH/01 Two changes to the default runtime configuration:
2939 (1) Move the checks for relay_from_hosts and authenticated clients from
2940 after to before the (commented out) DNS black list checks.
2942 (2) Add control=submission to the relay_from_hosts and authenticated
2943 clients checks, on the grounds that messages accepted by these
2944 statements are most likely to be submissions.
2946 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2948 (1) Generate an error if the third argument for the ${prvs expansion is
2951 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2954 (3) Reset the variables that are obtained from the first argument of
2955 ${prvscheck and used in the second argument before leaving the code,
2956 because their memory is reclaimed, so using them afterwards may do
2959 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2960 one (it's much easier than Tom thought :-).
2962 (5) Because of (4), we can now allow for the use of $prvscheck_result
2963 inside the third argument.
2965 PH/03 For some reason, the default setting of PATH when running a command from
2966 a pipe transport was just "/usr/bin". I have changed it to
2969 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2970 anything to be listed in the output from -bV.
2972 PH/05 When a filter generated an autoreply, the entire To: header line was
2973 quoted in the delivery log line, like this:
2975 => >A.N.Other <ano@some.domain> <original@ddress> ...
2977 This has been changed so that it extracts the operative address. There
2978 may be more than one such address. If so, they are comma-separated, like
2981 => >ano@some.domain,ona@other.domain <original@ddress> ...
2983 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2984 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2985 not being looked up in the rDNS to get a host name, Exim was showing the
2986 IP address twice in Received: lines, even though the IP addresses were
2987 identical. For example:
2989 Received: from [1.2.3.4] (helo=[1.2.3.4])
2991 However, if the real host name was known, it was omitting the HELO data
2992 if it matched the actual IP address. This has been tidied up so that it
2993 doesn't show the same IP address twice.
2995 PH/07 When both +timestamp and +memory debugging was on, the value given by
2996 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2997 called by the debug printing, thereby overwriting the timestamp buffer.
2998 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3000 PH/08 When the original message was included in an autoreply transport, it
3001 always said "this is a copy of the message, including all the headers",
3002 even if body_only or headers_only was set. It now gives an appropriate
3005 PH/09 Applied a patch from the Sieve maintainer which:
3007 o fixes some comments
3008 o adds the (disabled) notify extension core
3009 o adds some debug output for the result of if/elsif tests
3010 o points to the current vacation draft in the documentation
3011 and documents the missing references header update
3015 o fixes a bug in processing the envelope test (when testing
3016 multiple envelope elements, the last element determinted the
3019 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3020 Electronic Mail") by including:
3022 Auto-submitted: auto-generated
3024 in the messages that it generates (bounce messages and others, such as
3025 warnings). In the case of bounce messages for non-SMTP mesages, there was
3026 also a typo: it was using "Auto_submitted" (underscore instead of
3027 hyphen). Since every message generated by Exim is necessarily in response
3028 to another message, thes have all been changed to:
3030 Auto-Submitted: auto-replied
3032 in accordance with these statements in the RFC:
3034 The auto-replied keyword:
3036 - SHOULD be used on messages sent in direct response to another
3037 message by an automatic process,
3039 - MUST NOT be used on manually-generated messages,
3041 - MAY be used on Delivery Status Notifications (DSNs) and Message
3042 Disposition Notifications (MDNs),
3044 - MUST NOT be used on messages generated by automatic or periodic
3045 processes, except for messages which are automatic responses to
3048 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3049 to the default Received: header definition.
3051 PH/12 Added log selector acl_warn_skipped (default on).
3053 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3054 variables because (a) they are in the wrong storage pool and (b) even if
3055 they were copied, it wouldn't work properly because of the caching.
3057 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3058 checking when decoding. Apparently there are clients that generate
3059 overlong encoded strings. Why am I not surprised?
3061 PH/15 If the first argument of "${if match_address" was not empty, but did not
3062 contain an "@" character, Exim crashed. Now it writes a panic log message
3063 and treats the condition as false.
3065 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3067 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3068 extension "envelope-auth". The code is finished and in agreement with
3069 other implementations, but there is no documentation so far and in fact,
3070 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3071 not changing the active code.
3073 Print executed "if" and "elsif" statements when debugging is used. This
3074 helps a great deal to understand what a filter does.
3076 Document more things not specified clearly in RFC3028. I had all this
3077 sorted out, when out of a sudden new issues came to my mind. Oops."
3079 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3082 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3083 canonical form (as documented). However, after a host name lookup from
3084 the IP address, check_host() was doing a simple string comparison with
3085 addresses acquired from the DNS when checking that the found name did
3086 have the original IP as one of its addresses. Since any found IPv6
3087 addresses are likely to be in abbreviated form, the comparison could
3088 fail. Luckily, there already exists a function for doing the comparison
3089 by converting both addresses to binary, so now that is used instead of
3090 the text comparison.
3092 PH/20 There was another similar case to PH/19, when a complete host name was
3093 given in a host list; looking up its IP address could give an abbreviated
3094 form, whereas the current host's name might or might not be abbreviated.
3095 The same fix has been applied.
3101 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3102 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3105 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3106 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3108 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3110 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3111 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3112 Also, for "Auto-Submitted", treat anything other than "no" as
3113 non-personal, in accordance with RFC 3834. (Previously it treated
3114 anything starting "auto-" as non-personal.)
3116 TF/01 The control=submission/name=... option had a problem with syntax
3117 errors if the name included a slash character. The /name= option
3118 now slurps the rest of the string, so it can include any characters
3119 but it must come last in the list of options (after /sender_retain
3122 PH/05 Some modifications to the interface to the fake nameserver for the new
3130 TK/01 Added the "success_on_redirect" address verification option. See
3131 NewStuff for rationale and an example.
3133 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3135 PH/02 Patch to exigrep to allow it to work on syslog lines.
3137 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3138 fread() to skip over the body file's header line, because in Cygwin the
3139 header line is locked and is inaccessible.
3141 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3142 co-exist for some time) to make it clear that it is the Exim ID that is
3143 referenced, not the Message-ID: header line.
3145 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3146 string_format() function, because snprintf() does not exist on all
3149 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3150 ${stat: expansion item.
3152 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3153 protocol synchronization error", to keep the pedants happy.
3155 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3156 well as for IRIX systems, when gcc is being used. See the host.c source
3159 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3161 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3164 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3165 transport and to the smtp transport in LMTP mode.
3167 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3169 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3170 run a filter in a subprocess. This could lead to confusion in subsequent
3171 lookups in the parent process. There should also be a search_tidyup() at
3172 the end of the subprocess.
3174 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3175 only if the host matched helo_try_verify_hosts, which caused the
3176 verification to occur when the EHLO/HELO command was issued. The ACL just
3177 tested the remembered result. Now, if a previous verification attempt has
3178 not happened, "verify = helo" does it there and then.
3180 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3182 TK/03 Fix log output including CR from clamd.
3184 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3185 exist provoked a memory error which could cause a segfault.
3187 PH/15 Installed PCRE 6.2
3189 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3191 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3192 of the problem. Specifically, suggested +O2 rather than +O1 for the
3195 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3197 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3198 record, the port was not correctly logged when the outgoing_port log
3199 selector was set (it logged the transort's default port).
3201 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3202 fallback_hosts, and "hosts" in the smtp transport.
3204 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3205 host errors such as "Connection refused".
3207 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3208 authentication with radiusclient 0.4.9:
3210 - Error returned from rc_read_config was caught wrongly
3211 - Username/password not passed on to radius server due to wrong length.
3213 The presumption is that some radiusclient API changes for 4.51/PH/17
3214 were not taken care of correctly. The code is still untested by me (my
3215 Linux distribution still has 0.3.2 of radiusclient), but it was
3216 contributed by a Radius user.
3218 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3219 expanding the "port" option of the smtp transport.
3221 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3222 while reading a MIME header. Thanks to Tom Hughes for a patch.
3224 PH/24 Include config.h inside local_scan.h so that configuration settings are
3227 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3228 that in an ACL for RCPT (for example), you can examine exactly what was
3231 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3232 commands, but it was not correctly comparing the address with the actual
3233 client host address. Thus, it would show the EHLO address in Received:
3234 header lines when this was not necessary.
3236 PH/27 Added the % operator to ${eval:}.
3238 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3239 it should be ignoring failures (because with -C, for example, it has lost
3240 privilege). It wasn't ignoring creation failures other than "already
3243 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3246 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3247 because an input line was too long, either on its own, or by virtue of
3248 too many continuations, the temporary file was not being removed, and the
3249 return code was incorrect.
3251 PH/31 Missing "BOOL" in function definition in filtertest.c.
3253 PH/32 Applied Sieve patches from the maintainer.
3255 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3257 PH/33 Added "verify = not_blind".
3259 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3260 Local/Makefile (with some defaults set). These are used in built scripts
3261 such as exicyclog, but they have never been used in the exim_install
3262 script (though there are many overriding facilities there). I have
3263 arranged that the exim_install script now takes note of these two
3266 PH/35 Installed configuration files for Dragonfly.
3268 PH/36 When a locally submitted message by a trusted user did not contain a
3269 From: header, and the sender address was obtained from -f or from an SMTP
3270 MAIL command, and the trusted user did not use -F to supply a sender
3271 name, $originator_name was incorrectly used when constructing a From:
3272 header. Furthermore, $originator_name was used for submission mode
3273 messages from external hosts without From: headers in a similar way,
3274 which is clearly wrong.
3276 PH/37 Added control=suppress_local_fixups.
3278 PH/38 When log_selector = +received_sender was set, and the addition of the
3279 sender made the log line's construction buffer exactly full, or one byte
3280 less than full, an overflow happened when the terminating "\n" was
3283 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3284 when the result of a list match is failure because a DNS lookup failed.
3286 PH/40 RM_COMMAND is now used in the building process.
3288 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3289 the "build-* directories that it finds.
3291 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3292 address in a domain literal was a prefix of an interface address.
3294 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3295 when verifying a sender address, unless rewrite_headers is false.
3297 PH/44 Wrote a long comment about why errors_to addresses are verified as
3298 recipients, not senders.
3300 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3301 the ratelimit ACL was added.
3303 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3305 PH/46 Added extra information about PostgreSQL errors to the error string.
3307 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3308 avoiding the need to install special zones in a real server. This is
3309 backwards compatible; if it can't find the fake resolver, it drops back.
3310 Thus, both old and new test suites can be run.
3312 TF/02 Added util/ratelimit.pl
3314 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3315 clock is set back in time.
3317 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3318 Candler <B.Candler@pobox.com>.
3320 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3321 OK for addresses that are the result of redirections.
3323 PH/48 A number of further additions for the benefit of the new test suite,
3324 including a fake gethostbyname() that interfaces to the fake DNS resolver
3327 TF/06 The fix for widen_domains has also been applied to qualify_single and
3328 search_parents which are the other dnslookup options that can cause
3331 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3334 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3336 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3337 service" if it encountered an MX record with an empty target hostname.
3338 The message is now "an MX or SRV record indicated no SMTP service".
3340 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3341 if the DNS of the sending site is misconfigured. This is quite a
3342 common situation. This change restores the behaviour of treating a
3343 helo verification defer as a failure.
3345 PH/51 If self=fail was set on a router, the bounce message did not include the
3346 actual error message.
3352 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3354 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3355 command itself ended in error, the underlying message about the transport
3356 filter timeout was being overwritten with the pipe command error. Now the
3357 underlying error message should be appended to the second error message.
3359 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3361 PH/02 Reduce the amount of output that "make" produces by default. Full output
3362 can still be requested.
3364 PH/03 The warning log line about a condition test deferring for a "warn" verb
3365 was being output only once per connection, rather than after each
3366 occurrence (because it was using the same function as for successful
3367 "warn" verbs). This seems wrong, so I have changed it.
3369 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3370 it should not have, which might have caused a crash in the right
3371 circumstances, but probably never did.
3373 PH/04 Installed a modified version of Tony Finch's patch to make submission
3374 mode fix the return path as well as the Sender: header line, and to
3375 add a /name= option so that you can make the user's friendly name appear
3378 TF/03 Added the control = fakedefer ACL modifier.
3380 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3381 Mark Lowes for thorough testing.
3383 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3385 TK/03 Merged latest SRS patch from Miles Wilton.
3387 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3388 with the definition in sysexits.h (which is #included earlier).
3389 Fortunately, Exim does not actually use EX_OK. The code used to try to
3390 preserve the sysexits.h value, by assumimg that macro definitions were
3391 scanned for macro replacements. I have been disabused of this notion,
3392 so now the code just undefines EX_OK before #including unistd.h.
3394 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3395 in the smtp transport. When a block could not be written in a single
3396 write() function, the timeout was being re-applied to each part-write.
3397 This seems wrong - if the receiver was accepting one byte at a time it
3398 would take for ever. The timeout is now adjusted when this happens. It
3399 doesn't have to be particularly precise.
3401 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3402 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3404 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3405 without a domain if the check to <postmaster@domain> fails.
3407 SC/01 Eximstats: added -xls and the ability to specify output files
3408 (patch written by Frank Heydlauf).
3410 SC/02 Eximstats: use FileHandles for outputing results.
3412 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3414 SC/04 Eximstats: fixed display of large numbers with -nvr option
3416 SC/05 Eximstats: fixed merging of reports with empty tables.
3418 SC/06 Eximstats: added the -include_original_destination flag
3420 SC/07 Eximstats: removed tabs and trailing whitespace.
3422 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3424 TK/06 MBOX spool code: Add real "From " MBOX separator line
3425 so the .eml file is really in mbox format (even though
3426 most programs do not really care). Patch from Alex Miller.
3428 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3429 The latter is generated from $received_to and is only set if the
3430 message has one envelope recipient. SA can use these headers,
3431 obviously out-of-the-box. Patch from Alex Miller.
3433 PH/08 The ${def test on a variable was returning false if the variable's
3434 value was "0", contrary to what the specification has always said!
3435 The result should be true unless the variable is empty.
3437 PH/09 The syntax error of a character other than { following "${if
3438 def:variable_name" (after optional whitespace) was not being diagnosed.
3439 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3440 accidental colon was present, for example, could give incorrect results.
3442 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3443 result is used (not including appendfile, where other changes are about
3446 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3447 This involved changing a lot of size variables from int to off_t. It
3448 should work with maildirs and everything.
3450 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3451 spamd dying while we are connected to it.
3453 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3456 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3457 function for BDB 4.3.
3459 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3461 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3462 the sources for PCRE have been re-arranged and more files are now
3465 PH/15 The code I had for printing potentially long long variables in PH/11
3466 above was not the best (it lost precision). The length of off_t variables
3467 is now inspected at build time, and an appropriate printing format (%ld
3468 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3469 to be "long long int" or "long int". This is needed for the internal
3470 formatting function string_vformat().
3472 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3473 the configuration file to be ":syslog", then the script "guesses" where
3474 the logs files are, rather than using the compiled in default. In our
3475 case the guess is not the same as the compiled default, so the script
3476 suddenly stopped working when I started to use syslog. The patch checks
3477 to see if log_file_path is "". If so, it attempts to read it from exim
3478 with no configuration file to get the compiled in version, before it
3479 falls back to the previous guessing code."
3481 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3482 implementing BATV in an Exim configuration. See NewStuff for the gory
3485 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3486 Makefile that are specific to HP-UX.
3488 PH/18 If the "use_postmaster" option was set for a recipient callout together
3489 with the "random" option, the postmaster address was used as the MAIL
3490 FROM address for the random test, but not for the subsequent recipient
3491 test. It is now used for both.
3493 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3494 patch removes a few documentation additions to RFC 3028, because the
3495 latest draft now contains them. It adds the new en;ascii-case comparator
3496 and a new error check for 8bit text in MIME parts. Comparator and
3497 require names are now matched exactly. I enabled the subaddress
3498 extension, but it is not well tested yet (read: it works for me)."
3500 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3501 rework some of the code of TK/09 above to avoid the hardwired use of
3502 "%lld" and "long long". Replaced the call to snprintf() with a call to
3505 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3506 records point to non-existent hosts", "retry timeout exceeded", and
3507 "retry time not reached for any host after a long failure period".
3509 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3510 experimental DomainKeys support:
3512 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3513 (2) On an error such as an illegally used "control", the wrong name for
3514 the control was given.
3516 These problems did NOT occur unless DomainKeys support was compiled.
3518 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3520 PH/24 Added ${if match_ip condition.
3522 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3523 databases so that it will be absolutely obvious if a crash occurs in the
3524 DB library. This is a regular occurrence (often caused by mis-matched
3527 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3528 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3529 picked up on a user's system that detects such things. There doesn't seem
3530 to be a gcc warning option for this - only an attribute that has to be
3531 put on the function's prototype. It seems that in Fedora Core 4 they have
3532 set this on a number of new functions. No doubt there will be more in due
3535 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3536 specify a transport. However, if an address that was verified by such a
3537 router was the subject of a callout, Exim crashed because it tried to
3538 read the rcpt_include_affixes from the non-existent transport. Now it
3539 just assumes that the setting of that option is false. This bug was
3540 introduced by 4.51/PH/31.
3542 PH/28 Changed -d+all to exclude +memory, because that information is very
3543 rarely of interest, but it makes the output a lot bigger. People tend to
3544 do -d+all out of habit.
3546 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3547 code in os-type was giving problems when libc.so lives in lib64, like on
3550 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3551 aren't the modern standard, and it seems that some systems' include files
3552 don't always have them. Exim was already checking for some of the newer
3553 ones like T_AAAA, and defining it itself. I've added checks for all the
3554 record types that Exim uses.
3556 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3557 not automatically generating a new one, as it is supposed to. This
3558 prevented TLS from working. If the file did exist, but contained invalid
3559 data, a new version was generated, as expected. It was only the case of a
3560 non-existent file that was broken.
3562 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3563 with a change in libdomainkeys > 0.64.
3565 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3566 from DNS. If the selector record carries the flag, it now has
3567 precedence over the domain-wide flag.
3569 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3571 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3572 the use of a port name that isn't defined in /etc/services) occurred, the
3573 message was deferred as in a normal delivery, and thus remained on the
3574 spool, instead of being failed because of the mua_wrapper setting. This
3575 is now fixed, and I tidied up some of the mua_wrapper messages at the
3578 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3579 the messages in a hash of arrays rather than using individual hashes.
3580 This is a bit cleaner and results in dramatic memory savings, albeit
3581 at a slight CPU cost.
3583 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3584 as requested by Marc Sherman.
3586 SC/10 Eximstats: added histograms for user specified patterns as requested
3589 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3591 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3592 fopen() in the content-scanning modules that did not already have it.
3598 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3599 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3601 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3603 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3605 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3606 to test Sieve filters that use "vacation".
3608 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3609 that changes the way the GnuTLS parameters are stored in the cache file.
3610 The new format can be generated externally. For backward compatibility,
3611 if the data in the cache doesn't make sense, Exim assumes it has read an
3612 old-format file, and it generates new data and writes a new file. This
3613 means that you can't go back to an older release without removing the
3616 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3617 work if there are any delivery delays because "one_time" forces the
3618 parent to be marked "delivered", so its unseen clone is never tried
3619 again. For this reason, Exim now forbids the simultaneous setting of
3622 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3623 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3624 entry if you want to know the details. The fix, however, neglected to
3625 consider the case when local delivery batching is involved. The test for
3626 "previously delivered" was not happening when checking to see if an
3627 address could be batched with a previous (undelivered) one; under
3628 certain circumstances this could lead to multiple deliveries to the same
3631 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3632 in its include files, and this causes problems building Exim.
3634 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3635 header_syntax) but Exim was just ignoring anything given after a slash.
3636 In particular, this caused confusion with an attempt to use "verify =
3637 reverse_host_lookup/defer_ok". An error is now given when options are
3638 supplied for verify items that do not have them. (Maybe reverse_host_
3639 lookup should have a defer_ok option, but that's a different point.)
3641 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3642 defined by RFC 821) to 2048, because there were problems with some AUTH
3643 commands, and RFC 1869 says the size should be increased for extended
3644 SMTP commands that take arguments.
3646 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3649 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3650 "unknown" error; now it says that the functionality isn't in the binary.
3652 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3653 an address' error message when a string expansion fails (syntax or
3654 whatever). Otherwise the password may appear in the log. Following change
3655 PH/42 below, there is no longer a chance of it appearing in a bounce
3658 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3660 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3661 of its addresses was ever tried. (Bugzilla bug #2.)
3663 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3664 the result incorrectly in the debug output. (It correctly added a newline
3665 to what was transported.)
3667 TF/01 Added $received_time.
3669 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3670 commented out examples of how to interface to a virus scanner and to
3671 SpamAssassin. Also added commented examples of av_scanner and
3672 spamd_address settings.
3674 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3675 and controls are allowed in which ACLs. There were a couple of minor
3676 errors. Some of the entries in the conditions table (which is a table of
3677 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3678 negation of where the condition IS allowed.
3680 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3682 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3683 header file does not have a version number, so I've had to invent a new
3684 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3685 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3686 radiusclient), but it was contributed by a Radius user.
3688 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3689 files or directories, for OpenSSL.
3691 PH/19 When an Exim process that is running as root has to create an Exim log
3692 file, it does so in a subprocess that runs as exim:exim so as to get the
3693 ownership right at creation (otherwise, other Exim processes might see
3694 the file with the wrong ownership). There was no test for failure of this
3695 fork() call, which would lead to the process getting stuck as it waited
3696 for a non-existent subprocess. Forks do occasionally fail when resources
3697 run out. I reviewed all the other calls to fork(); they all seem to check
3700 PH/20 When checking for unexpected SMTP input at connect time (before writing
3701 the banner), Exim was not dealing correctly with a non-positive return
3702 from the read() function. If the client had disconnected by this time,
3703 the result was a log entry for a synchronization error with an empty
3704 string after "input=" when read() returned zero. If read() returned -1
3705 (an event I could not check), uninitialized data bytes were printed.
3706 There were reports of junk text (parts of files, etc) appearing after
3709 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3711 PH/22 Added support for macro redefinition, and (re)definition in between
3712 driver and ACL definitions.
3714 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3715 forgetting to use the resulting value; it was using the unexpanded value.
3717 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3718 hadn't been configured. The fix is from Juergen Kreileder, who
3719 understands it better than I do:
3721 "Here's what I see happening with three configured cyrus_sasl
3722 authenticators configured (plain, login, cram-md5):
3724 On startup auth_cyrus_sasl_init() gets called for each of these.
3725 This means three calls to sasl_listmech() without a specified mech_list.
3726 => SASL tests which mechs of all available mechs actually work
3727 => three warnings about OTP not working
3728 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3730 With the patch, sasl_listmech() also gets called three times. But now
3731 SASL's mech_list option is set to the server_mech specified in the the
3732 authenticator. Or in other words, the answer from sasl_listmech()
3733 gets limited to just the mech you're testing for (which is different
3735 => the return list contains just 'plain' or 'login', 'cram-md5' or
3736 nothing depending on the value of ob->server_mech.
3738 I've just tested the patch: Authentication still works fine,
3739 unavailable mechs specified in the exim configuration are still
3740 caught, and the auth.log warnings about OTP are gone."
3742 PH/25 When debugging is enabled, the contents of the command line are added
3743 to the debugging output, even when log_selector=+arguments is not
3746 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3747 answer is "GNU", and only if the return is "GNU/something" is the answer
3750 PH/27 $acl_verify_message is now set immediately after the failure of a
3751 verification in an ACL, and so is available in subsequent modifiers. In
3752 particular, the message can be preserved by coding like this:
3754 warn !verify = sender
3755 set acl_m0 = $acl_verify_message
3757 Previously, $acl_verify_message was set only while expanding "message"
3758 and "log_message" when a very denied access.
3760 PH/28 Modified OS/os.c-Linux with
3762 -#ifndef OS_LOAD_AVERAGE
3763 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3765 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3766 nomenclature these days.)
3768 PH/29 Installed patch from the Sieve maintainer that adds the options
3769 sieve_useraddress and sieve_subaddress to the redirect router.
3771 PH/30 In these circumstances:
3772 . Two addresses routed to the same list of hosts;
3773 . First host does not offer TLS;
3774 . First host accepts first address;
3775 . First host gives temporary error to second address;
3776 . Second host offers TLS and a TLS session is established;
3777 . Second host accepts second address.
3778 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3779 and peerdn, if requested) that were in fact used only for the second
3782 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3783 attention to any local part prefix or suffix that was matched by the
3784 router that accepted the address. It now behaves in the same way as it
3785 does for delivery: the affixes are removed from the local part unless
3786 rcpt_include_affixes is set on the transport.
3788 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3789 timeout during the DATA phase of an incoming message.
3791 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3792 applied a patch sanctioned by the Sieve maintainer.
3794 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3795 the uid or gid is negative. A case of a negative gid caused this to be
3796 noticed. The fix allows for either to be negative.
3798 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3799 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3802 PH/36 The change PH/12 above was broken. Fixed it.
3804 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3805 the grounds that routing the same address twice would always produce the
3806 same answer. This might have been true once, but it is certainly no
3807 longer true now. Routing a child address may depend on the previous
3808 routing that produced that child. Some complicated redirection strategies
3809 went wrong when messages had multiple recipients, and made Exim's
3810 behaviour dependent on the order in which the addresses were given.
3812 I have moved the duplicate checking until after the routing is complete.
3813 Exim scans the addresses that are assigned to local and remote
3814 transports, and removes any duplicates. This means that more work will be
3815 done, as duplicates will always all be routed, but duplicates are
3816 presumably rare, so I don't expect this is of any significance.
3818 For deliveries to pipes, files, and autoreplies, the duplicate checking
3819 still happens during the routing process, since they are not going to be
3822 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3823 It corrects a timeout issue with spamd. This is Ian's comment: "The
3824 background is that sometimes spamd either never reads data from a
3825 connection it has accepted, or it never writes response data. The exiscan
3826 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3827 blindly assumes that writes won't block so it may never time out."
3829 PH/39 Allow G after quota size as well as K and M.
3831 PH/40 The value set for $authenticated_id in an authenticator may not contain
3832 binary zeroes or newlines because the value is written to log lines and
3833 to spool files. There was no check on this. Now the value is run through
3834 the string_printing() function so that such characters are converted to
3835 printable escape sequences.
3837 PH/41 $message_linecount is a new variable that contains the total number of
3838 lines in the message. Compare $body_linecount, which is the count for the
3841 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3842 bounce and delay warning messages, except in certain special cases, which
3845 (a) An SMTP error message from a remote host;
3846 (b) A message specified in a :fail: redirection;
3847 (c) A message specified in a "fail" command in a system filter;
3848 (d) A message specified in a FAIL return from the queryprogram router;
3849 (e) A message specified by the cannot_route_message router option.
3851 In these cases only, Exim does include the error details in bounce and
3852 warning messages. There are also a few cases where bland messages such
3853 as "unrouteable address" or "local delivery error" are given.
3855 PH/43 $value is now also set for the "else" part of a ${run expansion.
3857 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3858 being worked on, but at least Exim now implements the latest version to
3861 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3862 process to complete was treated as a delivery failure, a timeout while
3863 writing the message to the pipe was logged, but erroneously treated as a
3864 successful delivery. Such timeouts include transport filter timeouts. For
3865 consistency with the overall process timeout, these timeouts are now
3866 treated as errors, giving rise to delivery failures by default. However,
3867 there is now a new Boolean option for the pipe transport called
3868 timeout_defer, which, if set TRUE, converts the failures into defers for
3869 both kinds of timeout. A transport filter timeout is now identified in
3872 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3873 systems where "make" and "gmake" are different, calling "gmake" at top
3874 level broke things. I've arranged for the value of $(MAKE) to be passed
3875 from the Makefile to this script so that it can call the same version of
3879 A note about Exim versions 4.44 and 4.50
3880 ----------------------------------------
3882 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3883 changes of various kinds. As a consequence, a big documentation update was
3884 needed. This delayed the release for rather longer than seemed good, especially
3885 in the light of a couple of (minor) security issues. Therefore, the changes
3886 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3887 release. So 4.44 and 4.50 are in effect two different branches that both start
3890 I have left the 4.50 change log unchanged; it contains all the changes since
3891 4.43. The change log for 4.44 is below; many of its items are identical to
3892 those for 4.50. This seems to be the most sensible way to preserve the
3893 historical information.
3899 1. Minor wording change to the doc/README.SIEVE file.
3901 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3902 computation of the current number of files was incorrect.
3904 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3905 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3908 4. Give more explanation in the error message when the command for a transport
3909 filter fails to execute.
3911 5. There are several places where Exim runs a non-Exim command in a
3912 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3913 was being done only for the command run by the queryprogram router. It is
3914 now done for all such subprocesses. The other cases are: ${run, transport
3915 filters, and the commands run by the lmtp and pipe transports.
3917 6. Added CONFIGURE_GROUP build-time option.
3919 7. Some older OS have a limit of 256 on the maximum number of file
3920 descriptors. Exim was using setrlimit() to set 1000 as a large value
3921 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3922 systems. I've change it so that if it can't get 1000, it tries for 256.
3924 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3925 was an oversight, and furthermore, ever since the addition of extra
3926 controls (e.g. 4.43/32), the checks on when to allow different forms of
3927 "control" were broken. There should now be diagnostics for all cases when a
3928 control that does not make sense is encountered.
3930 9. Added the /retain_sender option to "control=submission".
3932 10. $recipients is now available in the predata ACL (oversight).
3934 11. Tidy the search cache before the fork to do a delivery from a message
3935 received from the command line. Otherwise the child will trigger a lookup
3936 failure and thereby defer the delivery if it tries to use (for example) a
3937 cached ldap connection that the parent has called unbind on.
3939 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3940 of $address_data from the recipient verification was clobbered by the
3941 sender verification.
3943 13. The value of address_data from a sender verification is now available in
3944 $sender_address_data in subsequent conditions in the ACL statement.
3946 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3948 15. Added a new option "connect=<time>" to callout options, to set a different
3951 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3952 was its contents. (It was OK if the option was not defined at all.)
3954 17. A "Completed" log line is now written for messages that are removed from
3955 the spool by the -Mrm option.
3957 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3958 information about exactly what failed.
3960 19. Added -dd to debug only the daemon process.
3962 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3963 handles timeouts, both on the server side and network timeouts. Renamed the
3964 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3966 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3967 It is now set to "smtps".
3969 22. $host_address is now set to the target address during the checking of
3970 ignore_target_hosts.
3972 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3973 being passed; this would have caused $sender_host_name to have been used if
3974 matching the list had actually called for a host name (not very likely,
3975 since this list is usually IP addresses). A host name is now passed as
3978 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3979 code that specifies a non-restarting handler (typically sigaction() in
3980 modern systems) in an attempt to fix a rare and obscure crash bug.
3982 25. Narrowed the window for a race in the daemon that could cause it to ignore
3983 SIGCHLD signals. This is not a major problem, because they are used only to
3984 wake it up if nothing else does.
3986 26. A malformed maildirsize file could cause Exim to calculate negative values
3987 for the mailbox size or file count. Odd effects could occur as a result.
3988 The maildirsize information is now recalculated if the size or filecount
3991 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3992 support for a long time. Removed HAVE_SYS_VFS_H.
3994 28. Installed the latest version of exipick from John Jetmore.
3996 29. In an address list, if the pattern was not a regular expression, an empty
3997 subject address (from a bounce message) matched only if the pattern was an
3998 empty string. Non-empty patterns were not even tested. This was the wrong
3999 because it is perfectly reasonable to use an empty address as part of a
4000 database query. An empty address is now tested by patterns that are
4001 lookups. However, all the other forms of pattern expect the subject to
4002 contain a local part and a domain, and therefore, for them, an empty
4003 address still always fails if the pattern is not itself empty.
4005 30. Exim went into a mad DNS loop when attempting to do a callout where the
4006 host was specified on an smtp transport, and looking it up yielded more
4007 than one IP address.
4009 31. Re-factored the code for checking spool and log partition space into a
4010 function that finds that data and another that does the check. The former
4011 is then used to implement four new variables: $spool_space, $log_space,
4012 $spool_inodes, and $log_inodes.
4014 32. The RFC2047 encoding function was originally intended for short strings
4015 such as real names; it was not keeping to the 75-character limit for
4016 encoded words that the RFC imposes. It now respects the limit, and
4017 generates multiple encoded words if necessary. To be on the safe side, I
4018 have increased the buffer size for the ${rfc2047: expansion operator from
4021 33. It is now permitted to omit both strings after an "if" condition; if the
4022 condition is true, the result is "true". As before, when the second string
4023 is omitted, a false condition yields an empty string. This makes it less
4024 cumbersome to write custom ACL and router conditions.
4026 34. Failure to deliver a bounce message always caused it to be frozen, even if
4027 there was an errors_to setting on the router. The errors_to setting is now
4030 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4031 canonical form (fully expanded) before being placed in
4032 $sender_host_address.
4034 36. The table in the code that translates DNS record types into text (T_A to
4035 "A" for instance) was missing entries for NS and CNAME. It is just possible
4036 that this could have caused confusion if both these types were looked up
4037 for the same domain, because the text type is used as part of Exim's
4038 per-process caching. But the chance of anyone hitting this buglet seems
4041 37. The dnsdb lookup has been extended in a number of ways.
4043 (1) There is a new type, "zns", which walks up the domain tree until it
4044 finds some nameserver records. It should be used with care.
4046 (2) There is a new type, "mxh", which is like "mx" except that it returns
4047 just the host names, not the priorities.
4049 (3) It is now possible to give a list of domains (or IP addresses) to be
4050 looked up. The behaviour when one of the lookups defers can be
4051 controlled by a keyword.
4053 (4) It is now possible to specify the separator character for use when
4054 multiple records are returned.
4056 38. The dnslists ACL condition has been extended: it is now possible to supply
4057 a list of IP addresses and/or domains to be looked up in a particular DNS
4060 39. Added log_selector=+queue_time_overall.
4062 40. When running the queue in the test harness, wait just a tad after forking a
4063 delivery process, to get repeatability of debugging output.
4065 41. Include certificate and key file names in error message when GnuTLS fails
4066 to set them up, because the GnuTLS error message doesn't include the name
4067 of the failing file when there is a problem reading it.
4069 42. Allow both -bf and -bF in the same test run.
4071 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4073 44. The "Exiscan patch" is now merged into the mainline Exim source.
4075 45. Sometimes the final signoff response after QUIT could fail to get
4076 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4077 < 0 before doing a fflush(). This bug looks as though it goes back to the
4078 introduction of TLS in release 3.20, but "sometimes" must have been rare
4079 because the tests only now provoked it.
4081 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4082 (this can affect the format of dates).
4084 47. exim_tidydb, when checking for the continued existence of a message for
4085 which it has found a message-specific retry record, was not finding
4086 messages that were in split spool directories. Consequently, it was
4087 deleting retry records that should have stayed in existence.
4089 48. Steve fixed some bugs in eximstats.
4091 49. The SPA authentication driver was not abandoning authentication and moving
4092 on to the next authenticator when an expansion was forced to fail,
4093 contradicting the general specification for all authenticators. Instead it
4094 was generating a temporary error. It now behaves as specified.
4096 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4097 (the order specifies the preference for clients). The order is now AES256,
4098 AES128, 3DES, ARCFOUR128.
4100 51. Small patch to Sieve code - explicitly set From: when generating an
4103 52. Exim crashed if a remote delivery caused a very long error message to be
4104 recorded - for instance if somebody sent an entire SpamAssassin report back
4105 as a large number of 550 error lines. This bug was coincidentally fixed by
4106 increasing the size of one of Exim's internal buffers (big_buffer) that
4107 happened as part of the Exiscan merge. However, to be on the safe side, I
4108 have made the code more robust (and fixed the comments that describe what
4111 53. Now that there can be additional text after "Completed" in log lines (if
4112 the queue_time_overall log selector is set), a one-byte patch to exigrep
4113 was needed to allow it to recognize "Completed" as not the last thing in
4116 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4117 patch that reportedly fixes this has been added. I am not expert enough to
4118 create a test for it. This is what the patch creator wrote:
4120 "I found a little strange behaviour of ldap code when working with
4121 Windows 2003 AD Domain, where users was placed in more than one
4122 Organization Units. When I tried to give exim partial DN, the exit code
4123 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4124 But simultaneously result of request was absolutely normal ldap result,
4125 so I produce this patch..."
4127 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4128 REFERENCE, so I have modified the code to exclude the patch when that macro
4131 55. Some experimental protocols are using DNS PTR records for new purposes. The
4132 keys for these records are domain names, not reversed IP addresses. The
4133 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4134 leaves it alone. Component reversal etc. now happens only for IP addresses.
4137 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4139 57. Double the size of the debug message buffer (to 2048) so that more of very
4140 long debug lines gets shown.
4142 58. The exicyclog utility now does better if the number of log files to keep
4143 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4145 59. Two changes related to the smtp_active_hostname option:
4147 (1) $smtp_active_hostname is now available as a variable.
4148 (2) The default for smtp_banner uses $smtp_active_hostname instead
4149 of $primary_hostname.
4151 60. The host_aton() function is supposed to be passed a string that is known
4152 to be a valid IP address. However, in the case of IPv6 addresses, it was
4153 not checking this. This is a hostage to fortune. Exim now panics and dies
4154 if the condition is not met. A case was found where this could be provoked
4155 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4156 components; fortuitously, this particular loophole had already been fixed
4157 by change 4.50/55 above.
4159 If there are any other similar loopholes, the new check in host_aton()
4160 itself should stop them being exploited. The report I received stated that
4161 data on the command line could provoke the exploit when Exim was running as
4162 exim, but did not say which command line option was involved. All I could
4163 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4164 running as the user.
4167 61. There was a buffer overflow vulnerability in the SPA authentication code
4168 (which came originally from the Samba project). I have added a test to the
4169 spa_base64_to_bits() function which I hope fixes it.
4172 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4173 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4175 63. The daemon start-up calls getloadavg() while still root for those OS that
4176 need the first call to be done as root, but it missed one case: when
4177 deliver_queue_load_max is set with deliver_drop_privilege. This is
4178 necessary for the benefit of the queue runner, because there is no re-exec
4179 when deliver_drop_privilege is set.
4181 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4182 This has been fixed.
4184 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4185 was in use, was not putting the data itself into the right store pool;
4186 consequently, it could be overwritten for a subsequent message in the same
4187 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4190 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4192 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4193 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4194 the function were treating the return as a boolean value, which happened to
4195 work because 0=false and not-0=true, but is not correct code.
4197 68. The host_aton() function was not handling scoped IPv6 addresses (those
4198 with, for example, "%eth0" on the end) correctly.
4200 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4201 negated items (that is, ~something) in unsigned ints. Some compilers
4202 apparently mutter when there is no cast.
4204 70. If an address verification called from an ACL failed, and did not produce a
4205 user-specific message (i.e. there was only a "system" message), nothing was
4206 put in $acl_verify_message. In this situation, it now puts the system
4209 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4210 session; change 4.31/43 added the unwanted input to the log line - except
4211 that it did not do this in the start of session case. It now does.
4213 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4214 This is wrong and can cause the other end to generate a synchronization
4215 error if it is another Exim or anything else that does the synchronization
4216 check. A QUIT command is no longer sent after a timeout.
4218 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4219 during host lookups.
4221 74. The defer_ok option of callout verification was not working if it was used
4222 when verifying addresses in header lines, that is, for this case:
4224 verify = header_sender/callout=defer_ok
4226 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4227 those file descriptors could be used for SMTP connections. If anything
4228 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4229 could be sent to the SMTP client, causing chaos. The daemon now opens
4230 stdin, stdout, and stderr to /dev/null when it puts itself into the
4233 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4234 log by default. The user can override this with suitable Perl magic.
4236 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4237 the log message when discard triggers, was not working for the DATA ACL or
4238 for the non-SMTP ACL.
4240 78. Error message wording change in sieve.c.
4242 79. If smtp_accept_max_per_host was set, the number of connections could be
4243 restricted to fewer than expected, because the daemon was trying to set up
4244 a new connection before checking whether the processes handling previous
4245 connections had finished. The check for completed processes is now done
4246 earlier. On busy systems, this bug wouldn't be noticed because something
4247 else would have woken the daemon, and it would have reaped the completed
4250 80. If a message was submitted locally by a user whose login name contained one
4251 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4252 It caused a spool format error. I have fixed the spool reading code. A
4253 related problem was that the "from" clause in the Received: line became
4254 illegal because of the space(s). It is now covered by ${quote_local_part.
4256 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4259 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4261 83. Patch from Sieve maintainer to fix unterminated string problem in
4262 "vacation" handling.
4264 84. Some minor changes to the Linux configuration files to help with other
4265 OS variants using glibc.
4267 85. One more patch for Sieve to update vacation handling to latest spec.
4270 ----------------------------------------------------
4271 See the note above about the 4.44 and 4.50 releases.
4272 ----------------------------------------------------
4278 1. Change 4.43/35 introduced a bug that caused file counts to be
4279 incorrectly computed when quota_filecount was set in an appendfile
4282 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4283 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4286 3. Give more explanation in the error message when the command for a transport
4287 filter fails to execute.
4289 4. There are several places where Exim runs a non-Exim command in a
4290 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4291 was being done only for the command run by the queryprogram router. It is
4292 now done for all such subprocesses. The other cases are: ${run, transport
4293 filters, and the commands run by the lmtp and pipe transports.
4295 5. Some older OS have a limit of 256 on the maximum number of file
4296 descriptors. Exim was using setrlimit() to set 1000 as a large value
4297 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4298 systems. I've change it so that if it can't get 1000, it tries for 256.
4300 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4301 was an oversight, and furthermore, ever since the addition of extra
4302 controls (e.g. 4.43/32), the checks on when to allow different forms of
4303 "control" were broken. There should now be diagnostics for all cases when a
4304 control that does not make sense is encountered.
4306 7. $recipients is now available in the predata ACL (oversight).
4308 8. Tidy the search cache before the fork to do a delivery from a message
4309 received from the command line. Otherwise the child will trigger a lookup
4310 failure and thereby defer the delivery if it tries to use (for example) a
4311 cached ldap connection that the parent has called unbind on.
4313 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4314 of $address_data from the recipient verification was clobbered by the
4315 sender verification.
4317 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4318 was its contents. (It was OK if the option was not defined at all.)
4320 11. A "Completed" log line is now written for messages that are removed from
4321 the spool by the -Mrm option.
4323 12. $host_address is now set to the target address during the checking of
4324 ignore_target_hosts.
4326 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4327 being passed; this would have caused $sender_host_name to have been used if
4328 matching the list had actually called for a host name (not very likely,
4329 since this list is usually IP addresses). A host name is now passed as
4332 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4333 code that specifies a non-restarting handler (typically sigaction() in
4334 modern systems) in an attempt to fix a rare and obscure crash bug.
4336 15. Narrowed the window for a race in the daemon that could cause it to ignore
4337 SIGCHLD signals. This is not a major problem, because they are used only to
4338 wake it up if nothing else does.
4340 16. A malformed maildirsize file could cause Exim to calculate negative values
4341 for the mailbox size or file count. Odd effects could occur as a result.
4342 The maildirsize information is now recalculated if the size or filecount
4345 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4346 support for a long time. Removed HAVE_SYS_VFS_H.
4348 18. Updated exipick to current release from John Jetmore.
4350 19. Allow an empty sender to be matched against a lookup in an address list.
4351 Previously the only cases considered were a regular expression, or an
4354 20. Exim went into a mad DNS lookup loop when doing a callout where the
4355 host was specified on the transport, if the DNS lookup yielded more than
4358 21. The RFC2047 encoding function was originally intended for short strings
4359 such as real names; it was not keeping to the 75-character limit for
4360 encoded words that the RFC imposes. It now respects the limit, and
4361 generates multiple encoded words if necessary. To be on the safe side, I
4362 have increased the buffer size for the ${rfc2047: expansion operator from
4365 22. Failure to deliver a bounce message always caused it to be frozen, even if
4366 there was an errors_to setting on the router. The errors_to setting is now
4369 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4370 canonical form (fully expanded) before being placed in
4371 $sender_host_address.
4373 24. Updated eximstats to version 1.33
4375 25. Include certificate and key file names in error message when GnuTLS fails
4376 to set them up, because the GnuTLS error message doesn't include the name
4377 of the failing file when there is a problem reading it.
4379 26. Expand error message when OpenSSL has problems setting up cert/key files.
4382 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4383 (this can affect the format of dates).
4385 28. exim_tidydb, when checking for the continued existence of a message for
4386 which it has found a message-specific retry record, was not finding
4387 messages that were in split spool directories. Consequently, it was
4388 deleting retry records that should have stayed in existence.
4390 29. eximstats updated to version 1.35
4391 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4392 1.35 - bugfix such that pie charts by volume are generated correctly
4394 30. The SPA authentication driver was not abandoning authentication and moving
4395 on to the next authenticator when an expansion was forced to fail,
4396 contradicting the general specification for all authenticators. Instead it
4397 was generating a temporary error. It now behaves as specified.
4399 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4400 (the order specifies the preference for clients). The order is now AES256,
4401 AES128, 3DES, ARCFOUR128.
4403 31. Small patch to Sieve code - explicitly set From: when generating an
4406 32. Exim crashed if a remote delivery caused a very long error message to be
4407 recorded - for instance if somebody sent an entire SpamAssassin report back
4408 as a large number of 550 error lines. This bug was coincidentally fixed by
4409 increasing the size of one of Exim's internal buffers (big_buffer) that
4410 happened as part of the Exiscan merge. However, to be on the safe side, I
4411 have made the code more robust (and fixed the comments that describe what
4414 33. Some experimental protocols are using DNS PTR records for new purposes. The
4415 keys for these records are domain names, not reversed IP addresses. The
4416 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4417 leaves it alone. Component reversal etc. now happens only for IP addresses.
4420 34. The host_aton() function is supposed to be passed a string that is known
4421 to be a valid IP address. However, in the case of IPv6 addresses, it was
4422 not checking this. This is a hostage to fortune. Exim now panics and dies
4423 if the condition is not met. A case was found where this could be provoked
4424 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4425 components; fortuitously, this particular loophole had already been fixed
4426 by change 4.50/55 or 4.44/33 above.
4428 If there are any other similar loopholes, the new check in host_aton()
4429 itself should stop them being exploited. The report I received stated that
4430 data on the command line could provoke the exploit when Exim was running as
4431 exim, but did not say which command line option was involved. All I could
4432 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4433 running as the user.
4436 35. There was a buffer overflow vulnerability in the SPA authentication code
4437 (which came originally from the Samba project). I have added a test to the
4438 spa_base64_to_bits() function which I hope fixes it.
4441 36. The daemon start-up calls getloadavg() while still root for those OS that
4442 need the first call to be done as root, but it missed one case: when
4443 deliver_queue_load_max is set with deliver_drop_privilege. This is
4444 necessary for the benefit of the queue runner, because there is no re-exec
4445 when deliver_drop_privilege is set.
4447 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4448 was in use, was not putting the data itself into the right store pool;
4449 consequently, it could be overwritten for a subsequent message in the same
4450 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4453 38. Sometimes the final signoff response after QUIT could fail to get
4454 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4455 < 0 before doing a fflush(). This bug looks as though it goes back to the
4456 introduction of TLS in release 3.20, but "sometimes" must have been rare
4457 because the tests only now provoked it.
4463 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4464 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4465 Now it returns an integer. A number of calls were still expecting a T/F
4466 return. Fortuitously, in all cases, the tests worked in OK situations,
4467 which is the norm. However, things would have gone wrong on any write
4468 failures on the smtp file descriptor. This function is used when sending
4469 messages over SMTP and also when doing verify callouts.
4471 2. When Exim is called to do synchronous delivery of a locally submitted
4472 message (the -odf or -odi options), it no longer closes stderr before doing
4475 3. Implemented the mua_wrapper option.
4477 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4479 5. Implemented the functions header_remove(), header_testname(),
4480 header_add_at_position(), and receive_remove_recipient(), and exported them
4483 6. If an ACL "warn" statement specified the addition of headers, Exim already
4484 inserted X-ACL-Warn: at the start if there was no header name. However, it
4485 was not making this test for the second and subsequent header lines if
4486 there were newlines in the string. This meant that an invalid header could
4487 be inserted if Exim was badly configured.
4489 7. Allow an ACL "warn" statement to add header lines at the start or after all
4490 the Received: headers, as well as at the end.
4492 8. Added the rcpt_4xx retry error code.
4494 9. Added postmaster_mailfrom=xxx to callout verification option.
4496 10. Added mailfrom=xxxx to the callout verification option, for verify=
4499 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4500 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4502 12. Inserted some casts to stop certain compilers warning when using pointer
4503 differences as field lengths or precisions in printf-type calls (mostly
4504 affecting debugging statements).
4506 13. Added optional readline() support for -be (dynamically loaded).
4508 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4509 same clock tick as a message's arrival, so that its received time was the
4510 same as the "first fail" time on the retry record, and that message
4511 remained on the queue past the ultimate address timeout, every queue runner
4512 would try a delivery (because it was past the ultimate address timeout) but
4513 after another failure, the ultimate address timeout, which should have then
4514 bounced the address, did not kick in. This was a "< instead of <=" error;
4515 in most cases the first failure would have been in the next clock tick
4516 after the received time, and all would be well.
4518 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4519 being recognized when the domain list was tested by the match_domain
4520 condition in an expansion string.
4522 16. Added the ${str2b64: operator.
4524 17. Exim was always calling setrlimit() to set a large limit for the number of
4525 processes, without checking whether the existing limit was already
4526 adequate. (It did check for the limit on file descriptors.) Furthermore,
4527 errors from getrlimit() and setrlimit() were being ignored. Now they are
4528 logged to the main and panic logs, but Exim does carry on, to try to do its
4529 job under whatever limits there are.
4531 18. Imported PCRE 5.0.
4533 19. Trivial typo in log message " temporarily refused connection" (the leading
4536 20. If the log selector return_path_on_delivery was set and an address was
4537 redirected to /dev/null, the delivery process crashed because it assumed
4538 that a return path would always be set for a "successful" delivery. In this
4539 case, the whole delivery is bypassed as an optimization, and therefore no
4542 21. Internal re-arrangement: the function for sending a challenge and reading
4543 a response while authentication was assuming a zero-terminated challenge
4544 string. It's now changed to take a pointer and a length, to allow for
4545 binary data in such strings.
4547 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4549 23. Exim was not respecting finduser_retries when seeking the login of the
4550 uid under which it was called; it was always trying 10 times. (The default
4551 setting of finduser_retries is zero.) Also, it was sleeping after the final
4552 failure, which is pointless.
4554 24. Implemented tls_on_connect_ports.
4556 25. Implemented acl_smtp_predata.
4558 26. If the domain in control=submission is set empty, Exim assumes that the
4559 authenticated id is a complete email address when it generates From: or
4560 Sender: header lines.
4562 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4563 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4564 chown and chgrp in /bin and hostname in /usr/bin.
4566 28. Exim was keeping the "process log" file open after each use, just as it
4567 does for the main log. This opens the possibility of it remaining open for
4568 long periods when the USR1 signal hits a daemon. Occasional processlog
4569 errors were reported, that could have been caused by this. Anyway, it seems
4570 much more sensible not to leave this file open at all, so that is what now
4573 29. The long-running daemon process does not normally write to the log once it
4574 has entered its main loop, and it closes the log before doing so. This is
4575 so that log files can straightforwardly be renamed and moved. However,
4576 there are a couple of unusual error situations where the daemon does write
4577 log entries, and I had neglected to close the log afterwards.
4579 30. The text of an SMTP error response that was received during a remote
4580 delivery was being truncated at 512 bytes. This is too short for some of
4581 the long messages that one sometimes sees. I've increased the limit to
4584 31. It is now possible to make retry rules that apply only when a message has a
4585 specific sender, in particular, an empty sender.
4587 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4588 it possible to be selective about when SMTP synchronization is enforced.
4590 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4592 32. Implemented hosts_connection_nolog.
4594 33. Added an ACL for QUIT.
4596 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4599 35. Added mailbox_size and mailbox_filecount to appendfile.
4601 36. Added control = no_multiline_responses to ACLs.
4603 37. There was a bug in the logic of the code that waits for the clock to tick
4604 in the case where the clock went backwards by a substantial amount such
4605 that the microsecond fraction of "now" was more than the microsecond
4606 fraction of "then" (but the whole seconds number was less).
4608 38. Added support for the libradius Radius client library this is found on
4609 FreeBSD (previously only the radiusclient library was supported).
4615 1. When certain lookups returned multiple values in the form name=value, the
4616 quoting of the values was not always being done properly. Specifically:
4617 (a) If the value started with a double quote, but contained no whitespace,
4619 (b) If the value contained whitespace other than a space character (i.e.
4620 tabs or newlines or carriage returns) it was not quoted.
4621 This fix has been applied to the mysql and pgsql lookups by writing a
4622 separate quoting function and calling it from the lookup code. The fix
4623 should probably also be applied to nisplus, ibase and oracle lookups, but
4624 since I cannot test any of those, I have not disturbed their existing code.
4626 2. A hit in the callout cache for a specific address caused a log line with no
4627 reason for rejecting RCPT. Now it says "Previous (cached) callout
4628 verification failure".
4630 3. There was an off-by-one bug in the queryprogram router. An over-long
4631 return line was truncated at 256 instead of 255 characters, thereby
4632 overflowing its buffer with the terminating zero. As well as fixing this, I
4633 have increased the buffer size to 1024 (and made a note to document this).
4635 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4636 when Exim is waiting for an SMTP response from a remote server, Exim
4637 restarts its select() call on the socket, thereby resetting its timeout.
4638 This is not a problem when such interrupts are rare. Somebody set up a cron
4639 job to run exiwhat every 2 minutes, which is less than the normal select()
4640 timeout (5 or 10 minutes). This meant that the select() timeout never
4641 kicked in because it was always reset. I have fixed this by comparing the
4642 time when an interrupt arrives with the time at the start of the first call
4643 to select(). If more time than the timeout has elapsed, the interrupt is
4644 treated as a timeout.
4646 5. Some internal re-factoring in preparation for the addition of Sieve
4647 extensions (by MH). In particular, the "personal" test is moved to a
4648 separate function, and given an option for scanning Cc: and Bcc: (which is
4649 not set for Exim filters).
4651 6. When Exim created an email address using the login of the caller as the
4652 local part (e.g. when creating a From: or Sender: header line), it was not
4653 quoting the local part when it contained special characters such as @.
4655 7. Installed new OpenBSD configuration files.
4657 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4658 try to make them clearer.
4660 9. Callout options, other than the timeout value, were being ignored when
4661 verifying sender addresses in header lines. For example, when using
4663 verify = header_sender/callout=no_cache
4665 the cache was (incorrectly) being used.
4667 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4668 only the Cygwin environment.
4670 11. When return_path_on_delivery was set as a log selector, if different remote
4671 addresses in the same message used different return paths and parallel
4672 remote delivery occurred, the wrong values would sometimes be logged.
4673 (Whenever a remote delivery process finished, the return path value from
4674 the most recently started remote delivery process was logged.)
4676 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4677 lines when AUTH and/or TLS are in use. This is the "received protocol"
4678 field. Exim used to use "asmtp" for authenticated SMTP, without any
4679 indication (in the protocol name) for TLS use. Now it follows the RFC and
4680 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4681 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4682 names appear in log lines as well as in Received: header lines.
4684 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4685 extensions, and comparison tests, and to fix some bugs.
4687 14. Changes to the "personal" filter test:
4689 (1) The test was buggy in that it was just doing the equivalent of
4690 "contains" tests on header lines. For example, if a user's address was
4691 anne@some.where, the "personal" test would incorrectly be true for
4693 To: susanne@some.where
4695 This test is now done by extracting each address from the header in turn,
4696 and checking the entire address. Other tests that are part of "personal"
4697 are now done using regular expressions (for example, to check local parts
4698 of addresses in From: header lines).
4700 (2) The list of non-personal local parts in From: addresses has been
4701 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4702 taken from the Sieve specification recommendations.
4704 (3) If the message contains any header line starting with "List-" it is
4705 treated as non-personal.
4707 (4) The test for "circular" in the Subject: header line has been removed
4708 because it now seems ill-conceived.
4710 15. Minor typos in src/EDITME comments corrected.
4712 16. Installed latest exipick from John Jetmore.
4714 17. If headers_add on a router specified a text string that was too long for
4715 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4716 of string_sprintf() is now avoided.
4718 18. $message_body_size was not set (it was always zero) when running the DATA
4719 ACL and the local_scan() function.
4721 19. For the "mail" command in an Exim filter, no default was being set for
4722 the once_repeat time, causing a random time value to be used if "once" was
4723 specified. (If the value happened to be <= 0, no repeat happened.) The
4724 default is now 0s, meaning "never repeat". The "vacation" command was OK
4725 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4726 (I found it when inspecting the code).
4728 20. There is now an overall timeout for performing a callout verification. It
4729 defaults to 4 times the callout timeout, which applies to individual SMTP
4730 commands during the callout. The overall timeout applies when there is more
4731 than one host that can be tried. The timeout is checked before trying the
4732 next host. This prevents very long delays if there are a large number of
4733 hosts and all are timing out (e.g. when the network connections are timing
4734 out). The value of the overall timeout can be changed by specifying an
4735 additional sub-option for "callout", called "maxwait". For example:
4737 verify = sender/callout=5s,maxwait=20s
4739 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4740 to the end before writing, but this should make it even safer).
4742 22. Exim was forgetting that it had advertised PIPELINING for the second and
4743 subsequent messages on an SMTP connection. It was also not resetting its
4744 memory on STARTTLS and an internal HELO.
4746 23. When Exim logs an SMTP synchronization error within a session, it now
4747 records whether PIPELINING has been advertised or not.
4749 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4750 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4751 rather than long int.
4753 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4755 26. Added the never_mail option to autoreply.
4761 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4762 crash if the getsockname() call failed; this can happen if a connection is
4763 closed very soon after it is established. The problem was simply in the
4764 order in which certain operations were done, causing Exim to try to write
4765 to the SMTP stream before it had set up the file descriptor. The bug has
4766 been fixed by making things happen in the correct order.
4772 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4773 before the connection was closed, thus losing the rejection response.
4775 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4776 some early Solaris releases, but causes trouble in current releases where
4777 socklen_t is defined.
4779 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4782 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4785 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4786 that releasing the top of it at the end releases what was used for sub-
4787 expansions (unless the block got too big). However, discard this block if
4788 the first thing is a variable or header, so that we can use its block when
4789 it is dynamic (useful for very large $message_headers, for example).
4791 6. Lookups now cache *every* query, not just the most recent. A new, separate
4792 store pool is used for this. It can be recovered when all lookup caches are
4793 flushed. Lookups now release memory at the end of their result strings.
4794 This has involved some general refactoring of the lookup sources.
4796 7. Some code has been added to the store_xxx() functions to reduce the amount
4797 of flapping under certain conditions.
4799 8. log_incoming_interface used to affect only the <= reception log lines. Now
4800 it causes the local interface and port to be added to several more SMTP log
4801 lines, for example "SMTP connection from", and rejection lines.
4803 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4805 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4807 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4808 could be overwritten at the end of the current message (or the start of a
4809 new message if it was set in a HELO ACL). The value is now preserved for
4810 the duration of the SMTP connection.
4812 12. If a transport had a headers_rewrite setting, and a matching header line
4813 contained an unqualified address, that address was qualified, even if it
4814 did not match any rewriting rules. The underlying bug was that the values
4815 of the flags that permit the existence of unqualified sender and recipient
4816 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4817 non-local messages, and by -bnq for local messages) were not being
4818 preserved with the message after it was received.
4820 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4821 "next input=" as part of the text comprising the host identity instead of
4822 the correct text. The code was using the same buffer for two different
4823 strings. However, depending on which order the printing function evaluated
4824 its arguments, the bug did not always show up. Under Linux, for example, my
4825 test suite worked just fine.
4827 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4828 doesn't work with some older versions of Perl. It has been changed to "my",
4829 which in any case is probably the better facility to use.
4831 15. A really picky compiler found some instances of statements for creating
4832 error messages that either had too many or two few arguments for the format
4835 16. The size of the buffer for calls to the DNS resolver has been increased
4836 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4837 for addresses that have a lot of PTR records. This alleviates a problem; it
4838 does not fully solve it.
4840 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4841 buffer now truncates the list and logs the incident, which is the same
4842 action as happens when Exim is looking up a host name and its aliases.
4843 Previously in this situation something unpredictable would happen;
4844 sometimes it was "internal error: store_reset failed".
4846 18. If a server dropped the connection unexpectedly when an Exim client was
4847 using GnuTLS and trying to read a response, the client delivery process
4848 crashed while trying to generate an error log message.
4850 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4851 string, for example:
4853 warn message = H1: something\nH2: something
4855 the text was added as a single header line from Exim's point of view
4856 though it ended up OK in the delivered message. However, searching for the
4857 second and subsequent header lines using $h_h2: did not work. This has been
4858 fixed. Similarly, if a system filter added multiple headers in this way,
4859 the routers could not see them.
4861 20. Expanded the error message when iplsearch is called with an invalid key to
4862 suggest using net-iplsearch in a host list.
4864 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4865 ACLs are no longer actually imposed (and a message to that effect is
4868 22. If a "gecos" field in a passwd entry contained escaped characters, in
4869 particular, if it contained a \" sequence, Exim got it wrong when building
4870 a From: or a Sender: header from that name. A second bug also caused
4871 incorrect handling when an unquoted " was present following a character
4872 that needed quoting.
4874 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4875 was not being matched caselessly.
4877 24. Arranged for all hyphens in the exim.8 source to be escaped with
4880 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4881 properly. Recipient callout cache records were still being keyed to include
4882 the sender, even when use_sender was set false. This led to far more
4883 callouts that were necessary. The sender is no longer included in the key
4884 when use_sender is false.
4886 26. Added "control = submission" modifier to ACLs.
4888 27. Added the ${base62d: operator to decode base 62 numbers.
4890 28. dnsdb lookups can now access SRV records.
4892 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4893 the configuration file.
4895 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4896 (-v) mode. This makes the output for a verbose queue run more intelligible.
4898 31. Added a use_postmaster feature to recipient callouts.
4900 32. Added the $body_zerocount variable, containing the number of binary zero
4901 bytes in the message body.
4903 33. The time of last modification of the "new" subdirectory is now used as the
4904 "mailbox time last read" when there is a quota error for a maildir
4907 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4909 35. Added +ignore_unknown as a special item in host lists.
4911 36. Code for decoding IPv6 addresses in host lists is now included, even if
4912 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4913 address was recognized as an IP address, but was then not correctly decoded
4914 into binary, causing unexpected and incorrect effects when compared with
4921 1. Very minor rewording of debugging text in manualroute to say "list of
4922 hosts" instead of "hostlist".
4924 2. If verify=header_syntax was set, and a header line with an unqualified
4925 address (no domain) and a large number of spaces between the end of the
4926 name and the colon was received, the reception process suffered a buffer
4927 overflow, and (when I tested it) crashed. This was caused by some obsolete
4928 code that should have been removed. The fix is to remove it!
4930 3. When running in the test harness, delay a bit after writing a bounce
4931 message to get a bit more predictability in the log output.
4933 4. Added a call to search_tidyup() just before forking a reception process. In
4934 theory, someone could use a lookup in the expansion of smtp_accept_max_
4935 per_host which, without the tidyup, could leave open a database connection.
4937 5. Added the variables $recipient_data and $sender_data which get set from a
4938 lookup success in an ACL "recipients" or "senders" condition, or a router
4939 "senders" option, similar to $domain_data and $local_part_data.
4941 6. Moved the writing of debug_print from before to after the "senders" test
4944 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4945 problems for message scanning, either using a data ACL, or using
4946 local_scan() because the Received: header was not generated till after they
4947 were called (in order to set the time as the time of reception completion).
4948 I have revised the way this works. The header is now generated after the
4949 body is received, but before the ACL or local_scan() are called. After they
4950 are run, the timestamp in the header is updated.
4956 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4957 before starting a queue runner without re-exec. This happened only when
4958 deliver_drop_privilege was set or when the Exim user was set to root. The
4959 effect of the bug was that timeouts during subsequent deliveries caused
4960 crashes instead of being properly handled. The handler is now left at its
4961 default (and expected) setting.
4963 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4964 message, again when deliver_drop_privilege is set or Exim is run as root.
4965 The bug described in (1) was not present in this case, but the tidying up
4966 of the other signals was missing. I have made the two cases consistent.
4968 3. The ignore_target_hosts setting on a manualroute router was being ignored
4969 for hosts that were looked up using the /MX notation.
4971 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4974 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4975 operated on the sender address. After changing the $sender_address to <>
4976 for the sender address verify, Exim was re-instated it as the original
4977 (before rewriting) address, but remembering that it had rewritten it, so it
4978 wasn't rewriting it again. This bug also had the effect of breaking the
4979 sender address verification caching when the sender address was rewritten.
4981 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4982 This has been changed so that if the ip literal address matches
4983 ignore_target_hosts, the router declines.
4985 7. Added expansion conditions match_domain, match_address, and match_local_
4986 part (NOT match_host).
4988 8. The placeholder for the Received: header didn't have a length field set.
4990 9. Added code to Exim itself and to exim_lock to test for a specific race
4991 condition that could lead to file corruption when using MBX delivery. The
4992 issue is with the lockfile that is created in /tmp. If this file is removed
4993 after a process has opened it but before that process has acquired a lock,
4994 there is the potential for a second process to recreate the file and also
4995 acquire a lock. This could lead to two Exim processes writing to the file
4996 at the same time. The added code performs the same test as UW imapd; it
4997 checks after acquiring the lock that its file descriptor still refers to
4998 the same named file.
5000 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5001 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5004 11. Added the smtp_active_hostname option. If used, this will typically be made
5005 to depend on the incoming interface address. Because $interface_address is
5006 not set up until the daemon has forked a reception process, error responses
5007 that can happen earlier (such as "too many connections") no longer contain
5010 12. If an expansion in a condition on a "warn" statement fails because a lookup
5011 defers, the "warn" statement is abandoned, and the next ACL statement is
5012 processed. Previously this caused the whole ACL to be aborted.
5014 13. Added the iplsearch lookup type.
5016 14. Added ident_timeout as a log selector.
5018 15. Added tls_certificate_verified as a log selector.
5020 16. Added a global option tls_require_ciphers (compare the smtp transport
5021 option of the same name). This controls incoming TLS connections.
5023 17. I finally figured out how to make tls_require_ciphers do a similar thing
5024 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5025 before starting the TLS session.
5027 18. Tabs are now shown as \t in -bP output.
5029 19. If the log selector return_path_on_delivery was set, Exim crashed when
5030 bouncing a message because it had too many Received: header lines.
5032 20. If two routers both had headers_remove settings, and the first one included
5033 a superfluous trailing colon, the final name in the first list and the
5034 first name in the second list were incorrectly joined into one item (with a
5035 colon in the middle).
5041 1. Added -C and -D options to the exinext utility, mainly to make it easier
5042 to include in the automated testing, but these could be helpful when
5043 multiple configurations are in use.
5045 2. The exinext utility was not formatting the output nicely when there was
5046 an alternate port involved in the retry record key, nor when there was a
5047 message id as well (for retries that were specific to a specific message
5048 and a specific host). It was also confused by IPv6 addresses, because of
5049 the additional colons they contain. I have fixed the IPv4 problem, and
5050 patched it up to do a reasonable job for IPv6.
5052 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5053 delivery, the log line now contains "pipelined" if PIPELINING was used.
5055 4. An SMTP transport process used to panic and die if the bind() call to set
5056 an explicit outgoing interface failed. This has been changed; it is now
5057 treated in the same way as a connect() failure.
5059 5. A reference to $sender_host_name in the part of a conditional expansion
5060 that was being skipped was still causing a DNS lookup. This no longer
5063 6. The def: expansion condition was not recognizing references to header lines
5064 that used bh_ and bheader_.
5066 7. Added the _cache feature to named lists.
5068 8. The code for checking quota_filecount in the appendfile transport was
5069 allowing one more file than it should have been.
5071 9. For compatibility with Sendmail, the command line option
5079 and sets the incoming protocol and host name (for trusted callers). The
5080 host name and its colon can be omitted when only the protocol is to be set.
5081 Note the Exim already has two private options, -pd and -ps, that refer to
5082 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5083 "s", but I don't think that's a major issue.
5085 10. A number of refactoring changes to the code, none of which should affect
5088 (a) The number of logging options was getting close to filling up the
5089 32-bit word that was used as a bit map. I have split them into two classes:
5090 those that are passed in the argument to log_write(), and those that are
5091 only ever tested independently outside of that function. These are now in
5092 separate 32-bit words, so there is plenty of room for expansion again.
5093 There is no change in the user interface or the logging behaviour.
5095 (b) When building, for example, log lines, the code previously used a
5096 macro that called string_cat() twice, in order to add two strings. This is
5097 not really sufficiently general. Furthermore, there was one instance where
5098 it was actually wrong because one of the argument was used twice, and in
5099 one call a function was used. (As it happened, calling the function twice
5100 did not affect the overall behaviour.) The macro has been replaced by a
5101 function that can join an arbitrary number of extra strings onto a growing
5104 (c) The code for expansion conditions now uses a table and a binary chop
5105 instead of a serial search (which was left over from when there were very
5106 few conditions). Also, it now recognizes conditions like "pam" even when
5107 the relevant support is not compiled in: a suitably worded error message is
5108 given if an attempt is made to use such a condition.
5110 11. Added ${time_interval:xxxxx}.
5112 12. A bug was causing one of the ddress fields not to be passed back correctly
5113 from remote delivery subprocesses. The field in question was not being
5114 subsequently used, so this caused to problems in practice.
5116 13. Added new log selectors queue_time and deliver_time.
5118 14. Might have fixed a bug in maildirsizefile handling that threw up
5119 "unexpected character" debug warnings, and recalculated the data
5120 unnecessarily. In any case, I expanded the warning message to give more
5123 15. Added the message "Restricted characters in address" to the statements in
5124 the default ACL that block characters like @ and % in local parts.
5126 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5127 Three changes have been made:
5129 (a) There was a serious bug; a negative response to MAIL caused the whole
5130 recipient domain to be cached as invalid, thereby blocking all messages
5131 to all local parts at the same domain, from all senders. This bug has
5132 been fixed. The domain is no longer cached after a negative response to
5133 MAIL if the sender used is not empty.
5135 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5138 (c) A new callout option, "use_sender" has been added for people who want
5139 the modified behaviour.
5145 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5148 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5149 indeed breaks things for older releases.
5151 3. Added additional logging to the case where there is a problem reading data
5152 from a filter that is running in a subprocess using a pipe, in order to
5153 try to track down a specific problem.
5155 4. Testing facility fudge: when running in the test harness and attempting
5156 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5157 getting "No route to host". Convert this to a timeout.
5159 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5162 6. Some OS don't have socklen_t but use size_t instead. This affects the
5163 fifth argument of getsockopt() amongst other things. This is now
5164 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5165 can be set for individual OS. I have set it for SunOS5, OSF1, and
5166 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5167 some earlier ones do not.
5169 7. Change 4.30/15 was not doing the test caselessly.
5171 8. The standard form for an IPv6 address literal was being rejected by address
5172 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5173 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5174 this, as well as the form without the "IPv6" on the front (but only when
5175 address literals are enabled, of course).
5177 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5179 10. Exim crashed if a message with an empty sender address specified by -f
5180 encountered a router with an errors_to setting. This could be provoked only
5181 by a command such as
5185 where an empty string was supplied; "<>" did not hit this bug.
5187 11. Installed PCRE release 4.5.
5189 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5190 remained set. It is now erased.
5192 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5193 times from message ids (which are base 36 rather than the normal 62).
5195 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5196 were being counted as actual protocol errors, and logged if the log
5197 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5198 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5199 and DATA following a set of rejected RCPTs do not count as protocol errors.
5200 In other words, Exim assumes they were pipelined, though this may not
5201 actually be the case. Of course, in all cases the client gets an
5202 appropriate error code.
5204 15. If a lookup fails in an ACL condition, a message about the failure may
5205 be available; it is used if testing the ACL cannot continue, because most
5206 such messages specify what the cause of the deferral is. However, some
5207 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5208 that caused an old message to be retained and used if a later statement
5209 caused a defer, replacing the real cause of the deferral.
5211 16. If an IP address had so many PTR records that the DNS lookup buffer
5212 was not large enough to hold them, Exim could crash while trying to process
5213 the truncated data. It now detects and logs this case.
5215 17. Further to 4.21/58, another change has been made: if (and only if) the
5216 first line of a message (the first header line) ends with CRLF, a bare LF
5217 in a subsequent header line has a space inserted after it, so as not to
5218 terminate the header.
5220 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5221 unnecessarily, used atoi() instead of strtol(), and didn't check the
5222 termination when getting file sizes from file names by regex.
5224 19. Completely re-implemented the support for maildirsize files, in the light
5225 of a number of problems with the previous contributed implementation
5226 (4.30/29). In particular:
5228 . If the quota is zero, the maildirsize file is maintained, but no quota is
5231 . If the maildir directory does not exist, it is created before any attempt
5232 to write a maildirsize file.
5234 . The quota value in the file is just a cache; if the quota is changed in
5235 the transport, the new value overrides.
5237 . A regular expression is available for excluding directories from the
5240 20. The autoreply transport checks the characters in options that define the
5241 message's headers; it allows continued headers, but it was checking with
5242 isspace() after an embedded newline instead of explicitly looking for a
5245 21. If all the "regular" hosts to which an address was routed had passed their
5246 expiry times, and had not reached their retry times, the address was
5247 bounced, even if fallback hosts were defined. Now Exim should go on to try
5250 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5251 equivalent code in the SMTP transport. Some hosts send humungous responses
5252 to HELO/EHLO, more than 1024 it seems.
5254 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5255 gives compiler warnings in some environments. I've now done it "properly",
5258 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5259 (because of problems with the built-in one) was declared to return uschar *
5260 instead of char *, causing compiler failure.
5262 25. Fixed a file descriptor leak when processing alias/forward files.
5264 26. Fixed a minor format string issue in dbfn.c.
5266 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5268 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5269 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5272 29. When a sender address is verified, it is cached, to save repeating the test
5273 when there is more than one recipient in a message. However, when the
5274 verification involves a callout, it is possible for different callout
5275 options to be set for different recipients. It is too complicated to keep
5276 track of this in the cache, so now Exim always runs a verification when a
5277 callout is required, relying on the callout cache for the optimization.
5278 The overhead is duplication of the address routing, but this should not be
5281 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5282 to be verified with callout=postmaster, and the main callout worked but the
5283 postmaster check failed, the verification correctly failed. However, if a
5284 subsequent RCPT command asked for sender verification *without* the
5285 postmaster check, incorrect caching caused this verification also to fail,
5288 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5289 it was not caching the DNS options (qualify_single, search_parents) that
5290 were used when the lookup failed. A subsequent lookup with different
5291 options therefore always gave the same answer, though there were cases
5292 where it should not have. (Example: a "domains = !$mx_any" option on a
5293 dnslookup router: the "domains" option is always processed without any
5294 widening, but the router might have qualify_single set.) Now Exim uses the
5295 cached value only when the same options are set.
5297 32. Added John Jetmore's "exipick" utility to the distribution.
5299 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5300 than a timeout (e.g. a certificate is required, and is not provided), an
5301 Exim server now closes the connection immediately. Previously it waited for
5302 the client to close - but if the client is SSL, it seems that they each
5303 wait for each other, leading to a delay before one of them times out.
5305 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5306 maintained 0.8.x compatibility because I don't think many are using it, and
5307 it is clearly obsolete.
5309 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5312 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5313 client certificate was expired. A simple patch fixes this, though I don't
5314 understand the full logic of why the verify callback is called multiple
5317 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5318 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5319 which causes problems with some clients (such as the Certicom SSL Plus
5320 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5321 disables the coutermeasure allowing Eudora to connect."
5323 38. Exim was not checking that a write() to a log file succeeded. This could
5324 lead to Bad Things if a log got too big, in particular if it hit a file
5325 size limit. Exim now panics and dies if it cannot write to a log file, just
5326 as it does if it cannot open a log file.
5328 39. Modified OS/Makefile-Linux so that it now contains
5330 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5332 The two -D definitions ensure that Exim is compiled with large file
5333 support, which makes it possible to handle log files that are bigger than
5336 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5337 instance) a domain was checked against a named list that involved a lookup,
5338 causing $domain_data to be set, then another domain was checked against the
5339 same list, then the first domain was re-checked, the value of $domain_data
5340 after the final check could be wrong. In particular, if the second check
5341 failed, it could be set empty. This bug probably also applied to
5344 41. The strip_trailing_dot option was not being applied to the address given
5345 with the -f command-line option.
5347 42. The code for reading a message's header from the spool was incrementing
5348 $received_count, but never initializing it. This meant that the value was
5349 incorrect (doubled) while delivering a message in the same process in which
5350 it was received. In the most common configuration of Exim, this never
5351 happens - a fresh exec is done - but it can happen when
5352 deliver_drop_privilege is set.
5354 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5355 it now includes up to 150 characters of the unexpected data in the log
5358 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5359 and building data strings. The size of both of these buffers was 10 000
5360 bytes - far larger than anybody would *ever* want, thought I. Needless to
5361 say, somebody hit the limit. I have increased the maximum line length to
5362 20 000 and the maximum data length of concatenated lines to 100 000. I have
5363 also fixed two bugs, because there was no checking on these buffers. Tsk,
5364 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5365 buffer is too small.
5367 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5368 lsearch lookups. Now it does.
5370 46. When parsing a route_list item in a manualroute router, a fixed-length
5371 buffer was used for the list of hosts. I made this 1024 bytes long,
5372 thinking that nobody would ever have a list of hosts that long. Wrong.
5373 Somebody had a whole pile of complicated expansion conditions, and the
5374 string was silently truncated, leading to an expansion error. It turns out
5375 that it is easier to change to an unlimited length (owing to other changes
5376 that have happened since this code was originally written) than to build
5377 structure for giving a limitation error. The length of the item that
5378 expands into the list of hosts is now unlimited.
5380 47. The lsearch lookup could not handle data where the length of text line was
5381 more than 4095 characters. Such lines were truncated, leading to shortened
5382 data being returned. It should now handle lines of any length.
5384 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5385 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5388 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5389 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5390 no longer happens in comment lines. A list of replacements is now placed
5391 at the head of all of the source files, except those whose only change is
5392 to replace PERL_COMMAND in the very first #! line.
5394 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5395 messages on the queue, with a bottom-up merge sort, using code contributed
5396 by Michael Haardt. This should make operations like -bp somewhat faster on
5397 large queues. It won't affect queue runners, except when queue_run_in_order
5400 51. Installed eximstats 1.31 in the distribution.
5402 52. Added support for SRV lookups to the dnslookup router.
5404 53. If an ACL referred to $message_body or $message_body_end, the value was not
5405 reset for any messages that followed in the same SMTP session.
5407 54. The store-handling optimization for building very long strings was not
5408 differentiating between the different store pools. I don't think this
5409 actually made any difference in practice, but I've tidied it.
5411 55. While running the routers to verify a sender address, $sender_address
5412 was still set to the sender address. This is wrong, because when routing to
5413 send a bounce to the sender, it would be empty. Therefore, I have changed
5414 it so that, while verifying a sender address, $sender_address is set to <>.
5415 (There is no change to what happens when verifying a recipient address.)
5417 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5418 target A or AAAA records (if not already returned) without resetting the
5419 qualify_single or search_parents options of the DNS resolver. These are
5420 inappropriate in this case because the targets of MX and SRV records must
5421 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5422 target that, when qualified, matched something in the local domain. These
5423 two options are now turned off when doing these lookups.
5425 57. It seems that at least some releases of Reiserfs (which does not have the
5426 concept of a fixed number of inodes) returns zero and not -1 for the
5427 number of available inodes. This interacted badly with check_spool_inodes,
5428 which assumed that -1 was the "no such thing" setting. What I have done is
5429 to check that the total number of inodes is greater than zero before doing
5430 the test of how many are available.
5432 58. When a "warn" ACL statement has a log_message modifier, the message is
5433 remembered, and not repeated. This is to avoid a lot of repetition when a
5434 message has many recipients that cause the same warning to be written.
5435 Howewer, Exim was preserving the list of already written lines for an
5436 entire SMTP session, which doesn't seem right. The memory is now reset if a
5437 new message is started.
5439 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5440 debugging output unless log_rewrite was also set.
5442 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5443 of (int)(handle) when we know that handle contains (void *)(-1).
5445 61. The Exim daemon panic-logs an error return when it closes the incoming
5446 connection. However "connection reset by peer" seems to be common, and
5447 isn't really an error worthy of noting specially, so that particular error
5450 62. When Exim is trying to find all the local interfaces, it used to panic and
5451 die if the ioctl to get the interface flags failed. However, it seems that
5452 on at least one OS (Solaris 9) it is possible to have an interface that is
5453 included in the list of interfaces, but for which you get a failure error
5454 for this call. This happens when the interface is not "plumbed" into a
5455 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5456 failure of the "get flags" call assumes that the interface is down.
5458 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5459 makes life easier for people who are doing arithmetic on fields extracted
5460 from dates, where you often get leading zeros that should not be
5461 interpreted as octal.
5463 64. Added qualify_domain to the redirect router, to override the global
5466 65. If a pathologically long header line contained very many addresses (the
5467 report of this problem mentioned 10 000) and each of them was rewritten,
5468 Exim could use up a very large amount of memory. (It kept on making new
5469 copies of the header line as it rewrote, and never released the old ones.)
5470 At the expense of a bit more processing, the header rewriting function has
5471 been changed so that it no longer eats memory in this way.
5473 66. The generation of the Received: header has been moved from the time that a
5474 message starts to be received, to the time that it finishes. The timestamp
5475 in the Received: header should now be very close to that of the <= log
5476 line. There are two side-effects of this change:
5478 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5479 logged header lines no longer include the local Received: line, because
5480 it has not yet been created. The same applies to a copy of the message
5481 that is returned to a non-SMTP sender when a message is rejected.
5483 (b) When a filter file is tested using -bf, no additional Received: header
5484 is added to the test message. After some thought, I decided that this
5487 This change does not affect the value of $received_for. It is still set
5488 after address rewriting, but before local_scan() is called.
5490 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5492 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5493 gave an unhelpful panic error message, and a defer error. I have managed to
5494 change this behaviour so that it now rejects any supplied certificate,
5495 which seems right, as the list of acceptable certificates is empty.
5497 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5498 gave an unhelpful defer error. I have not managed to make this reject any
5499 supplied certificates, but the error message it gives is "no certificate
5500 supplied", which is not helpful.
5502 70. exigrep's output now also includes lines that are not associated with any
5503 message, but which match the given pattern. Implemented by a patch from
5504 Martin Sluka, which also tidied up the Perl a bit.
5506 71. Recipient callout verification, like sender verification, was using <> in
5507 the MAIL FROM command. This isn't really the right thing, since the actual
5508 sender may affect whether the remote host accepts the recipient or not. I
5509 have changed it to use the actual sender in the callout; this means that
5510 the cache record is now keyed on a recipient/sender pair, not just the
5511 recipient address. There doesn't seem to be a real danger of callout loops,
5512 since a callout by the remote host to check the sender would use <>.
5513 [SEE ABOVE: changed after hitting problems.]
5515 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5516 temporary errors. However, in the case of such a code being given after
5517 the end of a data transmission (i.e. after ".") Exim was failing to write
5518 a retry record for the message. (Yes, there was some broken host that was
5519 actually sending 8xx at this point.)
5521 73. An unknown lookup type in a host list could cause Exim to panic-die when
5522 the list was checked. (An example that provoked this was putting <; in the
5523 middle of a list instead of at the start.) If this happened during a DATA
5524 ACL check, a -D file could be left lying around. This kind of configuration
5525 error no longer causes Exim to die; instead it causes a defer errror. The
5526 incident is still logged to the main and panic logs.
5528 74. Buglet left over from Exim 3 conversion. The message "too many messages
5529 in one connection" was written to the rejectlog but not the mainlog, except
5530 when address rewriting (yes!) was being logged.
5532 75. Added write_rejectlog option.
5534 76. When a system filter was run not as root (that is, when system_filter_user
5535 was set), the values of the $n variables were not being returned to the
5536 main process; thus, they were not subsequently available in the $sn
5539 77. Added +return_path_on_delivery log selector.
5541 78. A connection timeout was being treated differently from recipients deferred
5542 when testing hosts_max_try with a message that was older than the host's
5543 retry timeout. (The host should not be counted, thus allowing all hosts to
5544 be tried at least once before bouncing.) This may have been the cause of an
5545 occasionally reported bug whereby a message would remain on the queue
5546 longer than the retry timeout, but would be bounced if a delivery was
5547 forced. I say "may" because I never totally pinned down the problem;
5548 setting up timeout/retry tests is difficult. See also the next item.
5550 79. The ultimate address timeout was not being applied to errors that involved
5551 a combination of host plus message (for example, a timeout on a MAIL
5552 command). When an address resolved to a number of possible hosts, and they
5553 were not all tried for each delivery (e.g. because of hosts_max_try), a
5554 message could remain on the queue longer than the retry timeout.
5556 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5559 81. Fixed an obscure SMTP outgoing bug which required at least the following
5560 conditions: (a) there was another message waiting for the same server;
5561 (b) the server returned 5xx to all RCPT commands in the first message so
5562 that the message was not completed; (c) the server dropped the connection
5563 or gave a negative response to the RSET that Exim sends to abort the
5564 transaction. The observed case was a dropped connection after DATA that had
5565 been sent in pipelining mode. That is, the server had advertised PIPELINING
5566 but was not implementing it correctly. The effect of the bug was incorrect
5567 behaviour, such as trying another host, and this could lead to a crash.
5573 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5574 and daemon.c were passed as pointers to ints; they should have been
5575 pointers to socklen_t variables (which are typically unsigned ints).
5577 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5580 3. Fixed a really odd bug that affected only the testing scheme; patching a
5581 certain fixed string in the binary changed the value of another string that
5582 happened to be identical to the end of the original first string.
5584 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5585 name", it returns that address as the IP address. On some operating
5586 systems (e.g. Solaris), it also passes back the IP address string as the
5587 "host name". However, on others (e.g. Linux), it passes back an empty
5588 string. Exim wasn't checking for this, and was changing the host name to an
5589 empty string, assuming it had been canonicized.
5591 5. Although rare, it is permitted to have more than one PTR record for a given
5592 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5593 all the names associated with an address, because they do in Solaris.
5594 However, it seems that they do not in Linux for data that comes from the
5595 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5596 I found this out when I moved to a new Linux workstation and tried to run
5597 the Exim test suite.
5599 To get round this problem I have changed the code so that it now does its
5600 own call to the DNS to look up PTR records when searching for a host name.
5601 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5602 addresses that are only in /etc/hosts are still found.
5604 This behaviour is, however, controlled by an option called host_lookup_
5605 order, which defaults to "bydns:byaddr". If people want to use the other
5606 order, or indeed, just use one or the other means of lookup, they can
5607 specify it in this variable.
5609 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5610 some operating systems, this comes back from gethostbyaddr() as an empty
5611 string, and this is what Exim used to test for. However, it seems that in
5612 other systems, "." is yielded. Exim now tests for this case too.
5614 7. The values of check_spool_space and check_log_space are now held internally
5615 as a number of kilobytes instead of an absolute number of bytes. If a
5616 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5617 kilobyte. This means that much larger values can be stored.
5619 8. Exim monitor: an attempt to get the action menu when not actually pointing
5620 at a message produces an empty menu entitled "No message selected". This
5621 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5622 no entries in it ("Shell widget menu has zero width and/or height"). So I
5623 have added a single, blank menu entry in this case.
5625 9. Added ${quote_local_part.
5627 10. MIME decoding is now applied to the contents of Subject: header lines when
5630 11. Now that a reference to $sender_host_address automatically causes a reverse
5631 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5632 host lookup before query-style lookups in lists that might use this
5633 variable. This has therefore been abolished, and the "net-" prefix is no
5634 longer necessary for query-style lookups.
5636 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5637 have been a typo for LFLAGS, so it has been changed.
5639 13. The install script calls Exim with "-C /dev/null" in order to find the
5640 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5641 to be output. Howeve, since Exim outputs its version number before the
5642 error, it didn't break the script. It just looked ugly. I fixed this by
5643 always allowing "-C /dev/null" if the caller is root.
5645 14. Ignore overlarge ACL variable number when reading spool file - insurance
5646 against a later release with more variables having written the file.
5648 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5649 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5650 this, as well as the form without the "IPv6" on the front.
5652 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5653 OS/Makefile-Darwin file.
5655 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5656 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5658 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5659 variable. However, some OS use ints for inodes. I've added cast to long int
5660 to get rid of the compiler warning.
5662 19. I had forgotten to lock out "/../" in configuration file names when
5663 ALT_CONFIG_PREFIX was set.
5665 20. Routers used for verification do not need to specify transports. However,
5666 if such a router generated a host list, and callout was configured, Exim
5667 crashed, because it could not find a port number from the (non-existent)
5668 transport. It now assumes port 25 in this circumstance.
5670 21. Added the -t option to exigrep.
5672 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5673 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5674 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5675 them is set, LOOKUP_LSEARCH is forced.
5677 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5679 24. Added sender and host information to the "rejected by local_scan()" log
5680 line; previously there was no indication of these.
5682 25. Added .include_if_exists.
5684 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5685 Linux. It turns out that not all file systems support this. Apparently some
5686 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5687 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5688 not supported on the file descriptor, is now ignored when Exim is trying to
5689 sync a directory. This applies only to Linux.
5691 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5693 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5694 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5697 29. Added support for maildirsize files from supplied patch (modified a bit).
5699 30. The use of :fail: followed by an empty string could lead Exim to respond to
5700 sender verification failures with (e.g.):
5702 550 Verification failed for <xxx>
5703 550 Sender verify failed
5705 where the first response line was missing the '-' that indicates it is not
5706 the final line of the response.
5708 31. The loop for finding the name of the user that called Exim had a hardwired
5709 limit of 10; it now uses the value of finduser_retries, which is used for
5710 all other user lookups.
5712 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5715 33. Exim was neglecting to zero errno before one call of strtol() when
5716 expanding a string and expecting an integer value. On some systems this
5717 resulted in spurious "integer overflow" errors. Also, it was casting the
5718 result into an int without checking.
5720 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5721 did not work. The code looks as if it has *never* worked, though it appears
5722 to have been documented since at least releast 1.62. I have made it work.
5724 35. The "timeout_DNS" error in retry rules, also documented since at least
5725 1.62, also never worked. As it isn't clear exactly what this means, and
5726 clearly it isn't a major issue, I have abolished the feature by treating it
5727 as "timeout", and writing a warning to the main and panic logs.
5729 36. The display of retry rules for -brt wasn't always showing the error code
5732 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5733 timeout_connect_A, timeout_connect_MX.
5735 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5736 to the empty sender.
5738 39. The daemon was not analysing the content of -oX till after it had closed
5739 stderr and disconnected from the controlling terminal. This meant that any
5740 syntax errors were only noted on the panic log, and the return code from
5741 the command was 0. By re-arranging the code a little, I've made the
5742 decoding happen first, so such errors now appear on stderr, and the return
5743 code is 1. However, the actual setting up of the sockets still happens in
5744 the disconnected process, so errors there are still only recorded on the
5747 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5748 connections (as happens on some IP stacks) was logged at start up time as
5749 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5750 it from "IPv6 and IPv4", which means that two separate sockets are being
5753 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5754 says whether AF_INET or AF_INET6 was passed as an argument.
5756 42. Exiwhat output was messed up when time zones were included in log
5759 43. Exiwhat now gives more information about the daemon's listening ports,
5760 and whether -tls-on-connect was used.
5762 44. The "port" option of the smtp transport is now expanded.
5764 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5765 silently ignored. Now an error message is written to the main and panic
5768 46. There's a new ACL modifier called "logwrite" which writes to a log file
5769 as soon as it is encountered.
5771 47. Added $local_user_uid and $local_user_gid at routing time.
5773 48. Exim crashed when trying to verify a sender address that was being
5776 49. Exim was recognizing only a space character after ".include". It now also
5777 recognizes a tab character.
5779 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5780 extracting the relevant information from the specification. The man page no
5781 longer contains scrambled data for the -d option, and I've added a section
5782 at the front about calling Exim under different names.
5784 51. Added "extra_headers" argument to the "mail" command in filter files.
5786 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5789 53. Installed eximstats 1.29.
5791 54. Added transport_filter_timeout as a generic transport option.
5793 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5794 Cc: header lines. This was required by RFC 822, but it not required by RFC
5797 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5798 incoming messages that did not have them. Now it does so only if the
5799 message originates locally, that is, if there is no associated remote host
5800 address. When Resent- header lines are present, this applies to the Resent-
5801 lines rather than the non-Resent- lines.
5803 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5804 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5806 58. Messages for configuration errors now include the name of the main
5807 configuration file - useful now that there may be more than one file in a
5808 list (.included file names were always shown).
5810 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5811 for those rare installations that do not start the daemon as root or run it
5812 setuid root. I've cut out the call to initgroups() if the daemon is not
5815 60. The Exim user and group can now be bound into the binary as text strings
5816 that are looked up at the start of Exim's processing.
5818 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5820 62. Added $mailstore_basename variable.
5822 63. Installed patch to sieve.c from Michael Haardt.
5824 64. When Exim failed to open the panic log after failing to open the main log,
5825 the original message it was trying to log was written to stderr and debug
5826 output, but if they were not available (the usual case in production), it
5827 was lost. Now it is written to syslog before the two lines that record the
5828 failures to open the logs.
5830 65. Users' Exim filters run in subprocesses under the user's uid. It is
5831 possible for a "deliver" command or an alias in a "personal" command to
5832 provoke an address rewrite. If logging of address rewriting is configured,
5833 this fails because the process is not running as root or exim. There may be
5834 a better way of dealing with this, but for the moment (because 4.30 needs
5835 to be released), I have disabled address rewrite logging when running a
5836 filter in a non-root, non-exim process.
5842 1. The buildconfig auxiliary program wasn't quoting the value set for
5843 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5844 not defined. This bug was masked in 4.22 by the effect that was fixed in
5847 2. Some messages that were rejected after a message id was allocated were
5848 shown as "incomplete" by exigrep. It no longer does this for messages that
5849 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5851 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5852 have allow_domain_literals set, the ID did not get logged in the <= line.
5853 Domain literals are now always recognized in Message-ID: header lines.
5855 4. The first argument for a ${extract expansion item is the key name or field
5856 number. Leading and trailing spaces in this item were not being ignored,
5857 causing some misleading effects.
5859 5. When deliver_drop_privilege was set, single queue runner processes started
5860 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5861 same command in the process it spins off) were not dropping privilege.
5863 6. When the daemon running as "exim" started a queue runner, it always
5864 re-executed Exim in the spun-off process. This is a waste of effort when
5865 deliver_drop_privilege is set. The new process now just calls the
5866 queue-runner function directly.
5872 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5875 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5876 ignored. Though the use of -oP was forcing the writing of a pid file, it
5877 was always written to the default place.
5879 3. If the message "no IP address found for host xxxx" is generated during
5880 incoming verification, it is now followed by identification of the incoming
5881 connection (so you can more easily find what provoked it).
5883 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5885 5. Added some features to "harden" Exim a bit more against certain attacks:
5887 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5888 be put in Local/Makefile. This is like the never_users runtime option,
5889 but it cannot be overridden. The default setting is "root".
5891 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5892 prefix string with which any file named in a -C command line option
5895 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5896 is retained for -C and -D only if the caller of Exim is root. Without
5897 it, the exim user may also use -C and -D and retain privilege.
5899 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5900 command line option is disabled.
5902 6. Macro names set by the -D option must start with an upper case letter, just
5903 like macro names defined in the configuration file.
5905 7. Added "dereference=" facility to LDAP.
5907 8. Two instances of the typo "uknown" in the source files are fixed.
5909 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5910 the Configure-Makefile script screwed up while processing it.
5912 10. Incorporated PCRE 4.4.
5914 11. The SMTP synchronization check was not operating right at the start of an
5915 SMTP session. For example, it could not catch a HELO sent before the client
5916 waited for the greeting. There is now a check for outstanding input at the
5917 point when the greeting is written. Because of the duplex, asynchronous
5918 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5919 way, but not yet received, when the check is performed.
5921 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5922 on TCP/IP sockets, because this apparently causes some broken clients to
5925 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5926 unchanged) from the Cygwin maintainer.
5928 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5929 was supported instead of testing for mbx. Effectively a typo.
5931 15. The spa authenticator server code was not checking that the input it
5932 received was valid base64.
5934 16. The debug output line for the "set" modifier in ACLs was not showing the
5935 name of the variable that was being set.
5937 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5939 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5940 lookup on its own. Something else was needed to trigger the lookup. For
5941 example, a match in host_lookup or the need for a host name in a host list.
5942 Now, if $sender_host_name is referenced and the host name has not yet been
5943 looked up, a lookup is performed. If the lookup fails, the variable remains
5944 empty, and $host_lookup_failed is set to "1".
5946 19. Added "eqi" as a case-independent comparison operator.
5948 20. The saslauthd authentication condition could segfault if neither service
5949 nor realm was specified.
5951 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5952 error message that was logged was misleading, and incoming SMTP
5953 connections were dropped. The message is now more accurate, and temporary
5954 errors are given to SMTP connections.
5956 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5957 (except RSET) with a 421 error, until QUIT is received. However, it was
5958 failing to send a response to QUIT.
5960 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5961 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5962 ACL is now run after the helo_try_verify_hosts code.
5964 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5965 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5966 case-independent, so other case variants are also recognized). Apparently
5967 some systems use these upper case variants.
5969 25. If more than two messages were waiting for the same host, and a transport
5970 filter was specified for the transport, Exim sent two messages over the
5971 same TCP/IP connection, and then failed with "socket operation on non-
5972 socket" when it tried to send the third.
5974 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5976 27. The extern definition of crypt16() in expand.c was not being excluded when
5977 the OS had its own crypt16() function.
5979 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5980 as a preferred synonym for return_size_limit, both as an option and as an
5983 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5985 30. Changed the default configuration ACL to relax the local part checking rule
5986 for addresses that are not in any local domains. For these addresses,
5987 slashes and pipe symbols are allowed within local parts, but the sequence
5988 /../ is explicitly forbidden.
5990 31. SPA server authentication was not clearing the challenge buffer before
5993 32. log_message in a "warn" ACL statement was writing to the reject log as
5994 well as to the main log, which contradicts the documentation and doesn't
5995 seem right (because no rejection is happening). So I have stopped it.
5997 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5998 However, I am unable to do any testing of this.
6000 34. Fixed an infelicity in the appendfile transport. When checking directories
6001 for a mailbox, to see if any needed to be created, it was accidentally
6002 using path names with one or more superfluous leading slashes; tracing
6003 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6005 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6006 contents are added to the log line that is written for every discarded
6007 recipient. (Previously a log_message setting was ignored.)
6009 36. The ${quote: operator now quotes the string if it is empty.
6011 37. The install script runs exim in order to find its version number. If for
6012 some reason other than non-existence or emptiness, which it checks, it
6013 could not run './exim', it was installing it with an empty version number,
6014 i.e. as "exim-". This error state is now caught, and the installation is
6017 38. An argument was missing from the function that creates an error message
6018 when Exim fails to connect to the socket for saslauthd authentication.
6019 This could cause Exim to crash, or give a corrupted message.
6021 39. Added isip, isip4, and isip6 to ${if conditions.
6023 40. The ACL variables $acl_xx are now saved with the message, and can be
6024 accessed later in routers, transports, and filters.
6026 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6027 strings in the file are not string-expanded.
6029 42. If a MAIL command specified a SIZE value that was too large to fit into an
6030 int variable, the check against message_size_limit failed. Such values are
6031 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6032 day this will have to be increased, but I don't think I want to be around
6033 when emails are that large.
6040 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6041 iconv() is not standard in FreeBSD.
6043 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6044 IPv6 enabled. The observed symptom was a segmentation fault on return from
6045 the function os_common_find_running_interfaces() in src/os.c.
6047 3. In the check_special_case() function in daemon.c I had used "errno" as an
6048 argument name, which causes warnings on some systems. This was basically a
6049 typo, since it was named "eno" in the comments!
6051 4. The code that waits for the clock to tick (at a resolution of some fraction
6052 of a second) so as to ensure message-id uniqueness was always waiting for
6053 at least one whole tick, when it could have waited for less. [This is
6054 almost certainly not relevant at current processor speeds, where it is
6055 unlikely to ever wait at all. But we try to future-proof.]
6057 5. The function that sleeps for a time interval that includes fractions of a
6058 second contained a race. It did not block SIGALRM between setting the
6059 timer, and suspending (a couple of lines later). If the interval was short
6060 and the sigsuspend() was delayed until after it had expired, the suspension
6061 never ended. On busy systems this could lead to processes getting stuck for
6064 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6065 process, before it forks any delivery processes. The open lookup caching
6066 mechanism meant that the open file or database connection was passed into
6067 the delivery process. The problem was that delivery processes always tidy
6068 up cached lookup data. This could cause a problem for the next delivery
6069 process started by the queue runner, because the external queue runner
6070 process does not know about the closure. So the next delivery process
6071 still has data in the lookup cache. In the case of a file lookup, there was
6072 no problem because closing a file descriptor in a subprocess doesn't affect
6073 the parent. However, if the lookup was caching a connection to a database,
6074 the connection was closed, and the second delivery process was likely to
6075 see errors such as "PGSQL: query failed: server closed the connection
6076 unexpectedly". The problem has been fixed by closing all cached lookups
6077 in a queue runner before running a delivery process.
6079 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6080 seem to have the "const" qualifier which it has on other OS. I've
6083 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6084 *of the same type* (client or server) with the same public name that an
6085 error should be diagnosed.
6087 9. When Exim looked up a host name for an IP address, but failed to find the
6088 original IP address when looking up the host name (a safety check), it
6089 output the message "<ip address> does not match any IP for NULL", which was
6090 confusing, to say the least. The bug was that the host name should have
6091 appeared instead of "NULL".
6093 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6094 user that is built into the binary, and the -C or -D options is used, root
6095 privilege is dropped before the configuration file is read. In addition,
6096 logging is switched to stderr instead of the normal log files. If the
6097 configuration then re-defines the Exim user, the unprivileged environment
6098 is probably not what is expected, so Exim logs a panic warning message (but
6101 However, if deliver_drop_privilege is set, the unprivileged state may well
6102 be exactly what is intended, so the warning has been cut out in that case,
6103 and Exim is allowed to try to write to its normal log files.
6109 1. smtp_return_error_details was not giving details for temporary sender
6110 or receiver verification errors.
6112 2. Diagnose a configuration error if two authenticators have the same public
6115 3. Exim used not to create the message log file for a message until the first
6116 delivery attempt. This could be confusing when incoming messages were held
6117 for policy or load reasons. The message log file is now created at the time
6118 the message is received, and an initial "Received" line is written to it.
6120 4. The automatically generated man page for command line options had a minor
6121 bug that caused no ill effects; however, a more serious problem was that
6122 the procedure for building the man page automatically didn't always
6123 operate. Consequently, release 4.20 contains an out-of-date version. This
6124 shouldn't happen again.
6126 5. When building Exim with embedded Perl support, the script that builds the
6127 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6128 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6130 6. The freeze_tell option was not being used for messages that were frozen on
6131 arrival, either by an ACL or by local_scan().
6133 7. Added the smtp_incomplete_transaction log selector.
6135 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6136 was accepting AUTH without a new EHLO.
6138 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6139 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6142 10. Logging of TCP/IP connections (when configured) now happens in the main
6143 daemon process instead of the child process, so that the TCP/IP connection
6144 count is more accurate (but it can never be perfect).
6146 11. The use of "drop" in a nested ACL was not being handled correctly in the
6147 outer ACL. Now, if condition failure induced by the nested "drop" causes
6148 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6149 or "require"), the connection is dropped.
6151 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6152 that yield "discard" can now be used with an "accept" or a "discard" verb,
6153 but an error is generated for any others (because I can't see a useful way
6154 to define what should happen).
6156 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6157 are now processed in the same way as lines in the Exim configuration file.
6158 In particular, continuation lines are supported.
6160 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6162 15. Added -ti meaning -t -i.
6164 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6165 domains, and warn by logging if others are found.
6167 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6168 returned from GIFCONF when Exim is trying to find the list of interfaces on
6169 a host. The code in os.c has been modified to copy each ifreq to an aligned
6170 structure in all cases.
6172 Also, in some cases, the returned ifreq's were being copied to a 'struct
6173 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6174 means the last couple of bytes of an IPv6 address could be chopped if the
6175 ifreq contained only a normal sockaddr (14 bytes storage).
6177 18. Named domain lists were not supported in the hosts_treat_as_local option.
6178 An entry such as +xxxx was not recognized, and was treated as a literal
6181 19. Ensure that header lines added by a DATA ACL are included in the reject log
6182 if the ACL subsequently rejects the message.
6184 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6185 MD5 (which is deprecated).
6187 21. When testing a filter file using -bf, Exim was writing a message when it
6188 took the sender from a "From " line in the message, but it was not doing so
6189 when it took $return_path from a Return-Path: header line. It now does.
6191 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6192 with a valid header line field name (a series of printing characters
6193 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6195 23. Changed "disc" in the source to "disk" to conform to the documentation and
6196 the book and for uniformity.
6198 24. Ignore Sendmail's -Ooption=value command line item.
6200 25. When execve() failed while trying to run a command in a pipe transport,
6201 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6202 could be confused with a return value of 69 from the command itself. This
6203 has been changed to 127, the value the shell returns if it is asked to run
6204 a non-existent command. The wording for the related log line suggests a
6205 non-existent command as the problem.
6207 26. If received_header_text expands to an empty string, do not add a Received:
6208 header line to the message. (Well, it adds a token one on the spool, but
6209 marks it "old" so that it doesn't get used or transmitted.)
6211 27. Installed eximstats 1.28 (addition of -nt option).
6213 28. There was no check for failure on the call to getsockname() in the daemon
6214 code. This can fail if there is a shortage of resources on the system, with
6215 ENOMEM, for example. A temporary error is now given on failure.
6217 29. Contrary to the C standard, it seems that in some environments, the
6218 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6219 program. Exim now does this explicitly; it affects the formatting of
6220 timestamps using strftime().
6222 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6223 complaints. I've now initialized all the variables, to avoid this.
6225 32. Header lines added by a system filter were not being "seen" during
6226 transport-time rewrites.
6228 33. The info_callback() function passed to OpenSSL is set up with type void
6229 (*)(SSL *, int, int), as described somewhere. However, when calling the
6230 function (actually a macro) that sets it up, the type void(*)() is
6231 expected. I've put in a cast to prevent warnings from picky compilers.
6233 34. If a DNS black list lookup found a CNAME record, but there were no A
6234 records associated with the domain it pointed at, Exim crashed.
6236 35. If a DNS black list lookup returned more than one A record, Exim ignored
6237 all but the first. It now scans all returned addresses if a particular IP
6238 value is being sought. In this situation, the contents of the
6239 $dnslist_value variable are a list of all the addresses, separated by a
6242 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6243 to accept a host name and all its aliases if the forward lookup for any of
6244 them yielded the IP address of the incoming connection. Now it accepts only
6245 those names whose forward lookup yields the correct IP address. Any other
6246 names are discarded. This closes a loophole whereby a rogue DNS
6247 administrator could create reverse DNS records to break through a
6248 wildcarded host restriction in an ACL.
6250 37. If a user filter or a system filter that ran in a subprocess used any of
6251 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6252 the wrong values were passed to the pipe command ($thisaddress had the
6253 value of $0, $0 had the value of $1, etc). This bug was introduced by
6254 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6256 38. Improved the line breaking for long SMTP error messages from ACLs.
6257 Previously, if there was no break point between 40 and 75 characters, Exim
6258 left the rest of the message alone. Two changes have been made: (a) I've
6259 reduced the minimum length to 35 characters; (b) if it can't find a break
6260 point between 35 and 75 characters, it looks ahead and uses the first one
6261 that it finds. This may give the occasional overlong line, but at least the
6262 remaining text gets split now.
6264 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6265 file descriptors might be low, and that setting 1000 would always raise it.
6266 It turns out that in some environments, the limit is already over 1000 and
6267 that lowering it causes trouble. So now Exim takes care not to decrease it.
6269 40. When delivering a message, the value of $return_path is set to $sender_
6270 address at the start of routing (routers may change the value). By an
6271 oversight, this default was not being set up when an address was tested by
6272 -bt or -bv, which affected the outcome if any router or filter referred to
6275 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6276 log only when "message" or "log_message" are set. However, if one of the
6277 conditions was an address verification, or a call to a nested ACL, the
6278 messages generated by the underlying test were being passed through. This
6279 no longer happens. The underlying message is available in $acl_verify_
6280 message for both "message" and "log_message" expansions, so it can be
6281 passed through if needed.
6283 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6284 new expansion $bh_ to give the encoded byte string without charset
6285 translation. Translation happens only if iconv() is available; HAVE_ICONV
6286 indicates this at build time. HEADERS_CHARSET gives the charset to
6287 translate to; headers_charset can change it in the configuration, and
6288 "headers charset" can change it in an individual filter file.
6290 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6291 that creates RFC 2047 encoded "words" labels them as that charset instead
6292 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6293 expansion operator; (ii) when Exim creates a From: line for a local
6294 message; (iii) when a header line is rewritten to include a "phrase" part.
6296 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6297 buggy, causing it to skip the first lines of messages whose message ID
6298 ended in 'D'. This would not have bitten before Exim release 4.14, because
6299 message IDs were unlikely to end in 'D' before then. The effect was to have
6300 incorrect size information for certain domains.
6302 45. #include "config.h" was missing at the start of the crypt16.c module. This
6303 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6306 46. If there was a timeout during a "random" callout check, Exim treated it as
6307 a failure of the random address, and carried on sending RSET and the real
6308 address. If the delay was just some slowness somewhere, the response to the
6309 original RCPT would be taken as a response to RSET and so on, causing
6310 mayhem of various kinds.
6312 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6313 when I implemented it. It didn't allow for the fact that some option values
6314 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6315 the right test for positive values.
6317 48. Domain names in DNS records are case-independent. Exim always looks them up
6318 in lower case. Some resolvers return domain names in exactly the case they
6319 appear in the zone file, that is, they may contain uppercase letters. Not
6320 all resolvers do this - some return always lower case. Exim was treating a
6321 change of case by a resolver as a change of domain, similar to a widening
6322 of a domain abbreviation. This triggered its re-routing code and so it was
6323 trying to route what was effectively the same domain again. This normally
6324 caused routing to fail (because the router wouldn't handle the domain
6325 twice). Now Exim checks for this case specially, and just changes the
6326 casing of the domain that it ultimately uses when it transmits the message
6329 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6332 50. If a filter generated a file delivery with a non-absolute name (possible if
6333 no home directory exists for the router), the forbid_file option was not
6336 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6337 the existing equality matching.
6339 52. Exim was using ints instead of ino_t variables in some places where it was
6340 dealing with inode numbers.
6342 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6343 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6344 called TMPDIR, and if it finds it is different, it changes its value.
6346 54. The smtp_printf() function is now made available to local_scan() so
6347 additional output lines can be written before returning. There is also an
6348 smtp_fflush() function to enable the detection of a dropped connection.
6349 The variables smtp_input and smtp_batched_input are exported to
6352 55. Changed the default runtime configuration: the message "Unknown user"
6353 has been removed from the ACL, and instead placed on the localuser router,
6354 using the cannot_route_message feature. This means that any verification
6355 failures that generate their own messages won't get overridden. Similarly,
6356 the "Unrouteable address" message that was in the ACL for unverifiable
6357 relay addresses has also been removed.
6359 56. Added hosts_avoid_esmtp to the smtp transport.
6361 57. The exicyclog script was not checking for the esoteric option
6362 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6363 will work only if exicyclog is run under the appropriate euid.
6365 58. Following a discussion on the list, the rules by which Exim recognises line
6366 endings on incoming messages have been changed. The -dropcr and drop_cr
6367 options are now no-ops, retained only for backwards compatibility. The
6368 following line terminators are recognized: LF CRLF CR. However, special
6369 processing applies to CR:
6371 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6372 nor a local message in the state where . is a terminator.
6374 (ii) If a bare CR is encountered in a header line, an extra space is added
6375 after the line terminator so as not to end the header. The reasoning
6376 behind this is that bare CRs in header lines are most likely either
6377 to be mistakes, or people trying to play silly games.
6379 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6380 was being incorrectly given as 18 bytes larger than it should have been.
6381 This is a VOB (very old bug).
6383 60. This may never have affected anything current, but just in case it has:
6384 When the local host is found other than at the start of a list of hosts,
6385 the local host, those with the same MX, and any that follow, are discarded.
6386 When the list in question was part of a longer list of hosts, the following
6387 hosts (not currently being processed) were also being discarded. This no
6388 longer happens. I'm not sure if this situation could ever has previously
6391 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6394 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6395 header. This is something that is recommended in a new Internet Draft, and
6396 is something that is documented as being done by Sendmail. There are two
6397 possible values. For messages generated by the autoreply transport, Exim
6400 Auto-Submitted: auto-replied
6402 whereas for all other generated messages (e.g. bounces) it adds
6404 Auto-Submitted: auto-generated
6406 63. The "personal" condition in filters now includes a test for the
6407 Auto-Submitted: header. If it contains the string "auto-" the message it
6408 not considered personal.
6410 64. Added rcpt_include_affixes as a generic transport option.
6412 65. Added queue_only_override (default true).
6414 66. Added the syslog_duplication option.
6416 67. If what should have been the first header line of a message consisted of
6417 a space followed by a colon, Exim was mis-interpreting it as a header line.
6418 It isn't of course - it is syntactically invalid and should therefore be
6419 treated as the start of the message body. The misbehaviour could have
6420 caused a number of strange effects, including loss of data in subsequent
6421 header lines, and spool format errors.
6423 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6424 client host had authenticated. This control can now be exercised by an ACL
6425 for more flexibility.
6427 69. By default, callouts do not happen when testing with -bh. There is now a
6428 variant, -bhc, which does actually run the callout code, including
6429 consulting and updating the callout cache.
6431 70. Added support for saslauthd authentication, courtesy of Alexander
6434 71. If statvfs() failed on the spool or log directories while checking their
6435 size for availability, Exim confusingly gave the error "space shortage".
6436 Furthermore, in debugging mode it crashed with a floating point exception.
6437 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6438 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6439 serious problem, Exim now writes to the main and panic logs when this
6440 happens, with details of the failure. It then refuses to accept the
6441 incoming message, giving the message "spool directory problem" or "log
6442 directory problem" with a 421 code for SMTP messages.
6444 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6445 0, 1, and 2 exist, because some OS complain for execs without them (see
6446 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6447 descriptors. However, the code omitted to check that the open succeeded,
6448 causing mysterious errors if for some reason the permissions on /dev/null
6449 got screwed. Now Exim writes a message to the main and panic logs, and
6450 bombs out if it can't open /dev/null.
6452 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6453 interact so that it is all more flexible. It is supposed to remain
6454 backwards compatible. Also added extra_local_interfaces.
6456 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6457 to bomb out with an assertion failure - to the client this appears as a
6458 connection drop. This problem occurs in the part of the code that was taken
6459 from the Samba project. Fortunately, the assertion is in a very simple
6460 function, so I have fixed this by reproducing the function inline in the
6461 one place where it is called, and arranging for authentication to fail
6462 instead of killing the process with assert().
6464 75. The SPA client code was not working when the server requested OEM rather
6465 than Unicode encoding.
6467 76. Added code to make require_files with a specific uid setting more usable in
6468 the case where statting the file as root fails - usually a non-root-mounted
6469 NFS file system. When this happens and the failure is EACCES, Exim now
6470 forks a subprocess and does the per-uid checking as the relevant uid.
6472 77. Added process_log_path.
6474 78. If log_file_path was not explicitly set, a setting of check_log_space or
6475 check_log_inodes was ignored.
6477 79. If a space check for the spool or log partitions fails, the incident is now
6478 logged. Of course, in the latter case the data may get lost...
6480 80. Added the %p formatting code to string_format() so that it can be used to
6481 print addresses in debug_print(). Adjusted all the address printing in the
6482 debugging in store.c to use %p rather than %d.
6484 81. There was a concern that a line of code in smtp_in.c could overflow a
6485 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6486 initially expressed, the concern was not well-founded, because trailing
6487 spaces are removed early. However, if the trailing spaces were followed by
6488 a NULL, they did not get removed, so the overflow was possible. Two fixes
6491 (a) I re-wrote the offending code in a cleaner fashion.
6492 (b) If an incoming SMTP command contains a NULL character, it is rejected
6495 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6496 runs initgroups(), so that if the Exim user is in any additional groups,
6497 they will be used during message reception.
6503 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / heiko / exim.git / blob