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.
181 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
184 JH/02 The smtp transport option "multi_domain" is now expanded.
186 JH/03 The smtp transport now requests PRDR by default, if the server offers
189 JH/04 Certificate name checking on server certificates, when exim is a client,
190 is now done by default. The transport option tls_verify_cert_hostnames
191 can be used to disable this per-host. The build option
192 EXPERIMENTAL_CERTNAMES is withdrawn.
194 JH/05 The value of the tls_verify_certificates smtp transport and main options
195 default to the word "system" to access the system default CA bundle.
196 For GnuTLS, only version 3.0.20 or later.
198 JH/06 Verification of the server certificate for a TLS connection is now tried
199 (but not required) by default. The verification status is now logged by
200 default, for both outbound TLS and client-certificate supplying inbound
203 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
206 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
207 Status Notification (bounce) messages are now MIME format per RFC 3464.
208 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
209 under the control of the dsn_advertise_hosts option, and routers may
210 have a dsn_lasthop option.
212 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
213 default, modifiable by a malware= option. The list separator for
214 the options can now be changed in the usual way. Bug 68.
216 JH/10 The smtp_receive_timeout main option is now expanded before use.
218 JH/11 The incoming_interface log option now also enables logging of the
219 local interface on delivery outgoing connections.
221 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
222 if the interface and destination host and port all match.
224 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
227 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
228 Patch from Andrew Lewis.
230 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
231 now supports optional time-restrictions, weighting, and priority
232 modifiers per server. Patch originally by <rommer@active.by>.
234 JH/16 The spamd_address main option now supports a mixed list of local
235 and remote servers. Remote servers can be IPv6 addresses, and
236 specify a port-range.
238 JH/17 Bug 68: The spamd_address main option now supports an optional
239 timeout value per server.
241 JH/18 Bug 1581: Router and transport options headers_add/remove can
242 now have the list separator specified.
244 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
247 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
250 JH/21 Support for the A6 type of dns record is withdrawn.
252 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
253 rather than the verbs used.
255 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
256 from 255 to 1024 chars.
258 JH/24 Verification callouts now attempt to use TLS by default.
260 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
261 are generic router options now. The defaults didn't change.
263 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
264 Original patch from Alexander Shikoff, worked over by JH.
266 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
267 files if ZCAT_COMMAND is not executable.
269 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
271 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
273 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
274 Normally benign, it bites when the pair was led to by a CNAME;
275 modern usage is to not canoicalize the domain to a CNAME target
276 (and we were inconsistent anyway for A-only vs AAAA+A).
278 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
280 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
281 when evaluating $sender_host_dnssec.
283 JH/31 Check the HELO verification lookup for DNSSEC, adding new
284 $sender_helo_dnssec variable.
286 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
288 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
290 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
292 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
293 documented as working, but never had. Support all but $spam_report.
295 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
296 added for tls authenticator.
301 TL/01 When running the test suite, the README says that variables such as
302 no_msglog_check are global and can be placed anywhere in a specific
303 test's script, however it was observed that placement needed to be near
304 the beginning for it to behave that way. Changed the runtest perl
305 script to read through the entire script once to detect and set these
306 variables, reset to the beginning of the script, and then run through
307 the script parsing/test process like normal.
309 TL/02 The BSD's have an arc4random API. One of the functions to induce
310 adding randomness was arc4random_stir(), but it has been removed in
311 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
312 function when detected.
314 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
315 cause callback expansion.
317 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
318 syntax errors in an expansion can be treated as a string instead of
319 logging or causing an error, due to the internal use of bool_lax
320 instead of bool when processing it.
322 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
323 server certificates when making smtp deliveries.
325 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
327 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
329 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
331 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
332 Merged patch from Sebastian Wiedenroth.
334 JH/05 Fix results-pipe from transport process. Several recipients, combined
335 with certificate use, exposed issues where response data items split
336 over buffer boundaries were not parsed properly. This eventually
337 resulted in duplicates being sent. This issue only became common enough
338 to notice due to the introduction of conection certificate information,
339 the item size being so much larger. Found and fixed by Wolfgang Breyha.
341 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
342 size buffer was used, resulting in syntax errors when an expansion
345 JH/07 Add support for directories of certificates when compiled with a GnuTLS
346 version 3.3.6 or later.
348 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
349 is EXPERIMENTAL_EVENT, the main-configuration and transport options
350 both become "event_action", the variables become $event_name, $event_data
351 and $event_defer_errno. There is a new variable $verify_mode, usable in
352 routers, transports and related events. The tls:cert event is now also
353 raised for inbound connections, if the main configuration event_action
356 TL/06 In test suite, disable OCSP for old versions of openssl which contained
357 early OCSP support, but no stapling (appears to be less than 1.0.0).
359 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
360 server certificate names available under the smtp transport option
361 "tls_verify_cert_hostname" now do not permit multi-component wildcard
364 JH/10 Time-related extraction expansions from certificates now use the main
365 option "timezone" setting for output formatting, and are consistent
366 between OpenSSL and GnuTLS compilations. Bug 1541.
368 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
369 encoded parameter in the incoming message. Bug 1558.
371 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
372 include certificate info, eximon was claiming there were spoolfile
375 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
377 JH/14 Log delivery-related information more consistently, using the sequence
378 "H=<name> [<ip>]" wherever possible.
380 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
381 are problematic for Debian distribution, omit them from the release
384 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
386 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
389 JH/17 Fix a null-indirection in certextract expansions when a nondefault
390 output list separator was used.
395 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
396 checkers that were complaining about end of non-void function with no
399 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
400 This was a regression intruduced in 4.83 by another bugfix.
402 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
404 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
405 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
411 TF/01 Correctly close the server side of TLS when forking for delivery.
413 When a message was received over SMTP with TLS, Exim failed to clear up
414 the incoming connection properly after forking off the child process to
415 deliver the message. In some situations the subsequent outgoing
416 delivery connection happened to have the same fd number as the incoming
417 connection previously had. Exim would try to use TLS and fail, logging
418 a "Bad file descriptor" error.
420 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
421 utilities have not been installed.
423 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
424 temporary space as the ACL may create new global variables.
426 TL/01 LDAP support uses per connection or global context settings, depending
427 upon the detected version of the libraries at build time.
429 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
430 to extract and use the src ip:port in logging and expansions as if it
431 were a direct connection from the outside internet. PPv2 support was
432 updated based on HAProxy spec change in May 2014.
434 JH/02 Add ${listextract {number}{list}{success}{fail}}.
436 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
437 Properly escape header and check for NULL return.
439 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
442 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
444 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
445 characters in header names, implemented as a verify condition.
446 Contributed by Michael Fischer v. Mollard.
448 TL/05 Rename SPF condition results err_perm and err_temp to standardized
449 results permerror and temperror. Previous values are deprecated but
450 still accepted. In a future release, err_perm and err_temp will be
451 completely removed, which will be a backward incompatibility if the
452 ACL tests for either of these two old results. Patch contributed by
453 user bes-internal on the mailing list.
455 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
457 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
458 selectors, in both main and reject logs.
460 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
463 JH/07 Add malware type "sock" for talking to simple daemon.
465 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
467 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
468 routers/transports under cutthrough routing.
470 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
471 numbers. Touch up "bool" conditional to keep the same definition.
473 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
475 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
477 JH/12 Expand items in router/transport headers_add or headers_remove lists
478 individually rather than the list as a whole. Bug 1452.
480 Required for reasonable handling of multiple headers_ options when
481 they may be empty; requires that headers_remove items with embedded
482 colons must have them doubled (or the list-separator changed).
484 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
485 view the policy declared in the DMARC record. Currently, $dmarc_status
486 is a combined value of both the record presence and the result of the
489 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
491 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
492 dnslookup router and the smtp transport (applying to the forward
495 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
496 of ldap servers used for a specific lookup. Patch provided by Heiko
499 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
500 New variable $lookup_dnssec_authenticated for observability.
502 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
503 Patch submitted by Lars Timman.
505 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
507 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
508 Requires trusted mode and valid format message id, aborts otherwise.
509 Patch contributed by Heiko Schlichting.
511 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
512 certextract with support for various fields. Bug 1358.
514 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
515 is requested by default, modifiable by smtp transport option
518 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
519 operate on certificate variables to give certificate fingerprints
520 Also new ${sha256:cert_variable}.
522 JH/23 The PRDR feature is moved from being Experimental into the mainline.
524 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
525 Christian Aistleitner.
527 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
529 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
530 file. Patch from Wolfgang Breyha.
532 JH/25 Expand the coverage of the delivery $host and $host_address to
533 client authenticators run in verify callout. Bug 1476.
535 JH/26 Port service names are now accepted for tls_on_connect_ports, to
536 align with daemon_smtp_ports. Bug 72.
538 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
539 support and error reporting did not work properly.
541 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
542 and is readable. Patch from Andrew Colin Kissa.
544 TL/14 Enhance documentation of ${run expansion and how it parses the
545 commandline after expansion, particularly in the case when an
546 unquoted variable expansion results in an empty value.
548 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
550 PP/02 Fix internal collision of T_APL on systems which support RFC3123
551 by renaming away from it. Addresses GH issue 15, reported by
554 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
556 TL/15 SECURITY: prevent double expansion in math comparison functions
557 (can expand unsanitized data). Not remotely exploitable.
564 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
566 PP/02 Make -n do something, by making it not do something.
567 When combined with -bP, the name of an option is not output.
569 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
572 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
573 $sender_host_name and config options to manage this, and basic check
576 PP/05 DSCP support for outbound connections and control modifier for inbound.
578 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
579 (Only plugin which currently uses this is kerberos4, which nobody should
580 be using, but we should make it available and other future plugins might
581 conceivably use it, even though it would break NAT; stuff *should* be
582 using channel bindings instead).
584 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
585 name; added for Sendmail compatibility; requires admin caller.
586 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
587 just ignore it); requires trusted caller.
588 Also parse but ignore: -Ac -Am -X<logfile>
591 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
593 TL/02 Add +smtp_confirmation as a default logging option.
595 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
596 Patch by Magnus Holmgren from 2007-02-20.
598 TL/04 Bugzilla 1281 - Spec typo.
599 Bugzilla 1283 - Spec typo.
600 Bugzilla 1290 - Spec grammar fixes.
602 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
604 TL/06 Add Experimental DMARC support using libopendmarc libraries.
606 TL/07 Fix an out of order global option causing a segfault. Reported to dev
607 mailing list by by Dmitry Isaikin.
609 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
611 JH/02 Support "G" suffix to numbers in ${if comparisons.
613 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
615 NM/01 Bugzilla 1197 - Spec typo
616 Bugzilla 1196 - Spec examples corrections
618 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
620 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
621 gnutls_enable_pkcs11, but renamed to more accurately indicate its
624 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
625 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
627 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
628 "acl {{name}{arg}...}", and optional args on acl condition
631 JH/05 Permit multiple router/transport headers_add/remove lines.
633 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
635 JH/07 Avoid using a waiting database for a single-message-only transport.
636 Performance patch from Paul Fisher. Bugzilla 1262.
638 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
641 JH/09 Add $headers_added variable, with content from use of ACL modifier
642 add_header (but not yet added to the message). Bugzilla 199.
644 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
645 Pulled from Bugzilla 817 by Wolfgang Breyha.
647 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
649 (nb: this is the same fix as in Exim 4.80.1)
651 JH/11 Add A= logging on delivery lines, and a client_set_id option on
654 JH/12 Add optional authenticated_sender logging to A= and a log_selector
657 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
659 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
660 advertise SMTP AUTH mechanism to us, instead of a generic
661 protocol violation error. Also, make Exim more robust to bad
662 data from the Dovecot auth socket.
664 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
666 When a queue runner is handling a message, Exim first routes the
667 recipient addresses, during which it prunes them based on the retry
668 hints database. After that it attempts to deliver the message to
669 any remaining recipients. It then updates the hints database using
672 So if a recipient address works intermittently, it can get repeatedly
673 deferred at routing time. The retry hints record remains fresh so the
674 address never reaches the final cutoff time.
676 This is a fairly common occurrence when a user is bumping up against
677 their storage quota. Exim had some logic in its local delivery code
678 to deal with this. However it did not apply to per-recipient defers
679 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
681 This change adds a proper retry rule check during routing so that the
682 final cutoff time is checked against the message's age. We only do
683 this check if there is an address retry record and there is not a
684 domain retry record; this implies that previous attempts to handle
685 the address had the retry_use_local_parts option turned on. We use
686 this as an approximation for the destination being like a local
687 delivery, as in LMTP.
689 I suspect this new check makes the old local delivery cutoff check
690 redundant, but I have not verified this so I left the code in place.
692 TF/02 Correct gecos expansion when From: is a prefix of the username.
694 Test 0254 submits a message to Exim with the header
698 When I ran the test suite under the user fanf2, Exim expanded
699 the header to contain my full name, whereas it should have added
700 a Resent-Sender: header. It erroneously treats any prefix of the
701 username as equal to the username.
703 This change corrects that bug.
705 GF/01 DCC debug and logging tidyup
706 Error conditions log to paniclog rather than rejectlog.
707 Debug lines prefixed by "DCC: " to remove any ambiguity.
709 TF/03 Avoid unnecessary rebuilds of lookup-related code.
711 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
712 Bug spotted by Jeremy Harris; was flawed since initial commit.
713 Would have resulted in OCSP responses post-SNI triggering an Exim
714 NULL dereference and crash.
716 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
718 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
719 Bug detection, analysis and fix by Samuel Thibault.
720 Bugzilla 1331, Debian bug #698092.
722 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
724 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
725 Server implementation by Todd Lyons, client by JH.
726 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
727 config variable "prdr_enable" controls whether the server
728 advertises the facility. If the client requests PRDR a new
729 acl_data_smtp_prdr ACL is called once for each recipient, after
730 the body content is received and before the acl_smtp_data ACL.
731 The client is controlled by bolth of: a hosts_try_prdr option
732 on the smtp transport, and the server advertisement.
733 Default client logging of deliveries and rejections involving
734 PRDR are flagged with the string "PRDR".
736 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
737 fclose(). Diagnosis by Todd Lyons.
739 PP/17 Update configure.default to handle IPv6 localhost better.
740 Patch by Alain Williams (plus minor tweaks).
743 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
744 This is now consistent with GnuTLS, and is now documented: the
745 previous undocumented portable approach to treating the option as
746 unset was to force an expansion failure. That still works, and
747 an empty string is now equivalent.
749 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
750 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
751 not performing validation itself.
753 PP/20 Added force_command boolean option to pipe transport.
754 Patch from Nick Koston, of cPanel Inc.
756 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
759 TF/04 Added udpsend ACL modifer and hexquote expansion operator
761 PP/21 Fix eximon continuous updating with timestamped log-files.
762 Broken in a format-string cleanup in 4.80, missed when I repaired the
763 other false fix of the same issue.
764 Report and fix from Heiko Schlichting.
767 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
768 Report from Prashanth Katuri.
770 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
771 It's SecureTransport, so affects any MacOS clients which use the
772 system-integrated TLS libraries, including email clients.
774 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
775 using a MIME ACL for non-SMTP local injection.
776 Report and assistance in diagnosis by Warren Baker.
778 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
780 JH/16 Fix comparisons for 64b. Bugzilla 1385.
782 TL/09 Add expansion variable $authenticated_fail_id to keep track of
783 last id that failed so it may be referenced in subsequent ACL's.
785 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
788 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
789 ldap library initialization, allowing self-signed CA's to be
790 used. Also properly sets require_cert option later in code by
791 using NULL (global ldap config) instead of ldap handle (per
792 session). Bug diagnosis and testing by alxgomz.
794 TL/12 Enhanced documentation in the ratelimit.pl script provided in
795 the src/util/ subdirectory.
797 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
798 renamed to Transport Post Delivery Action by Jeremy Harris, as
801 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
802 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
803 redis_servers = needs to be configured which will be used by the redis
804 lookup. Patch from Warren Baker, of The Packet Hub.
806 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
808 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
809 hostname or reverse DNS when processing a host list. Used suggestions
810 from multiple comments on this bug.
812 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
814 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
815 Missed a few lines, added it to make the runtest require no keyboard
818 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
819 contains upper case chars. Make router use caseful_local_part.
821 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
822 support when GnuTLS has been built with p11-kit.
828 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
830 This, or similar/improved, will also be change PP/11 of 4.82.
836 PP/01 Handle short writes when writing local log-files.
837 In practice, only affects FreeBSD (8 onwards).
838 Bugzilla 1053, with thanks to Dmitry Isaikin.
840 NM/01 Bugzilla 949 - Documentation tweak
842 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
845 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
847 PP/02 Implemented gsasl authenticator.
849 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
851 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
852 `pkg-config foo` for cflags/libs.
854 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
855 with rest of GSASL and with heimdal_gssapi.
857 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
858 `pkg-config foo` for cflags/libs for the TLS implementation.
860 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
861 properties get this fed in as external SSF. A number of robustness
862 and debugging improvements to the cyrus_sasl authenticator.
864 PP/08 cyrus_sasl server now expands the server_realm option.
866 PP/09 Bugzilla 1214 - Log authentication information in reject log.
867 Patch by Jeremy Harris.
869 PP/10 Added dbmjz lookup type.
871 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
873 PP/12 MAIL args handles TAB as well as SP, for better interop with
874 non-compliant senders.
875 Analysis and variant patch by Todd Lyons.
877 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
878 Bug report from Lars Müller <lars@samba.org> (via SUSE),
879 Patch from Dirk Mueller <dmueller@suse.com>
881 PP/13 tls_peerdn now print-escaped for spool files.
882 Observed some $tls_peerdn in wild which contained \n, which resulted
883 in spool file corruption.
885 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
886 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
887 or write after TLS renegotiation, which otherwise led to messages
890 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
891 as a tracking header (ie: a signed header comes before the signature).
892 Patch from Wolfgang Breyha.
894 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
895 comma-sep list; embedded commas doubled.
897 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
899 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
901 Report and patch from Dmitry Banschikov.
903 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
904 Removed SSL_clear() after SSL_new() which led to protocol negotiation
905 failures. We appear to now support TLS1.1+ with Exim.
907 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
908 lets Exim select keys and certificates based upon TLS SNI from client.
909 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
910 before an outbound SMTP session. New log_selector, +tls_sni.
912 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
913 NULL dereference. Report and patch from Alun Jones.
915 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
916 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
917 Not seeing resolver debug output on NetBSD, but suspect this is a
918 resolver implementation change.
920 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
921 Left warnings. Added "eximon gdb" invocation mode.
923 PP/21 Defaulting "accept_8bitmime" to true, not false.
925 PP/22 Added -bw for inetd wait mode support.
927 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
928 locate the relevant includes and libraries. Made this the default.
930 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
931 Bugzilla 1246, report and most of solution from Tomasz Kusy.
933 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
934 This may cause build issues on older platforms.
936 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
937 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
938 gnutls_require_mac & gnutls_require_protocols (no longer supported).
939 Added SNI support via GnuTLS too.
940 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
942 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
944 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
945 Applied second patch from Janne, implementing suggestion to default
946 multiple-strings-in-record handling to match SPF spec.
948 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
950 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
951 read-only, out of scope).
952 Patch from Wolfgang Breyha, report from Stuart Northfield.
954 PP/29 Fix three issues highlighted by clang analyser static analysis.
955 Only crash-plausible issue would require the Cambridge-specific
956 iplookup router and a misconfiguration.
957 Report from Marcin Mirosław.
959 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
961 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
962 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
963 As part of this, removing so much warning spew let me fix some minor
964 real issues in debug logging.
966 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
967 assignment on my part. Fixed.
969 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
970 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
971 Janne Snabb (who went above and beyond: thank you).
973 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
974 string otherwise requires a connection and a bunch more work and it's
975 relatively easy to get wrong. Should also expose TLS library linkage
978 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
979 64-bit ${eval} (JH/03).
981 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
982 GNU libc to support some of the 64-bit stuff, should not lead to
983 conflicts. Defined before os.h is pulled in, so if a given platform
984 needs to override this, it can.
986 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
987 protection layer was required, which is not implemented.
988 Bugzilla 1254, patch from Wolfgang Breyha.
990 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
991 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
992 tls_dhparam take prime identifiers. Also unbreak combination of
993 OpenSSL+DH_params+TLSSNI.
995 PP/39 Disable SSLv2 by default in OpenSSL support.
1001 PP/01 Solaris build fix for Oracle's LDAP libraries.
1002 Bugzilla 1109, patch from Stephen Usher.
1004 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1006 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1009 TF/02 Fix a couple more cases where we did not log the error message
1010 when unlink() failed. See also change 4.74-TF/03.
1012 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1013 lock up or crash if it happened to be inside a call to libc when it
1014 got a SIGUSR1 from exiwhat.
1016 The SIGUSR1 handler appends the current process status to the process
1017 log which is later printed by exiwhat. It used to use the general
1018 purpose logging code to do this, but several functions it calls are
1019 not safe for signals.
1021 The new output code in the SIGUSR1 handler is specific to the process
1022 log, and simple enough that it's easy to inspect for signal safety.
1023 Removing some special cases also simplifies the general logging code.
1024 Removing the spurious timestamps from the process log simplifies
1027 TF/04 Improved ratelimit ACL condition.
1029 The /noupdate option has been deprecated in favour of /readonly which
1030 has clearer semantics. The /leaky, /strict, and /readonly update modes
1031 are mutually exclusive. The update mode is no longer included in the
1032 database key; it just determines when the database is updated. (This
1033 means that when you upgrde Exim will forget old rate measurements.)
1035 Exim now checks that the per_* options are used with an update mode that
1036 makes sense for the current ACL. For example, when Exim is processing a
1037 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1038 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1039 must specify per_mail/readonly. If you omit the update mode it defaults to
1040 /leaky where that makes sense (as before) or /readonly where required.
1042 The /noupdate option is now undocumented but still supported for
1043 backwards compatibility. It is equivalent to /readonly except that in
1044 ACLs where /readonly is required you may specify /leaky/noupdate or
1045 /strict/noupdate which are treated the same as /readonly.
1047 A useful new feature is the /count= option. This is a generalization
1048 of the per_byte option, so that you can measure the throughput of other
1049 aggregate values. For example, the per_byte option is now equivalent
1050 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1052 The per_rcpt option has been generalized using the /count= mechanism
1053 (though it's more complicated than the per_byte equivalence). When it is
1054 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1055 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1056 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1057 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1058 Note that using per_rcpt with a non-readonly update mode in more than
1059 one ACL will cause the recipients to be double-counted. (The per_mail
1060 and per_byte options don't have this problem.)
1062 The handling of very low rates has changed slightly. If the computed rate
1063 is less than the event's count (usually one) then this event is the first
1064 after a long gap. In this case the rate is set to the same as this event's
1065 count, so that the first message of a spam run is counted properly.
1067 The major new feature is a mechanism for counting the rate of unique
1068 events. The new per_addr option counts the number of different
1069 recipients that someone has sent messages to in the last time period. It
1070 behaves like per_rcpt if all the recipient addresses are different, but
1071 duplicate recipient addresses do not increase the measured rate. Like
1072 the /count= option this is a general mechanism, so the per_addr option
1073 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1074 example, measure the rate that a client uses different sender addresses
1075 with the options per_mail/unique=$sender_address. There are further
1076 details in the main documentation.
1078 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1080 TF/06 Removed a few PCRE remnants.
1082 TF/07 Automatically extract Exim's version number from tags in the git
1083 repository when doing development or release builds.
1085 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1086 Bugzilla 879. Patch from Paul Fisher.
1088 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1089 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1092 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1094 PP/05 Variable $av_failed, true if the AV scanner deferred.
1095 Bugzilla 1078. Patch from John Horne.
1097 PP/06 Stop make process more reliably on build failure.
1098 Bugzilla 1087. Patch from Heiko Schlittermann.
1100 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1101 Bugzilla 1089. Patch from Heiko Schlittermann.
1103 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1104 Bugzilla 1131. Patch from Holger Weiß.
1106 PP/09 Handle IPv6 addresses with SPF.
1107 Bugzilla 860. Patch from Wolfgang Breyha.
1109 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1111 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1114 PP/11 match_* no longer expand right-hand-side by default.
1115 New compile-time build option, EXPAND_LISTMATCH_RHS.
1116 New expansion conditions, "inlist", "inlisti".
1118 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1120 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1122 PP/14 fix log_write() format string regression from TF/03.
1123 Bugzilla 1152. Patch from Dmitry Isaikin.
1129 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1131 PP/02 Harmonised TLS library version reporting; only show if debugging.
1132 Layout now matches that introduced for other libraries in 4.74 PP/03.
1134 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1136 PP/04 New "dns_use_edns0" global option.
1138 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1141 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1142 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1144 TK/01 Updated PolarSSL code to 0.14.2.
1145 Bugzilla 1097. Patch from Andreas Metzler.
1147 PP/07 Catch divide-by-zero in ${eval:...}.
1148 Fixes bugzilla 1102.
1150 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1153 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1154 format-string attack -- SECURITY: remote arbitrary code execution.
1156 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1157 time unintentionally subject to list matching rules, letting the header
1158 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1159 arbitrary string expansion). This allowed for information disclosure.
1161 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1162 INT_MIN/-1 -- value coerced to INT_MAX.
1168 NM/01 Workround for PCRE version dependency in version reporting
1171 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1172 This fixes portability to compilers other than gcc, notably
1173 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1175 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1176 makefiles for portability to HP-UX and POSIX correctness.
1178 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1179 Also via indented variable definition in the Makefile.
1180 (Debugging by Oliver Heesakkers).
1182 PP/02 Restore caching of spamd results with expanded spamd_address.
1183 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1185 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1186 Improves build reliability. Fix from: Frank Elsner
1188 NM/02 Fix wide character breakage in the rfc2047 coding
1189 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1191 NM/03 Allow underscore in dnslist lookups
1192 Fixes bug 1026. Patch from Graeme Fowler
1194 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1195 Code patches from Adam Ciarcinski of NetBSD.
1197 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1200 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1201 is logged, to avoid truncation. Patch from John Horne.
1203 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1204 Patch from Jakob Hirsch.
1206 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1207 SQL string expansion failure details.
1208 Patch from Andrey Oktyabrski.
1210 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1211 Patch from Simon Arlott.
1213 PP/09 New lookups functionality failed to compile on old gcc which rejects
1214 extern declarations in function scope.
1215 Patch from Oliver Fleischmann
1217 PP/10 Use sig_atomic_t for flags set from signal handlers.
1218 Check getgroups() return and improve debugging.
1219 Fixed developed for diagnosis in bug 927 (which turned out to be
1222 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1223 Patch from Mark Zealey.
1225 PP/12 Bugzilla 1056: Improved spamd server selection.
1226 Patch from Mark Zealey.
1228 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1229 Based on patch from Heiko Schlittermann.
1231 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1232 Patch from Uwe Doering, sign-off by Michael Haardt.
1234 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1235 variable declaration deep within a block. Bug and patch from
1238 PP/15 lookups-Makefile IRIX compatibilty coercion.
1240 PP/16 Make DISABLE_DKIM build knob functional.
1242 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1243 Patch by Simon Arlott
1245 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1246 variable argument macros. Our copy now differs from upstream.
1252 TF/01 Failure to get a lock on a hints database can have serious
1253 consequences so log it to the panic log.
1255 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1256 controlled using the smtp_confirmation log selector.
1258 TF/03 Include the error message when we fail to unlink a spool file.
1260 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1261 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1262 for maintaining out-of-tree patches for some time.
1264 PP/01 Bugzilla 139: Documentation and portability issues.
1265 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1266 Handle per-OS dynamic-module compilation flags.
1268 PP/02 Let /dev/null have normal permissions.
1269 The 4.73 fixes were a little too stringent and complained about the
1270 permissions on /dev/null. Exempt it from some checks.
1271 Reported by Andreas M. Kirchwitz.
1273 PP/03 Report version information for many libraries, including
1274 Exim version information for dynamically loaded libraries. Created
1275 version.h, now support a version extension string for distributors
1276 who patch heavily. Dynamic module ABI change.
1278 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1279 privilege escalation vulnerability whereby the Exim run-time user
1280 can cause root to append content of the attacker's choosing to
1283 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1286 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1287 If dropping privileges for untrusted macros, we disabled normal logging
1288 on the basis that it would fail; for the Exim run-time user, this is not
1289 the case, and it resulted in successful deliveries going unlogged.
1290 Fixed. Reported by Andreas Metzler.
1296 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1297 only prepend for the Resent-* case. Fixes regression introduced in
1298 Exim 4.70 by NM/22 for Bugzilla 607.
1300 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1301 increasing numbers of administrators be bitten by this.
1303 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1305 PP/03 Bugzilla 994: added openssl_options main configuration option.
1307 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1309 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1311 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1313 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1314 without a peer certificate, leading to a segfault because of an
1315 assumption that peers always have certificates. Be a little more
1316 paranoid. Problem reported by Martin Tscholak.
1318 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1319 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1320 NB: ClamAV planning to remove STREAM in "middle of 2010".
1321 CL also introduces -bmalware, various -d+acl logging additions and
1322 more caution in buffer sizes.
1324 PP/09 Implemented reverse_ip expansion operator.
1326 PP/10 Bugzilla 937: provide a "debug" ACL control.
1328 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1330 PP/12 Bugzilla 973: Implement --version.
1332 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1334 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1336 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1338 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1339 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1340 ignore trailing whitespace.
1342 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1344 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1347 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1348 Notification from Dr Andrew Aitchison.
1350 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1351 ExtendedDetectionInfo response format.
1352 Notification from John Horne.
1354 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1357 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1358 XSL and documented dependency on system catalogs, with examples of how
1361 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1364 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1365 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1366 configuration file which is writeable by the Exim user or group.
1368 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1369 of configuration files to cover files specified with the -C option if
1370 they are going to be used with root privileges, not just the default
1373 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1374 option (effectively making it always true).
1376 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1377 files to be used while preserving root privileges.
1379 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1380 that rogue child processes cannot use them.
1382 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1383 run-time user, instead of root.
1385 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1386 Exim run-time user without dropping privileges.
1388 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1389 result string, instead of calling string_vformat() twice with the same
1392 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1393 for other users. Others should always drop root privileges if they use
1394 -C on the command line, even for a whitelisted configure file.
1396 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1398 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1404 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1405 $data_path, and $header_path variables; fixed documentation bugs and
1408 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1409 exipick to access non-standard spools, including the "frozen" queue
1412 NM/01 Bugzilla 965: Support mysql stored procedures.
1413 Patch from Alain Williams
1415 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1417 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1418 Patch from Andreas Metzler
1420 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1421 Patch from Kirill Miazine
1423 NM/05 Bugzilla 671: Added umask to procmail example.
1425 JJ/03 installed exipick 20100323.0, fixing doc bug
1427 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1428 directory. Notification and patch from Dan Rosenberg.
1430 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1432 TK/02 Improve log output when DKIM signing operation fails.
1434 MH/01 Treat the transport option dkim_domain as a colon separated
1435 list, not as a single string, and sign the message with each element,
1436 omitting multiple occurences of the same signer.
1438 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1439 Bugzilla 985, 986. Patch by Simon Arlott
1441 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1442 Patch by Simon Arlott
1444 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1445 MBX locking. Notification from Dan Rosenberg.
1451 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1453 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1455 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1457 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1459 NM/04 Bugzilla 915: Flag broken perl installation during build.
1465 TK/01 Added patch by Johannes Berg that expands the main option
1466 "spamd_address" if it starts with a dollar sign.
1468 TK/02 Write list of recipients to X-Envelope-Sender header when building
1469 the mbox-format spool file for content scanning (suggested by Jakob
1472 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1473 (http://www.dcc-servers.net/) support via dccifd. Activated by
1474 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1476 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1477 by Mark Daniel Reidel <mr@df.eu>.
1479 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1480 When building exim an external PCRE library is now needed -
1481 PCRE is a system library on the majority of modern systems.
1482 See entry on PCRE_LIBS in EDITME file.
1484 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1485 conversation. Added nologin parameter to request.
1486 Patch contributed by Kirill Miazine.
1488 TF/01 Do not log submission mode rewrites if they do not change the address.
1490 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1492 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1493 log files in place. Contributed by Roberto Lima.
1495 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1497 TF/03 Bugzilla 615: When checking the local_parts router precondition
1498 after a local_part_suffix or local_part_prefix option, Exim now
1499 does not use the address's named list lookup cache, since this
1500 contains cached lookups for the whole local part.
1502 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1503 Robert Millan. Documentation is in experimental-spec.txt.
1505 TF/04 Bugzilla 668: Fix parallel build (make -j).
1507 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1509 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1510 Patch provided by Jan Srzednicki.
1512 TF/05 Leading white space used to be stripped from $spam_report which
1513 wrecked the formatting. Now it is preserved.
1515 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1516 that they are available at delivery time.
1518 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1520 TF/08 TLS error reporting now respects the incoming_interface and
1521 incoming_port log selectors.
1523 TF/09 Produce a more useful error message if an SMTP transport's hosts
1524 setting expands to an empty string.
1526 NM/06 Bugzilla 744: EXPN did not work under TLS.
1527 Patch provided by Phil Pennock.
1529 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1530 Patch provided by Richard Godbee.
1532 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1533 acl_smtp_notquit, added index entry.
1535 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1536 Patch provided by Eugene Bujak.
1538 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1539 accept(). Patch provided by Maxim Dounin.
1541 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1542 Patch provided by Phil Pennock.
1544 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1546 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1547 Patch provided by Brad "anomie" Jorsch.
1549 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1550 Patch provided by Dean Brooks.
1552 TK/05 Add native DKIM support (does not depend on external libraries).
1554 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1555 Patch provided by Graeme Fowler.
1557 NM/16 Bugzilla 851: Documentation example syntax fix.
1559 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1561 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1564 NM/19 Bugzilla 745: TLS version reporting.
1565 Patch provided by Phil Pennock.
1567 NM/20 Bugzilla 167: bool: condition support.
1568 Patch provided by Phil Pennock.
1570 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1571 clients. Patch provided by Phil Pennock.
1573 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1574 Patch provided by Brad "anomie" Jorsch.
1576 NM/23 Bugzilla 687: Fix misparses in eximstats.
1577 Patch provided by Heiko Schlittermann.
1579 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1580 Patch provided by Heiko Schlittermann.
1582 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1583 plus update to original patch.
1585 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1587 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1588 Patch provided by David Brownlee.
1590 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1592 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1594 NM/30 Bugzilla 888: TLS documentation bugfixes.
1596 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1598 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1599 Unlike the original bugzilla I have changed all shell scripts in src tree.
1601 NM/33 Bugzilla 898: Transport filter timeout fix.
1602 Patch by Todd Rinaldo.
1604 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1605 Patch by Serge Demonchaux.
1607 NM/35 Bugzilla 39: Base64 decode bug fixes.
1608 Patch by Jakob Hirsch.
1610 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1612 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1614 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1616 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1622 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1623 ALT-N's libdkim that I have put here:
1624 http://duncanthrax.net/exim-experimental/
1626 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1627 were called 'true' and it seems that C99 defines that as a reserved
1628 keyword to be used with 'bool' variable types. That means you could
1629 not include C99-style headers which use bools without triggering
1630 build errors in sieve.c.
1632 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1633 as mailq or other aliases. Changed the --help handling significantly
1634 to do whats expected. exim_usage() emits usage/help information.
1636 SC/01 Added the -bylocaldomain option to eximstats.
1638 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1640 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1642 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1648 PH/01 Another patch from the Sieve maintainer.
1650 PH/02 When an IPv6 address is converted to a string for single-key lookup
1651 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1652 dots are used instead of colons so that keys in lsearch files need not
1653 contain colons. This was done some time before quoting was made available
1654 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1655 (notated using the quote facility) so as to distinguish them from IPv4
1656 keys. This meant that lookups for IP addresses in host lists did not work
1657 for iplsearch lookups.
1659 This has been fixed by arranging for IPv6 addresses to be expressed with
1660 colons if the lookup type is iplsearch. This is not incompatible, because
1661 previously such lookups could never work.
1663 The situation is now rather anomolous, since one *can* have colons in
1664 ordinary lsearch keys. However, making the change in all cases is
1665 incompatible and would probably break a number of configurations.
1667 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1670 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1671 conversion specification without a maximum field width, thereby enabling
1672 a rogue spamd server to cause a buffer overflow. While nobody in their
1673 right mind would setup Exim to query an untrusted spamd server, an
1674 attacker that gains access to a server running spamd could potentially
1675 exploit this vulnerability to run arbitrary code as the Exim user.
1677 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1678 $primary_hostname instead of what libspf2 thinks the hosts name is.
1680 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1681 a directory entry by the name of the lookup key. Previously, if a
1682 symlink pointed to a non-existing file or a file in a directory that
1683 Exim lacked permissions to read, a lookup for a key matching that
1684 symlink would fail. Now it is enough that a matching directory entry
1685 exists, symlink or not. (Bugzilla 503.)
1687 PH/03 The body_linecount and body_zerocount variables are now exported in the
1690 PH/04 Added the $dnslist_matched variable.
1692 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1693 This means they are set thereafter only if the connection becomes
1696 PH/06 Added the client_condition to authenticators so that some can be skipped
1697 by clients under certain conditions.
1699 PH/07 The error message for a badly-placed control=no_multiline_responses left
1700 "_responses" off the end of the name.
1702 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1704 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1705 (without spaces) instead of just copying the configuration text.
1707 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1709 PH/11 Added $max_received_linelength.
1711 PH/12 Added +ignore_defer and +include_defer to host lists.
1713 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1714 way in which PCRE > 7.0 is built.
1716 PH/14 Implemented queue_only_load_latch.
1718 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1719 MAIL command. The effect was to mangle the value on 64-bit systems.
1721 PH/16 Another patch from the Sieve maintainer.
1723 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1725 PH/18 If a system quota error occurred while trying to create the file for
1726 a maildir delivery, the message "Mailbox is full" was not appended to the
1727 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1728 only to a quota excession during the actual writing of the file.
1730 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1731 characters?) which causes problems in log lines. The DN values are now
1732 passed through string_printing() before being added to log lines.
1734 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1735 and InterBase are left for another time.)
1737 PH/21 Added message_body_newlines option.
1739 PH/22 Guard against possible overflow in moan_check_errorcopy().
1741 PH/23 POSIX allows open() to be a macro; guard against that.
1743 PH/24 If the recipient of an error message contained an @ in the local part
1744 (suitably quoted, of course), incorrect values were put in $domain and
1745 $local_part during the evaluation of errors_copy.
1751 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1752 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1755 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1756 issue a MAIL command.
1758 PH/02 In an ACL statement such as
1760 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1762 if a client was not listed at all, or was listed with a value other than
1763 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1764 the condition was not true (as it should be), so access was not denied.
1765 The bug was that the ! inversion was incorrectly passed on to the second
1766 item. This has been fixed.
1768 PH/03 Added additional dnslists conditions == and =& which are different from
1769 = and & when the dns lookup returns more than one IP address.
1771 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1772 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1774 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1775 FSYNC, which compiles an option called disable_fsync that allows for
1776 bypassing fsync(). The documentation is heavily laced with warnings.
1778 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1780 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1781 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1782 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1783 including adding "make clean"; (3) Added -fPIC when compiling the test
1784 dynamically loaded module, to get rid of a warning.
1786 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1787 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1788 = 0s. The bug is otherwise harmless.
1790 PH/07 There was a bug in the dovecot authenticator such that the value of
1791 $auth1 could be overwritten, and so not correctly preserved, after a
1792 successful authentication. This usually meant that the value preserved by
1793 the server_setid option was incorrect.
1795 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1797 PH/09 Installed PCRE release 7.0.
1799 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1800 run for batched SMTP input. It is now run at the start of every message
1801 in the batch. While fixing this I discovered that the process information
1802 (output by running exiwhat) was not always getting set for -bs and -bS
1803 input. This is fixed, and it now also says "batched" for BSMTP.
1805 PH/11 Added control=no_pipelining.
1807 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1808 patch, slightly modified), and move the expansion of helo_data till after
1809 the connection is made in the smtp transport (so it can use these
1812 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1814 PH/14 Added log_selector = +pid.
1816 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1818 PH/16 Add ${if forany and ${if forall.
1820 PH/17 Added dsn_from option to vary the From: line in DSNs.
1822 PH/18 Flush SMTP output before performing a callout, unless control =
1823 no_callout_flush is set.
1825 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1826 was true (the default) a successful delivery failed to delete the retry
1827 item, thus causing premature timeout of the address. The bug is now
1830 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1832 PH/21 Long custom messages for fakedefer and fakereject are now split up
1833 into multiline reponses in the same way that messages for "deny" and
1834 other ACL rejections are.
1836 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1837 with slight modification.
1839 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1840 draft, changing the syntax and factoring some duplicate code".
1842 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1843 for deliveries of the second and subsequent messages over the same SMTP
1846 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1847 ${reduce, with only minor "tidies".
1849 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1851 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1852 expansion side effects.
1854 PH/27 When a message times out after an over-quota error from an Exim-imposed
1855 quota, the bounce message says "mailbox is full". This message was not
1856 being given when it was a system quota that was exceeded. It now should
1859 MH/03 Made $recipients available in local_scan(). local_scan() already has
1860 better access to the recipient list through recipients_list[], but
1861 $recipients can be useful in postmaster-provided expansion strings.
1863 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1864 in the case of a MAIL command with additional options following the
1865 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1866 were accidentally chopped off.
1868 PH/29 SMTP synchronization checks are implemented when a command is read -
1869 there is a check that no more input is waiting when there shouldn't be
1870 any. However, for some commands, a delay in an ACL can mean that it is
1871 some time before the response is written. In this time, more input might
1872 arrive, invalidly. So now there are extra checks after an ACL has run for
1873 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1874 pipelining has not been advertised.
1876 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1878 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1879 returned in the SMTP response when smtp_return_error_details was false.
1880 This has been fixed.
1882 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1883 instead of the C I/O that was originally supplied, because problems were
1884 reported on Solaris.
1886 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1887 Exim which did not show up earlier: it was assuming that a call to
1888 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1889 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1890 was a macro that became an assignment, so it seemed to work. This has
1891 changed to a proper function call with a void return, hence the compile
1892 error. Exim's code has been fixed.
1894 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1897 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1899 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1901 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1902 because it turns out to be too big or there is a timeout) there is no
1903 "Completed" line in the log. When some messages of this type were
1904 selected by exigrep, they were listed as "not completed". Others were
1905 picked up by some special patterns. I have improved the selection
1906 criteria to be more general.
1908 PH/38 The host_find_failed option in the manualroute router can now be set
1909 to "ignore", to completely ignore a host whose IP address cannot be
1910 found. If all hosts are ignored, the behaviour is controlled by the new
1911 host_all_ignored option.
1913 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1914 homing or because of multiple MX records with /mx) generated more than
1915 one IP address, and the following item turned out to be the local host,
1916 all the secondary addresses of the first item were incorrectly removed
1917 from the list, along with the local host and any following hosts (which
1918 is what is supposed to happen).
1920 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1921 whoever called Exim into the -H file. In the case of the daemon it was
1922 behaving confusingly. When first started, it used values for whoever
1923 started the daemon, but after a SIGHUP it used the Exim user (because it
1924 calls itself on a restart). I have changed the code so that it now always
1927 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1928 message are rejected with the same error (e.g. no authentication or bad
1929 sender address), and a DATA command is nevertheless sent (as can happen
1930 with PIPELINING or a stupid MUA), the error message that was given to the
1931 RCPT commands is included in the rejection of the DATA command. This is
1932 intended to be helpful for MUAs that show only the final error to their
1935 PH/42 Another patch from the Sieve maintainer.
1937 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1938 Eximstats - Fixed some broken HTML links and added missing column headers
1940 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1941 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1943 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1945 PH/43 Yet another patch from the Sieve maintainer.
1947 PH/44 I found a way to check for a TCP/IP connection going away before sending
1948 the response to the final '.' that terminates a message, but only in the
1949 case where the client has not sent further data following the '.'
1950 (unfortunately, this is allowed). However, in many cases there won't be
1951 any further data because there won't be any more messages to send. A call
1952 to select() can be used: if it shows that the input is "ready", there is
1953 either input waiting, or the socket has been closed. An attempt to read
1954 the next input character can distinguish the two cases. Previously, Exim
1955 would have sent an OK response which the client would never have see.
1956 This could lead to message repetition. This fix should cure that, at
1957 least in a lot of common cases.
1959 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1960 advertised in response to EHLO.
1966 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1967 fixed by 4.65/MH/01 (is this a record?) are fixed:
1969 (i) An empty string was always treated as zero by the numeric comparison
1970 operators. This behaviour has been restored.
1972 (ii) It is documented that the numeric comparison operators always treat
1973 their arguments as decimal numbers. This was broken in that numbers
1974 starting with 0 were being interpreted as octal.
1976 While fixing these problems I realized that there was another issue that
1977 hadn't been noticed. Values of message_size_limit (both the global option
1978 and the transport option) were treated as octal if they started with 0.
1979 The documentation was vague. These values are now always treated as
1980 decimal, and I will make that clear in the documentation.
1986 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1987 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1990 MH/01 Don't check that the operands of numeric comparison operators are
1991 integers when their expansion is in "skipping" mode (fixes bug
1992 introduced by 4.64-PH/07).
1994 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1995 child addresses, Exim now panics and dies. Previously, because the count
1996 is held in a short int, deliveries were likely to be lost. As such a
1997 large number of recipients for a single message is ridiculous
1998 (performance will be very, very poor), I have chosen to impose a limit
1999 rather than extend the field.
2005 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2006 leftover -K file (the existence of which was triggered by #402).
2007 While we were at it, introduced process PID as part of the -K
2008 filename. This should rule out race conditions when creating
2011 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2012 processing considerably. Previous code took too long for large mails,
2013 triggering a timeout which in turn triggers #401.
2015 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2016 in the DK code in transports.c. sendfile() is not really portable,
2017 hence the _LINUX specificness.
2019 TF/01 In the add_headers option to the mail command in an Exim filter,
2020 there was a bug that Exim would claim a syntax error in any
2021 header after the first one which had an odd number of characters
2024 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2025 callout verification, Exim cached a "reject" for the entire domain. This
2026 is correct for most verifications, but it is not correct for a recipient
2027 verification with use_sender or use_postmaster set, because in that case
2028 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2029 case of MAIL FROM:<> rejection from other early rejections (e.g.
2030 rejection of HELO). When verifying a recipient using a non-null MAIL
2031 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2032 Whatever the result of the callout, the value of the domain cache is
2033 left unchanged (for any other kind of callout, getting as far as trying
2034 RCPT means that the domain itself is ok).
2036 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2039 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2040 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2041 session, the daemon ignores ECONNECT errors and logs others; it now
2042 ignores EPIPE as well.
2044 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2045 (quoted-printable decoding).
2047 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2048 later the small subsequent patch to fix an introduced bug.
2050 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2052 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2054 PH/08 An error is now given if message_size_limit is specified negative.
2056 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2057 to be given (somewhat) arbitrary names.
2059 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2062 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2063 miscellaneous code fixes
2065 PH/10 Added the log_reject_target ACL modifier to specify where to log
2068 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2069 hostname. This is wrong, because it relates to the incoming message (and
2070 probably the interface on which it is arriving) and not to the outgoing
2071 callout (which could be using a different interface). This has been
2072 changed to use the value of the helo_data option from the smtp transport
2073 instead - this is what is used when a message is actually being sent. If
2074 there is no remote transport (possible with a router that sets up host
2075 addresses), $smtp_active_hostname is used.
2077 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2078 tweaks were necessary in order to get it to work (see also 21 below):
2079 (a) The code assumed that strncpy() returns a negative number on buffer
2080 overflow, which isn't the case. Replaced with Exim's string_format()
2082 (b) There were several signed/unsigned issues. I just did the minimum
2083 hacking in of casts. There is scope for a larger refactoring.
2084 (c) The code used strcasecmp() which is not a standard C function.
2085 Replaced with Exim's strcmpic() function.
2086 (d) The code set only $1; it now sets $auth1 as well.
2087 (e) A simple test gave the error "authentication client didn't specify
2088 service in request". It would seem that Dovecot has changed its
2089 interface. Fortunately there's a specification; I followed it and
2090 changed what the client sends and it appears to be working now.
2092 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2095 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2096 address A is aliased to B and C, where B exists and C does not. Without
2097 -v the output is "A verified" because verification stops after a
2098 successful redirection if more than one address is generated. However,
2099 with -v the child addresses are also verified. Exim was outputting "A
2100 failed to verify" and then showing the successful verification for C,
2101 with its parentage. It now outputs "B failed to verify", showing B's
2102 parentage before showing the successful verification of C.
2104 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2105 look up a TXT record in a specific list after matching in a combined
2108 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2109 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2110 they consult the DNS. I had assumed they would set it the way they
2111 wanted; and indeed my experiments on Linux seem to show that in some
2112 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2113 To be on the safe side, however, I have now made the interface to
2114 host_find_byname() similar to host_find_bydns(), with an argument
2115 containing the DNS resolver options. The host_find_byname() function now
2116 sets these options at its start, just as host_find_bydns() does. The smtp
2117 transport options dns_qualify_single and dns_search_parents are passed to
2118 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2119 of host_find_byname() use the default settings of RES_DEFNAMES
2120 (qualify_single) but not RES_DNSRCH (search_parents).
2122 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2123 spool_read_header() do less string testing, by means of a preliminary
2124 switch on the second character of optional "-foo" lines. (This is
2125 overdue, caused by the large number of possibilities that now exist.
2126 Originally there were few.) While I was there, I also converted the
2127 str(n)cmp tests so they don't re-test the leading "-" and the first
2128 character, in the hope this might squeeze out yet more improvement.
2130 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2131 flag allowing group syntax was set by the header_syntax check but not
2132 turned off, possible causing trouble later; (2) The flag was not being
2133 set at all for the header_verify test, causing "group"-style headers to
2134 be rejected. I have now set it in this case, and also caused header_
2135 verify to ignore an empty address taken from a group. While doing this, I
2136 came across some other cases where the code for allowing group syntax
2137 while scanning a header line wasn't quite right (mostly, not resetting
2138 the flag correctly in the right place). These bugs could have caused
2139 trouble for malformed header lines. I hope it is now all correct.
2141 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2142 with the "reply" argument non-NULL. The code, however (which originally
2143 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2144 but it didn't always do it. This confused somebody who was copying the
2145 code for some other use. I have removed all the tests.
2147 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2148 feature that was used to support insecure browsers during the U.S. crypto
2149 embargo. It requires special client support, and Exim is probably the
2150 only MTA that supported it -- and would never use it because real RSA is
2151 always available. This code has been removed, because it had the bad
2152 effect of slowing Exim down by computing (never used) parameters for the
2153 RSA_EXPORT functionality.
2155 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2156 authenticator to fail if there's a tab character in the incoming data
2157 (there should never be unless someone is messing about, as it's supposed
2158 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2159 if the connection is using TLS or if the remote IP is the same as the
2160 local IP, and the "valid-client-cert option" if a client certificate has
2163 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2164 authenticators. This can be used for authorization after authentication
2165 succeeds. (In the case of plaintext, it servers for both authentication
2168 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2169 if any retry times were supplied.
2171 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2172 connection, where there is no client IP address to check. In this
2173 situation, the verify now always succeeds.
2175 PH/25 Applied John Jetmore's -Mset patch.
2177 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2179 PH/27 In a string expansion for a processed (not raw) header when multiple
2180 headers of the same name were present, leading whitespace was being
2181 removed from all of them, but trailing whitespace was being removed only
2182 from the last one. Now trailing whitespace is removed from each header
2183 before concatenation. Completely empty headers in a concatenation (as
2184 before) are ignored.
2186 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2187 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2189 PH/29 [Removed. This was a change that I later backed out, and forgot to
2190 correct the ChangeLog entry (that I had efficiently created) before
2191 committing the later change.]
2193 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2194 address errors (4xx response to RCPT) over the same connection as other
2195 messages routed to the same hosts. Such deliveries are always "forced",
2196 so retry times are not inspected. This resulted in far too many retries
2197 for the affected addresses. The effect occurred only when there were more
2198 hosts than the hosts_max_try setting in the smtp transport when it had
2199 the 4xx errors. Those hosts that it had tried were not added to the list
2200 of hosts for which the message was waiting, so if all were tried, there
2201 was no problem. Two fixes have been applied:
2203 (i) If there are any address or message errors in an SMTP delivery, none
2204 of the hosts (tried or untried) are now added to the list of hosts
2205 for which the message is waiting, so the message should not be a
2206 candidate for sending over the same connection that was used for a
2207 successful delivery of some other message. This seems entirely
2208 reasonable: after all the message is NOT "waiting for some host".
2209 This is so "obvious" that I'm not sure why it wasn't done
2210 previously. Hope I haven't missed anything, but it can't do any
2211 harm, as the worst effect is to miss an optimization.
2213 (ii) If, despite (i), such a delivery is accidentally attempted, the
2214 routing retry time is respected, so at least it doesn't keep
2215 hammering the server.
2217 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2218 in ${readsocket because some servers need this prod.
2220 PH/32 Added some extra debug output when updating a wait-xxx database.
2222 PH/33 The hint "could be header name not terminated by colon", which has been
2223 given for certain expansion errors for a long time, was not being given
2224 for the ${if def:h_colon_omitted{... case.
2226 PH/34 The spec says: "With one important exception, whenever a domain list is
2227 being scanned, $domain contains the subject domain." There was at least
2228 one case where this was not true.
2230 PH/35 The error "getsockname() failed: connection reset by peer" was being
2231 written to the panic log as well as the main log, but it isn't really
2232 panic-worthy as it just means the connection died rather early on. I have
2233 removed the panic log writing for the ECONNRESET error when getsockname()
2236 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2237 runs only) independently of the message's sender address. This meant
2238 that, if the 4xx error was in fact related to the sender, a different
2239 message to the same recipient with a different sender could confuse
2240 things. In particualar, this can happen when sending to a greylisting
2241 server, but other circumstances could also provoke similar problems.
2242 I have changed the default so that the retry time for these errors is now
2243 based a combination of the sender and recipient addresses. This change
2244 can be overridden by setting address_retry_include_sender=false in the
2247 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2248 remote server are returned as part of bounce messages. This was not
2249 happening for LMTP over a pipe (the lmtp transport), but now it is the
2250 same for both kinds of LMTP.
2252 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2253 in header lines that were in fact CNAMEs. This is no longer the case.
2255 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2256 and queue runs started by the daemon processed all messages. This has
2257 been fixed so that -R and -S can now usefully be given with -q<time>.
2259 PH/40 Import PCRE release 6.7 (fixes some bugs).
2261 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2263 PH/42 Give an error if -q is specified more than once.
2265 PH/43 Renamed the variables $interface_address and $interface_port as
2266 $received_ip_address and $received_port, to make it clear that these
2267 values apply to message reception, and not to the outgoing interface when
2268 a message is delivered. (The old names remain recognized, of course.)
2270 PH/44 There was no timeout on the connect() call when using a Unix domain
2271 socket in the ${readsocket expansion. There now is.
2273 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2274 be meaningful with "accept".
2276 SC/01 Eximstats V1.43
2277 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2279 SC/02 Eximstats V1.44
2280 Use a glob alias rather than an array ref in the generated
2281 parser. This improves both readability and performance.
2283 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2284 Collect SpamAssassin and rejection statistics.
2285 Don't display local sender or destination tables unless
2286 there is data to show.
2287 Added average volumes into the top table text output.
2289 SC/04 Eximstats V1.46
2290 Collect data on the number of addresses (recipients)
2291 as well as the number of messages.
2293 SC/05 Eximstats V1.47
2294 Added 'Message too big' to the list of mail rejection
2295 reasons (thanks to Marco Gaiarin).
2297 SC/06 Eximstats V1.48
2298 Mainlog lines which have GMT offsets and are too short to
2299 have a flag are now skipped.
2301 SC/07 Eximstats V1.49 (Alain Williams)
2302 Added the -emptyok flag.
2304 SC/08 Eximstats V1.50
2305 Fixes for obtaining the IP address from reject messages.
2307 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2308 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2309 whitesspace changes from 4.64-PH/27
2311 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2314 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2315 are found, allow negative numbers in numeric criteria)
2317 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2319 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2320 to match changes made in 4.64-PH/43
2322 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2324 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2325 "domain" is actually an IP address, and force a failure. This locks out
2326 those revolvers/nameservers that support "A-for-A" lookups, in
2327 contravention of the specifications.
2329 PH/48 When a host name was looked up from an IP address, and the subsequent
2330 forward lookup of the name timed out, the host name was left in
2331 $sender_host_name, contrary to the specification.
2333 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2334 restricted to single-key lookups, Exim was not diagnosing an error if
2335 * or *@ was used with a query-style lookup.
2337 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2339 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2340 long ago, but noone interested enough thought of it. Let's just say that
2341 the "1.1" means that there are some new functions that weren't there at
2342 some point in the past.
2344 PH/51 Error processing for expansion failure of helo_data from an smtp
2345 transport during callout processing was broken.
2347 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2348 tested/used via the -bh/-bhc/-bs options.
2350 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2351 bug, fixed in subsequent PCRE releases).
2353 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2354 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2356 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2362 SC/01 Use a glob alias rather than an array ref in eximstats generated
2363 parser. This improves both readability and performance.
2365 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2366 Don't display local sender or destination tables in eximstats unless
2367 there is data to show.
2368 Added average volumes into the eximstats top table text output.
2370 SC/03 Collect data on the number of addresses (recipients) as well
2371 as the number of messages in eximstats.
2373 TF/01 Correct an error in the documentation for the redirect router. Exim
2374 does (usually) call initgroups() when daemonizing.
2376 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2377 with consistent privilege compared to when running as a daemon.
2379 TF/03 Note in the spec that $authenticated_id is not set for local
2380 submissions from trusted users.
2382 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2383 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2385 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2386 by adding some example configuration directives to the default
2387 configuration file. A little bit of work is required to uncomment the
2388 directives and define how usernames and passwords are checked, but
2389 there is now a framework to start from.
2391 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2392 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2393 without this. I don't know how relevant this is to other LDAP libraries.
2395 PH/02 Add the verb name to the "unknown ACL verb" error.
2397 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2399 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2401 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2402 directory not expanded when it should be if an expanded home directory
2403 was set for the address (which is overridden by the transport).
2405 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2408 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2409 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2410 because it is too late at that time, and has no effect.
2412 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2413 security issue with \' (bugzilla #107). I could not use the
2414 PQescapeStringConn() function, because it needs a PGconn value as one of
2417 PH/08 When testing addresses using -bt, indicate those final addresses that
2418 are duplicates that would not cause an additional delivery. At least one
2419 person was confused, thinking that -bt output corresponded to deliveries.
2420 (Suppressing duplicates isn't a good idea as you lose the information
2421 about possibly different redirections that led to the duplicates.)
2423 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2424 systems where poll() doesn't work, in particular OS X.
2426 PH/10 Added more information to debugging output for retry time not reached.
2428 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2429 operations in malware.c.
2431 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2434 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2435 syslog_duplication set false, log lines that would normally be written
2436 both the the main log and to the reject log were not written to syslog at
2439 PH/14 In the default configuration, change the use of "message" in ACL warn
2440 statements to "add_header".
2442 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2443 not followed by a command (e.g. "seen endif").
2445 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2446 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2449 PH/17 Added extra conditions to the default value of delay_warning_condition
2453 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2454 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2455 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2458 The Auto-Submitted: and various List- headers are standardised, whereas I
2459 don't think Precedence: ever was.
2461 PH/18 Refactored debugging code in route_finduser() to show more information,
2462 in particular, the error code if getpwnam() issues one.
2464 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2465 This is apparently needed in addition to the PH/07 change above to avoid
2466 any possible encoding problems.
2468 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2469 but not after initializing Perl.
2471 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2472 output them only if debugging. By default they are written stderr,
2473 apparently, which is not desirable.
2475 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2478 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2481 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2483 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2484 authenticated or an ident call has been made. Suppress the default
2485 values for $authenticated_id and $authenticated_sender (but permit -oMai
2486 and -oMas) when testing with -bh.
2488 PH/24 Re-jigged the order of the tests in the default configuration so that the
2489 tests for valid domains and recipients precede the DNS black list and CSA
2490 tests, on the grounds that those ones are more expensive.
2492 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2493 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2494 This bug exists in every version of Exim that I still have, right back to
2497 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2498 However, an attempt to turn on case-sensitivity in a regex key by
2499 including (?-i) didn't work because the subject string was already
2500 lowercased, and the effects were non-intuitive. It turns out that a
2501 one-line patch can be used to allow (?-i) to work as expected.
2507 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2508 other effects) broke the use of negated acl sub-conditions.
2510 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2513 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2514 "Deny" causes Exim to reject the incoming connection with a 554 error.
2515 Unfortunately, if there is a major crisis, such as a disk failure,
2516 tcp-wrappers gives "deny", whereas what one would like would be some
2517 kind of temporary error. A kludge has been added to help with this.
2518 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2519 554 error is used if errno is still zero or contains ENOENT (which occurs
2520 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2523 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2525 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2526 errors. Otherwise a message that provokes a temporary error (when other
2527 messages do not) can cause a whole host to time out.
2529 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2530 addresses were routed directly to files or pipes from a redirect router.
2531 File deliveries just didn't batch; pipe deliveries might have suffered
2534 PH/06 A failure to get a lock for a hints database would erroneously always say
2535 "Failed to get write lock", even when it was really a read lock.
2537 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2538 of 0600. This has been changed to use the value of the lockfile_mode
2539 option (which defaults to 0600).
2541 PH/08 Applied small patch from the Sieve maintainer.
2543 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2544 folder from quota calculations, a direct delivery into this folder messed
2545 up the contents of the maildirsize file. This was because the regex was
2546 used only to exclude .Trash (or whatever) when the size of the mailbox
2547 was calculated. There was no check that a delivery was happening into an
2548 excluded directory. This bug has been fixed by ignoring all quota
2549 processing for deliveries into excluded directories.
2551 PH/10 Added the maildirfolder_create_regex option to appendfile.
2557 PH/01 The code for finding all the local interface addresses on a FreeBSD
2558 system running IPv6 was broken. This may well have applied to all BSD
2559 systems, as well as to others that have similar system calls. The broken
2560 code found IPv4 interfaces correctly, but gave incorrect values for the
2561 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2562 that it would not match correctly against @[] and not recognize the IPv6
2565 PH/02 The ipliteral router was not recognizing addresses of the form user@
2566 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2568 PH/03 Added disable_ipv6.
2570 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2571 decoded form, because it is most often used to construct To: headers
2572 lines in autoreplies, and the decoded form may well be syntactically
2573 invalid. However, $reply_address has leading white space removed, and all
2574 newlines turned into spaces so that the autoreply transport does not
2577 PH/05 If group was specified without a user on a router, and no group or user
2578 was specified on a transport, the group from the router was ignored.
2580 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2581 for visible compile-time settings that can be used to change these
2582 numbers, for those that want even more. Backwards compatibility with old
2583 spool files has been maintained. However, going back to a previous Exim
2584 release will lost any variables that are in spool files.
2586 PH/07 Two small changes when running in the test harness: increase delay when
2587 passing a TCP/IP connection to a new process, in case the original
2588 process has to generate a bounce, and remove special handling of
2589 127.0.0.2 (sic), which is no longer necessary.
2591 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2592 be the same on different OS.
2594 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2597 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2598 whether --show-vars was specified or not
2600 JJ/02 exipick: Added support for new ACL variable spool format introduced
2603 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2604 syntactically invalid From: or Reply-to: line, and a filter used this to
2605 generate an autoreply, and therefore failed to obtain an address for the
2606 autoreply, Exim could try to deliver to a non-existent relative file
2607 name, causing unrelated and misleading errors. What now happens is that
2608 it logs this as a hard delivery error, but does not attempt to create a
2611 PH/11 The exinext utility has a -C option for testing purposes, but although
2612 the given file was scanned by exinext itself; it wasn't being passed on
2613 when Exim was called.
2615 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2616 an end-of-file indication when reading a command response.
2618 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2619 compiled. In many other places in Exim, IPv6 addresses are always
2620 recognized, so I have changed this. It also means that IPv4 domain
2621 literals of the form [IPV4:n.n.n.n] are now always recognized.
2623 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2624 used if the router is not running as root, for example, when verifying at
2625 ACL time, or when using -bh. The debugging output from this situation was
2626 non-existent - all you got was a failure to exec. I have made two
2629 (a) Failures to set uid/gid, the current directory, or a process leader
2630 in a subprocess such as that created by queryprogram now generate
2631 suitable debugging ouput when -d is set.
2633 (b) The queryprogram router detects when it is not running as root,
2634 outputs suitable debugging information if -d is set, and then runs
2635 the subprocess without attempting to change uid/gid.
2637 PH/15 Minor change to Makefile for building test_host (undocumented testing
2640 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2641 additional section of a DNS packet that returns MX or SRV records.
2642 Instead, it always explicitly searches for A/AAAA records. This avoids
2643 major problems that occur when a DNS server includes only records of one
2644 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2645 fixed another bug: if SRV records were looked up and the corresponding
2646 address records were *not* found in the additional section, the port
2647 values from the SRV records were lost.
2649 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2650 using the correct key (the original address) when searching the retry
2651 rules in order to find which one to use for generating the retry hint.
2653 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2654 adding the default one. Similarly, if it contains a Reply-To: header, the
2655 errors_reply_to option, if set, is not used.
2657 PH/19 When calculating a retry time, Exim used to measure the "time since
2658 failure" by looking at the "first failed" field in the retry record. Now
2659 it does not use this if it is later than than the arrival time of the
2660 message. Instead it uses the arrival time. This makes for better
2661 behaviour in cases where some deliveries succeed, thus re-setting the
2662 "first failed" field. An example is a quota failure for a huge message
2663 when small messages continue to be delivered. Without this change, the
2664 "time since failure" will always be short, possible causing more frequent
2665 delivery attempts for the huge message than are intended.
2666 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2668 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2669 $1, $2, $3) because the numerical variables can be reset during some
2670 expansion items (e.g. "match"), thereby losing the authentication data.
2672 PH/21 Make -bV show the size of off_t variables so that the test suite can
2673 decide whether to run tests for quotas > 2G.
2675 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2676 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2677 filecount value is greater than 2G or if a quota value is greater than 2G
2678 on a system where the size of off_t is not greater than 4, a panic error
2681 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2682 never match. The debug and -bh output now contains an explicit error
2683 message indicating a malformed IPv4 address or mask.
2685 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2686 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2687 PH/23 above applies.
2689 PH/25 Do not write to syslog when running in the test harness. The only
2690 occasion when this arises is a failure to open the main or panic logs
2691 (for which there is an explicit test).
2693 PH/26 Added the /no_tell option to "control=freeze".
2695 PH/27 If a host name lookup failed very early in a connection, for example, if
2696 the IP address matched host_lookup and the reverse lookup yielded a name
2697 that did not have a forward lookup, an error message of the form "no IP
2698 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2699 could be logged. Now it outputs the IP address instead of "NULL".
2701 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2702 allows the sender and the authenticated sender to be set when
2703 submitting a message from within Exim. Since child_open_exim() is
2704 documented for local_scan(), the new function should be too.
2706 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2707 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2708 results in an empty string is now treated as unset.
2710 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2712 PH/31 Added sender_verify_fail logging option.
2714 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2715 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2716 tidied the source and removed it altogether.
2718 PH/33 When a queue run was abandoned because the load average was too high, a
2719 log line was always written; now it is written only if the queue_run log
2720 selector is set. In addition, the log line for abandonment now contains
2721 information about the queue run such as the pid. This is always present
2722 in "start" and "stop" lines but was omitted from the "abandon" line.
2724 PH/34 Omit spaces between a header name and the colon in the error message that
2725 is given when verify = headers_syntax fails (if there are lots of them,
2726 the message gets confusing).
2728 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2729 names, as there are now some PTR records that contain slashes. This check
2730 is only to protect against broken name servers that fall over on strange
2731 characters, so the fact that it applies to all lookups doesn't matter.
2733 PH/36 Now that the new test suite is complete, we can remove some of the
2734 special code in Exim that was needed for the old test suite. For example,
2735 sorting DNS records because real resolvers return them in an arbitrary
2736 order. The new test suite's fake resolver always returns records in the
2739 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2740 bounces) except when queue_only is set, to avoid logging races between
2741 the different processes.
2743 PH/38 Panic-die if .include specifies a non-absolute path.
2745 PH/39 A tweak to the "H" retry rule from its user.
2747 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2748 a label. They prevented compilation on older perls.
2750 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2751 a warning to be raised on newish perls.
2753 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2754 on queue. Changes to match documented behaviour of showing count of
2755 messages matching specified criteria.
2757 PH/40 Changed the default ident timeout from 30s to 5s.
2759 PH/41 Added support for the use of login_cap features, on those BSD systems
2760 that have them, for controlling the resources used by pipe deliveries.
2762 PH/42 The content-scanning code uses fopen() to create files in which to put
2763 message data. Previously it was not paying any attention to the mode of
2764 the files. Exim runs with umask(0) because the rest of the code creates
2765 files with open(), and sets the required mode explicitly. Thus, these
2766 files were ending up world-writeable. This was not a big issue, because,
2767 being within the spool directory, they were not world-accessible. I have
2768 created a function called modefopen, which takes an additional mode
2769 argument. It sets umask(777), creates the file, chmods it to the required
2770 mode, then resets the umask. All the relevant calls to fopen() in the
2771 content scanning code have been changed to use this function.
2773 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2774 to 24 hours. This avoids potential overflow problems when processing G
2775 and H retry rules. I suspect nobody ever tinkers with this value.
2777 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2779 PH/45 When the plaintext authenticator is running as a client, the server's
2780 challenges are checked to ensure they are valid base64 strings. By
2781 default, the authentication attempt is cancelled if an invalid string is
2782 received. Setting client_ignore_invalid_base64 true ignores these errors.
2783 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2784 they are received. Thus, the responses can be made to depend on the
2785 challenges. If an invalid string is ignored, an empty string is placed in
2788 PH/46 Messages that are created by the autoreply transport now contains a
2789 References: header, in accordance with RFCs 2822 and 3834.
2791 PH/47 Added authenticated_sender_force to the smtp transport.
2793 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2795 PH/49 Installed latest patch from the Sieve maintainer.
2797 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2798 was also set, the appendfile transport was unnecessarily scanning a
2799 directory of message files (e.g. for maildir delivery) to find the count
2800 of files (along with the size), even though it did not need this
2801 information. It now does the scan only if it needs to find either the
2802 size of the count of files.
2804 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2806 PH/52 Two bugs concerned with error handling when the smtp transport is
2809 (i) Exim was not creating retry information for temporary errors given
2810 for individual recipients after the DATA command when the smtp transport
2811 was used in LMTP mode. This meant that they could be retried too
2812 frequently, and not timed out correctly.
2814 (ii) Exim was setting the flag that allows error details to be returned
2815 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2816 recipients that were returned after the DATA command.
2818 PH/53 This is related to PH/52, but is more general: for any failing address,
2819 when detailed error information was permitted to be returned to the
2820 sender, but the error was temporary, then after the final timeout, only
2821 "retry timeout exceeded" was returned. Now it returns the full error as
2822 well as "retry timeout exceeded".
2824 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2825 do this, and (what is worse) MTAs that accept it.
2827 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2828 will now be deprecated.
2830 PH/56 New os.c-cygwin from the Cygwin maintainer.
2832 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2833 formats (previously only available in exim formats via -bpr, -bpru,
2834 and -bpra. Now also available in native and exiqgrep formats)
2836 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2837 with very large, slow to parse queues
2839 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2841 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2843 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2844 responses to authentication challenges, though it was showing the
2845 challenges; (ii) I've removed the CR characters from the debug output for
2848 PH/58 Allow for the insertion of a newline as well as a space when a string
2849 is turned into more than one encoded-word during RFC 2047 encoding. The
2850 Sieve code now uses this.
2852 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2853 data_4xx, lost_connection, tls_required.
2855 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2856 message was being sent as an SMTP response.
2858 PH/61 Add -l and -k options to exicyclog.
2860 PH/62 When verifying, if an address was redirected to one new address, so that
2861 verification continued, and the new address failed or deferred after
2862 having set something in $address_data, the value of $address_data was not
2863 passed back to the ACL. This was different to the case when no
2864 redirection occurred. The value is now passed back in both cases.
2866 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2867 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2868 use login_cap.h, so on its own it isn't the distinguishing feature. The
2869 new name refers directly to the setclassresources() function.
2871 PH/65 Added configuration files for NetBSD3.
2873 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2875 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2876 is preferred over IPv4.
2878 PH/68 The bounce_return_message and bounce_return_body options were not being
2879 honoured for bounces generated during the reception of non-SMTP messages.
2880 In particular, this applied to messages rejected by the ACL. This bug has
2881 been fixed. However, if bounce_return_message is true and bounce_return_
2882 body is false, the headers that are returned for a non-SMTP message
2883 include only those that have been read before the error was detected.
2884 (In the case of an ACL rejection, they have all been read.)
2886 PH/69 The HTML version of the specification is now built in a directory called
2887 spec_html instead of spec.html, because the latter looks like a path with
2888 a MIME-type, and this confuses some software.
2890 PH/70 Catch two compiler warnings in sieve.c.
2892 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2893 function verify_get_ident() calls ip_connect() to connect a socket, but
2894 if the "connect()" function timed out, ip_connect() used to close the
2895 socket. However, verify_get_ident() also closes the socket later, and in
2896 between Exim writes to the log, which may get opened at this point. When
2897 the socket was closed in ip_connect(), the log could get the same file
2898 descriptor number as the socket. This naturally causes chaos. The fix is
2899 not to close the socket in ip_connect(); the socket should be closed by
2900 the function that creates it. There was only one place in the code where
2901 this was missing, in the iplookup router, which I don't think anybody now
2902 uses, but I've fixed it anyway.
2904 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2905 well as to direct DNS lookups. Otherwise the handling of names in host
2906 lists is inconsistent and therefore confusing.
2912 PH/01 Two changes to the default runtime configuration:
2914 (1) Move the checks for relay_from_hosts and authenticated clients from
2915 after to before the (commented out) DNS black list checks.
2917 (2) Add control=submission to the relay_from_hosts and authenticated
2918 clients checks, on the grounds that messages accepted by these
2919 statements are most likely to be submissions.
2921 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2923 (1) Generate an error if the third argument for the ${prvs expansion is
2926 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2929 (3) Reset the variables that are obtained from the first argument of
2930 ${prvscheck and used in the second argument before leaving the code,
2931 because their memory is reclaimed, so using them afterwards may do
2934 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2935 one (it's much easier than Tom thought :-).
2937 (5) Because of (4), we can now allow for the use of $prvscheck_result
2938 inside the third argument.
2940 PH/03 For some reason, the default setting of PATH when running a command from
2941 a pipe transport was just "/usr/bin". I have changed it to
2944 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2945 anything to be listed in the output from -bV.
2947 PH/05 When a filter generated an autoreply, the entire To: header line was
2948 quoted in the delivery log line, like this:
2950 => >A.N.Other <ano@some.domain> <original@ddress> ...
2952 This has been changed so that it extracts the operative address. There
2953 may be more than one such address. If so, they are comma-separated, like
2956 => >ano@some.domain,ona@other.domain <original@ddress> ...
2958 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2959 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2960 not being looked up in the rDNS to get a host name, Exim was showing the
2961 IP address twice in Received: lines, even though the IP addresses were
2962 identical. For example:
2964 Received: from [1.2.3.4] (helo=[1.2.3.4])
2966 However, if the real host name was known, it was omitting the HELO data
2967 if it matched the actual IP address. This has been tidied up so that it
2968 doesn't show the same IP address twice.
2970 PH/07 When both +timestamp and +memory debugging was on, the value given by
2971 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2972 called by the debug printing, thereby overwriting the timestamp buffer.
2973 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2975 PH/08 When the original message was included in an autoreply transport, it
2976 always said "this is a copy of the message, including all the headers",
2977 even if body_only or headers_only was set. It now gives an appropriate
2980 PH/09 Applied a patch from the Sieve maintainer which:
2982 o fixes some comments
2983 o adds the (disabled) notify extension core
2984 o adds some debug output for the result of if/elsif tests
2985 o points to the current vacation draft in the documentation
2986 and documents the missing references header update
2990 o fixes a bug in processing the envelope test (when testing
2991 multiple envelope elements, the last element determinted the
2994 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2995 Electronic Mail") by including:
2997 Auto-submitted: auto-generated
2999 in the messages that it generates (bounce messages and others, such as
3000 warnings). In the case of bounce messages for non-SMTP mesages, there was
3001 also a typo: it was using "Auto_submitted" (underscore instead of
3002 hyphen). Since every message generated by Exim is necessarily in response
3003 to another message, thes have all been changed to:
3005 Auto-Submitted: auto-replied
3007 in accordance with these statements in the RFC:
3009 The auto-replied keyword:
3011 - SHOULD be used on messages sent in direct response to another
3012 message by an automatic process,
3014 - MUST NOT be used on manually-generated messages,
3016 - MAY be used on Delivery Status Notifications (DSNs) and Message
3017 Disposition Notifications (MDNs),
3019 - MUST NOT be used on messages generated by automatic or periodic
3020 processes, except for messages which are automatic responses to
3023 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3024 to the default Received: header definition.
3026 PH/12 Added log selector acl_warn_skipped (default on).
3028 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3029 variables because (a) they are in the wrong storage pool and (b) even if
3030 they were copied, it wouldn't work properly because of the caching.
3032 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3033 checking when decoding. Apparently there are clients that generate
3034 overlong encoded strings. Why am I not surprised?
3036 PH/15 If the first argument of "${if match_address" was not empty, but did not
3037 contain an "@" character, Exim crashed. Now it writes a panic log message
3038 and treats the condition as false.
3040 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3042 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3043 extension "envelope-auth". The code is finished and in agreement with
3044 other implementations, but there is no documentation so far and in fact,
3045 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3046 not changing the active code.
3048 Print executed "if" and "elsif" statements when debugging is used. This
3049 helps a great deal to understand what a filter does.
3051 Document more things not specified clearly in RFC3028. I had all this
3052 sorted out, when out of a sudden new issues came to my mind. Oops."
3054 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3057 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3058 canonical form (as documented). However, after a host name lookup from
3059 the IP address, check_host() was doing a simple string comparison with
3060 addresses acquired from the DNS when checking that the found name did
3061 have the original IP as one of its addresses. Since any found IPv6
3062 addresses are likely to be in abbreviated form, the comparison could
3063 fail. Luckily, there already exists a function for doing the comparison
3064 by converting both addresses to binary, so now that is used instead of
3065 the text comparison.
3067 PH/20 There was another similar case to PH/19, when a complete host name was
3068 given in a host list; looking up its IP address could give an abbreviated
3069 form, whereas the current host's name might or might not be abbreviated.
3070 The same fix has been applied.
3076 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3077 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3080 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3081 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3083 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3085 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3086 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3087 Also, for "Auto-Submitted", treat anything other than "no" as
3088 non-personal, in accordance with RFC 3834. (Previously it treated
3089 anything starting "auto-" as non-personal.)
3091 TF/01 The control=submission/name=... option had a problem with syntax
3092 errors if the name included a slash character. The /name= option
3093 now slurps the rest of the string, so it can include any characters
3094 but it must come last in the list of options (after /sender_retain
3097 PH/05 Some modifications to the interface to the fake nameserver for the new
3105 TK/01 Added the "success_on_redirect" address verification option. See
3106 NewStuff for rationale and an example.
3108 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3110 PH/02 Patch to exigrep to allow it to work on syslog lines.
3112 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3113 fread() to skip over the body file's header line, because in Cygwin the
3114 header line is locked and is inaccessible.
3116 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3117 co-exist for some time) to make it clear that it is the Exim ID that is
3118 referenced, not the Message-ID: header line.
3120 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3121 string_format() function, because snprintf() does not exist on all
3124 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3125 ${stat: expansion item.
3127 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3128 protocol synchronization error", to keep the pedants happy.
3130 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3131 well as for IRIX systems, when gcc is being used. See the host.c source
3134 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3136 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3139 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3140 transport and to the smtp transport in LMTP mode.
3142 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3144 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3145 run a filter in a subprocess. This could lead to confusion in subsequent
3146 lookups in the parent process. There should also be a search_tidyup() at
3147 the end of the subprocess.
3149 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3150 only if the host matched helo_try_verify_hosts, which caused the
3151 verification to occur when the EHLO/HELO command was issued. The ACL just
3152 tested the remembered result. Now, if a previous verification attempt has
3153 not happened, "verify = helo" does it there and then.
3155 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3157 TK/03 Fix log output including CR from clamd.
3159 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3160 exist provoked a memory error which could cause a segfault.
3162 PH/15 Installed PCRE 6.2
3164 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3166 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3167 of the problem. Specifically, suggested +O2 rather than +O1 for the
3170 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3172 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3173 record, the port was not correctly logged when the outgoing_port log
3174 selector was set (it logged the transort's default port).
3176 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3177 fallback_hosts, and "hosts" in the smtp transport.
3179 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3180 host errors such as "Connection refused".
3182 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3183 authentication with radiusclient 0.4.9:
3185 - Error returned from rc_read_config was caught wrongly
3186 - Username/password not passed on to radius server due to wrong length.
3188 The presumption is that some radiusclient API changes for 4.51/PH/17
3189 were not taken care of correctly. The code is still untested by me (my
3190 Linux distribution still has 0.3.2 of radiusclient), but it was
3191 contributed by a Radius user.
3193 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3194 expanding the "port" option of the smtp transport.
3196 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3197 while reading a MIME header. Thanks to Tom Hughes for a patch.
3199 PH/24 Include config.h inside local_scan.h so that configuration settings are
3202 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3203 that in an ACL for RCPT (for example), you can examine exactly what was
3206 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3207 commands, but it was not correctly comparing the address with the actual
3208 client host address. Thus, it would show the EHLO address in Received:
3209 header lines when this was not necessary.
3211 PH/27 Added the % operator to ${eval:}.
3213 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3214 it should be ignoring failures (because with -C, for example, it has lost
3215 privilege). It wasn't ignoring creation failures other than "already
3218 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3221 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3222 because an input line was too long, either on its own, or by virtue of
3223 too many continuations, the temporary file was not being removed, and the
3224 return code was incorrect.
3226 PH/31 Missing "BOOL" in function definition in filtertest.c.
3228 PH/32 Applied Sieve patches from the maintainer.
3230 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3232 PH/33 Added "verify = not_blind".
3234 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3235 Local/Makefile (with some defaults set). These are used in built scripts
3236 such as exicyclog, but they have never been used in the exim_install
3237 script (though there are many overriding facilities there). I have
3238 arranged that the exim_install script now takes note of these two
3241 PH/35 Installed configuration files for Dragonfly.
3243 PH/36 When a locally submitted message by a trusted user did not contain a
3244 From: header, and the sender address was obtained from -f or from an SMTP
3245 MAIL command, and the trusted user did not use -F to supply a sender
3246 name, $originator_name was incorrectly used when constructing a From:
3247 header. Furthermore, $originator_name was used for submission mode
3248 messages from external hosts without From: headers in a similar way,
3249 which is clearly wrong.
3251 PH/37 Added control=suppress_local_fixups.
3253 PH/38 When log_selector = +received_sender was set, and the addition of the
3254 sender made the log line's construction buffer exactly full, or one byte
3255 less than full, an overflow happened when the terminating "\n" was
3258 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3259 when the result of a list match is failure because a DNS lookup failed.
3261 PH/40 RM_COMMAND is now used in the building process.
3263 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3264 the "build-* directories that it finds.
3266 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3267 address in a domain literal was a prefix of an interface address.
3269 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3270 when verifying a sender address, unless rewrite_headers is false.
3272 PH/44 Wrote a long comment about why errors_to addresses are verified as
3273 recipients, not senders.
3275 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3276 the ratelimit ACL was added.
3278 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3280 PH/46 Added extra information about PostgreSQL errors to the error string.
3282 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3283 avoiding the need to install special zones in a real server. This is
3284 backwards compatible; if it can't find the fake resolver, it drops back.
3285 Thus, both old and new test suites can be run.
3287 TF/02 Added util/ratelimit.pl
3289 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3290 clock is set back in time.
3292 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3293 Candler <B.Candler@pobox.com>.
3295 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3296 OK for addresses that are the result of redirections.
3298 PH/48 A number of further additions for the benefit of the new test suite,
3299 including a fake gethostbyname() that interfaces to the fake DNS resolver
3302 TF/06 The fix for widen_domains has also been applied to qualify_single and
3303 search_parents which are the other dnslookup options that can cause
3306 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3309 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3311 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3312 service" if it encountered an MX record with an empty target hostname.
3313 The message is now "an MX or SRV record indicated no SMTP service".
3315 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3316 if the DNS of the sending site is misconfigured. This is quite a
3317 common situation. This change restores the behaviour of treating a
3318 helo verification defer as a failure.
3320 PH/51 If self=fail was set on a router, the bounce message did not include the
3321 actual error message.
3327 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3329 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3330 command itself ended in error, the underlying message about the transport
3331 filter timeout was being overwritten with the pipe command error. Now the
3332 underlying error message should be appended to the second error message.
3334 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3336 PH/02 Reduce the amount of output that "make" produces by default. Full output
3337 can still be requested.
3339 PH/03 The warning log line about a condition test deferring for a "warn" verb
3340 was being output only once per connection, rather than after each
3341 occurrence (because it was using the same function as for successful
3342 "warn" verbs). This seems wrong, so I have changed it.
3344 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3345 it should not have, which might have caused a crash in the right
3346 circumstances, but probably never did.
3348 PH/04 Installed a modified version of Tony Finch's patch to make submission
3349 mode fix the return path as well as the Sender: header line, and to
3350 add a /name= option so that you can make the user's friendly name appear
3353 TF/03 Added the control = fakedefer ACL modifier.
3355 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3356 Mark Lowes for thorough testing.
3358 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3360 TK/03 Merged latest SRS patch from Miles Wilton.
3362 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3363 with the definition in sysexits.h (which is #included earlier).
3364 Fortunately, Exim does not actually use EX_OK. The code used to try to
3365 preserve the sysexits.h value, by assumimg that macro definitions were
3366 scanned for macro replacements. I have been disabused of this notion,
3367 so now the code just undefines EX_OK before #including unistd.h.
3369 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3370 in the smtp transport. When a block could not be written in a single
3371 write() function, the timeout was being re-applied to each part-write.
3372 This seems wrong - if the receiver was accepting one byte at a time it
3373 would take for ever. The timeout is now adjusted when this happens. It
3374 doesn't have to be particularly precise.
3376 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3377 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3379 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3380 without a domain if the check to <postmaster@domain> fails.
3382 SC/01 Eximstats: added -xls and the ability to specify output files
3383 (patch written by Frank Heydlauf).
3385 SC/02 Eximstats: use FileHandles for outputing results.
3387 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3389 SC/04 Eximstats: fixed display of large numbers with -nvr option
3391 SC/05 Eximstats: fixed merging of reports with empty tables.
3393 SC/06 Eximstats: added the -include_original_destination flag
3395 SC/07 Eximstats: removed tabs and trailing whitespace.
3397 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3399 TK/06 MBOX spool code: Add real "From " MBOX separator line
3400 so the .eml file is really in mbox format (even though
3401 most programs do not really care). Patch from Alex Miller.
3403 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3404 The latter is generated from $received_to and is only set if the
3405 message has one envelope recipient. SA can use these headers,
3406 obviously out-of-the-box. Patch from Alex Miller.
3408 PH/08 The ${def test on a variable was returning false if the variable's
3409 value was "0", contrary to what the specification has always said!
3410 The result should be true unless the variable is empty.
3412 PH/09 The syntax error of a character other than { following "${if
3413 def:variable_name" (after optional whitespace) was not being diagnosed.
3414 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3415 accidental colon was present, for example, could give incorrect results.
3417 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3418 result is used (not including appendfile, where other changes are about
3421 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3422 This involved changing a lot of size variables from int to off_t. It
3423 should work with maildirs and everything.
3425 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3426 spamd dying while we are connected to it.
3428 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3431 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3432 function for BDB 4.3.
3434 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3436 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3437 the sources for PCRE have been re-arranged and more files are now
3440 PH/15 The code I had for printing potentially long long variables in PH/11
3441 above was not the best (it lost precision). The length of off_t variables
3442 is now inspected at build time, and an appropriate printing format (%ld
3443 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3444 to be "long long int" or "long int". This is needed for the internal
3445 formatting function string_vformat().
3447 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3448 the configuration file to be ":syslog", then the script "guesses" where
3449 the logs files are, rather than using the compiled in default. In our
3450 case the guess is not the same as the compiled default, so the script
3451 suddenly stopped working when I started to use syslog. The patch checks
3452 to see if log_file_path is "". If so, it attempts to read it from exim
3453 with no configuration file to get the compiled in version, before it
3454 falls back to the previous guessing code."
3456 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3457 implementing BATV in an Exim configuration. See NewStuff for the gory
3460 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3461 Makefile that are specific to HP-UX.
3463 PH/18 If the "use_postmaster" option was set for a recipient callout together
3464 with the "random" option, the postmaster address was used as the MAIL
3465 FROM address for the random test, but not for the subsequent recipient
3466 test. It is now used for both.
3468 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3469 patch removes a few documentation additions to RFC 3028, because the
3470 latest draft now contains them. It adds the new en;ascii-case comparator
3471 and a new error check for 8bit text in MIME parts. Comparator and
3472 require names are now matched exactly. I enabled the subaddress
3473 extension, but it is not well tested yet (read: it works for me)."
3475 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3476 rework some of the code of TK/09 above to avoid the hardwired use of
3477 "%lld" and "long long". Replaced the call to snprintf() with a call to
3480 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3481 records point to non-existent hosts", "retry timeout exceeded", and
3482 "retry time not reached for any host after a long failure period".
3484 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3485 experimental DomainKeys support:
3487 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3488 (2) On an error such as an illegally used "control", the wrong name for
3489 the control was given.
3491 These problems did NOT occur unless DomainKeys support was compiled.
3493 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3495 PH/24 Added ${if match_ip condition.
3497 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3498 databases so that it will be absolutely obvious if a crash occurs in the
3499 DB library. This is a regular occurrence (often caused by mis-matched
3502 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3503 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3504 picked up on a user's system that detects such things. There doesn't seem
3505 to be a gcc warning option for this - only an attribute that has to be
3506 put on the function's prototype. It seems that in Fedora Core 4 they have
3507 set this on a number of new functions. No doubt there will be more in due
3510 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3511 specify a transport. However, if an address that was verified by such a
3512 router was the subject of a callout, Exim crashed because it tried to
3513 read the rcpt_include_affixes from the non-existent transport. Now it
3514 just assumes that the setting of that option is false. This bug was
3515 introduced by 4.51/PH/31.
3517 PH/28 Changed -d+all to exclude +memory, because that information is very
3518 rarely of interest, but it makes the output a lot bigger. People tend to
3519 do -d+all out of habit.
3521 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3522 code in os-type was giving problems when libc.so lives in lib64, like on
3525 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3526 aren't the modern standard, and it seems that some systems' include files
3527 don't always have them. Exim was already checking for some of the newer
3528 ones like T_AAAA, and defining it itself. I've added checks for all the
3529 record types that Exim uses.
3531 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3532 not automatically generating a new one, as it is supposed to. This
3533 prevented TLS from working. If the file did exist, but contained invalid
3534 data, a new version was generated, as expected. It was only the case of a
3535 non-existent file that was broken.
3537 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3538 with a change in libdomainkeys > 0.64.
3540 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3541 from DNS. If the selector record carries the flag, it now has
3542 precedence over the domain-wide flag.
3544 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3546 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3547 the use of a port name that isn't defined in /etc/services) occurred, the
3548 message was deferred as in a normal delivery, and thus remained on the
3549 spool, instead of being failed because of the mua_wrapper setting. This
3550 is now fixed, and I tidied up some of the mua_wrapper messages at the
3553 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3554 the messages in a hash of arrays rather than using individual hashes.
3555 This is a bit cleaner and results in dramatic memory savings, albeit
3556 at a slight CPU cost.
3558 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3559 as requested by Marc Sherman.
3561 SC/10 Eximstats: added histograms for user specified patterns as requested
3564 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3566 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3567 fopen() in the content-scanning modules that did not already have it.
3573 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3574 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3576 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3578 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3580 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3581 to test Sieve filters that use "vacation".
3583 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3584 that changes the way the GnuTLS parameters are stored in the cache file.
3585 The new format can be generated externally. For backward compatibility,
3586 if the data in the cache doesn't make sense, Exim assumes it has read an
3587 old-format file, and it generates new data and writes a new file. This
3588 means that you can't go back to an older release without removing the
3591 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3592 work if there are any delivery delays because "one_time" forces the
3593 parent to be marked "delivered", so its unseen clone is never tried
3594 again. For this reason, Exim now forbids the simultaneous setting of
3597 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3598 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3599 entry if you want to know the details. The fix, however, neglected to
3600 consider the case when local delivery batching is involved. The test for
3601 "previously delivered" was not happening when checking to see if an
3602 address could be batched with a previous (undelivered) one; under
3603 certain circumstances this could lead to multiple deliveries to the same
3606 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3607 in its include files, and this causes problems building Exim.
3609 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3610 header_syntax) but Exim was just ignoring anything given after a slash.
3611 In particular, this caused confusion with an attempt to use "verify =
3612 reverse_host_lookup/defer_ok". An error is now given when options are
3613 supplied for verify items that do not have them. (Maybe reverse_host_
3614 lookup should have a defer_ok option, but that's a different point.)
3616 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3617 defined by RFC 821) to 2048, because there were problems with some AUTH
3618 commands, and RFC 1869 says the size should be increased for extended
3619 SMTP commands that take arguments.
3621 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3624 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3625 "unknown" error; now it says that the functionality isn't in the binary.
3627 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3628 an address' error message when a string expansion fails (syntax or
3629 whatever). Otherwise the password may appear in the log. Following change
3630 PH/42 below, there is no longer a chance of it appearing in a bounce
3633 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3635 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3636 of its addresses was ever tried. (Bugzilla bug #2.)
3638 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3639 the result incorrectly in the debug output. (It correctly added a newline
3640 to what was transported.)
3642 TF/01 Added $received_time.
3644 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3645 commented out examples of how to interface to a virus scanner and to
3646 SpamAssassin. Also added commented examples of av_scanner and
3647 spamd_address settings.
3649 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3650 and controls are allowed in which ACLs. There were a couple of minor
3651 errors. Some of the entries in the conditions table (which is a table of
3652 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3653 negation of where the condition IS allowed.
3655 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3657 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3658 header file does not have a version number, so I've had to invent a new
3659 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3660 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3661 radiusclient), but it was contributed by a Radius user.
3663 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3664 files or directories, for OpenSSL.
3666 PH/19 When an Exim process that is running as root has to create an Exim log
3667 file, it does so in a subprocess that runs as exim:exim so as to get the
3668 ownership right at creation (otherwise, other Exim processes might see
3669 the file with the wrong ownership). There was no test for failure of this
3670 fork() call, which would lead to the process getting stuck as it waited
3671 for a non-existent subprocess. Forks do occasionally fail when resources
3672 run out. I reviewed all the other calls to fork(); they all seem to check
3675 PH/20 When checking for unexpected SMTP input at connect time (before writing
3676 the banner), Exim was not dealing correctly with a non-positive return
3677 from the read() function. If the client had disconnected by this time,
3678 the result was a log entry for a synchronization error with an empty
3679 string after "input=" when read() returned zero. If read() returned -1
3680 (an event I could not check), uninitialized data bytes were printed.
3681 There were reports of junk text (parts of files, etc) appearing after
3684 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3686 PH/22 Added support for macro redefinition, and (re)definition in between
3687 driver and ACL definitions.
3689 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3690 forgetting to use the resulting value; it was using the unexpanded value.
3692 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3693 hadn't been configured. The fix is from Juergen Kreileder, who
3694 understands it better than I do:
3696 "Here's what I see happening with three configured cyrus_sasl
3697 authenticators configured (plain, login, cram-md5):
3699 On startup auth_cyrus_sasl_init() gets called for each of these.
3700 This means three calls to sasl_listmech() without a specified mech_list.
3701 => SASL tests which mechs of all available mechs actually work
3702 => three warnings about OTP not working
3703 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3705 With the patch, sasl_listmech() also gets called three times. But now
3706 SASL's mech_list option is set to the server_mech specified in the the
3707 authenticator. Or in other words, the answer from sasl_listmech()
3708 gets limited to just the mech you're testing for (which is different
3710 => the return list contains just 'plain' or 'login', 'cram-md5' or
3711 nothing depending on the value of ob->server_mech.
3713 I've just tested the patch: Authentication still works fine,
3714 unavailable mechs specified in the exim configuration are still
3715 caught, and the auth.log warnings about OTP are gone."
3717 PH/25 When debugging is enabled, the contents of the command line are added
3718 to the debugging output, even when log_selector=+arguments is not
3721 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3722 answer is "GNU", and only if the return is "GNU/something" is the answer
3725 PH/27 $acl_verify_message is now set immediately after the failure of a
3726 verification in an ACL, and so is available in subsequent modifiers. In
3727 particular, the message can be preserved by coding like this:
3729 warn !verify = sender
3730 set acl_m0 = $acl_verify_message
3732 Previously, $acl_verify_message was set only while expanding "message"
3733 and "log_message" when a very denied access.
3735 PH/28 Modified OS/os.c-Linux with
3737 -#ifndef OS_LOAD_AVERAGE
3738 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3740 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3741 nomenclature these days.)
3743 PH/29 Installed patch from the Sieve maintainer that adds the options
3744 sieve_useraddress and sieve_subaddress to the redirect router.
3746 PH/30 In these circumstances:
3747 . Two addresses routed to the same list of hosts;
3748 . First host does not offer TLS;
3749 . First host accepts first address;
3750 . First host gives temporary error to second address;
3751 . Second host offers TLS and a TLS session is established;
3752 . Second host accepts second address.
3753 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3754 and peerdn, if requested) that were in fact used only for the second
3757 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3758 attention to any local part prefix or suffix that was matched by the
3759 router that accepted the address. It now behaves in the same way as it
3760 does for delivery: the affixes are removed from the local part unless
3761 rcpt_include_affixes is set on the transport.
3763 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3764 timeout during the DATA phase of an incoming message.
3766 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3767 applied a patch sanctioned by the Sieve maintainer.
3769 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3770 the uid or gid is negative. A case of a negative gid caused this to be
3771 noticed. The fix allows for either to be negative.
3773 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3774 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3777 PH/36 The change PH/12 above was broken. Fixed it.
3779 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3780 the grounds that routing the same address twice would always produce the
3781 same answer. This might have been true once, but it is certainly no
3782 longer true now. Routing a child address may depend on the previous
3783 routing that produced that child. Some complicated redirection strategies
3784 went wrong when messages had multiple recipients, and made Exim's
3785 behaviour dependent on the order in which the addresses were given.
3787 I have moved the duplicate checking until after the routing is complete.
3788 Exim scans the addresses that are assigned to local and remote
3789 transports, and removes any duplicates. This means that more work will be
3790 done, as duplicates will always all be routed, but duplicates are
3791 presumably rare, so I don't expect this is of any significance.
3793 For deliveries to pipes, files, and autoreplies, the duplicate checking
3794 still happens during the routing process, since they are not going to be
3797 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3798 It corrects a timeout issue with spamd. This is Ian's comment: "The
3799 background is that sometimes spamd either never reads data from a
3800 connection it has accepted, or it never writes response data. The exiscan
3801 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3802 blindly assumes that writes won't block so it may never time out."
3804 PH/39 Allow G after quota size as well as K and M.
3806 PH/40 The value set for $authenticated_id in an authenticator may not contain
3807 binary zeroes or newlines because the value is written to log lines and
3808 to spool files. There was no check on this. Now the value is run through
3809 the string_printing() function so that such characters are converted to
3810 printable escape sequences.
3812 PH/41 $message_linecount is a new variable that contains the total number of
3813 lines in the message. Compare $body_linecount, which is the count for the
3816 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3817 bounce and delay warning messages, except in certain special cases, which
3820 (a) An SMTP error message from a remote host;
3821 (b) A message specified in a :fail: redirection;
3822 (c) A message specified in a "fail" command in a system filter;
3823 (d) A message specified in a FAIL return from the queryprogram router;
3824 (e) A message specified by the cannot_route_message router option.
3826 In these cases only, Exim does include the error details in bounce and
3827 warning messages. There are also a few cases where bland messages such
3828 as "unrouteable address" or "local delivery error" are given.
3830 PH/43 $value is now also set for the "else" part of a ${run expansion.
3832 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3833 being worked on, but at least Exim now implements the latest version to
3836 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3837 process to complete was treated as a delivery failure, a timeout while
3838 writing the message to the pipe was logged, but erroneously treated as a
3839 successful delivery. Such timeouts include transport filter timeouts. For
3840 consistency with the overall process timeout, these timeouts are now
3841 treated as errors, giving rise to delivery failures by default. However,
3842 there is now a new Boolean option for the pipe transport called
3843 timeout_defer, which, if set TRUE, converts the failures into defers for
3844 both kinds of timeout. A transport filter timeout is now identified in
3847 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3848 systems where "make" and "gmake" are different, calling "gmake" at top
3849 level broke things. I've arranged for the value of $(MAKE) to be passed
3850 from the Makefile to this script so that it can call the same version of
3854 A note about Exim versions 4.44 and 4.50
3855 ----------------------------------------
3857 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3858 changes of various kinds. As a consequence, a big documentation update was
3859 needed. This delayed the release for rather longer than seemed good, especially
3860 in the light of a couple of (minor) security issues. Therefore, the changes
3861 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3862 release. So 4.44 and 4.50 are in effect two different branches that both start
3865 I have left the 4.50 change log unchanged; it contains all the changes since
3866 4.43. The change log for 4.44 is below; many of its items are identical to
3867 those for 4.50. This seems to be the most sensible way to preserve the
3868 historical information.
3874 1. Minor wording change to the doc/README.SIEVE file.
3876 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3877 computation of the current number of files was incorrect.
3879 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3880 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3883 4. Give more explanation in the error message when the command for a transport
3884 filter fails to execute.
3886 5. There are several places where Exim runs a non-Exim command in a
3887 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3888 was being done only for the command run by the queryprogram router. It is
3889 now done for all such subprocesses. The other cases are: ${run, transport
3890 filters, and the commands run by the lmtp and pipe transports.
3892 6. Added CONFIGURE_GROUP build-time option.
3894 7. Some older OS have a limit of 256 on the maximum number of file
3895 descriptors. Exim was using setrlimit() to set 1000 as a large value
3896 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3897 systems. I've change it so that if it can't get 1000, it tries for 256.
3899 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3900 was an oversight, and furthermore, ever since the addition of extra
3901 controls (e.g. 4.43/32), the checks on when to allow different forms of
3902 "control" were broken. There should now be diagnostics for all cases when a
3903 control that does not make sense is encountered.
3905 9. Added the /retain_sender option to "control=submission".
3907 10. $recipients is now available in the predata ACL (oversight).
3909 11. Tidy the search cache before the fork to do a delivery from a message
3910 received from the command line. Otherwise the child will trigger a lookup
3911 failure and thereby defer the delivery if it tries to use (for example) a
3912 cached ldap connection that the parent has called unbind on.
3914 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3915 of $address_data from the recipient verification was clobbered by the
3916 sender verification.
3918 13. The value of address_data from a sender verification is now available in
3919 $sender_address_data in subsequent conditions in the ACL statement.
3921 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3923 15. Added a new option "connect=<time>" to callout options, to set a different
3926 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3927 was its contents. (It was OK if the option was not defined at all.)
3929 17. A "Completed" log line is now written for messages that are removed from
3930 the spool by the -Mrm option.
3932 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3933 information about exactly what failed.
3935 19. Added -dd to debug only the daemon process.
3937 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3938 handles timeouts, both on the server side and network timeouts. Renamed the
3939 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3941 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3942 It is now set to "smtps".
3944 22. $host_address is now set to the target address during the checking of
3945 ignore_target_hosts.
3947 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3948 being passed; this would have caused $sender_host_name to have been used if
3949 matching the list had actually called for a host name (not very likely,
3950 since this list is usually IP addresses). A host name is now passed as
3953 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3954 code that specifies a non-restarting handler (typically sigaction() in
3955 modern systems) in an attempt to fix a rare and obscure crash bug.
3957 25. Narrowed the window for a race in the daemon that could cause it to ignore
3958 SIGCHLD signals. This is not a major problem, because they are used only to
3959 wake it up if nothing else does.
3961 26. A malformed maildirsize file could cause Exim to calculate negative values
3962 for the mailbox size or file count. Odd effects could occur as a result.
3963 The maildirsize information is now recalculated if the size or filecount
3966 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3967 support for a long time. Removed HAVE_SYS_VFS_H.
3969 28. Installed the latest version of exipick from John Jetmore.
3971 29. In an address list, if the pattern was not a regular expression, an empty
3972 subject address (from a bounce message) matched only if the pattern was an
3973 empty string. Non-empty patterns were not even tested. This was the wrong
3974 because it is perfectly reasonable to use an empty address as part of a
3975 database query. An empty address is now tested by patterns that are
3976 lookups. However, all the other forms of pattern expect the subject to
3977 contain a local part and a domain, and therefore, for them, an empty
3978 address still always fails if the pattern is not itself empty.
3980 30. Exim went into a mad DNS loop when attempting to do a callout where the
3981 host was specified on an smtp transport, and looking it up yielded more
3982 than one IP address.
3984 31. Re-factored the code for checking spool and log partition space into a
3985 function that finds that data and another that does the check. The former
3986 is then used to implement four new variables: $spool_space, $log_space,
3987 $spool_inodes, and $log_inodes.
3989 32. The RFC2047 encoding function was originally intended for short strings
3990 such as real names; it was not keeping to the 75-character limit for
3991 encoded words that the RFC imposes. It now respects the limit, and
3992 generates multiple encoded words if necessary. To be on the safe side, I
3993 have increased the buffer size for the ${rfc2047: expansion operator from
3996 33. It is now permitted to omit both strings after an "if" condition; if the
3997 condition is true, the result is "true". As before, when the second string
3998 is omitted, a false condition yields an empty string. This makes it less
3999 cumbersome to write custom ACL and router conditions.
4001 34. Failure to deliver a bounce message always caused it to be frozen, even if
4002 there was an errors_to setting on the router. The errors_to setting is now
4005 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4006 canonical form (fully expanded) before being placed in
4007 $sender_host_address.
4009 36. The table in the code that translates DNS record types into text (T_A to
4010 "A" for instance) was missing entries for NS and CNAME. It is just possible
4011 that this could have caused confusion if both these types were looked up
4012 for the same domain, because the text type is used as part of Exim's
4013 per-process caching. But the chance of anyone hitting this buglet seems
4016 37. The dnsdb lookup has been extended in a number of ways.
4018 (1) There is a new type, "zns", which walks up the domain tree until it
4019 finds some nameserver records. It should be used with care.
4021 (2) There is a new type, "mxh", which is like "mx" except that it returns
4022 just the host names, not the priorities.
4024 (3) It is now possible to give a list of domains (or IP addresses) to be
4025 looked up. The behaviour when one of the lookups defers can be
4026 controlled by a keyword.
4028 (4) It is now possible to specify the separator character for use when
4029 multiple records are returned.
4031 38. The dnslists ACL condition has been extended: it is now possible to supply
4032 a list of IP addresses and/or domains to be looked up in a particular DNS
4035 39. Added log_selector=+queue_time_overall.
4037 40. When running the queue in the test harness, wait just a tad after forking a
4038 delivery process, to get repeatability of debugging output.
4040 41. Include certificate and key file names in error message when GnuTLS fails
4041 to set them up, because the GnuTLS error message doesn't include the name
4042 of the failing file when there is a problem reading it.
4044 42. Allow both -bf and -bF in the same test run.
4046 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4048 44. The "Exiscan patch" is now merged into the mainline Exim source.
4050 45. Sometimes the final signoff response after QUIT could fail to get
4051 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4052 < 0 before doing a fflush(). This bug looks as though it goes back to the
4053 introduction of TLS in release 3.20, but "sometimes" must have been rare
4054 because the tests only now provoked it.
4056 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4057 (this can affect the format of dates).
4059 47. exim_tidydb, when checking for the continued existence of a message for
4060 which it has found a message-specific retry record, was not finding
4061 messages that were in split spool directories. Consequently, it was
4062 deleting retry records that should have stayed in existence.
4064 48. Steve fixed some bugs in eximstats.
4066 49. The SPA authentication driver was not abandoning authentication and moving
4067 on to the next authenticator when an expansion was forced to fail,
4068 contradicting the general specification for all authenticators. Instead it
4069 was generating a temporary error. It now behaves as specified.
4071 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4072 (the order specifies the preference for clients). The order is now AES256,
4073 AES128, 3DES, ARCFOUR128.
4075 51. Small patch to Sieve code - explicitly set From: when generating an
4078 52. Exim crashed if a remote delivery caused a very long error message to be
4079 recorded - for instance if somebody sent an entire SpamAssassin report back
4080 as a large number of 550 error lines. This bug was coincidentally fixed by
4081 increasing the size of one of Exim's internal buffers (big_buffer) that
4082 happened as part of the Exiscan merge. However, to be on the safe side, I
4083 have made the code more robust (and fixed the comments that describe what
4086 53. Now that there can be additional text after "Completed" in log lines (if
4087 the queue_time_overall log selector is set), a one-byte patch to exigrep
4088 was needed to allow it to recognize "Completed" as not the last thing in
4091 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4092 patch that reportedly fixes this has been added. I am not expert enough to
4093 create a test for it. This is what the patch creator wrote:
4095 "I found a little strange behaviour of ldap code when working with
4096 Windows 2003 AD Domain, where users was placed in more than one
4097 Organization Units. When I tried to give exim partial DN, the exit code
4098 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4099 But simultaneously result of request was absolutely normal ldap result,
4100 so I produce this patch..."
4102 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4103 REFERENCE, so I have modified the code to exclude the patch when that macro
4106 55. Some experimental protocols are using DNS PTR records for new purposes. The
4107 keys for these records are domain names, not reversed IP addresses. The
4108 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4109 leaves it alone. Component reversal etc. now happens only for IP addresses.
4112 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4114 57. Double the size of the debug message buffer (to 2048) so that more of very
4115 long debug lines gets shown.
4117 58. The exicyclog utility now does better if the number of log files to keep
4118 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4120 59. Two changes related to the smtp_active_hostname option:
4122 (1) $smtp_active_hostname is now available as a variable.
4123 (2) The default for smtp_banner uses $smtp_active_hostname instead
4124 of $primary_hostname.
4126 60. The host_aton() function is supposed to be passed a string that is known
4127 to be a valid IP address. However, in the case of IPv6 addresses, it was
4128 not checking this. This is a hostage to fortune. Exim now panics and dies
4129 if the condition is not met. A case was found where this could be provoked
4130 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4131 components; fortuitously, this particular loophole had already been fixed
4132 by change 4.50/55 above.
4134 If there are any other similar loopholes, the new check in host_aton()
4135 itself should stop them being exploited. The report I received stated that
4136 data on the command line could provoke the exploit when Exim was running as
4137 exim, but did not say which command line option was involved. All I could
4138 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4139 running as the user.
4142 61. There was a buffer overflow vulnerability in the SPA authentication code
4143 (which came originally from the Samba project). I have added a test to the
4144 spa_base64_to_bits() function which I hope fixes it.
4147 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4148 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4150 63. The daemon start-up calls getloadavg() while still root for those OS that
4151 need the first call to be done as root, but it missed one case: when
4152 deliver_queue_load_max is set with deliver_drop_privilege. This is
4153 necessary for the benefit of the queue runner, because there is no re-exec
4154 when deliver_drop_privilege is set.
4156 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4157 This has been fixed.
4159 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4160 was in use, was not putting the data itself into the right store pool;
4161 consequently, it could be overwritten for a subsequent message in the same
4162 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4165 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4167 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4168 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4169 the function were treating the return as a boolean value, which happened to
4170 work because 0=false and not-0=true, but is not correct code.
4172 68. The host_aton() function was not handling scoped IPv6 addresses (those
4173 with, for example, "%eth0" on the end) correctly.
4175 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4176 negated items (that is, ~something) in unsigned ints. Some compilers
4177 apparently mutter when there is no cast.
4179 70. If an address verification called from an ACL failed, and did not produce a
4180 user-specific message (i.e. there was only a "system" message), nothing was
4181 put in $acl_verify_message. In this situation, it now puts the system
4184 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4185 session; change 4.31/43 added the unwanted input to the log line - except
4186 that it did not do this in the start of session case. It now does.
4188 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4189 This is wrong and can cause the other end to generate a synchronization
4190 error if it is another Exim or anything else that does the synchronization
4191 check. A QUIT command is no longer sent after a timeout.
4193 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4194 during host lookups.
4196 74. The defer_ok option of callout verification was not working if it was used
4197 when verifying addresses in header lines, that is, for this case:
4199 verify = header_sender/callout=defer_ok
4201 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4202 those file descriptors could be used for SMTP connections. If anything
4203 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4204 could be sent to the SMTP client, causing chaos. The daemon now opens
4205 stdin, stdout, and stderr to /dev/null when it puts itself into the
4208 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4209 log by default. The user can override this with suitable Perl magic.
4211 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4212 the log message when discard triggers, was not working for the DATA ACL or
4213 for the non-SMTP ACL.
4215 78. Error message wording change in sieve.c.
4217 79. If smtp_accept_max_per_host was set, the number of connections could be
4218 restricted to fewer than expected, because the daemon was trying to set up
4219 a new connection before checking whether the processes handling previous
4220 connections had finished. The check for completed processes is now done
4221 earlier. On busy systems, this bug wouldn't be noticed because something
4222 else would have woken the daemon, and it would have reaped the completed
4225 80. If a message was submitted locally by a user whose login name contained one
4226 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4227 It caused a spool format error. I have fixed the spool reading code. A
4228 related problem was that the "from" clause in the Received: line became
4229 illegal because of the space(s). It is now covered by ${quote_local_part.
4231 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4234 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4236 83. Patch from Sieve maintainer to fix unterminated string problem in
4237 "vacation" handling.
4239 84. Some minor changes to the Linux configuration files to help with other
4240 OS variants using glibc.
4242 85. One more patch for Sieve to update vacation handling to latest spec.
4245 ----------------------------------------------------
4246 See the note above about the 4.44 and 4.50 releases.
4247 ----------------------------------------------------
4253 1. Change 4.43/35 introduced a bug that caused file counts to be
4254 incorrectly computed when quota_filecount was set in an appendfile
4257 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4258 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4261 3. Give more explanation in the error message when the command for a transport
4262 filter fails to execute.
4264 4. There are several places where Exim runs a non-Exim command in a
4265 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4266 was being done only for the command run by the queryprogram router. It is
4267 now done for all such subprocesses. The other cases are: ${run, transport
4268 filters, and the commands run by the lmtp and pipe transports.
4270 5. Some older OS have a limit of 256 on the maximum number of file
4271 descriptors. Exim was using setrlimit() to set 1000 as a large value
4272 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4273 systems. I've change it so that if it can't get 1000, it tries for 256.
4275 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4276 was an oversight, and furthermore, ever since the addition of extra
4277 controls (e.g. 4.43/32), the checks on when to allow different forms of
4278 "control" were broken. There should now be diagnostics for all cases when a
4279 control that does not make sense is encountered.
4281 7. $recipients is now available in the predata ACL (oversight).
4283 8. Tidy the search cache before the fork to do a delivery from a message
4284 received from the command line. Otherwise the child will trigger a lookup
4285 failure and thereby defer the delivery if it tries to use (for example) a
4286 cached ldap connection that the parent has called unbind on.
4288 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4289 of $address_data from the recipient verification was clobbered by the
4290 sender verification.
4292 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4293 was its contents. (It was OK if the option was not defined at all.)
4295 11. A "Completed" log line is now written for messages that are removed from
4296 the spool by the -Mrm option.
4298 12. $host_address is now set to the target address during the checking of
4299 ignore_target_hosts.
4301 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4302 being passed; this would have caused $sender_host_name to have been used if
4303 matching the list had actually called for a host name (not very likely,
4304 since this list is usually IP addresses). A host name is now passed as
4307 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4308 code that specifies a non-restarting handler (typically sigaction() in
4309 modern systems) in an attempt to fix a rare and obscure crash bug.
4311 15. Narrowed the window for a race in the daemon that could cause it to ignore
4312 SIGCHLD signals. This is not a major problem, because they are used only to
4313 wake it up if nothing else does.
4315 16. A malformed maildirsize file could cause Exim to calculate negative values
4316 for the mailbox size or file count. Odd effects could occur as a result.
4317 The maildirsize information is now recalculated if the size or filecount
4320 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4321 support for a long time. Removed HAVE_SYS_VFS_H.
4323 18. Updated exipick to current release from John Jetmore.
4325 19. Allow an empty sender to be matched against a lookup in an address list.
4326 Previously the only cases considered were a regular expression, or an
4329 20. Exim went into a mad DNS lookup loop when doing a callout where the
4330 host was specified on the transport, if the DNS lookup yielded more than
4333 21. The RFC2047 encoding function was originally intended for short strings
4334 such as real names; it was not keeping to the 75-character limit for
4335 encoded words that the RFC imposes. It now respects the limit, and
4336 generates multiple encoded words if necessary. To be on the safe side, I
4337 have increased the buffer size for the ${rfc2047: expansion operator from
4340 22. Failure to deliver a bounce message always caused it to be frozen, even if
4341 there was an errors_to setting on the router. The errors_to setting is now
4344 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4345 canonical form (fully expanded) before being placed in
4346 $sender_host_address.
4348 24. Updated eximstats to version 1.33
4350 25. Include certificate and key file names in error message when GnuTLS fails
4351 to set them up, because the GnuTLS error message doesn't include the name
4352 of the failing file when there is a problem reading it.
4354 26. Expand error message when OpenSSL has problems setting up cert/key files.
4357 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4358 (this can affect the format of dates).
4360 28. exim_tidydb, when checking for the continued existence of a message for
4361 which it has found a message-specific retry record, was not finding
4362 messages that were in split spool directories. Consequently, it was
4363 deleting retry records that should have stayed in existence.
4365 29. eximstats updated to version 1.35
4366 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4367 1.35 - bugfix such that pie charts by volume are generated correctly
4369 30. The SPA authentication driver was not abandoning authentication and moving
4370 on to the next authenticator when an expansion was forced to fail,
4371 contradicting the general specification for all authenticators. Instead it
4372 was generating a temporary error. It now behaves as specified.
4374 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4375 (the order specifies the preference for clients). The order is now AES256,
4376 AES128, 3DES, ARCFOUR128.
4378 31. Small patch to Sieve code - explicitly set From: when generating an
4381 32. Exim crashed if a remote delivery caused a very long error message to be
4382 recorded - for instance if somebody sent an entire SpamAssassin report back
4383 as a large number of 550 error lines. This bug was coincidentally fixed by
4384 increasing the size of one of Exim's internal buffers (big_buffer) that
4385 happened as part of the Exiscan merge. However, to be on the safe side, I
4386 have made the code more robust (and fixed the comments that describe what
4389 33. Some experimental protocols are using DNS PTR records for new purposes. The
4390 keys for these records are domain names, not reversed IP addresses. The
4391 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4392 leaves it alone. Component reversal etc. now happens only for IP addresses.
4395 34. The host_aton() function is supposed to be passed a string that is known
4396 to be a valid IP address. However, in the case of IPv6 addresses, it was
4397 not checking this. This is a hostage to fortune. Exim now panics and dies
4398 if the condition is not met. A case was found where this could be provoked
4399 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4400 components; fortuitously, this particular loophole had already been fixed
4401 by change 4.50/55 or 4.44/33 above.
4403 If there are any other similar loopholes, the new check in host_aton()
4404 itself should stop them being exploited. The report I received stated that
4405 data on the command line could provoke the exploit when Exim was running as
4406 exim, but did not say which command line option was involved. All I could
4407 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4408 running as the user.
4411 35. There was a buffer overflow vulnerability in the SPA authentication code
4412 (which came originally from the Samba project). I have added a test to the
4413 spa_base64_to_bits() function which I hope fixes it.
4416 36. The daemon start-up calls getloadavg() while still root for those OS that
4417 need the first call to be done as root, but it missed one case: when
4418 deliver_queue_load_max is set with deliver_drop_privilege. This is
4419 necessary for the benefit of the queue runner, because there is no re-exec
4420 when deliver_drop_privilege is set.
4422 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4423 was in use, was not putting the data itself into the right store pool;
4424 consequently, it could be overwritten for a subsequent message in the same
4425 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4428 38. Sometimes the final signoff response after QUIT could fail to get
4429 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4430 < 0 before doing a fflush(). This bug looks as though it goes back to the
4431 introduction of TLS in release 3.20, but "sometimes" must have been rare
4432 because the tests only now provoked it.
4438 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4439 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4440 Now it returns an integer. A number of calls were still expecting a T/F
4441 return. Fortuitously, in all cases, the tests worked in OK situations,
4442 which is the norm. However, things would have gone wrong on any write
4443 failures on the smtp file descriptor. This function is used when sending
4444 messages over SMTP and also when doing verify callouts.
4446 2. When Exim is called to do synchronous delivery of a locally submitted
4447 message (the -odf or -odi options), it no longer closes stderr before doing
4450 3. Implemented the mua_wrapper option.
4452 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4454 5. Implemented the functions header_remove(), header_testname(),
4455 header_add_at_position(), and receive_remove_recipient(), and exported them
4458 6. If an ACL "warn" statement specified the addition of headers, Exim already
4459 inserted X-ACL-Warn: at the start if there was no header name. However, it
4460 was not making this test for the second and subsequent header lines if
4461 there were newlines in the string. This meant that an invalid header could
4462 be inserted if Exim was badly configured.
4464 7. Allow an ACL "warn" statement to add header lines at the start or after all
4465 the Received: headers, as well as at the end.
4467 8. Added the rcpt_4xx retry error code.
4469 9. Added postmaster_mailfrom=xxx to callout verification option.
4471 10. Added mailfrom=xxxx to the callout verification option, for verify=
4474 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4475 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4477 12. Inserted some casts to stop certain compilers warning when using pointer
4478 differences as field lengths or precisions in printf-type calls (mostly
4479 affecting debugging statements).
4481 13. Added optional readline() support for -be (dynamically loaded).
4483 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4484 same clock tick as a message's arrival, so that its received time was the
4485 same as the "first fail" time on the retry record, and that message
4486 remained on the queue past the ultimate address timeout, every queue runner
4487 would try a delivery (because it was past the ultimate address timeout) but
4488 after another failure, the ultimate address timeout, which should have then
4489 bounced the address, did not kick in. This was a "< instead of <=" error;
4490 in most cases the first failure would have been in the next clock tick
4491 after the received time, and all would be well.
4493 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4494 being recognized when the domain list was tested by the match_domain
4495 condition in an expansion string.
4497 16. Added the ${str2b64: operator.
4499 17. Exim was always calling setrlimit() to set a large limit for the number of
4500 processes, without checking whether the existing limit was already
4501 adequate. (It did check for the limit on file descriptors.) Furthermore,
4502 errors from getrlimit() and setrlimit() were being ignored. Now they are
4503 logged to the main and panic logs, but Exim does carry on, to try to do its
4504 job under whatever limits there are.
4506 18. Imported PCRE 5.0.
4508 19. Trivial typo in log message " temporarily refused connection" (the leading
4511 20. If the log selector return_path_on_delivery was set and an address was
4512 redirected to /dev/null, the delivery process crashed because it assumed
4513 that a return path would always be set for a "successful" delivery. In this
4514 case, the whole delivery is bypassed as an optimization, and therefore no
4517 21. Internal re-arrangement: the function for sending a challenge and reading
4518 a response while authentication was assuming a zero-terminated challenge
4519 string. It's now changed to take a pointer and a length, to allow for
4520 binary data in such strings.
4522 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4524 23. Exim was not respecting finduser_retries when seeking the login of the
4525 uid under which it was called; it was always trying 10 times. (The default
4526 setting of finduser_retries is zero.) Also, it was sleeping after the final
4527 failure, which is pointless.
4529 24. Implemented tls_on_connect_ports.
4531 25. Implemented acl_smtp_predata.
4533 26. If the domain in control=submission is set empty, Exim assumes that the
4534 authenticated id is a complete email address when it generates From: or
4535 Sender: header lines.
4537 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4538 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4539 chown and chgrp in /bin and hostname in /usr/bin.
4541 28. Exim was keeping the "process log" file open after each use, just as it
4542 does for the main log. This opens the possibility of it remaining open for
4543 long periods when the USR1 signal hits a daemon. Occasional processlog
4544 errors were reported, that could have been caused by this. Anyway, it seems
4545 much more sensible not to leave this file open at all, so that is what now
4548 29. The long-running daemon process does not normally write to the log once it
4549 has entered its main loop, and it closes the log before doing so. This is
4550 so that log files can straightforwardly be renamed and moved. However,
4551 there are a couple of unusual error situations where the daemon does write
4552 log entries, and I had neglected to close the log afterwards.
4554 30. The text of an SMTP error response that was received during a remote
4555 delivery was being truncated at 512 bytes. This is too short for some of
4556 the long messages that one sometimes sees. I've increased the limit to
4559 31. It is now possible to make retry rules that apply only when a message has a
4560 specific sender, in particular, an empty sender.
4562 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4563 it possible to be selective about when SMTP synchronization is enforced.
4565 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4567 32. Implemented hosts_connection_nolog.
4569 33. Added an ACL for QUIT.
4571 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4574 35. Added mailbox_size and mailbox_filecount to appendfile.
4576 36. Added control = no_multiline_responses to ACLs.
4578 37. There was a bug in the logic of the code that waits for the clock to tick
4579 in the case where the clock went backwards by a substantial amount such
4580 that the microsecond fraction of "now" was more than the microsecond
4581 fraction of "then" (but the whole seconds number was less).
4583 38. Added support for the libradius Radius client library this is found on
4584 FreeBSD (previously only the radiusclient library was supported).
4590 1. When certain lookups returned multiple values in the form name=value, the
4591 quoting of the values was not always being done properly. Specifically:
4592 (a) If the value started with a double quote, but contained no whitespace,
4594 (b) If the value contained whitespace other than a space character (i.e.
4595 tabs or newlines or carriage returns) it was not quoted.
4596 This fix has been applied to the mysql and pgsql lookups by writing a
4597 separate quoting function and calling it from the lookup code. The fix
4598 should probably also be applied to nisplus, ibase and oracle lookups, but
4599 since I cannot test any of those, I have not disturbed their existing code.
4601 2. A hit in the callout cache for a specific address caused a log line with no
4602 reason for rejecting RCPT. Now it says "Previous (cached) callout
4603 verification failure".
4605 3. There was an off-by-one bug in the queryprogram router. An over-long
4606 return line was truncated at 256 instead of 255 characters, thereby
4607 overflowing its buffer with the terminating zero. As well as fixing this, I
4608 have increased the buffer size to 1024 (and made a note to document this).
4610 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4611 when Exim is waiting for an SMTP response from a remote server, Exim
4612 restarts its select() call on the socket, thereby resetting its timeout.
4613 This is not a problem when such interrupts are rare. Somebody set up a cron
4614 job to run exiwhat every 2 minutes, which is less than the normal select()
4615 timeout (5 or 10 minutes). This meant that the select() timeout never
4616 kicked in because it was always reset. I have fixed this by comparing the
4617 time when an interrupt arrives with the time at the start of the first call
4618 to select(). If more time than the timeout has elapsed, the interrupt is
4619 treated as a timeout.
4621 5. Some internal re-factoring in preparation for the addition of Sieve
4622 extensions (by MH). In particular, the "personal" test is moved to a
4623 separate function, and given an option for scanning Cc: and Bcc: (which is
4624 not set for Exim filters).
4626 6. When Exim created an email address using the login of the caller as the
4627 local part (e.g. when creating a From: or Sender: header line), it was not
4628 quoting the local part when it contained special characters such as @.
4630 7. Installed new OpenBSD configuration files.
4632 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4633 try to make them clearer.
4635 9. Callout options, other than the timeout value, were being ignored when
4636 verifying sender addresses in header lines. For example, when using
4638 verify = header_sender/callout=no_cache
4640 the cache was (incorrectly) being used.
4642 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4643 only the Cygwin environment.
4645 11. When return_path_on_delivery was set as a log selector, if different remote
4646 addresses in the same message used different return paths and parallel
4647 remote delivery occurred, the wrong values would sometimes be logged.
4648 (Whenever a remote delivery process finished, the return path value from
4649 the most recently started remote delivery process was logged.)
4651 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4652 lines when AUTH and/or TLS are in use. This is the "received protocol"
4653 field. Exim used to use "asmtp" for authenticated SMTP, without any
4654 indication (in the protocol name) for TLS use. Now it follows the RFC and
4655 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4656 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4657 names appear in log lines as well as in Received: header lines.
4659 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4660 extensions, and comparison tests, and to fix some bugs.
4662 14. Changes to the "personal" filter test:
4664 (1) The test was buggy in that it was just doing the equivalent of
4665 "contains" tests on header lines. For example, if a user's address was
4666 anne@some.where, the "personal" test would incorrectly be true for
4668 To: susanne@some.where
4670 This test is now done by extracting each address from the header in turn,
4671 and checking the entire address. Other tests that are part of "personal"
4672 are now done using regular expressions (for example, to check local parts
4673 of addresses in From: header lines).
4675 (2) The list of non-personal local parts in From: addresses has been
4676 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4677 taken from the Sieve specification recommendations.
4679 (3) If the message contains any header line starting with "List-" it is
4680 treated as non-personal.
4682 (4) The test for "circular" in the Subject: header line has been removed
4683 because it now seems ill-conceived.
4685 15. Minor typos in src/EDITME comments corrected.
4687 16. Installed latest exipick from John Jetmore.
4689 17. If headers_add on a router specified a text string that was too long for
4690 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4691 of string_sprintf() is now avoided.
4693 18. $message_body_size was not set (it was always zero) when running the DATA
4694 ACL and the local_scan() function.
4696 19. For the "mail" command in an Exim filter, no default was being set for
4697 the once_repeat time, causing a random time value to be used if "once" was
4698 specified. (If the value happened to be <= 0, no repeat happened.) The
4699 default is now 0s, meaning "never repeat". The "vacation" command was OK
4700 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4701 (I found it when inspecting the code).
4703 20. There is now an overall timeout for performing a callout verification. It
4704 defaults to 4 times the callout timeout, which applies to individual SMTP
4705 commands during the callout. The overall timeout applies when there is more
4706 than one host that can be tried. The timeout is checked before trying the
4707 next host. This prevents very long delays if there are a large number of
4708 hosts and all are timing out (e.g. when the network connections are timing
4709 out). The value of the overall timeout can be changed by specifying an
4710 additional sub-option for "callout", called "maxwait". For example:
4712 verify = sender/callout=5s,maxwait=20s
4714 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4715 to the end before writing, but this should make it even safer).
4717 22. Exim was forgetting that it had advertised PIPELINING for the second and
4718 subsequent messages on an SMTP connection. It was also not resetting its
4719 memory on STARTTLS and an internal HELO.
4721 23. When Exim logs an SMTP synchronization error within a session, it now
4722 records whether PIPELINING has been advertised or not.
4724 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4725 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4726 rather than long int.
4728 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4730 26. Added the never_mail option to autoreply.
4736 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4737 crash if the getsockname() call failed; this can happen if a connection is
4738 closed very soon after it is established. The problem was simply in the
4739 order in which certain operations were done, causing Exim to try to write
4740 to the SMTP stream before it had set up the file descriptor. The bug has
4741 been fixed by making things happen in the correct order.
4747 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4748 before the connection was closed, thus losing the rejection response.
4750 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4751 some early Solaris releases, but causes trouble in current releases where
4752 socklen_t is defined.
4754 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4757 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4760 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4761 that releasing the top of it at the end releases what was used for sub-
4762 expansions (unless the block got too big). However, discard this block if
4763 the first thing is a variable or header, so that we can use its block when
4764 it is dynamic (useful for very large $message_headers, for example).
4766 6. Lookups now cache *every* query, not just the most recent. A new, separate
4767 store pool is used for this. It can be recovered when all lookup caches are
4768 flushed. Lookups now release memory at the end of their result strings.
4769 This has involved some general refactoring of the lookup sources.
4771 7. Some code has been added to the store_xxx() functions to reduce the amount
4772 of flapping under certain conditions.
4774 8. log_incoming_interface used to affect only the <= reception log lines. Now
4775 it causes the local interface and port to be added to several more SMTP log
4776 lines, for example "SMTP connection from", and rejection lines.
4778 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4780 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4782 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4783 could be overwritten at the end of the current message (or the start of a
4784 new message if it was set in a HELO ACL). The value is now preserved for
4785 the duration of the SMTP connection.
4787 12. If a transport had a headers_rewrite setting, and a matching header line
4788 contained an unqualified address, that address was qualified, even if it
4789 did not match any rewriting rules. The underlying bug was that the values
4790 of the flags that permit the existence of unqualified sender and recipient
4791 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4792 non-local messages, and by -bnq for local messages) were not being
4793 preserved with the message after it was received.
4795 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4796 "next input=" as part of the text comprising the host identity instead of
4797 the correct text. The code was using the same buffer for two different
4798 strings. However, depending on which order the printing function evaluated
4799 its arguments, the bug did not always show up. Under Linux, for example, my
4800 test suite worked just fine.
4802 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4803 doesn't work with some older versions of Perl. It has been changed to "my",
4804 which in any case is probably the better facility to use.
4806 15. A really picky compiler found some instances of statements for creating
4807 error messages that either had too many or two few arguments for the format
4810 16. The size of the buffer for calls to the DNS resolver has been increased
4811 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4812 for addresses that have a lot of PTR records. This alleviates a problem; it
4813 does not fully solve it.
4815 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4816 buffer now truncates the list and logs the incident, which is the same
4817 action as happens when Exim is looking up a host name and its aliases.
4818 Previously in this situation something unpredictable would happen;
4819 sometimes it was "internal error: store_reset failed".
4821 18. If a server dropped the connection unexpectedly when an Exim client was
4822 using GnuTLS and trying to read a response, the client delivery process
4823 crashed while trying to generate an error log message.
4825 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4826 string, for example:
4828 warn message = H1: something\nH2: something
4830 the text was added as a single header line from Exim's point of view
4831 though it ended up OK in the delivered message. However, searching for the
4832 second and subsequent header lines using $h_h2: did not work. This has been
4833 fixed. Similarly, if a system filter added multiple headers in this way,
4834 the routers could not see them.
4836 20. Expanded the error message when iplsearch is called with an invalid key to
4837 suggest using net-iplsearch in a host list.
4839 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4840 ACLs are no longer actually imposed (and a message to that effect is
4843 22. If a "gecos" field in a passwd entry contained escaped characters, in
4844 particular, if it contained a \" sequence, Exim got it wrong when building
4845 a From: or a Sender: header from that name. A second bug also caused
4846 incorrect handling when an unquoted " was present following a character
4847 that needed quoting.
4849 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4850 was not being matched caselessly.
4852 24. Arranged for all hyphens in the exim.8 source to be escaped with
4855 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4856 properly. Recipient callout cache records were still being keyed to include
4857 the sender, even when use_sender was set false. This led to far more
4858 callouts that were necessary. The sender is no longer included in the key
4859 when use_sender is false.
4861 26. Added "control = submission" modifier to ACLs.
4863 27. Added the ${base62d: operator to decode base 62 numbers.
4865 28. dnsdb lookups can now access SRV records.
4867 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4868 the configuration file.
4870 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4871 (-v) mode. This makes the output for a verbose queue run more intelligible.
4873 31. Added a use_postmaster feature to recipient callouts.
4875 32. Added the $body_zerocount variable, containing the number of binary zero
4876 bytes in the message body.
4878 33. The time of last modification of the "new" subdirectory is now used as the
4879 "mailbox time last read" when there is a quota error for a maildir
4882 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4884 35. Added +ignore_unknown as a special item in host lists.
4886 36. Code for decoding IPv6 addresses in host lists is now included, even if
4887 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4888 address was recognized as an IP address, but was then not correctly decoded
4889 into binary, causing unexpected and incorrect effects when compared with
4896 1. Very minor rewording of debugging text in manualroute to say "list of
4897 hosts" instead of "hostlist".
4899 2. If verify=header_syntax was set, and a header line with an unqualified
4900 address (no domain) and a large number of spaces between the end of the
4901 name and the colon was received, the reception process suffered a buffer
4902 overflow, and (when I tested it) crashed. This was caused by some obsolete
4903 code that should have been removed. The fix is to remove it!
4905 3. When running in the test harness, delay a bit after writing a bounce
4906 message to get a bit more predictability in the log output.
4908 4. Added a call to search_tidyup() just before forking a reception process. In
4909 theory, someone could use a lookup in the expansion of smtp_accept_max_
4910 per_host which, without the tidyup, could leave open a database connection.
4912 5. Added the variables $recipient_data and $sender_data which get set from a
4913 lookup success in an ACL "recipients" or "senders" condition, or a router
4914 "senders" option, similar to $domain_data and $local_part_data.
4916 6. Moved the writing of debug_print from before to after the "senders" test
4919 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4920 problems for message scanning, either using a data ACL, or using
4921 local_scan() because the Received: header was not generated till after they
4922 were called (in order to set the time as the time of reception completion).
4923 I have revised the way this works. The header is now generated after the
4924 body is received, but before the ACL or local_scan() are called. After they
4925 are run, the timestamp in the header is updated.
4931 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4932 before starting a queue runner without re-exec. This happened only when
4933 deliver_drop_privilege was set or when the Exim user was set to root. The
4934 effect of the bug was that timeouts during subsequent deliveries caused
4935 crashes instead of being properly handled. The handler is now left at its
4936 default (and expected) setting.
4938 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4939 message, again when deliver_drop_privilege is set or Exim is run as root.
4940 The bug described in (1) was not present in this case, but the tidying up
4941 of the other signals was missing. I have made the two cases consistent.
4943 3. The ignore_target_hosts setting on a manualroute router was being ignored
4944 for hosts that were looked up using the /MX notation.
4946 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4949 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4950 operated on the sender address. After changing the $sender_address to <>
4951 for the sender address verify, Exim was re-instated it as the original
4952 (before rewriting) address, but remembering that it had rewritten it, so it
4953 wasn't rewriting it again. This bug also had the effect of breaking the
4954 sender address verification caching when the sender address was rewritten.
4956 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4957 This has been changed so that if the ip literal address matches
4958 ignore_target_hosts, the router declines.
4960 7. Added expansion conditions match_domain, match_address, and match_local_
4961 part (NOT match_host).
4963 8. The placeholder for the Received: header didn't have a length field set.
4965 9. Added code to Exim itself and to exim_lock to test for a specific race
4966 condition that could lead to file corruption when using MBX delivery. The
4967 issue is with the lockfile that is created in /tmp. If this file is removed
4968 after a process has opened it but before that process has acquired a lock,
4969 there is the potential for a second process to recreate the file and also
4970 acquire a lock. This could lead to two Exim processes writing to the file
4971 at the same time. The added code performs the same test as UW imapd; it
4972 checks after acquiring the lock that its file descriptor still refers to
4973 the same named file.
4975 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4976 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4979 11. Added the smtp_active_hostname option. If used, this will typically be made
4980 to depend on the incoming interface address. Because $interface_address is
4981 not set up until the daemon has forked a reception process, error responses
4982 that can happen earlier (such as "too many connections") no longer contain
4985 12. If an expansion in a condition on a "warn" statement fails because a lookup
4986 defers, the "warn" statement is abandoned, and the next ACL statement is
4987 processed. Previously this caused the whole ACL to be aborted.
4989 13. Added the iplsearch lookup type.
4991 14. Added ident_timeout as a log selector.
4993 15. Added tls_certificate_verified as a log selector.
4995 16. Added a global option tls_require_ciphers (compare the smtp transport
4996 option of the same name). This controls incoming TLS connections.
4998 17. I finally figured out how to make tls_require_ciphers do a similar thing
4999 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5000 before starting the TLS session.
5002 18. Tabs are now shown as \t in -bP output.
5004 19. If the log selector return_path_on_delivery was set, Exim crashed when
5005 bouncing a message because it had too many Received: header lines.
5007 20. If two routers both had headers_remove settings, and the first one included
5008 a superfluous trailing colon, the final name in the first list and the
5009 first name in the second list were incorrectly joined into one item (with a
5010 colon in the middle).
5016 1. Added -C and -D options to the exinext utility, mainly to make it easier
5017 to include in the automated testing, but these could be helpful when
5018 multiple configurations are in use.
5020 2. The exinext utility was not formatting the output nicely when there was
5021 an alternate port involved in the retry record key, nor when there was a
5022 message id as well (for retries that were specific to a specific message
5023 and a specific host). It was also confused by IPv6 addresses, because of
5024 the additional colons they contain. I have fixed the IPv4 problem, and
5025 patched it up to do a reasonable job for IPv6.
5027 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5028 delivery, the log line now contains "pipelined" if PIPELINING was used.
5030 4. An SMTP transport process used to panic and die if the bind() call to set
5031 an explicit outgoing interface failed. This has been changed; it is now
5032 treated in the same way as a connect() failure.
5034 5. A reference to $sender_host_name in the part of a conditional expansion
5035 that was being skipped was still causing a DNS lookup. This no longer
5038 6. The def: expansion condition was not recognizing references to header lines
5039 that used bh_ and bheader_.
5041 7. Added the _cache feature to named lists.
5043 8. The code for checking quota_filecount in the appendfile transport was
5044 allowing one more file than it should have been.
5046 9. For compatibility with Sendmail, the command line option
5054 and sets the incoming protocol and host name (for trusted callers). The
5055 host name and its colon can be omitted when only the protocol is to be set.
5056 Note the Exim already has two private options, -pd and -ps, that refer to
5057 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5058 "s", but I don't think that's a major issue.
5060 10. A number of refactoring changes to the code, none of which should affect
5063 (a) The number of logging options was getting close to filling up the
5064 32-bit word that was used as a bit map. I have split them into two classes:
5065 those that are passed in the argument to log_write(), and those that are
5066 only ever tested independently outside of that function. These are now in
5067 separate 32-bit words, so there is plenty of room for expansion again.
5068 There is no change in the user interface or the logging behaviour.
5070 (b) When building, for example, log lines, the code previously used a
5071 macro that called string_cat() twice, in order to add two strings. This is
5072 not really sufficiently general. Furthermore, there was one instance where
5073 it was actually wrong because one of the argument was used twice, and in
5074 one call a function was used. (As it happened, calling the function twice
5075 did not affect the overall behaviour.) The macro has been replaced by a
5076 function that can join an arbitrary number of extra strings onto a growing
5079 (c) The code for expansion conditions now uses a table and a binary chop
5080 instead of a serial search (which was left over from when there were very
5081 few conditions). Also, it now recognizes conditions like "pam" even when
5082 the relevant support is not compiled in: a suitably worded error message is
5083 given if an attempt is made to use such a condition.
5085 11. Added ${time_interval:xxxxx}.
5087 12. A bug was causing one of the ddress fields not to be passed back correctly
5088 from remote delivery subprocesses. The field in question was not being
5089 subsequently used, so this caused to problems in practice.
5091 13. Added new log selectors queue_time and deliver_time.
5093 14. Might have fixed a bug in maildirsizefile handling that threw up
5094 "unexpected character" debug warnings, and recalculated the data
5095 unnecessarily. In any case, I expanded the warning message to give more
5098 15. Added the message "Restricted characters in address" to the statements in
5099 the default ACL that block characters like @ and % in local parts.
5101 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5102 Three changes have been made:
5104 (a) There was a serious bug; a negative response to MAIL caused the whole
5105 recipient domain to be cached as invalid, thereby blocking all messages
5106 to all local parts at the same domain, from all senders. This bug has
5107 been fixed. The domain is no longer cached after a negative response to
5108 MAIL if the sender used is not empty.
5110 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5113 (c) A new callout option, "use_sender" has been added for people who want
5114 the modified behaviour.
5120 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5123 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5124 indeed breaks things for older releases.
5126 3. Added additional logging to the case where there is a problem reading data
5127 from a filter that is running in a subprocess using a pipe, in order to
5128 try to track down a specific problem.
5130 4. Testing facility fudge: when running in the test harness and attempting
5131 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5132 getting "No route to host". Convert this to a timeout.
5134 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5137 6. Some OS don't have socklen_t but use size_t instead. This affects the
5138 fifth argument of getsockopt() amongst other things. This is now
5139 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5140 can be set for individual OS. I have set it for SunOS5, OSF1, and
5141 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5142 some earlier ones do not.
5144 7. Change 4.30/15 was not doing the test caselessly.
5146 8. The standard form for an IPv6 address literal was being rejected by address
5147 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5148 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5149 this, as well as the form without the "IPv6" on the front (but only when
5150 address literals are enabled, of course).
5152 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5154 10. Exim crashed if a message with an empty sender address specified by -f
5155 encountered a router with an errors_to setting. This could be provoked only
5156 by a command such as
5160 where an empty string was supplied; "<>" did not hit this bug.
5162 11. Installed PCRE release 4.5.
5164 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5165 remained set. It is now erased.
5167 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5168 times from message ids (which are base 36 rather than the normal 62).
5170 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5171 were being counted as actual protocol errors, and logged if the log
5172 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5173 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5174 and DATA following a set of rejected RCPTs do not count as protocol errors.
5175 In other words, Exim assumes they were pipelined, though this may not
5176 actually be the case. Of course, in all cases the client gets an
5177 appropriate error code.
5179 15. If a lookup fails in an ACL condition, a message about the failure may
5180 be available; it is used if testing the ACL cannot continue, because most
5181 such messages specify what the cause of the deferral is. However, some
5182 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5183 that caused an old message to be retained and used if a later statement
5184 caused a defer, replacing the real cause of the deferral.
5186 16. If an IP address had so many PTR records that the DNS lookup buffer
5187 was not large enough to hold them, Exim could crash while trying to process
5188 the truncated data. It now detects and logs this case.
5190 17. Further to 4.21/58, another change has been made: if (and only if) the
5191 first line of a message (the first header line) ends with CRLF, a bare LF
5192 in a subsequent header line has a space inserted after it, so as not to
5193 terminate the header.
5195 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5196 unnecessarily, used atoi() instead of strtol(), and didn't check the
5197 termination when getting file sizes from file names by regex.
5199 19. Completely re-implemented the support for maildirsize files, in the light
5200 of a number of problems with the previous contributed implementation
5201 (4.30/29). In particular:
5203 . If the quota is zero, the maildirsize file is maintained, but no quota is
5206 . If the maildir directory does not exist, it is created before any attempt
5207 to write a maildirsize file.
5209 . The quota value in the file is just a cache; if the quota is changed in
5210 the transport, the new value overrides.
5212 . A regular expression is available for excluding directories from the
5215 20. The autoreply transport checks the characters in options that define the
5216 message's headers; it allows continued headers, but it was checking with
5217 isspace() after an embedded newline instead of explicitly looking for a
5220 21. If all the "regular" hosts to which an address was routed had passed their
5221 expiry times, and had not reached their retry times, the address was
5222 bounced, even if fallback hosts were defined. Now Exim should go on to try
5225 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5226 equivalent code in the SMTP transport. Some hosts send humungous responses
5227 to HELO/EHLO, more than 1024 it seems.
5229 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5230 gives compiler warnings in some environments. I've now done it "properly",
5233 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5234 (because of problems with the built-in one) was declared to return uschar *
5235 instead of char *, causing compiler failure.
5237 25. Fixed a file descriptor leak when processing alias/forward files.
5239 26. Fixed a minor format string issue in dbfn.c.
5241 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5243 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5244 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5247 29. When a sender address is verified, it is cached, to save repeating the test
5248 when there is more than one recipient in a message. However, when the
5249 verification involves a callout, it is possible for different callout
5250 options to be set for different recipients. It is too complicated to keep
5251 track of this in the cache, so now Exim always runs a verification when a
5252 callout is required, relying on the callout cache for the optimization.
5253 The overhead is duplication of the address routing, but this should not be
5256 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5257 to be verified with callout=postmaster, and the main callout worked but the
5258 postmaster check failed, the verification correctly failed. However, if a
5259 subsequent RCPT command asked for sender verification *without* the
5260 postmaster check, incorrect caching caused this verification also to fail,
5263 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5264 it was not caching the DNS options (qualify_single, search_parents) that
5265 were used when the lookup failed. A subsequent lookup with different
5266 options therefore always gave the same answer, though there were cases
5267 where it should not have. (Example: a "domains = !$mx_any" option on a
5268 dnslookup router: the "domains" option is always processed without any
5269 widening, but the router might have qualify_single set.) Now Exim uses the
5270 cached value only when the same options are set.
5272 32. Added John Jetmore's "exipick" utility to the distribution.
5274 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5275 than a timeout (e.g. a certificate is required, and is not provided), an
5276 Exim server now closes the connection immediately. Previously it waited for
5277 the client to close - but if the client is SSL, it seems that they each
5278 wait for each other, leading to a delay before one of them times out.
5280 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5281 maintained 0.8.x compatibility because I don't think many are using it, and
5282 it is clearly obsolete.
5284 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5287 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5288 client certificate was expired. A simple patch fixes this, though I don't
5289 understand the full logic of why the verify callback is called multiple
5292 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5293 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5294 which causes problems with some clients (such as the Certicom SSL Plus
5295 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5296 disables the coutermeasure allowing Eudora to connect."
5298 38. Exim was not checking that a write() to a log file succeeded. This could
5299 lead to Bad Things if a log got too big, in particular if it hit a file
5300 size limit. Exim now panics and dies if it cannot write to a log file, just
5301 as it does if it cannot open a log file.
5303 39. Modified OS/Makefile-Linux so that it now contains
5305 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5307 The two -D definitions ensure that Exim is compiled with large file
5308 support, which makes it possible to handle log files that are bigger than
5311 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5312 instance) a domain was checked against a named list that involved a lookup,
5313 causing $domain_data to be set, then another domain was checked against the
5314 same list, then the first domain was re-checked, the value of $domain_data
5315 after the final check could be wrong. In particular, if the second check
5316 failed, it could be set empty. This bug probably also applied to
5319 41. The strip_trailing_dot option was not being applied to the address given
5320 with the -f command-line option.
5322 42. The code for reading a message's header from the spool was incrementing
5323 $received_count, but never initializing it. This meant that the value was
5324 incorrect (doubled) while delivering a message in the same process in which
5325 it was received. In the most common configuration of Exim, this never
5326 happens - a fresh exec is done - but it can happen when
5327 deliver_drop_privilege is set.
5329 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5330 it now includes up to 150 characters of the unexpected data in the log
5333 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5334 and building data strings. The size of both of these buffers was 10 000
5335 bytes - far larger than anybody would *ever* want, thought I. Needless to
5336 say, somebody hit the limit. I have increased the maximum line length to
5337 20 000 and the maximum data length of concatenated lines to 100 000. I have
5338 also fixed two bugs, because there was no checking on these buffers. Tsk,
5339 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5340 buffer is too small.
5342 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5343 lsearch lookups. Now it does.
5345 46. When parsing a route_list item in a manualroute router, a fixed-length
5346 buffer was used for the list of hosts. I made this 1024 bytes long,
5347 thinking that nobody would ever have a list of hosts that long. Wrong.
5348 Somebody had a whole pile of complicated expansion conditions, and the
5349 string was silently truncated, leading to an expansion error. It turns out
5350 that it is easier to change to an unlimited length (owing to other changes
5351 that have happened since this code was originally written) than to build
5352 structure for giving a limitation error. The length of the item that
5353 expands into the list of hosts is now unlimited.
5355 47. The lsearch lookup could not handle data where the length of text line was
5356 more than 4095 characters. Such lines were truncated, leading to shortened
5357 data being returned. It should now handle lines of any length.
5359 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5360 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5363 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5364 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5365 no longer happens in comment lines. A list of replacements is now placed
5366 at the head of all of the source files, except those whose only change is
5367 to replace PERL_COMMAND in the very first #! line.
5369 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5370 messages on the queue, with a bottom-up merge sort, using code contributed
5371 by Michael Haardt. This should make operations like -bp somewhat faster on
5372 large queues. It won't affect queue runners, except when queue_run_in_order
5375 51. Installed eximstats 1.31 in the distribution.
5377 52. Added support for SRV lookups to the dnslookup router.
5379 53. If an ACL referred to $message_body or $message_body_end, the value was not
5380 reset for any messages that followed in the same SMTP session.
5382 54. The store-handling optimization for building very long strings was not
5383 differentiating between the different store pools. I don't think this
5384 actually made any difference in practice, but I've tidied it.
5386 55. While running the routers to verify a sender address, $sender_address
5387 was still set to the sender address. This is wrong, because when routing to
5388 send a bounce to the sender, it would be empty. Therefore, I have changed
5389 it so that, while verifying a sender address, $sender_address is set to <>.
5390 (There is no change to what happens when verifying a recipient address.)
5392 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5393 target A or AAAA records (if not already returned) without resetting the
5394 qualify_single or search_parents options of the DNS resolver. These are
5395 inappropriate in this case because the targets of MX and SRV records must
5396 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5397 target that, when qualified, matched something in the local domain. These
5398 two options are now turned off when doing these lookups.
5400 57. It seems that at least some releases of Reiserfs (which does not have the
5401 concept of a fixed number of inodes) returns zero and not -1 for the
5402 number of available inodes. This interacted badly with check_spool_inodes,
5403 which assumed that -1 was the "no such thing" setting. What I have done is
5404 to check that the total number of inodes is greater than zero before doing
5405 the test of how many are available.
5407 58. When a "warn" ACL statement has a log_message modifier, the message is
5408 remembered, and not repeated. This is to avoid a lot of repetition when a
5409 message has many recipients that cause the same warning to be written.
5410 Howewer, Exim was preserving the list of already written lines for an
5411 entire SMTP session, which doesn't seem right. The memory is now reset if a
5412 new message is started.
5414 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5415 debugging output unless log_rewrite was also set.
5417 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5418 of (int)(handle) when we know that handle contains (void *)(-1).
5420 61. The Exim daemon panic-logs an error return when it closes the incoming
5421 connection. However "connection reset by peer" seems to be common, and
5422 isn't really an error worthy of noting specially, so that particular error
5425 62. When Exim is trying to find all the local interfaces, it used to panic and
5426 die if the ioctl to get the interface flags failed. However, it seems that
5427 on at least one OS (Solaris 9) it is possible to have an interface that is
5428 included in the list of interfaces, but for which you get a failure error
5429 for this call. This happens when the interface is not "plumbed" into a
5430 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5431 failure of the "get flags" call assumes that the interface is down.
5433 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5434 makes life easier for people who are doing arithmetic on fields extracted
5435 from dates, where you often get leading zeros that should not be
5436 interpreted as octal.
5438 64. Added qualify_domain to the redirect router, to override the global
5441 65. If a pathologically long header line contained very many addresses (the
5442 report of this problem mentioned 10 000) and each of them was rewritten,
5443 Exim could use up a very large amount of memory. (It kept on making new
5444 copies of the header line as it rewrote, and never released the old ones.)
5445 At the expense of a bit more processing, the header rewriting function has
5446 been changed so that it no longer eats memory in this way.
5448 66. The generation of the Received: header has been moved from the time that a
5449 message starts to be received, to the time that it finishes. The timestamp
5450 in the Received: header should now be very close to that of the <= log
5451 line. There are two side-effects of this change:
5453 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5454 logged header lines no longer include the local Received: line, because
5455 it has not yet been created. The same applies to a copy of the message
5456 that is returned to a non-SMTP sender when a message is rejected.
5458 (b) When a filter file is tested using -bf, no additional Received: header
5459 is added to the test message. After some thought, I decided that this
5462 This change does not affect the value of $received_for. It is still set
5463 after address rewriting, but before local_scan() is called.
5465 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5467 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5468 gave an unhelpful panic error message, and a defer error. I have managed to
5469 change this behaviour so that it now rejects any supplied certificate,
5470 which seems right, as the list of acceptable certificates is empty.
5472 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5473 gave an unhelpful defer error. I have not managed to make this reject any
5474 supplied certificates, but the error message it gives is "no certificate
5475 supplied", which is not helpful.
5477 70. exigrep's output now also includes lines that are not associated with any
5478 message, but which match the given pattern. Implemented by a patch from
5479 Martin Sluka, which also tidied up the Perl a bit.
5481 71. Recipient callout verification, like sender verification, was using <> in
5482 the MAIL FROM command. This isn't really the right thing, since the actual
5483 sender may affect whether the remote host accepts the recipient or not. I
5484 have changed it to use the actual sender in the callout; this means that
5485 the cache record is now keyed on a recipient/sender pair, not just the
5486 recipient address. There doesn't seem to be a real danger of callout loops,
5487 since a callout by the remote host to check the sender would use <>.
5488 [SEE ABOVE: changed after hitting problems.]
5490 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5491 temporary errors. However, in the case of such a code being given after
5492 the end of a data transmission (i.e. after ".") Exim was failing to write
5493 a retry record for the message. (Yes, there was some broken host that was
5494 actually sending 8xx at this point.)
5496 73. An unknown lookup type in a host list could cause Exim to panic-die when
5497 the list was checked. (An example that provoked this was putting <; in the
5498 middle of a list instead of at the start.) If this happened during a DATA
5499 ACL check, a -D file could be left lying around. This kind of configuration
5500 error no longer causes Exim to die; instead it causes a defer errror. The
5501 incident is still logged to the main and panic logs.
5503 74. Buglet left over from Exim 3 conversion. The message "too many messages
5504 in one connection" was written to the rejectlog but not the mainlog, except
5505 when address rewriting (yes!) was being logged.
5507 75. Added write_rejectlog option.
5509 76. When a system filter was run not as root (that is, when system_filter_user
5510 was set), the values of the $n variables were not being returned to the
5511 main process; thus, they were not subsequently available in the $sn
5514 77. Added +return_path_on_delivery log selector.
5516 78. A connection timeout was being treated differently from recipients deferred
5517 when testing hosts_max_try with a message that was older than the host's
5518 retry timeout. (The host should not be counted, thus allowing all hosts to
5519 be tried at least once before bouncing.) This may have been the cause of an
5520 occasionally reported bug whereby a message would remain on the queue
5521 longer than the retry timeout, but would be bounced if a delivery was
5522 forced. I say "may" because I never totally pinned down the problem;
5523 setting up timeout/retry tests is difficult. See also the next item.
5525 79. The ultimate address timeout was not being applied to errors that involved
5526 a combination of host plus message (for example, a timeout on a MAIL
5527 command). When an address resolved to a number of possible hosts, and they
5528 were not all tried for each delivery (e.g. because of hosts_max_try), a
5529 message could remain on the queue longer than the retry timeout.
5531 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5534 81. Fixed an obscure SMTP outgoing bug which required at least the following
5535 conditions: (a) there was another message waiting for the same server;
5536 (b) the server returned 5xx to all RCPT commands in the first message so
5537 that the message was not completed; (c) the server dropped the connection
5538 or gave a negative response to the RSET that Exim sends to abort the
5539 transaction. The observed case was a dropped connection after DATA that had
5540 been sent in pipelining mode. That is, the server had advertised PIPELINING
5541 but was not implementing it correctly. The effect of the bug was incorrect
5542 behaviour, such as trying another host, and this could lead to a crash.
5548 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5549 and daemon.c were passed as pointers to ints; they should have been
5550 pointers to socklen_t variables (which are typically unsigned ints).
5552 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5555 3. Fixed a really odd bug that affected only the testing scheme; patching a
5556 certain fixed string in the binary changed the value of another string that
5557 happened to be identical to the end of the original first string.
5559 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5560 name", it returns that address as the IP address. On some operating
5561 systems (e.g. Solaris), it also passes back the IP address string as the
5562 "host name". However, on others (e.g. Linux), it passes back an empty
5563 string. Exim wasn't checking for this, and was changing the host name to an
5564 empty string, assuming it had been canonicized.
5566 5. Although rare, it is permitted to have more than one PTR record for a given
5567 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5568 all the names associated with an address, because they do in Solaris.
5569 However, it seems that they do not in Linux for data that comes from the
5570 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5571 I found this out when I moved to a new Linux workstation and tried to run
5572 the Exim test suite.
5574 To get round this problem I have changed the code so that it now does its
5575 own call to the DNS to look up PTR records when searching for a host name.
5576 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5577 addresses that are only in /etc/hosts are still found.
5579 This behaviour is, however, controlled by an option called host_lookup_
5580 order, which defaults to "bydns:byaddr". If people want to use the other
5581 order, or indeed, just use one or the other means of lookup, they can
5582 specify it in this variable.
5584 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5585 some operating systems, this comes back from gethostbyaddr() as an empty
5586 string, and this is what Exim used to test for. However, it seems that in
5587 other systems, "." is yielded. Exim now tests for this case too.
5589 7. The values of check_spool_space and check_log_space are now held internally
5590 as a number of kilobytes instead of an absolute number of bytes. If a
5591 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5592 kilobyte. This means that much larger values can be stored.
5594 8. Exim monitor: an attempt to get the action menu when not actually pointing
5595 at a message produces an empty menu entitled "No message selected". This
5596 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5597 no entries in it ("Shell widget menu has zero width and/or height"). So I
5598 have added a single, blank menu entry in this case.
5600 9. Added ${quote_local_part.
5602 10. MIME decoding is now applied to the contents of Subject: header lines when
5605 11. Now that a reference to $sender_host_address automatically causes a reverse
5606 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5607 host lookup before query-style lookups in lists that might use this
5608 variable. This has therefore been abolished, and the "net-" prefix is no
5609 longer necessary for query-style lookups.
5611 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5612 have been a typo for LFLAGS, so it has been changed.
5614 13. The install script calls Exim with "-C /dev/null" in order to find the
5615 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5616 to be output. Howeve, since Exim outputs its version number before the
5617 error, it didn't break the script. It just looked ugly. I fixed this by
5618 always allowing "-C /dev/null" if the caller is root.
5620 14. Ignore overlarge ACL variable number when reading spool file - insurance
5621 against a later release with more variables having written the file.
5623 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5624 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5625 this, as well as the form without the "IPv6" on the front.
5627 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5628 OS/Makefile-Darwin file.
5630 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5631 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5633 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5634 variable. However, some OS use ints for inodes. I've added cast to long int
5635 to get rid of the compiler warning.
5637 19. I had forgotten to lock out "/../" in configuration file names when
5638 ALT_CONFIG_PREFIX was set.
5640 20. Routers used for verification do not need to specify transports. However,
5641 if such a router generated a host list, and callout was configured, Exim
5642 crashed, because it could not find a port number from the (non-existent)
5643 transport. It now assumes port 25 in this circumstance.
5645 21. Added the -t option to exigrep.
5647 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5648 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5649 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5650 them is set, LOOKUP_LSEARCH is forced.
5652 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5654 24. Added sender and host information to the "rejected by local_scan()" log
5655 line; previously there was no indication of these.
5657 25. Added .include_if_exists.
5659 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5660 Linux. It turns out that not all file systems support this. Apparently some
5661 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5662 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5663 not supported on the file descriptor, is now ignored when Exim is trying to
5664 sync a directory. This applies only to Linux.
5666 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5668 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5669 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5672 29. Added support for maildirsize files from supplied patch (modified a bit).
5674 30. The use of :fail: followed by an empty string could lead Exim to respond to
5675 sender verification failures with (e.g.):
5677 550 Verification failed for <xxx>
5678 550 Sender verify failed
5680 where the first response line was missing the '-' that indicates it is not
5681 the final line of the response.
5683 31. The loop for finding the name of the user that called Exim had a hardwired
5684 limit of 10; it now uses the value of finduser_retries, which is used for
5685 all other user lookups.
5687 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5690 33. Exim was neglecting to zero errno before one call of strtol() when
5691 expanding a string and expecting an integer value. On some systems this
5692 resulted in spurious "integer overflow" errors. Also, it was casting the
5693 result into an int without checking.
5695 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5696 did not work. The code looks as if it has *never* worked, though it appears
5697 to have been documented since at least releast 1.62. I have made it work.
5699 35. The "timeout_DNS" error in retry rules, also documented since at least
5700 1.62, also never worked. As it isn't clear exactly what this means, and
5701 clearly it isn't a major issue, I have abolished the feature by treating it
5702 as "timeout", and writing a warning to the main and panic logs.
5704 36. The display of retry rules for -brt wasn't always showing the error code
5707 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5708 timeout_connect_A, timeout_connect_MX.
5710 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5711 to the empty sender.
5713 39. The daemon was not analysing the content of -oX till after it had closed
5714 stderr and disconnected from the controlling terminal. This meant that any
5715 syntax errors were only noted on the panic log, and the return code from
5716 the command was 0. By re-arranging the code a little, I've made the
5717 decoding happen first, so such errors now appear on stderr, and the return
5718 code is 1. However, the actual setting up of the sockets still happens in
5719 the disconnected process, so errors there are still only recorded on the
5722 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5723 connections (as happens on some IP stacks) was logged at start up time as
5724 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5725 it from "IPv6 and IPv4", which means that two separate sockets are being
5728 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5729 says whether AF_INET or AF_INET6 was passed as an argument.
5731 42. Exiwhat output was messed up when time zones were included in log
5734 43. Exiwhat now gives more information about the daemon's listening ports,
5735 and whether -tls-on-connect was used.
5737 44. The "port" option of the smtp transport is now expanded.
5739 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5740 silently ignored. Now an error message is written to the main and panic
5743 46. There's a new ACL modifier called "logwrite" which writes to a log file
5744 as soon as it is encountered.
5746 47. Added $local_user_uid and $local_user_gid at routing time.
5748 48. Exim crashed when trying to verify a sender address that was being
5751 49. Exim was recognizing only a space character after ".include". It now also
5752 recognizes a tab character.
5754 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5755 extracting the relevant information from the specification. The man page no
5756 longer contains scrambled data for the -d option, and I've added a section
5757 at the front about calling Exim under different names.
5759 51. Added "extra_headers" argument to the "mail" command in filter files.
5761 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5764 53. Installed eximstats 1.29.
5766 54. Added transport_filter_timeout as a generic transport option.
5768 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5769 Cc: header lines. This was required by RFC 822, but it not required by RFC
5772 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5773 incoming messages that did not have them. Now it does so only if the
5774 message originates locally, that is, if there is no associated remote host
5775 address. When Resent- header lines are present, this applies to the Resent-
5776 lines rather than the non-Resent- lines.
5778 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5779 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5781 58. Messages for configuration errors now include the name of the main
5782 configuration file - useful now that there may be more than one file in a
5783 list (.included file names were always shown).
5785 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5786 for those rare installations that do not start the daemon as root or run it
5787 setuid root. I've cut out the call to initgroups() if the daemon is not
5790 60. The Exim user and group can now be bound into the binary as text strings
5791 that are looked up at the start of Exim's processing.
5793 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5795 62. Added $mailstore_basename variable.
5797 63. Installed patch to sieve.c from Michael Haardt.
5799 64. When Exim failed to open the panic log after failing to open the main log,
5800 the original message it was trying to log was written to stderr and debug
5801 output, but if they were not available (the usual case in production), it
5802 was lost. Now it is written to syslog before the two lines that record the
5803 failures to open the logs.
5805 65. Users' Exim filters run in subprocesses under the user's uid. It is
5806 possible for a "deliver" command or an alias in a "personal" command to
5807 provoke an address rewrite. If logging of address rewriting is configured,
5808 this fails because the process is not running as root or exim. There may be
5809 a better way of dealing with this, but for the moment (because 4.30 needs
5810 to be released), I have disabled address rewrite logging when running a
5811 filter in a non-root, non-exim process.
5817 1. The buildconfig auxiliary program wasn't quoting the value set for
5818 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5819 not defined. This bug was masked in 4.22 by the effect that was fixed in
5822 2. Some messages that were rejected after a message id was allocated were
5823 shown as "incomplete" by exigrep. It no longer does this for messages that
5824 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5826 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5827 have allow_domain_literals set, the ID did not get logged in the <= line.
5828 Domain literals are now always recognized in Message-ID: header lines.
5830 4. The first argument for a ${extract expansion item is the key name or field
5831 number. Leading and trailing spaces in this item were not being ignored,
5832 causing some misleading effects.
5834 5. When deliver_drop_privilege was set, single queue runner processes started
5835 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5836 same command in the process it spins off) were not dropping privilege.
5838 6. When the daemon running as "exim" started a queue runner, it always
5839 re-executed Exim in the spun-off process. This is a waste of effort when
5840 deliver_drop_privilege is set. The new process now just calls the
5841 queue-runner function directly.
5847 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5850 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5851 ignored. Though the use of -oP was forcing the writing of a pid file, it
5852 was always written to the default place.
5854 3. If the message "no IP address found for host xxxx" is generated during
5855 incoming verification, it is now followed by identification of the incoming
5856 connection (so you can more easily find what provoked it).
5858 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5860 5. Added some features to "harden" Exim a bit more against certain attacks:
5862 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5863 be put in Local/Makefile. This is like the never_users runtime option,
5864 but it cannot be overridden. The default setting is "root".
5866 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5867 prefix string with which any file named in a -C command line option
5870 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5871 is retained for -C and -D only if the caller of Exim is root. Without
5872 it, the exim user may also use -C and -D and retain privilege.
5874 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5875 command line option is disabled.
5877 6. Macro names set by the -D option must start with an upper case letter, just
5878 like macro names defined in the configuration file.
5880 7. Added "dereference=" facility to LDAP.
5882 8. Two instances of the typo "uknown" in the source files are fixed.
5884 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5885 the Configure-Makefile script screwed up while processing it.
5887 10. Incorporated PCRE 4.4.
5889 11. The SMTP synchronization check was not operating right at the start of an
5890 SMTP session. For example, it could not catch a HELO sent before the client
5891 waited for the greeting. There is now a check for outstanding input at the
5892 point when the greeting is written. Because of the duplex, asynchronous
5893 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5894 way, but not yet received, when the check is performed.
5896 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5897 on TCP/IP sockets, because this apparently causes some broken clients to
5900 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5901 unchanged) from the Cygwin maintainer.
5903 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5904 was supported instead of testing for mbx. Effectively a typo.
5906 15. The spa authenticator server code was not checking that the input it
5907 received was valid base64.
5909 16. The debug output line for the "set" modifier in ACLs was not showing the
5910 name of the variable that was being set.
5912 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5914 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5915 lookup on its own. Something else was needed to trigger the lookup. For
5916 example, a match in host_lookup or the need for a host name in a host list.
5917 Now, if $sender_host_name is referenced and the host name has not yet been
5918 looked up, a lookup is performed. If the lookup fails, the variable remains
5919 empty, and $host_lookup_failed is set to "1".
5921 19. Added "eqi" as a case-independent comparison operator.
5923 20. The saslauthd authentication condition could segfault if neither service
5924 nor realm was specified.
5926 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5927 error message that was logged was misleading, and incoming SMTP
5928 connections were dropped. The message is now more accurate, and temporary
5929 errors are given to SMTP connections.
5931 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5932 (except RSET) with a 421 error, until QUIT is received. However, it was
5933 failing to send a response to QUIT.
5935 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5936 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5937 ACL is now run after the helo_try_verify_hosts code.
5939 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5940 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5941 case-independent, so other case variants are also recognized). Apparently
5942 some systems use these upper case variants.
5944 25. If more than two messages were waiting for the same host, and a transport
5945 filter was specified for the transport, Exim sent two messages over the
5946 same TCP/IP connection, and then failed with "socket operation on non-
5947 socket" when it tried to send the third.
5949 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5951 27. The extern definition of crypt16() in expand.c was not being excluded when
5952 the OS had its own crypt16() function.
5954 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5955 as a preferred synonym for return_size_limit, both as an option and as an
5958 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5960 30. Changed the default configuration ACL to relax the local part checking rule
5961 for addresses that are not in any local domains. For these addresses,
5962 slashes and pipe symbols are allowed within local parts, but the sequence
5963 /../ is explicitly forbidden.
5965 31. SPA server authentication was not clearing the challenge buffer before
5968 32. log_message in a "warn" ACL statement was writing to the reject log as
5969 well as to the main log, which contradicts the documentation and doesn't
5970 seem right (because no rejection is happening). So I have stopped it.
5972 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5973 However, I am unable to do any testing of this.
5975 34. Fixed an infelicity in the appendfile transport. When checking directories
5976 for a mailbox, to see if any needed to be created, it was accidentally
5977 using path names with one or more superfluous leading slashes; tracing
5978 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5980 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5981 contents are added to the log line that is written for every discarded
5982 recipient. (Previously a log_message setting was ignored.)
5984 36. The ${quote: operator now quotes the string if it is empty.
5986 37. The install script runs exim in order to find its version number. If for
5987 some reason other than non-existence or emptiness, which it checks, it
5988 could not run './exim', it was installing it with an empty version number,
5989 i.e. as "exim-". This error state is now caught, and the installation is
5992 38. An argument was missing from the function that creates an error message
5993 when Exim fails to connect to the socket for saslauthd authentication.
5994 This could cause Exim to crash, or give a corrupted message.
5996 39. Added isip, isip4, and isip6 to ${if conditions.
5998 40. The ACL variables $acl_xx are now saved with the message, and can be
5999 accessed later in routers, transports, and filters.
6001 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6002 strings in the file are not string-expanded.
6004 42. If a MAIL command specified a SIZE value that was too large to fit into an
6005 int variable, the check against message_size_limit failed. Such values are
6006 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6007 day this will have to be increased, but I don't think I want to be around
6008 when emails are that large.
6015 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6016 iconv() is not standard in FreeBSD.
6018 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6019 IPv6 enabled. The observed symptom was a segmentation fault on return from
6020 the function os_common_find_running_interfaces() in src/os.c.
6022 3. In the check_special_case() function in daemon.c I had used "errno" as an
6023 argument name, which causes warnings on some systems. This was basically a
6024 typo, since it was named "eno" in the comments!
6026 4. The code that waits for the clock to tick (at a resolution of some fraction
6027 of a second) so as to ensure message-id uniqueness was always waiting for
6028 at least one whole tick, when it could have waited for less. [This is
6029 almost certainly not relevant at current processor speeds, where it is
6030 unlikely to ever wait at all. But we try to future-proof.]
6032 5. The function that sleeps for a time interval that includes fractions of a
6033 second contained a race. It did not block SIGALRM between setting the
6034 timer, and suspending (a couple of lines later). If the interval was short
6035 and the sigsuspend() was delayed until after it had expired, the suspension
6036 never ended. On busy systems this could lead to processes getting stuck for
6039 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6040 process, before it forks any delivery processes. The open lookup caching
6041 mechanism meant that the open file or database connection was passed into
6042 the delivery process. The problem was that delivery processes always tidy
6043 up cached lookup data. This could cause a problem for the next delivery
6044 process started by the queue runner, because the external queue runner
6045 process does not know about the closure. So the next delivery process
6046 still has data in the lookup cache. In the case of a file lookup, there was
6047 no problem because closing a file descriptor in a subprocess doesn't affect
6048 the parent. However, if the lookup was caching a connection to a database,
6049 the connection was closed, and the second delivery process was likely to
6050 see errors such as "PGSQL: query failed: server closed the connection
6051 unexpectedly". The problem has been fixed by closing all cached lookups
6052 in a queue runner before running a delivery process.
6054 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6055 seem to have the "const" qualifier which it has on other OS. I've
6058 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6059 *of the same type* (client or server) with the same public name that an
6060 error should be diagnosed.
6062 9. When Exim looked up a host name for an IP address, but failed to find the
6063 original IP address when looking up the host name (a safety check), it
6064 output the message "<ip address> does not match any IP for NULL", which was
6065 confusing, to say the least. The bug was that the host name should have
6066 appeared instead of "NULL".
6068 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6069 user that is built into the binary, and the -C or -D options is used, root
6070 privilege is dropped before the configuration file is read. In addition,
6071 logging is switched to stderr instead of the normal log files. If the
6072 configuration then re-defines the Exim user, the unprivileged environment
6073 is probably not what is expected, so Exim logs a panic warning message (but
6076 However, if deliver_drop_privilege is set, the unprivileged state may well
6077 be exactly what is intended, so the warning has been cut out in that case,
6078 and Exim is allowed to try to write to its normal log files.
6084 1. smtp_return_error_details was not giving details for temporary sender
6085 or receiver verification errors.
6087 2. Diagnose a configuration error if two authenticators have the same public
6090 3. Exim used not to create the message log file for a message until the first
6091 delivery attempt. This could be confusing when incoming messages were held
6092 for policy or load reasons. The message log file is now created at the time
6093 the message is received, and an initial "Received" line is written to it.
6095 4. The automatically generated man page for command line options had a minor
6096 bug that caused no ill effects; however, a more serious problem was that
6097 the procedure for building the man page automatically didn't always
6098 operate. Consequently, release 4.20 contains an out-of-date version. This
6099 shouldn't happen again.
6101 5. When building Exim with embedded Perl support, the script that builds the
6102 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6103 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6105 6. The freeze_tell option was not being used for messages that were frozen on
6106 arrival, either by an ACL or by local_scan().
6108 7. Added the smtp_incomplete_transaction log selector.
6110 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6111 was accepting AUTH without a new EHLO.
6113 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6114 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6117 10. Logging of TCP/IP connections (when configured) now happens in the main
6118 daemon process instead of the child process, so that the TCP/IP connection
6119 count is more accurate (but it can never be perfect).
6121 11. The use of "drop" in a nested ACL was not being handled correctly in the
6122 outer ACL. Now, if condition failure induced by the nested "drop" causes
6123 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6124 or "require"), the connection is dropped.
6126 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6127 that yield "discard" can now be used with an "accept" or a "discard" verb,
6128 but an error is generated for any others (because I can't see a useful way
6129 to define what should happen).
6131 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6132 are now processed in the same way as lines in the Exim configuration file.
6133 In particular, continuation lines are supported.
6135 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6137 15. Added -ti meaning -t -i.
6139 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6140 domains, and warn by logging if others are found.
6142 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6143 returned from GIFCONF when Exim is trying to find the list of interfaces on
6144 a host. The code in os.c has been modified to copy each ifreq to an aligned
6145 structure in all cases.
6147 Also, in some cases, the returned ifreq's were being copied to a 'struct
6148 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6149 means the last couple of bytes of an IPv6 address could be chopped if the
6150 ifreq contained only a normal sockaddr (14 bytes storage).
6152 18. Named domain lists were not supported in the hosts_treat_as_local option.
6153 An entry such as +xxxx was not recognized, and was treated as a literal
6156 19. Ensure that header lines added by a DATA ACL are included in the reject log
6157 if the ACL subsequently rejects the message.
6159 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6160 MD5 (which is deprecated).
6162 21. When testing a filter file using -bf, Exim was writing a message when it
6163 took the sender from a "From " line in the message, but it was not doing so
6164 when it took $return_path from a Return-Path: header line. It now does.
6166 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6167 with a valid header line field name (a series of printing characters
6168 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6170 23. Changed "disc" in the source to "disk" to conform to the documentation and
6171 the book and for uniformity.
6173 24. Ignore Sendmail's -Ooption=value command line item.
6175 25. When execve() failed while trying to run a command in a pipe transport,
6176 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6177 could be confused with a return value of 69 from the command itself. This
6178 has been changed to 127, the value the shell returns if it is asked to run
6179 a non-existent command. The wording for the related log line suggests a
6180 non-existent command as the problem.
6182 26. If received_header_text expands to an empty string, do not add a Received:
6183 header line to the message. (Well, it adds a token one on the spool, but
6184 marks it "old" so that it doesn't get used or transmitted.)
6186 27. Installed eximstats 1.28 (addition of -nt option).
6188 28. There was no check for failure on the call to getsockname() in the daemon
6189 code. This can fail if there is a shortage of resources on the system, with
6190 ENOMEM, for example. A temporary error is now given on failure.
6192 29. Contrary to the C standard, it seems that in some environments, the
6193 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6194 program. Exim now does this explicitly; it affects the formatting of
6195 timestamps using strftime().
6197 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6198 complaints. I've now initialized all the variables, to avoid this.
6200 32. Header lines added by a system filter were not being "seen" during
6201 transport-time rewrites.
6203 33. The info_callback() function passed to OpenSSL is set up with type void
6204 (*)(SSL *, int, int), as described somewhere. However, when calling the
6205 function (actually a macro) that sets it up, the type void(*)() is
6206 expected. I've put in a cast to prevent warnings from picky compilers.
6208 34. If a DNS black list lookup found a CNAME record, but there were no A
6209 records associated with the domain it pointed at, Exim crashed.
6211 35. If a DNS black list lookup returned more than one A record, Exim ignored
6212 all but the first. It now scans all returned addresses if a particular IP
6213 value is being sought. In this situation, the contents of the
6214 $dnslist_value variable are a list of all the addresses, separated by a
6217 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6218 to accept a host name and all its aliases if the forward lookup for any of
6219 them yielded the IP address of the incoming connection. Now it accepts only
6220 those names whose forward lookup yields the correct IP address. Any other
6221 names are discarded. This closes a loophole whereby a rogue DNS
6222 administrator could create reverse DNS records to break through a
6223 wildcarded host restriction in an ACL.
6225 37. If a user filter or a system filter that ran in a subprocess used any of
6226 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6227 the wrong values were passed to the pipe command ($thisaddress had the
6228 value of $0, $0 had the value of $1, etc). This bug was introduced by
6229 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6231 38. Improved the line breaking for long SMTP error messages from ACLs.
6232 Previously, if there was no break point between 40 and 75 characters, Exim
6233 left the rest of the message alone. Two changes have been made: (a) I've
6234 reduced the minimum length to 35 characters; (b) if it can't find a break
6235 point between 35 and 75 characters, it looks ahead and uses the first one
6236 that it finds. This may give the occasional overlong line, but at least the
6237 remaining text gets split now.
6239 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6240 file descriptors might be low, and that setting 1000 would always raise it.
6241 It turns out that in some environments, the limit is already over 1000 and
6242 that lowering it causes trouble. So now Exim takes care not to decrease it.
6244 40. When delivering a message, the value of $return_path is set to $sender_
6245 address at the start of routing (routers may change the value). By an
6246 oversight, this default was not being set up when an address was tested by
6247 -bt or -bv, which affected the outcome if any router or filter referred to
6250 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6251 log only when "message" or "log_message" are set. However, if one of the
6252 conditions was an address verification, or a call to a nested ACL, the
6253 messages generated by the underlying test were being passed through. This
6254 no longer happens. The underlying message is available in $acl_verify_
6255 message for both "message" and "log_message" expansions, so it can be
6256 passed through if needed.
6258 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6259 new expansion $bh_ to give the encoded byte string without charset
6260 translation. Translation happens only if iconv() is available; HAVE_ICONV
6261 indicates this at build time. HEADERS_CHARSET gives the charset to
6262 translate to; headers_charset can change it in the configuration, and
6263 "headers charset" can change it in an individual filter file.
6265 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6266 that creates RFC 2047 encoded "words" labels them as that charset instead
6267 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6268 expansion operator; (ii) when Exim creates a From: line for a local
6269 message; (iii) when a header line is rewritten to include a "phrase" part.
6271 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6272 buggy, causing it to skip the first lines of messages whose message ID
6273 ended in 'D'. This would not have bitten before Exim release 4.14, because
6274 message IDs were unlikely to end in 'D' before then. The effect was to have
6275 incorrect size information for certain domains.
6277 45. #include "config.h" was missing at the start of the crypt16.c module. This
6278 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6281 46. If there was a timeout during a "random" callout check, Exim treated it as
6282 a failure of the random address, and carried on sending RSET and the real
6283 address. If the delay was just some slowness somewhere, the response to the
6284 original RCPT would be taken as a response to RSET and so on, causing
6285 mayhem of various kinds.
6287 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6288 when I implemented it. It didn't allow for the fact that some option values
6289 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6290 the right test for positive values.
6292 48. Domain names in DNS records are case-independent. Exim always looks them up
6293 in lower case. Some resolvers return domain names in exactly the case they
6294 appear in the zone file, that is, they may contain uppercase letters. Not
6295 all resolvers do this - some return always lower case. Exim was treating a
6296 change of case by a resolver as a change of domain, similar to a widening
6297 of a domain abbreviation. This triggered its re-routing code and so it was
6298 trying to route what was effectively the same domain again. This normally
6299 caused routing to fail (because the router wouldn't handle the domain
6300 twice). Now Exim checks for this case specially, and just changes the
6301 casing of the domain that it ultimately uses when it transmits the message
6304 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6307 50. If a filter generated a file delivery with a non-absolute name (possible if
6308 no home directory exists for the router), the forbid_file option was not
6311 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6312 the existing equality matching.
6314 52. Exim was using ints instead of ino_t variables in some places where it was
6315 dealing with inode numbers.
6317 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6318 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6319 called TMPDIR, and if it finds it is different, it changes its value.
6321 54. The smtp_printf() function is now made available to local_scan() so
6322 additional output lines can be written before returning. There is also an
6323 smtp_fflush() function to enable the detection of a dropped connection.
6324 The variables smtp_input and smtp_batched_input are exported to
6327 55. Changed the default runtime configuration: the message "Unknown user"
6328 has been removed from the ACL, and instead placed on the localuser router,
6329 using the cannot_route_message feature. This means that any verification
6330 failures that generate their own messages won't get overridden. Similarly,
6331 the "Unrouteable address" message that was in the ACL for unverifiable
6332 relay addresses has also been removed.
6334 56. Added hosts_avoid_esmtp to the smtp transport.
6336 57. The exicyclog script was not checking for the esoteric option
6337 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6338 will work only if exicyclog is run under the appropriate euid.
6340 58. Following a discussion on the list, the rules by which Exim recognises line
6341 endings on incoming messages have been changed. The -dropcr and drop_cr
6342 options are now no-ops, retained only for backwards compatibility. The
6343 following line terminators are recognized: LF CRLF CR. However, special
6344 processing applies to CR:
6346 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6347 nor a local message in the state where . is a terminator.
6349 (ii) If a bare CR is encountered in a header line, an extra space is added
6350 after the line terminator so as not to end the header. The reasoning
6351 behind this is that bare CRs in header lines are most likely either
6352 to be mistakes, or people trying to play silly games.
6354 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6355 was being incorrectly given as 18 bytes larger than it should have been.
6356 This is a VOB (very old bug).
6358 60. This may never have affected anything current, but just in case it has:
6359 When the local host is found other than at the start of a list of hosts,
6360 the local host, those with the same MX, and any that follow, are discarded.
6361 When the list in question was part of a longer list of hosts, the following
6362 hosts (not currently being processed) were also being discarded. This no
6363 longer happens. I'm not sure if this situation could ever has previously
6366 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6369 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6370 header. This is something that is recommended in a new Internet Draft, and
6371 is something that is documented as being done by Sendmail. There are two
6372 possible values. For messages generated by the autoreply transport, Exim
6375 Auto-Submitted: auto-replied
6377 whereas for all other generated messages (e.g. bounces) it adds
6379 Auto-Submitted: auto-generated
6381 63. The "personal" condition in filters now includes a test for the
6382 Auto-Submitted: header. If it contains the string "auto-" the message it
6383 not considered personal.
6385 64. Added rcpt_include_affixes as a generic transport option.
6387 65. Added queue_only_override (default true).
6389 66. Added the syslog_duplication option.
6391 67. If what should have been the first header line of a message consisted of
6392 a space followed by a colon, Exim was mis-interpreting it as a header line.
6393 It isn't of course - it is syntactically invalid and should therefore be
6394 treated as the start of the message body. The misbehaviour could have
6395 caused a number of strange effects, including loss of data in subsequent
6396 header lines, and spool format errors.
6398 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6399 client host had authenticated. This control can now be exercised by an ACL
6400 for more flexibility.
6402 69. By default, callouts do not happen when testing with -bh. There is now a
6403 variant, -bhc, which does actually run the callout code, including
6404 consulting and updating the callout cache.
6406 70. Added support for saslauthd authentication, courtesy of Alexander
6409 71. If statvfs() failed on the spool or log directories while checking their
6410 size for availability, Exim confusingly gave the error "space shortage".
6411 Furthermore, in debugging mode it crashed with a floating point exception.
6412 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6413 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6414 serious problem, Exim now writes to the main and panic logs when this
6415 happens, with details of the failure. It then refuses to accept the
6416 incoming message, giving the message "spool directory problem" or "log
6417 directory problem" with a 421 code for SMTP messages.
6419 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6420 0, 1, and 2 exist, because some OS complain for execs without them (see
6421 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6422 descriptors. However, the code omitted to check that the open succeeded,
6423 causing mysterious errors if for some reason the permissions on /dev/null
6424 got screwed. Now Exim writes a message to the main and panic logs, and
6425 bombs out if it can't open /dev/null.
6427 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6428 interact so that it is all more flexible. It is supposed to remain
6429 backwards compatible. Also added extra_local_interfaces.
6431 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6432 to bomb out with an assertion failure - to the client this appears as a
6433 connection drop. This problem occurs in the part of the code that was taken
6434 from the Samba project. Fortunately, the assertion is in a very simple
6435 function, so I have fixed this by reproducing the function inline in the
6436 one place where it is called, and arranging for authentication to fail
6437 instead of killing the process with assert().
6439 75. The SPA client code was not working when the server requested OEM rather
6440 than Unicode encoding.
6442 76. Added code to make require_files with a specific uid setting more usable in
6443 the case where statting the file as root fails - usually a non-root-mounted
6444 NFS file system. When this happens and the failure is EACCES, Exim now
6445 forks a subprocess and does the per-uid checking as the relevant uid.
6447 77. Added process_log_path.
6449 78. If log_file_path was not explicitly set, a setting of check_log_space or
6450 check_log_inodes was ignored.
6452 79. If a space check for the spool or log partitions fails, the incident is now
6453 logged. Of course, in the latter case the data may get lost...
6455 80. Added the %p formatting code to string_format() so that it can be used to
6456 print addresses in debug_print(). Adjusted all the address printing in the
6457 debugging in store.c to use %p rather than %d.
6459 81. There was a concern that a line of code in smtp_in.c could overflow a
6460 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6461 initially expressed, the concern was not well-founded, because trailing
6462 spaces are removed early. However, if the trailing spaces were followed by
6463 a NULL, they did not get removed, so the overflow was possible. Two fixes
6466 (a) I re-wrote the offending code in a cleaner fashion.
6467 (b) If an incoming SMTP command contains a NULL character, it is rejected
6470 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6471 runs initgroups(), so that if the Exim user is in any additional groups,
6472 they will be used during message reception.
6478 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob