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 nessage
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.
170 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
173 JH/02 The smtp transport option "multi_domain" is now expanded.
175 JH/03 The smtp transport now requests PRDR by default, if the server offers
178 JH/04 Certificate name checking on server certificates, when exim is a client,
179 is now done by default. The transport option tls_verify_cert_hostnames
180 can be used to disable this per-host. The build option
181 EXPERIMENTAL_CERTNAMES is withdrawn.
183 JH/05 The value of the tls_verify_certificates smtp transport and main options
184 default to the word "system" to access the system default CA bundle.
185 For GnuTLS, only version 3.0.20 or later.
187 JH/06 Verification of the server certificate for a TLS connection is now tried
188 (but not required) by default. The verification status is now logged by
189 default, for both outbound TLS and client-certificate supplying inbound
192 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
195 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
196 Status Notification (bounce) messages are now MIME format per RFC 3464.
197 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
198 under the control of the dsn_advertise_hosts option, and routers may
199 have a dsn_lasthop option.
201 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
202 default, modifiable by a malware= option. The list separator for
203 the options can now be changed in the usual way. Bug 68.
205 JH/10 The smtp_receive_timeout main option is now expanded before use.
207 JH/11 The incoming_interface log option now also enables logging of the
208 local interface on delivery outgoing connections.
210 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
211 if the interface and destination host and port all match.
213 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
216 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
217 Patch from Andrew Lewis.
219 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
220 now supports optional time-restrictions, weighting, and priority
221 modifiers per server. Patch originally by <rommer@active.by>.
223 JH/16 The spamd_address main option now supports a mixed list of local
224 and remote servers. Remote servers can be IPv6 addresses, and
225 specify a port-range.
227 JH/17 Bug 68: The spamd_address main option now supports an optional
228 timeout value per server.
230 JH/18 Bug 1581: Router and transport options headers_add/remove can
231 now have the list separator specified.
233 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
236 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
239 JH/21 Support for the A6 type of dns record is withdrawn.
241 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
242 rather than the verbs used.
244 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
245 from 255 to 1024 chars.
247 JH/24 Verification callouts now attempt to use TLS by default.
249 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
250 are generic router options now. The defaults didn't change.
252 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
253 Original patch from Alexander Shikoff, worked over by JH.
255 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
256 files if ZCAT_COMMAND is not executable.
258 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
260 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
262 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
263 Normally benign, it bites when the pair was led to by a CNAME;
264 modern usage is to not canoicalize the domain to a CNAME target
265 (and we were inconsistent anyway for A-only vs AAAA+A).
267 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
269 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
270 when evaluating $sender_host_dnssec.
272 JH/31 Check the HELO verification lookup for DNSSEC, adding new
273 $sender_helo_dnssec variable.
275 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
277 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
279 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
281 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
282 documented as working, but never had. Support all but $spam_report.
284 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
285 added for tls authenticator.
290 TL/01 When running the test suite, the README says that variables such as
291 no_msglog_check are global and can be placed anywhere in a specific
292 test's script, however it was observed that placement needed to be near
293 the beginning for it to behave that way. Changed the runtest perl
294 script to read through the entire script once to detect and set these
295 variables, reset to the beginning of the script, and then run through
296 the script parsing/test process like normal.
298 TL/02 The BSD's have an arc4random API. One of the functions to induce
299 adding randomness was arc4random_stir(), but it has been removed in
300 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
301 function when detected.
303 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
304 cause callback expansion.
306 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
307 syntax errors in an expansion can be treated as a string instead of
308 logging or causing an error, due to the internal use of bool_lax
309 instead of bool when processing it.
311 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
312 server certificates when making smtp deliveries.
314 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
316 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
318 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
320 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
321 Merged patch from Sebastian Wiedenroth.
323 JH/05 Fix results-pipe from transport process. Several recipients, combined
324 with certificate use, exposed issues where response data items split
325 over buffer boundaries were not parsed properly. This eventually
326 resulted in duplicates being sent. This issue only became common enough
327 to notice due to the introduction of conection certificate information,
328 the item size being so much larger. Found and fixed by Wolfgang Breyha.
330 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
331 size buffer was used, resulting in syntax errors when an expansion
334 JH/07 Add support for directories of certificates when compiled with a GnuTLS
335 version 3.3.6 or later.
337 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
338 is EXPERIMENTAL_EVENT, the main-configuration and transport options
339 both become "event_action", the variables become $event_name, $event_data
340 and $event_defer_errno. There is a new variable $verify_mode, usable in
341 routers, transports and related events. The tls:cert event is now also
342 raised for inbound connections, if the main configuration event_action
345 TL/06 In test suite, disable OCSP for old versions of openssl which contained
346 early OCSP support, but no stapling (appears to be less than 1.0.0).
348 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
349 server certificate names available under the smtp transport option
350 "tls_verify_cert_hostname" now do not permit multi-component wildcard
353 JH/10 Time-related extraction expansions from certificates now use the main
354 option "timezone" setting for output formatting, and are consistent
355 between OpenSSL and GnuTLS compilations. Bug 1541.
357 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
358 encoded parameter in the incoming message. Bug 1558.
360 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
361 include certificate info, eximon was claiming there were spoolfile
364 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
366 JH/14 Log delivery-related information more consistently, using the sequence
367 "H=<name> [<ip>]" wherever possible.
369 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
370 are problematic for Debian distribution, omit them from the release
373 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
375 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
378 JH/17 Fix a null-indirection in certextract expansions when a nondefault
379 output list separator was used.
384 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
385 checkers that were complaining about end of non-void function with no
388 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
389 This was a regression intruduced in 4.83 by another bugfix.
391 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
393 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
394 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
400 TF/01 Correctly close the server side of TLS when forking for delivery.
402 When a message was received over SMTP with TLS, Exim failed to clear up
403 the incoming connection properly after forking off the child process to
404 deliver the message. In some situations the subsequent outgoing
405 delivery connection happened to have the same fd number as the incoming
406 connection previously had. Exim would try to use TLS and fail, logging
407 a "Bad file descriptor" error.
409 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
410 utilities have not been installed.
412 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
413 temporary space as the ACL may create new global variables.
415 TL/01 LDAP support uses per connection or global context settings, depending
416 upon the detected version of the libraries at build time.
418 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
419 to extract and use the src ip:port in logging and expansions as if it
420 were a direct connection from the outside internet. PPv2 support was
421 updated based on HAProxy spec change in May 2014.
423 JH/02 Add ${listextract {number}{list}{success}{fail}}.
425 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
426 Properly escape header and check for NULL return.
428 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
431 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
433 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
434 characters in header names, implemented as a verify condition.
435 Contributed by Michael Fischer v. Mollard.
437 TL/05 Rename SPF condition results err_perm and err_temp to standardized
438 results permerror and temperror. Previous values are deprecated but
439 still accepted. In a future release, err_perm and err_temp will be
440 completely removed, which will be a backward incompatibility if the
441 ACL tests for either of these two old results. Patch contributed by
442 user bes-internal on the mailing list.
444 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
446 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
447 selectors, in both main and reject logs.
449 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
452 JH/07 Add malware type "sock" for talking to simple daemon.
454 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
456 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
457 routers/transports under cutthrough routing.
459 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
460 numbers. Touch up "bool" conditional to keep the same definition.
462 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
464 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
466 JH/12 Expand items in router/transport headers_add or headers_remove lists
467 individually rather than the list as a whole. Bug 1452.
469 Required for reasonable handling of multiple headers_ options when
470 they may be empty; requires that headers_remove items with embedded
471 colons must have them doubled (or the list-separator changed).
473 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
474 view the policy declared in the DMARC record. Currently, $dmarc_status
475 is a combined value of both the record presence and the result of the
478 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
480 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
481 dnslookup router and the smtp transport (applying to the forward
484 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
485 of ldap servers used for a specific lookup. Patch provided by Heiko
488 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
489 New variable $lookup_dnssec_authenticated for observability.
491 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
492 Patch submitted by Lars Timman.
494 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
496 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
497 Requires trusted mode and valid format message id, aborts otherwise.
498 Patch contributed by Heiko Schlichting.
500 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
501 certextract with support for various fields. Bug 1358.
503 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
504 is requested by default, modifiable by smtp transport option
507 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
508 operate on certificate variables to give certificate fingerprints
509 Also new ${sha256:cert_variable}.
511 JH/23 The PRDR feature is moved from being Experimental into the mainline.
513 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
514 Christian Aistleitner.
516 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
518 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
519 file. Patch from Wolfgang Breyha.
521 JH/25 Expand the coverage of the delivery $host and $host_address to
522 client authenticators run in verify callout. Bug 1476.
524 JH/26 Port service names are now accepted for tls_on_connect_ports, to
525 align with daemon_smtp_ports. Bug 72.
527 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
528 support and error reporting did not work properly.
530 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
531 and is readable. Patch from Andrew Colin Kissa.
533 TL/14 Enhance documentation of ${run expansion and how it parses the
534 commandline after expansion, particularly in the case when an
535 unquoted variable expansion results in an empty value.
537 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
539 PP/02 Fix internal collision of T_APL on systems which support RFC3123
540 by renaming away from it. Addresses GH issue 15, reported by
543 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
545 TL/15 SECURITY: prevent double expansion in math comparison functions
546 (can expand unsanitized data). Not remotely exploitable.
553 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
555 PP/02 Make -n do something, by making it not do something.
556 When combined with -bP, the name of an option is not output.
558 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
561 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
562 $sender_host_name and config options to manage this, and basic check
565 PP/05 DSCP support for outbound connections and control modifier for inbound.
567 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
568 (Only plugin which currently uses this is kerberos4, which nobody should
569 be using, but we should make it available and other future plugins might
570 conceivably use it, even though it would break NAT; stuff *should* be
571 using channel bindings instead).
573 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
574 name; added for Sendmail compatibility; requires admin caller.
575 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
576 just ignore it); requires trusted caller.
577 Also parse but ignore: -Ac -Am -X<logfile>
580 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
582 TL/02 Add +smtp_confirmation as a default logging option.
584 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
585 Patch by Magnus Holmgren from 2007-02-20.
587 TL/04 Bugzilla 1281 - Spec typo.
588 Bugzilla 1283 - Spec typo.
589 Bugzilla 1290 - Spec grammar fixes.
591 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
593 TL/06 Add Experimental DMARC support using libopendmarc libraries.
595 TL/07 Fix an out of order global option causing a segfault. Reported to dev
596 mailing list by by Dmitry Isaikin.
598 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
600 JH/02 Support "G" suffix to numbers in ${if comparisons.
602 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
604 NM/01 Bugzilla 1197 - Spec typo
605 Bugzilla 1196 - Spec examples corrections
607 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
609 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
610 gnutls_enable_pkcs11, but renamed to more accurately indicate its
613 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
614 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
616 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
617 "acl {{name}{arg}...}", and optional args on acl condition
620 JH/05 Permit multiple router/transport headers_add/remove lines.
622 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
624 JH/07 Avoid using a waiting database for a single-message-only transport.
625 Performance patch from Paul Fisher. Bugzilla 1262.
627 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
630 JH/09 Add $headers_added variable, with content from use of ACL modifier
631 add_header (but not yet added to the message). Bugzilla 199.
633 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
634 Pulled from Bugzilla 817 by Wolfgang Breyha.
636 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
638 (nb: this is the same fix as in Exim 4.80.1)
640 JH/11 Add A= logging on delivery lines, and a client_set_id option on
643 JH/12 Add optional authenticated_sender logging to A= and a log_selector
646 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
648 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
649 advertise SMTP AUTH mechanism to us, instead of a generic
650 protocol violation error. Also, make Exim more robust to bad
651 data from the Dovecot auth socket.
653 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
655 When a queue runner is handling a message, Exim first routes the
656 recipient addresses, during which it prunes them based on the retry
657 hints database. After that it attempts to deliver the message to
658 any remaining recipients. It then updates the hints database using
661 So if a recipient address works intermittently, it can get repeatedly
662 deferred at routing time. The retry hints record remains fresh so the
663 address never reaches the final cutoff time.
665 This is a fairly common occurrence when a user is bumping up against
666 their storage quota. Exim had some logic in its local delivery code
667 to deal with this. However it did not apply to per-recipient defers
668 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
670 This change adds a proper retry rule check during routing so that the
671 final cutoff time is checked against the message's age. We only do
672 this check if there is an address retry record and there is not a
673 domain retry record; this implies that previous attempts to handle
674 the address had the retry_use_local_parts option turned on. We use
675 this as an approximation for the destination being like a local
676 delivery, as in LMTP.
678 I suspect this new check makes the old local delivery cutoff check
679 redundant, but I have not verified this so I left the code in place.
681 TF/02 Correct gecos expansion when From: is a prefix of the username.
683 Test 0254 submits a message to Exim with the header
687 When I ran the test suite under the user fanf2, Exim expanded
688 the header to contain my full name, whereas it should have added
689 a Resent-Sender: header. It erroneously treats any prefix of the
690 username as equal to the username.
692 This change corrects that bug.
694 GF/01 DCC debug and logging tidyup
695 Error conditions log to paniclog rather than rejectlog.
696 Debug lines prefixed by "DCC: " to remove any ambiguity.
698 TF/03 Avoid unnecessary rebuilds of lookup-related code.
700 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
701 Bug spotted by Jeremy Harris; was flawed since initial commit.
702 Would have resulted in OCSP responses post-SNI triggering an Exim
703 NULL dereference and crash.
705 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
707 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
708 Bug detection, analysis and fix by Samuel Thibault.
709 Bugzilla 1331, Debian bug #698092.
711 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
713 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
714 Server implementation by Todd Lyons, client by JH.
715 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
716 config variable "prdr_enable" controls whether the server
717 advertises the facility. If the client requests PRDR a new
718 acl_data_smtp_prdr ACL is called once for each recipient, after
719 the body content is received and before the acl_smtp_data ACL.
720 The client is controlled by bolth of: a hosts_try_prdr option
721 on the smtp transport, and the server advertisement.
722 Default client logging of deliveries and rejections involving
723 PRDR are flagged with the string "PRDR".
725 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
726 fclose(). Diagnosis by Todd Lyons.
728 PP/17 Update configure.default to handle IPv6 localhost better.
729 Patch by Alain Williams (plus minor tweaks).
732 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
733 This is now consistent with GnuTLS, and is now documented: the
734 previous undocumented portable approach to treating the option as
735 unset was to force an expansion failure. That still works, and
736 an empty string is now equivalent.
738 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
739 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
740 not performing validation itself.
742 PP/20 Added force_command boolean option to pipe transport.
743 Patch from Nick Koston, of cPanel Inc.
745 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
748 TF/04 Added udpsend ACL modifer and hexquote expansion operator
750 PP/21 Fix eximon continuous updating with timestamped log-files.
751 Broken in a format-string cleanup in 4.80, missed when I repaired the
752 other false fix of the same issue.
753 Report and fix from Heiko Schlichting.
756 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
757 Report from Prashanth Katuri.
759 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
760 It's SecureTransport, so affects any MacOS clients which use the
761 system-integrated TLS libraries, including email clients.
763 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
764 using a MIME ACL for non-SMTP local injection.
765 Report and assistance in diagnosis by Warren Baker.
767 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
769 JH/16 Fix comparisons for 64b. Bugzilla 1385.
771 TL/09 Add expansion variable $authenticated_fail_id to keep track of
772 last id that failed so it may be referenced in subsequent ACL's.
774 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
777 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
778 ldap library initialization, allowing self-signed CA's to be
779 used. Also properly sets require_cert option later in code by
780 using NULL (global ldap config) instead of ldap handle (per
781 session). Bug diagnosis and testing by alxgomz.
783 TL/12 Enhanced documentation in the ratelimit.pl script provided in
784 the src/util/ subdirectory.
786 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
787 renamed to Transport Post Delivery Action by Jeremy Harris, as
790 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
791 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
792 redis_servers = needs to be configured which will be used by the redis
793 lookup. Patch from Warren Baker, of The Packet Hub.
795 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
797 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
798 hostname or reverse DNS when processing a host list. Used suggestions
799 from multiple comments on this bug.
801 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
803 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
804 Missed a few lines, added it to make the runtest require no keyboard
807 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
808 contains upper case chars. Make router use caseful_local_part.
810 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
811 support when GnuTLS has been built with p11-kit.
817 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
819 This, or similar/improved, will also be change PP/11 of 4.82.
825 PP/01 Handle short writes when writing local log-files.
826 In practice, only affects FreeBSD (8 onwards).
827 Bugzilla 1053, with thanks to Dmitry Isaikin.
829 NM/01 Bugzilla 949 - Documentation tweak
831 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
834 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
836 PP/02 Implemented gsasl authenticator.
838 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
840 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
841 `pkg-config foo` for cflags/libs.
843 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
844 with rest of GSASL and with heimdal_gssapi.
846 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
847 `pkg-config foo` for cflags/libs for the TLS implementation.
849 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
850 properties get this fed in as external SSF. A number of robustness
851 and debugging improvements to the cyrus_sasl authenticator.
853 PP/08 cyrus_sasl server now expands the server_realm option.
855 PP/09 Bugzilla 1214 - Log authentication information in reject log.
856 Patch by Jeremy Harris.
858 PP/10 Added dbmjz lookup type.
860 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
862 PP/12 MAIL args handles TAB as well as SP, for better interop with
863 non-compliant senders.
864 Analysis and variant patch by Todd Lyons.
866 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
867 Bug report from Lars Müller <lars@samba.org> (via SUSE),
868 Patch from Dirk Mueller <dmueller@suse.com>
870 PP/13 tls_peerdn now print-escaped for spool files.
871 Observed some $tls_peerdn in wild which contained \n, which resulted
872 in spool file corruption.
874 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
875 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
876 or write after TLS renegotiation, which otherwise led to messages
879 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
880 as a tracking header (ie: a signed header comes before the signature).
881 Patch from Wolfgang Breyha.
883 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
884 comma-sep list; embedded commas doubled.
886 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
888 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
890 Report and patch from Dmitry Banschikov.
892 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
893 Removed SSL_clear() after SSL_new() which led to protocol negotiation
894 failures. We appear to now support TLS1.1+ with Exim.
896 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
897 lets Exim select keys and certificates based upon TLS SNI from client.
898 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
899 before an outbound SMTP session. New log_selector, +tls_sni.
901 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
902 NULL dereference. Report and patch from Alun Jones.
904 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
905 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
906 Not seeing resolver debug output on NetBSD, but suspect this is a
907 resolver implementation change.
909 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
910 Left warnings. Added "eximon gdb" invocation mode.
912 PP/21 Defaulting "accept_8bitmime" to true, not false.
914 PP/22 Added -bw for inetd wait mode support.
916 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
917 locate the relevant includes and libraries. Made this the default.
919 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
920 Bugzilla 1246, report and most of solution from Tomasz Kusy.
922 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
923 This may cause build issues on older platforms.
925 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
926 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
927 gnutls_require_mac & gnutls_require_protocols (no longer supported).
928 Added SNI support via GnuTLS too.
929 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
931 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
933 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
934 Applied second patch from Janne, implementing suggestion to default
935 multiple-strings-in-record handling to match SPF spec.
937 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
939 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
940 read-only, out of scope).
941 Patch from Wolfgang Breyha, report from Stuart Northfield.
943 PP/29 Fix three issues highlighted by clang analyser static analysis.
944 Only crash-plausible issue would require the Cambridge-specific
945 iplookup router and a misconfiguration.
946 Report from Marcin Mirosław.
948 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
950 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
951 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
952 As part of this, removing so much warning spew let me fix some minor
953 real issues in debug logging.
955 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
956 assignment on my part. Fixed.
958 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
959 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
960 Janne Snabb (who went above and beyond: thank you).
962 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
963 string otherwise requires a connection and a bunch more work and it's
964 relatively easy to get wrong. Should also expose TLS library linkage
967 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
968 64-bit ${eval} (JH/03).
970 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
971 GNU libc to support some of the 64-bit stuff, should not lead to
972 conflicts. Defined before os.h is pulled in, so if a given platform
973 needs to override this, it can.
975 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
976 protection layer was required, which is not implemented.
977 Bugzilla 1254, patch from Wolfgang Breyha.
979 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
980 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
981 tls_dhparam take prime identifiers. Also unbreak combination of
982 OpenSSL+DH_params+TLSSNI.
984 PP/39 Disable SSLv2 by default in OpenSSL support.
990 PP/01 Solaris build fix for Oracle's LDAP libraries.
991 Bugzilla 1109, patch from Stephen Usher.
993 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
995 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
998 TF/02 Fix a couple more cases where we did not log the error message
999 when unlink() failed. See also change 4.74-TF/03.
1001 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1002 lock up or crash if it happened to be inside a call to libc when it
1003 got a SIGUSR1 from exiwhat.
1005 The SIGUSR1 handler appends the current process status to the process
1006 log which is later printed by exiwhat. It used to use the general
1007 purpose logging code to do this, but several functions it calls are
1008 not safe for signals.
1010 The new output code in the SIGUSR1 handler is specific to the process
1011 log, and simple enough that it's easy to inspect for signal safety.
1012 Removing some special cases also simplifies the general logging code.
1013 Removing the spurious timestamps from the process log simplifies
1016 TF/04 Improved ratelimit ACL condition.
1018 The /noupdate option has been deprecated in favour of /readonly which
1019 has clearer semantics. The /leaky, /strict, and /readonly update modes
1020 are mutually exclusive. The update mode is no longer included in the
1021 database key; it just determines when the database is updated. (This
1022 means that when you upgrde Exim will forget old rate measurements.)
1024 Exim now checks that the per_* options are used with an update mode that
1025 makes sense for the current ACL. For example, when Exim is processing a
1026 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1027 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1028 must specify per_mail/readonly. If you omit the update mode it defaults to
1029 /leaky where that makes sense (as before) or /readonly where required.
1031 The /noupdate option is now undocumented but still supported for
1032 backwards compatibility. It is equivalent to /readonly except that in
1033 ACLs where /readonly is required you may specify /leaky/noupdate or
1034 /strict/noupdate which are treated the same as /readonly.
1036 A useful new feature is the /count= option. This is a generalization
1037 of the per_byte option, so that you can measure the throughput of other
1038 aggregate values. For example, the per_byte option is now equivalent
1039 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1041 The per_rcpt option has been generalized using the /count= mechanism
1042 (though it's more complicated than the per_byte equivalence). When it is
1043 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1044 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1045 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1046 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1047 Note that using per_rcpt with a non-readonly update mode in more than
1048 one ACL will cause the recipients to be double-counted. (The per_mail
1049 and per_byte options don't have this problem.)
1051 The handling of very low rates has changed slightly. If the computed rate
1052 is less than the event's count (usually one) then this event is the first
1053 after a long gap. In this case the rate is set to the same as this event's
1054 count, so that the first message of a spam run is counted properly.
1056 The major new feature is a mechanism for counting the rate of unique
1057 events. The new per_addr option counts the number of different
1058 recipients that someone has sent messages to in the last time period. It
1059 behaves like per_rcpt if all the recipient addresses are different, but
1060 duplicate recipient addresses do not increase the measured rate. Like
1061 the /count= option this is a general mechanism, so the per_addr option
1062 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1063 example, measure the rate that a client uses different sender addresses
1064 with the options per_mail/unique=$sender_address. There are further
1065 details in the main documentation.
1067 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1069 TF/06 Removed a few PCRE remnants.
1071 TF/07 Automatically extract Exim's version number from tags in the git
1072 repository when doing development or release builds.
1074 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1075 Bugzilla 879. Patch from Paul Fisher.
1077 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1078 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1081 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1083 PP/05 Variable $av_failed, true if the AV scanner deferred.
1084 Bugzilla 1078. Patch from John Horne.
1086 PP/06 Stop make process more reliably on build failure.
1087 Bugzilla 1087. Patch from Heiko Schlittermann.
1089 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1090 Bugzilla 1089. Patch from Heiko Schlittermann.
1092 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1093 Bugzilla 1131. Patch from Holger Weiß.
1095 PP/09 Handle IPv6 addresses with SPF.
1096 Bugzilla 860. Patch from Wolfgang Breyha.
1098 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1100 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1103 PP/11 match_* no longer expand right-hand-side by default.
1104 New compile-time build option, EXPAND_LISTMATCH_RHS.
1105 New expansion conditions, "inlist", "inlisti".
1107 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1109 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1111 PP/14 fix log_write() format string regression from TF/03.
1112 Bugzilla 1152. Patch from Dmitry Isaikin.
1118 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1120 PP/02 Harmonised TLS library version reporting; only show if debugging.
1121 Layout now matches that introduced for other libraries in 4.74 PP/03.
1123 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1125 PP/04 New "dns_use_edns0" global option.
1127 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1130 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1131 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1133 TK/01 Updated PolarSSL code to 0.14.2.
1134 Bugzilla 1097. Patch from Andreas Metzler.
1136 PP/07 Catch divide-by-zero in ${eval:...}.
1137 Fixes bugzilla 1102.
1139 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1142 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1143 format-string attack -- SECURITY: remote arbitrary code execution.
1145 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1146 time unintentionally subject to list matching rules, letting the header
1147 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1148 arbitrary string expansion). This allowed for information disclosure.
1150 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1151 INT_MIN/-1 -- value coerced to INT_MAX.
1157 NM/01 Workround for PCRE version dependency in version reporting
1160 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1161 This fixes portability to compilers other than gcc, notably
1162 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1164 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1165 makefiles for portability to HP-UX and POSIX correctness.
1167 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1168 Also via indented variable definition in the Makefile.
1169 (Debugging by Oliver Heesakkers).
1171 PP/02 Restore caching of spamd results with expanded spamd_address.
1172 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1174 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1175 Improves build reliability. Fix from: Frank Elsner
1177 NM/02 Fix wide character breakage in the rfc2047 coding
1178 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1180 NM/03 Allow underscore in dnslist lookups
1181 Fixes bug 1026. Patch from Graeme Fowler
1183 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1184 Code patches from Adam Ciarcinski of NetBSD.
1186 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1189 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1190 is logged, to avoid truncation. Patch from John Horne.
1192 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1193 Patch from Jakob Hirsch.
1195 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1196 SQL string expansion failure details.
1197 Patch from Andrey Oktyabrski.
1199 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1200 Patch from Simon Arlott.
1202 PP/09 New lookups functionality failed to compile on old gcc which rejects
1203 extern declarations in function scope.
1204 Patch from Oliver Fleischmann
1206 PP/10 Use sig_atomic_t for flags set from signal handlers.
1207 Check getgroups() return and improve debugging.
1208 Fixed developed for diagnosis in bug 927 (which turned out to be
1211 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1212 Patch from Mark Zealey.
1214 PP/12 Bugzilla 1056: Improved spamd server selection.
1215 Patch from Mark Zealey.
1217 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1218 Based on patch from Heiko Schlittermann.
1220 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1221 Patch from Uwe Doering, sign-off by Michael Haardt.
1223 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1224 variable declaration deep within a block. Bug and patch from
1227 PP/15 lookups-Makefile IRIX compatibilty coercion.
1229 PP/16 Make DISABLE_DKIM build knob functional.
1231 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1232 Patch by Simon Arlott
1234 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1235 variable argument macros. Our copy now differs from upstream.
1241 TF/01 Failure to get a lock on a hints database can have serious
1242 consequences so log it to the panic log.
1244 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1245 controlled using the smtp_confirmation log selector.
1247 TF/03 Include the error message when we fail to unlink a spool file.
1249 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1250 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1251 for maintaining out-of-tree patches for some time.
1253 PP/01 Bugzilla 139: Documentation and portability issues.
1254 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1255 Handle per-OS dynamic-module compilation flags.
1257 PP/02 Let /dev/null have normal permissions.
1258 The 4.73 fixes were a little too stringent and complained about the
1259 permissions on /dev/null. Exempt it from some checks.
1260 Reported by Andreas M. Kirchwitz.
1262 PP/03 Report version information for many libraries, including
1263 Exim version information for dynamically loaded libraries. Created
1264 version.h, now support a version extension string for distributors
1265 who patch heavily. Dynamic module ABI change.
1267 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1268 privilege escalation vulnerability whereby the Exim run-time user
1269 can cause root to append content of the attacker's choosing to
1272 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1275 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1276 If dropping privileges for untrusted macros, we disabled normal logging
1277 on the basis that it would fail; for the Exim run-time user, this is not
1278 the case, and it resulted in successful deliveries going unlogged.
1279 Fixed. Reported by Andreas Metzler.
1285 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1286 only prepend for the Resent-* case. Fixes regression introduced in
1287 Exim 4.70 by NM/22 for Bugzilla 607.
1289 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1290 increasing numbers of administrators be bitten by this.
1292 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1294 PP/03 Bugzilla 994: added openssl_options main configuration option.
1296 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1298 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1300 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1302 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1303 without a peer certificate, leading to a segfault because of an
1304 assumption that peers always have certificates. Be a little more
1305 paranoid. Problem reported by Martin Tscholak.
1307 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1308 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1309 NB: ClamAV planning to remove STREAM in "middle of 2010".
1310 CL also introduces -bmalware, various -d+acl logging additions and
1311 more caution in buffer sizes.
1313 PP/09 Implemented reverse_ip expansion operator.
1315 PP/10 Bugzilla 937: provide a "debug" ACL control.
1317 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1319 PP/12 Bugzilla 973: Implement --version.
1321 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1323 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1325 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1327 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1328 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1329 ignore trailing whitespace.
1331 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1333 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1336 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1337 Notification from Dr Andrew Aitchison.
1339 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1340 ExtendedDetectionInfo response format.
1341 Notification from John Horne.
1343 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1346 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1347 XSL and documented dependency on system catalogs, with examples of how
1350 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1353 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1354 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1355 configuration file which is writeable by the Exim user or group.
1357 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1358 of configuration files to cover files specified with the -C option if
1359 they are going to be used with root privileges, not just the default
1362 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1363 option (effectively making it always true).
1365 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1366 files to be used while preserving root privileges.
1368 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1369 that rogue child processes cannot use them.
1371 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1372 run-time user, instead of root.
1374 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1375 Exim run-time user without dropping privileges.
1377 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1378 result string, instead of calling string_vformat() twice with the same
1381 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1382 for other users. Others should always drop root privileges if they use
1383 -C on the command line, even for a whitelisted configure file.
1385 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1387 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1393 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1394 $data_path, and $header_path variables; fixed documentation bugs and
1397 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1398 exipick to access non-standard spools, including the "frozen" queue
1401 NM/01 Bugzilla 965: Support mysql stored procedures.
1402 Patch from Alain Williams
1404 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1406 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1407 Patch from Andreas Metzler
1409 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1410 Patch from Kirill Miazine
1412 NM/05 Bugzilla 671: Added umask to procmail example.
1414 JJ/03 installed exipick 20100323.0, fixing doc bug
1416 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1417 directory. Notification and patch from Dan Rosenberg.
1419 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1421 TK/02 Improve log output when DKIM signing operation fails.
1423 MH/01 Treat the transport option dkim_domain as a colon separated
1424 list, not as a single string, and sign the message with each element,
1425 omitting multiple occurences of the same signer.
1427 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1428 Bugzilla 985, 986. Patch by Simon Arlott
1430 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1431 Patch by Simon Arlott
1433 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1434 MBX locking. Notification from Dan Rosenberg.
1440 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1442 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1444 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1446 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1448 NM/04 Bugzilla 915: Flag broken perl installation during build.
1454 TK/01 Added patch by Johannes Berg that expands the main option
1455 "spamd_address" if it starts with a dollar sign.
1457 TK/02 Write list of recipients to X-Envelope-Sender header when building
1458 the mbox-format spool file for content scanning (suggested by Jakob
1461 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1462 (http://www.dcc-servers.net/) support via dccifd. Activated by
1463 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1465 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1466 by Mark Daniel Reidel <mr@df.eu>.
1468 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1469 When building exim an external PCRE library is now needed -
1470 PCRE is a system library on the majority of modern systems.
1471 See entry on PCRE_LIBS in EDITME file.
1473 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1474 conversation. Added nologin parameter to request.
1475 Patch contributed by Kirill Miazine.
1477 TF/01 Do not log submission mode rewrites if they do not change the address.
1479 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1481 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1482 log files in place. Contributed by Roberto Lima.
1484 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1486 TF/03 Bugzilla 615: When checking the local_parts router precondition
1487 after a local_part_suffix or local_part_prefix option, Exim now
1488 does not use the address's named list lookup cache, since this
1489 contains cached lookups for the whole local part.
1491 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1492 Robert Millan. Documentation is in experimental-spec.txt.
1494 TF/04 Bugzilla 668: Fix parallel build (make -j).
1496 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1498 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1499 Patch provided by Jan Srzednicki.
1501 TF/05 Leading white space used to be stripped from $spam_report which
1502 wrecked the formatting. Now it is preserved.
1504 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1505 that they are available at delivery time.
1507 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1509 TF/08 TLS error reporting now respects the incoming_interface and
1510 incoming_port log selectors.
1512 TF/09 Produce a more useful error message if an SMTP transport's hosts
1513 setting expands to an empty string.
1515 NM/06 Bugzilla 744: EXPN did not work under TLS.
1516 Patch provided by Phil Pennock.
1518 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1519 Patch provided by Richard Godbee.
1521 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1522 acl_smtp_notquit, added index entry.
1524 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1525 Patch provided by Eugene Bujak.
1527 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1528 accept(). Patch provided by Maxim Dounin.
1530 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1531 Patch provided by Phil Pennock.
1533 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1535 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1536 Patch provided by Brad "anomie" Jorsch.
1538 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1539 Patch provided by Dean Brooks.
1541 TK/05 Add native DKIM support (does not depend on external libraries).
1543 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1544 Patch provided by Graeme Fowler.
1546 NM/16 Bugzilla 851: Documentation example syntax fix.
1548 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1550 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1553 NM/19 Bugzilla 745: TLS version reporting.
1554 Patch provided by Phil Pennock.
1556 NM/20 Bugzilla 167: bool: condition support.
1557 Patch provided by Phil Pennock.
1559 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1560 clients. Patch provided by Phil Pennock.
1562 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1563 Patch provided by Brad "anomie" Jorsch.
1565 NM/23 Bugzilla 687: Fix misparses in eximstats.
1566 Patch provided by Heiko Schlittermann.
1568 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1569 Patch provided by Heiko Schlittermann.
1571 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1572 plus update to original patch.
1574 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1576 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1577 Patch provided by David Brownlee.
1579 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1581 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1583 NM/30 Bugzilla 888: TLS documentation bugfixes.
1585 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1587 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1588 Unlike the original bugzilla I have changed all shell scripts in src tree.
1590 NM/33 Bugzilla 898: Transport filter timeout fix.
1591 Patch by Todd Rinaldo.
1593 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1594 Patch by Serge Demonchaux.
1596 NM/35 Bugzilla 39: Base64 decode bug fixes.
1597 Patch by Jakob Hirsch.
1599 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1601 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1603 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1605 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1611 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1612 ALT-N's libdkim that I have put here:
1613 http://duncanthrax.net/exim-experimental/
1615 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1616 were called 'true' and it seems that C99 defines that as a reserved
1617 keyword to be used with 'bool' variable types. That means you could
1618 not include C99-style headers which use bools without triggering
1619 build errors in sieve.c.
1621 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1622 as mailq or other aliases. Changed the --help handling significantly
1623 to do whats expected. exim_usage() emits usage/help information.
1625 SC/01 Added the -bylocaldomain option to eximstats.
1627 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1629 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1631 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1637 PH/01 Another patch from the Sieve maintainer.
1639 PH/02 When an IPv6 address is converted to a string for single-key lookup
1640 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1641 dots are used instead of colons so that keys in lsearch files need not
1642 contain colons. This was done some time before quoting was made available
1643 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1644 (notated using the quote facility) so as to distinguish them from IPv4
1645 keys. This meant that lookups for IP addresses in host lists did not work
1646 for iplsearch lookups.
1648 This has been fixed by arranging for IPv6 addresses to be expressed with
1649 colons if the lookup type is iplsearch. This is not incompatible, because
1650 previously such lookups could never work.
1652 The situation is now rather anomolous, since one *can* have colons in
1653 ordinary lsearch keys. However, making the change in all cases is
1654 incompatible and would probably break a number of configurations.
1656 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1659 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1660 conversion specification without a maximum field width, thereby enabling
1661 a rogue spamd server to cause a buffer overflow. While nobody in their
1662 right mind would setup Exim to query an untrusted spamd server, an
1663 attacker that gains access to a server running spamd could potentially
1664 exploit this vulnerability to run arbitrary code as the Exim user.
1666 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1667 $primary_hostname instead of what libspf2 thinks the hosts name is.
1669 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1670 a directory entry by the name of the lookup key. Previously, if a
1671 symlink pointed to a non-existing file or a file in a directory that
1672 Exim lacked permissions to read, a lookup for a key matching that
1673 symlink would fail. Now it is enough that a matching directory entry
1674 exists, symlink or not. (Bugzilla 503.)
1676 PH/03 The body_linecount and body_zerocount variables are now exported in the
1679 PH/04 Added the $dnslist_matched variable.
1681 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1682 This means they are set thereafter only if the connection becomes
1685 PH/06 Added the client_condition to authenticators so that some can be skipped
1686 by clients under certain conditions.
1688 PH/07 The error message for a badly-placed control=no_multiline_responses left
1689 "_responses" off the end of the name.
1691 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1693 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1694 (without spaces) instead of just copying the configuration text.
1696 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1698 PH/11 Added $max_received_linelength.
1700 PH/12 Added +ignore_defer and +include_defer to host lists.
1702 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1703 way in which PCRE > 7.0 is built.
1705 PH/14 Implemented queue_only_load_latch.
1707 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1708 MAIL command. The effect was to mangle the value on 64-bit systems.
1710 PH/16 Another patch from the Sieve maintainer.
1712 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1714 PH/18 If a system quota error occurred while trying to create the file for
1715 a maildir delivery, the message "Mailbox is full" was not appended to the
1716 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1717 only to a quota excession during the actual writing of the file.
1719 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1720 characters?) which causes problems in log lines. The DN values are now
1721 passed through string_printing() before being added to log lines.
1723 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1724 and InterBase are left for another time.)
1726 PH/21 Added message_body_newlines option.
1728 PH/22 Guard against possible overflow in moan_check_errorcopy().
1730 PH/23 POSIX allows open() to be a macro; guard against that.
1732 PH/24 If the recipient of an error message contained an @ in the local part
1733 (suitably quoted, of course), incorrect values were put in $domain and
1734 $local_part during the evaluation of errors_copy.
1740 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1741 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1744 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1745 issue a MAIL command.
1747 PH/02 In an ACL statement such as
1749 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1751 if a client was not listed at all, or was listed with a value other than
1752 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1753 the condition was not true (as it should be), so access was not denied.
1754 The bug was that the ! inversion was incorrectly passed on to the second
1755 item. This has been fixed.
1757 PH/03 Added additional dnslists conditions == and =& which are different from
1758 = and & when the dns lookup returns more than one IP address.
1760 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1761 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1763 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1764 FSYNC, which compiles an option called disable_fsync that allows for
1765 bypassing fsync(). The documentation is heavily laced with warnings.
1767 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1769 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1770 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1771 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1772 including adding "make clean"; (3) Added -fPIC when compiling the test
1773 dynamically loaded module, to get rid of a warning.
1775 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1776 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1777 = 0s. The bug is otherwise harmless.
1779 PH/07 There was a bug in the dovecot authenticator such that the value of
1780 $auth1 could be overwritten, and so not correctly preserved, after a
1781 successful authentication. This usually meant that the value preserved by
1782 the server_setid option was incorrect.
1784 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1786 PH/09 Installed PCRE release 7.0.
1788 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1789 run for batched SMTP input. It is now run at the start of every message
1790 in the batch. While fixing this I discovered that the process information
1791 (output by running exiwhat) was not always getting set for -bs and -bS
1792 input. This is fixed, and it now also says "batched" for BSMTP.
1794 PH/11 Added control=no_pipelining.
1796 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1797 patch, slightly modified), and move the expansion of helo_data till after
1798 the connection is made in the smtp transport (so it can use these
1801 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1803 PH/14 Added log_selector = +pid.
1805 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1807 PH/16 Add ${if forany and ${if forall.
1809 PH/17 Added dsn_from option to vary the From: line in DSNs.
1811 PH/18 Flush SMTP output before performing a callout, unless control =
1812 no_callout_flush is set.
1814 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1815 was true (the default) a successful delivery failed to delete the retry
1816 item, thus causing premature timeout of the address. The bug is now
1819 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1821 PH/21 Long custom messages for fakedefer and fakereject are now split up
1822 into multiline reponses in the same way that messages for "deny" and
1823 other ACL rejections are.
1825 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1826 with slight modification.
1828 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1829 draft, changing the syntax and factoring some duplicate code".
1831 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1832 for deliveries of the second and subsequent messages over the same SMTP
1835 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1836 ${reduce, with only minor "tidies".
1838 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1840 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1841 expansion side effects.
1843 PH/27 When a message times out after an over-quota error from an Exim-imposed
1844 quota, the bounce message says "mailbox is full". This message was not
1845 being given when it was a system quota that was exceeded. It now should
1848 MH/03 Made $recipients available in local_scan(). local_scan() already has
1849 better access to the recipient list through recipients_list[], but
1850 $recipients can be useful in postmaster-provided expansion strings.
1852 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1853 in the case of a MAIL command with additional options following the
1854 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1855 were accidentally chopped off.
1857 PH/29 SMTP synchronization checks are implemented when a command is read -
1858 there is a check that no more input is waiting when there shouldn't be
1859 any. However, for some commands, a delay in an ACL can mean that it is
1860 some time before the response is written. In this time, more input might
1861 arrive, invalidly. So now there are extra checks after an ACL has run for
1862 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1863 pipelining has not been advertised.
1865 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1867 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1868 returned in the SMTP response when smtp_return_error_details was false.
1869 This has been fixed.
1871 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1872 instead of the C I/O that was originally supplied, because problems were
1873 reported on Solaris.
1875 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1876 Exim which did not show up earlier: it was assuming that a call to
1877 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1878 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1879 was a macro that became an assignment, so it seemed to work. This has
1880 changed to a proper function call with a void return, hence the compile
1881 error. Exim's code has been fixed.
1883 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1886 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1888 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1890 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1891 because it turns out to be too big or there is a timeout) there is no
1892 "Completed" line in the log. When some messages of this type were
1893 selected by exigrep, they were listed as "not completed". Others were
1894 picked up by some special patterns. I have improved the selection
1895 criteria to be more general.
1897 PH/38 The host_find_failed option in the manualroute router can now be set
1898 to "ignore", to completely ignore a host whose IP address cannot be
1899 found. If all hosts are ignored, the behaviour is controlled by the new
1900 host_all_ignored option.
1902 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1903 homing or because of multiple MX records with /mx) generated more than
1904 one IP address, and the following item turned out to be the local host,
1905 all the secondary addresses of the first item were incorrectly removed
1906 from the list, along with the local host and any following hosts (which
1907 is what is supposed to happen).
1909 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1910 whoever called Exim into the -H file. In the case of the daemon it was
1911 behaving confusingly. When first started, it used values for whoever
1912 started the daemon, but after a SIGHUP it used the Exim user (because it
1913 calls itself on a restart). I have changed the code so that it now always
1916 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1917 message are rejected with the same error (e.g. no authentication or bad
1918 sender address), and a DATA command is nevertheless sent (as can happen
1919 with PIPELINING or a stupid MUA), the error message that was given to the
1920 RCPT commands is included in the rejection of the DATA command. This is
1921 intended to be helpful for MUAs that show only the final error to their
1924 PH/42 Another patch from the Sieve maintainer.
1926 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1927 Eximstats - Fixed some broken HTML links and added missing column headers
1929 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1930 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1932 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1934 PH/43 Yet another patch from the Sieve maintainer.
1936 PH/44 I found a way to check for a TCP/IP connection going away before sending
1937 the response to the final '.' that terminates a message, but only in the
1938 case where the client has not sent further data following the '.'
1939 (unfortunately, this is allowed). However, in many cases there won't be
1940 any further data because there won't be any more messages to send. A call
1941 to select() can be used: if it shows that the input is "ready", there is
1942 either input waiting, or the socket has been closed. An attempt to read
1943 the next input character can distinguish the two cases. Previously, Exim
1944 would have sent an OK response which the client would never have see.
1945 This could lead to message repetition. This fix should cure that, at
1946 least in a lot of common cases.
1948 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1949 advertised in response to EHLO.
1955 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1956 fixed by 4.65/MH/01 (is this a record?) are fixed:
1958 (i) An empty string was always treated as zero by the numeric comparison
1959 operators. This behaviour has been restored.
1961 (ii) It is documented that the numeric comparison operators always treat
1962 their arguments as decimal numbers. This was broken in that numbers
1963 starting with 0 were being interpreted as octal.
1965 While fixing these problems I realized that there was another issue that
1966 hadn't been noticed. Values of message_size_limit (both the global option
1967 and the transport option) were treated as octal if they started with 0.
1968 The documentation was vague. These values are now always treated as
1969 decimal, and I will make that clear in the documentation.
1975 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1976 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1979 MH/01 Don't check that the operands of numeric comparison operators are
1980 integers when their expansion is in "skipping" mode (fixes bug
1981 introduced by 4.64-PH/07).
1983 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1984 child addresses, Exim now panics and dies. Previously, because the count
1985 is held in a short int, deliveries were likely to be lost. As such a
1986 large number of recipients for a single message is ridiculous
1987 (performance will be very, very poor), I have chosen to impose a limit
1988 rather than extend the field.
1994 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1995 leftover -K file (the existence of which was triggered by #402).
1996 While we were at it, introduced process PID as part of the -K
1997 filename. This should rule out race conditions when creating
2000 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2001 processing considerably. Previous code took too long for large mails,
2002 triggering a timeout which in turn triggers #401.
2004 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2005 in the DK code in transports.c. sendfile() is not really portable,
2006 hence the _LINUX specificness.
2008 TF/01 In the add_headers option to the mail command in an Exim filter,
2009 there was a bug that Exim would claim a syntax error in any
2010 header after the first one which had an odd number of characters
2013 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2014 callout verification, Exim cached a "reject" for the entire domain. This
2015 is correct for most verifications, but it is not correct for a recipient
2016 verification with use_sender or use_postmaster set, because in that case
2017 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2018 case of MAIL FROM:<> rejection from other early rejections (e.g.
2019 rejection of HELO). When verifying a recipient using a non-null MAIL
2020 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2021 Whatever the result of the callout, the value of the domain cache is
2022 left unchanged (for any other kind of callout, getting as far as trying
2023 RCPT means that the domain itself is ok).
2025 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2028 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2029 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2030 session, the daemon ignores ECONNECT errors and logs others; it now
2031 ignores EPIPE as well.
2033 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2034 (quoted-printable decoding).
2036 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2037 later the small subsequent patch to fix an introduced bug.
2039 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2041 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2043 PH/08 An error is now given if message_size_limit is specified negative.
2045 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2046 to be given (somewhat) arbitrary names.
2048 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2051 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2052 miscellaneous code fixes
2054 PH/10 Added the log_reject_target ACL modifier to specify where to log
2057 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2058 hostname. This is wrong, because it relates to the incoming message (and
2059 probably the interface on which it is arriving) and not to the outgoing
2060 callout (which could be using a different interface). This has been
2061 changed to use the value of the helo_data option from the smtp transport
2062 instead - this is what is used when a message is actually being sent. If
2063 there is no remote transport (possible with a router that sets up host
2064 addresses), $smtp_active_hostname is used.
2066 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2067 tweaks were necessary in order to get it to work (see also 21 below):
2068 (a) The code assumed that strncpy() returns a negative number on buffer
2069 overflow, which isn't the case. Replaced with Exim's string_format()
2071 (b) There were several signed/unsigned issues. I just did the minimum
2072 hacking in of casts. There is scope for a larger refactoring.
2073 (c) The code used strcasecmp() which is not a standard C function.
2074 Replaced with Exim's strcmpic() function.
2075 (d) The code set only $1; it now sets $auth1 as well.
2076 (e) A simple test gave the error "authentication client didn't specify
2077 service in request". It would seem that Dovecot has changed its
2078 interface. Fortunately there's a specification; I followed it and
2079 changed what the client sends and it appears to be working now.
2081 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2084 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2085 address A is aliased to B and C, where B exists and C does not. Without
2086 -v the output is "A verified" because verification stops after a
2087 successful redirection if more than one address is generated. However,
2088 with -v the child addresses are also verified. Exim was outputting "A
2089 failed to verify" and then showing the successful verification for C,
2090 with its parentage. It now outputs "B failed to verify", showing B's
2091 parentage before showing the successful verification of C.
2093 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2094 look up a TXT record in a specific list after matching in a combined
2097 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2098 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2099 they consult the DNS. I had assumed they would set it the way they
2100 wanted; and indeed my experiments on Linux seem to show that in some
2101 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2102 To be on the safe side, however, I have now made the interface to
2103 host_find_byname() similar to host_find_bydns(), with an argument
2104 containing the DNS resolver options. The host_find_byname() function now
2105 sets these options at its start, just as host_find_bydns() does. The smtp
2106 transport options dns_qualify_single and dns_search_parents are passed to
2107 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2108 of host_find_byname() use the default settings of RES_DEFNAMES
2109 (qualify_single) but not RES_DNSRCH (search_parents).
2111 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2112 spool_read_header() do less string testing, by means of a preliminary
2113 switch on the second character of optional "-foo" lines. (This is
2114 overdue, caused by the large number of possibilities that now exist.
2115 Originally there were few.) While I was there, I also converted the
2116 str(n)cmp tests so they don't re-test the leading "-" and the first
2117 character, in the hope this might squeeze out yet more improvement.
2119 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2120 flag allowing group syntax was set by the header_syntax check but not
2121 turned off, possible causing trouble later; (2) The flag was not being
2122 set at all for the header_verify test, causing "group"-style headers to
2123 be rejected. I have now set it in this case, and also caused header_
2124 verify to ignore an empty address taken from a group. While doing this, I
2125 came across some other cases where the code for allowing group syntax
2126 while scanning a header line wasn't quite right (mostly, not resetting
2127 the flag correctly in the right place). These bugs could have caused
2128 trouble for malformed header lines. I hope it is now all correct.
2130 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2131 with the "reply" argument non-NULL. The code, however (which originally
2132 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2133 but it didn't always do it. This confused somebody who was copying the
2134 code for some other use. I have removed all the tests.
2136 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2137 feature that was used to support insecure browsers during the U.S. crypto
2138 embargo. It requires special client support, and Exim is probably the
2139 only MTA that supported it -- and would never use it because real RSA is
2140 always available. This code has been removed, because it had the bad
2141 effect of slowing Exim down by computing (never used) parameters for the
2142 RSA_EXPORT functionality.
2144 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2145 authenticator to fail if there's a tab character in the incoming data
2146 (there should never be unless someone is messing about, as it's supposed
2147 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2148 if the connection is using TLS or if the remote IP is the same as the
2149 local IP, and the "valid-client-cert option" if a client certificate has
2152 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2153 authenticators. This can be used for authorization after authentication
2154 succeeds. (In the case of plaintext, it servers for both authentication
2157 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2158 if any retry times were supplied.
2160 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2161 connection, where there is no client IP address to check. In this
2162 situation, the verify now always succeeds.
2164 PH/25 Applied John Jetmore's -Mset patch.
2166 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2168 PH/27 In a string expansion for a processed (not raw) header when multiple
2169 headers of the same name were present, leading whitespace was being
2170 removed from all of them, but trailing whitespace was being removed only
2171 from the last one. Now trailing whitespace is removed from each header
2172 before concatenation. Completely empty headers in a concatenation (as
2173 before) are ignored.
2175 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2176 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2178 PH/29 [Removed. This was a change that I later backed out, and forgot to
2179 correct the ChangeLog entry (that I had efficiently created) before
2180 committing the later change.]
2182 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2183 address errors (4xx response to RCPT) over the same connection as other
2184 messages routed to the same hosts. Such deliveries are always "forced",
2185 so retry times are not inspected. This resulted in far too many retries
2186 for the affected addresses. The effect occurred only when there were more
2187 hosts than the hosts_max_try setting in the smtp transport when it had
2188 the 4xx errors. Those hosts that it had tried were not added to the list
2189 of hosts for which the message was waiting, so if all were tried, there
2190 was no problem. Two fixes have been applied:
2192 (i) If there are any address or message errors in an SMTP delivery, none
2193 of the hosts (tried or untried) are now added to the list of hosts
2194 for which the message is waiting, so the message should not be a
2195 candidate for sending over the same connection that was used for a
2196 successful delivery of some other message. This seems entirely
2197 reasonable: after all the message is NOT "waiting for some host".
2198 This is so "obvious" that I'm not sure why it wasn't done
2199 previously. Hope I haven't missed anything, but it can't do any
2200 harm, as the worst effect is to miss an optimization.
2202 (ii) If, despite (i), such a delivery is accidentally attempted, the
2203 routing retry time is respected, so at least it doesn't keep
2204 hammering the server.
2206 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2207 in ${readsocket because some servers need this prod.
2209 PH/32 Added some extra debug output when updating a wait-xxx database.
2211 PH/33 The hint "could be header name not terminated by colon", which has been
2212 given for certain expansion errors for a long time, was not being given
2213 for the ${if def:h_colon_omitted{... case.
2215 PH/34 The spec says: "With one important exception, whenever a domain list is
2216 being scanned, $domain contains the subject domain." There was at least
2217 one case where this was not true.
2219 PH/35 The error "getsockname() failed: connection reset by peer" was being
2220 written to the panic log as well as the main log, but it isn't really
2221 panic-worthy as it just means the connection died rather early on. I have
2222 removed the panic log writing for the ECONNRESET error when getsockname()
2225 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2226 runs only) independently of the message's sender address. This meant
2227 that, if the 4xx error was in fact related to the sender, a different
2228 message to the same recipient with a different sender could confuse
2229 things. In particualar, this can happen when sending to a greylisting
2230 server, but other circumstances could also provoke similar problems.
2231 I have changed the default so that the retry time for these errors is now
2232 based a combination of the sender and recipient addresses. This change
2233 can be overridden by setting address_retry_include_sender=false in the
2236 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2237 remote server are returned as part of bounce messages. This was not
2238 happening for LMTP over a pipe (the lmtp transport), but now it is the
2239 same for both kinds of LMTP.
2241 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2242 in header lines that were in fact CNAMEs. This is no longer the case.
2244 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2245 and queue runs started by the daemon processed all messages. This has
2246 been fixed so that -R and -S can now usefully be given with -q<time>.
2248 PH/40 Import PCRE release 6.7 (fixes some bugs).
2250 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2252 PH/42 Give an error if -q is specified more than once.
2254 PH/43 Renamed the variables $interface_address and $interface_port as
2255 $received_ip_address and $received_port, to make it clear that these
2256 values apply to message reception, and not to the outgoing interface when
2257 a message is delivered. (The old names remain recognized, of course.)
2259 PH/44 There was no timeout on the connect() call when using a Unix domain
2260 socket in the ${readsocket expansion. There now is.
2262 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2263 be meaningful with "accept".
2265 SC/01 Eximstats V1.43
2266 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2268 SC/02 Eximstats V1.44
2269 Use a glob alias rather than an array ref in the generated
2270 parser. This improves both readability and performance.
2272 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2273 Collect SpamAssassin and rejection statistics.
2274 Don't display local sender or destination tables unless
2275 there is data to show.
2276 Added average volumes into the top table text output.
2278 SC/04 Eximstats V1.46
2279 Collect data on the number of addresses (recipients)
2280 as well as the number of messages.
2282 SC/05 Eximstats V1.47
2283 Added 'Message too big' to the list of mail rejection
2284 reasons (thanks to Marco Gaiarin).
2286 SC/06 Eximstats V1.48
2287 Mainlog lines which have GMT offsets and are too short to
2288 have a flag are now skipped.
2290 SC/07 Eximstats V1.49 (Alain Williams)
2291 Added the -emptyok flag.
2293 SC/08 Eximstats V1.50
2294 Fixes for obtaining the IP address from reject messages.
2296 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2297 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2298 whitesspace changes from 4.64-PH/27
2300 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2303 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2304 are found, allow negative numbers in numeric criteria)
2306 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2308 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2309 to match changes made in 4.64-PH/43
2311 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2313 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2314 "domain" is actually an IP address, and force a failure. This locks out
2315 those revolvers/nameservers that support "A-for-A" lookups, in
2316 contravention of the specifications.
2318 PH/48 When a host name was looked up from an IP address, and the subsequent
2319 forward lookup of the name timed out, the host name was left in
2320 $sender_host_name, contrary to the specification.
2322 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2323 restricted to single-key lookups, Exim was not diagnosing an error if
2324 * or *@ was used with a query-style lookup.
2326 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2328 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2329 long ago, but noone interested enough thought of it. Let's just say that
2330 the "1.1" means that there are some new functions that weren't there at
2331 some point in the past.
2333 PH/51 Error processing for expansion failure of helo_data from an smtp
2334 transport during callout processing was broken.
2336 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2337 tested/used via the -bh/-bhc/-bs options.
2339 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2340 bug, fixed in subsequent PCRE releases).
2342 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2343 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2345 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2351 SC/01 Use a glob alias rather than an array ref in eximstats generated
2352 parser. This improves both readability and performance.
2354 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2355 Don't display local sender or destination tables in eximstats unless
2356 there is data to show.
2357 Added average volumes into the eximstats top table text output.
2359 SC/03 Collect data on the number of addresses (recipients) as well
2360 as the number of messages in eximstats.
2362 TF/01 Correct an error in the documentation for the redirect router. Exim
2363 does (usually) call initgroups() when daemonizing.
2365 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2366 with consistent privilege compared to when running as a daemon.
2368 TF/03 Note in the spec that $authenticated_id is not set for local
2369 submissions from trusted users.
2371 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2372 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2374 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2375 by adding some example configuration directives to the default
2376 configuration file. A little bit of work is required to uncomment the
2377 directives and define how usernames and passwords are checked, but
2378 there is now a framework to start from.
2380 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2381 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2382 without this. I don't know how relevant this is to other LDAP libraries.
2384 PH/02 Add the verb name to the "unknown ACL verb" error.
2386 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2388 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2390 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2391 directory not expanded when it should be if an expanded home directory
2392 was set for the address (which is overridden by the transport).
2394 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2397 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2398 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2399 because it is too late at that time, and has no effect.
2401 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2402 security issue with \' (bugzilla #107). I could not use the
2403 PQescapeStringConn() function, because it needs a PGconn value as one of
2406 PH/08 When testing addresses using -bt, indicate those final addresses that
2407 are duplicates that would not cause an additional delivery. At least one
2408 person was confused, thinking that -bt output corresponded to deliveries.
2409 (Suppressing duplicates isn't a good idea as you lose the information
2410 about possibly different redirections that led to the duplicates.)
2412 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2413 systems where poll() doesn't work, in particular OS X.
2415 PH/10 Added more information to debugging output for retry time not reached.
2417 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2418 operations in malware.c.
2420 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2423 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2424 syslog_duplication set false, log lines that would normally be written
2425 both the the main log and to the reject log were not written to syslog at
2428 PH/14 In the default configuration, change the use of "message" in ACL warn
2429 statements to "add_header".
2431 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2432 not followed by a command (e.g. "seen endif").
2434 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2435 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2438 PH/17 Added extra conditions to the default value of delay_warning_condition
2442 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2443 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2444 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2447 The Auto-Submitted: and various List- headers are standardised, whereas I
2448 don't think Precedence: ever was.
2450 PH/18 Refactored debugging code in route_finduser() to show more information,
2451 in particular, the error code if getpwnam() issues one.
2453 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2454 This is apparently needed in addition to the PH/07 change above to avoid
2455 any possible encoding problems.
2457 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2458 but not after initializing Perl.
2460 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2461 output them only if debugging. By default they are written stderr,
2462 apparently, which is not desirable.
2464 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2467 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2470 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2472 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2473 authenticated or an ident call has been made. Suppress the default
2474 values for $authenticated_id and $authenticated_sender (but permit -oMai
2475 and -oMas) when testing with -bh.
2477 PH/24 Re-jigged the order of the tests in the default configuration so that the
2478 tests for valid domains and recipients precede the DNS black list and CSA
2479 tests, on the grounds that those ones are more expensive.
2481 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2482 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2483 This bug exists in every version of Exim that I still have, right back to
2486 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2487 However, an attempt to turn on case-sensitivity in a regex key by
2488 including (?-i) didn't work because the subject string was already
2489 lowercased, and the effects were non-intuitive. It turns out that a
2490 one-line patch can be used to allow (?-i) to work as expected.
2496 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2497 other effects) broke the use of negated acl sub-conditions.
2499 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2502 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2503 "Deny" causes Exim to reject the incoming connection with a 554 error.
2504 Unfortunately, if there is a major crisis, such as a disk failure,
2505 tcp-wrappers gives "deny", whereas what one would like would be some
2506 kind of temporary error. A kludge has been added to help with this.
2507 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2508 554 error is used if errno is still zero or contains ENOENT (which occurs
2509 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2512 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2514 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2515 errors. Otherwise a message that provokes a temporary error (when other
2516 messages do not) can cause a whole host to time out.
2518 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2519 addresses were routed directly to files or pipes from a redirect router.
2520 File deliveries just didn't batch; pipe deliveries might have suffered
2523 PH/06 A failure to get a lock for a hints database would erroneously always say
2524 "Failed to get write lock", even when it was really a read lock.
2526 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2527 of 0600. This has been changed to use the value of the lockfile_mode
2528 option (which defaults to 0600).
2530 PH/08 Applied small patch from the Sieve maintainer.
2532 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2533 folder from quota calculations, a direct delivery into this folder messed
2534 up the contents of the maildirsize file. This was because the regex was
2535 used only to exclude .Trash (or whatever) when the size of the mailbox
2536 was calculated. There was no check that a delivery was happening into an
2537 excluded directory. This bug has been fixed by ignoring all quota
2538 processing for deliveries into excluded directories.
2540 PH/10 Added the maildirfolder_create_regex option to appendfile.
2546 PH/01 The code for finding all the local interface addresses on a FreeBSD
2547 system running IPv6 was broken. This may well have applied to all BSD
2548 systems, as well as to others that have similar system calls. The broken
2549 code found IPv4 interfaces correctly, but gave incorrect values for the
2550 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2551 that it would not match correctly against @[] and not recognize the IPv6
2554 PH/02 The ipliteral router was not recognizing addresses of the form user@
2555 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2557 PH/03 Added disable_ipv6.
2559 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2560 decoded form, because it is most often used to construct To: headers
2561 lines in autoreplies, and the decoded form may well be syntactically
2562 invalid. However, $reply_address has leading white space removed, and all
2563 newlines turned into spaces so that the autoreply transport does not
2566 PH/05 If group was specified without a user on a router, and no group or user
2567 was specified on a transport, the group from the router was ignored.
2569 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2570 for visible compile-time settings that can be used to change these
2571 numbers, for those that want even more. Backwards compatibility with old
2572 spool files has been maintained. However, going back to a previous Exim
2573 release will lost any variables that are in spool files.
2575 PH/07 Two small changes when running in the test harness: increase delay when
2576 passing a TCP/IP connection to a new process, in case the original
2577 process has to generate a bounce, and remove special handling of
2578 127.0.0.2 (sic), which is no longer necessary.
2580 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2581 be the same on different OS.
2583 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2586 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2587 whether --show-vars was specified or not
2589 JJ/02 exipick: Added support for new ACL variable spool format introduced
2592 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2593 syntactically invalid From: or Reply-to: line, and a filter used this to
2594 generate an autoreply, and therefore failed to obtain an address for the
2595 autoreply, Exim could try to deliver to a non-existent relative file
2596 name, causing unrelated and misleading errors. What now happens is that
2597 it logs this as a hard delivery error, but does not attempt to create a
2600 PH/11 The exinext utility has a -C option for testing purposes, but although
2601 the given file was scanned by exinext itself; it wasn't being passed on
2602 when Exim was called.
2604 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2605 an end-of-file indication when reading a command response.
2607 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2608 compiled. In many other places in Exim, IPv6 addresses are always
2609 recognized, so I have changed this. It also means that IPv4 domain
2610 literals of the form [IPV4:n.n.n.n] are now always recognized.
2612 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2613 used if the router is not running as root, for example, when verifying at
2614 ACL time, or when using -bh. The debugging output from this situation was
2615 non-existent - all you got was a failure to exec. I have made two
2618 (a) Failures to set uid/gid, the current directory, or a process leader
2619 in a subprocess such as that created by queryprogram now generate
2620 suitable debugging ouput when -d is set.
2622 (b) The queryprogram router detects when it is not running as root,
2623 outputs suitable debugging information if -d is set, and then runs
2624 the subprocess without attempting to change uid/gid.
2626 PH/15 Minor change to Makefile for building test_host (undocumented testing
2629 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2630 additional section of a DNS packet that returns MX or SRV records.
2631 Instead, it always explicitly searches for A/AAAA records. This avoids
2632 major problems that occur when a DNS server includes only records of one
2633 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2634 fixed another bug: if SRV records were looked up and the corresponding
2635 address records were *not* found in the additional section, the port
2636 values from the SRV records were lost.
2638 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2639 using the correct key (the original address) when searching the retry
2640 rules in order to find which one to use for generating the retry hint.
2642 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2643 adding the default one. Similarly, if it contains a Reply-To: header, the
2644 errors_reply_to option, if set, is not used.
2646 PH/19 When calculating a retry time, Exim used to measure the "time since
2647 failure" by looking at the "first failed" field in the retry record. Now
2648 it does not use this if it is later than than the arrival time of the
2649 message. Instead it uses the arrival time. This makes for better
2650 behaviour in cases where some deliveries succeed, thus re-setting the
2651 "first failed" field. An example is a quota failure for a huge message
2652 when small messages continue to be delivered. Without this change, the
2653 "time since failure" will always be short, possible causing more frequent
2654 delivery attempts for the huge message than are intended.
2655 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2657 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2658 $1, $2, $3) because the numerical variables can be reset during some
2659 expansion items (e.g. "match"), thereby losing the authentication data.
2661 PH/21 Make -bV show the size of off_t variables so that the test suite can
2662 decide whether to run tests for quotas > 2G.
2664 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2665 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2666 filecount value is greater than 2G or if a quota value is greater than 2G
2667 on a system where the size of off_t is not greater than 4, a panic error
2670 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2671 never match. The debug and -bh output now contains an explicit error
2672 message indicating a malformed IPv4 address or mask.
2674 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2675 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2676 PH/23 above applies.
2678 PH/25 Do not write to syslog when running in the test harness. The only
2679 occasion when this arises is a failure to open the main or panic logs
2680 (for which there is an explicit test).
2682 PH/26 Added the /no_tell option to "control=freeze".
2684 PH/27 If a host name lookup failed very early in a connection, for example, if
2685 the IP address matched host_lookup and the reverse lookup yielded a name
2686 that did not have a forward lookup, an error message of the form "no IP
2687 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2688 could be logged. Now it outputs the IP address instead of "NULL".
2690 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2691 allows the sender and the authenticated sender to be set when
2692 submitting a message from within Exim. Since child_open_exim() is
2693 documented for local_scan(), the new function should be too.
2695 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2696 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2697 results in an empty string is now treated as unset.
2699 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2701 PH/31 Added sender_verify_fail logging option.
2703 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2704 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2705 tidied the source and removed it altogether.
2707 PH/33 When a queue run was abandoned because the load average was too high, a
2708 log line was always written; now it is written only if the queue_run log
2709 selector is set. In addition, the log line for abandonment now contains
2710 information about the queue run such as the pid. This is always present
2711 in "start" and "stop" lines but was omitted from the "abandon" line.
2713 PH/34 Omit spaces between a header name and the colon in the error message that
2714 is given when verify = headers_syntax fails (if there are lots of them,
2715 the message gets confusing).
2717 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2718 names, as there are now some PTR records that contain slashes. This check
2719 is only to protect against broken name servers that fall over on strange
2720 characters, so the fact that it applies to all lookups doesn't matter.
2722 PH/36 Now that the new test suite is complete, we can remove some of the
2723 special code in Exim that was needed for the old test suite. For example,
2724 sorting DNS records because real resolvers return them in an arbitrary
2725 order. The new test suite's fake resolver always returns records in the
2728 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2729 bounces) except when queue_only is set, to avoid logging races between
2730 the different processes.
2732 PH/38 Panic-die if .include specifies a non-absolute path.
2734 PH/39 A tweak to the "H" retry rule from its user.
2736 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2737 a label. They prevented compilation on older perls.
2739 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2740 a warning to be raised on newish perls.
2742 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2743 on queue. Changes to match documented behaviour of showing count of
2744 messages matching specified criteria.
2746 PH/40 Changed the default ident timeout from 30s to 5s.
2748 PH/41 Added support for the use of login_cap features, on those BSD systems
2749 that have them, for controlling the resources used by pipe deliveries.
2751 PH/42 The content-scanning code uses fopen() to create files in which to put
2752 message data. Previously it was not paying any attention to the mode of
2753 the files. Exim runs with umask(0) because the rest of the code creates
2754 files with open(), and sets the required mode explicitly. Thus, these
2755 files were ending up world-writeable. This was not a big issue, because,
2756 being within the spool directory, they were not world-accessible. I have
2757 created a function called modefopen, which takes an additional mode
2758 argument. It sets umask(777), creates the file, chmods it to the required
2759 mode, then resets the umask. All the relevant calls to fopen() in the
2760 content scanning code have been changed to use this function.
2762 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2763 to 24 hours. This avoids potential overflow problems when processing G
2764 and H retry rules. I suspect nobody ever tinkers with this value.
2766 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2768 PH/45 When the plaintext authenticator is running as a client, the server's
2769 challenges are checked to ensure they are valid base64 strings. By
2770 default, the authentication attempt is cancelled if an invalid string is
2771 received. Setting client_ignore_invalid_base64 true ignores these errors.
2772 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2773 they are received. Thus, the responses can be made to depend on the
2774 challenges. If an invalid string is ignored, an empty string is placed in
2777 PH/46 Messages that are created by the autoreply transport now contains a
2778 References: header, in accordance with RFCs 2822 and 3834.
2780 PH/47 Added authenticated_sender_force to the smtp transport.
2782 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2784 PH/49 Installed latest patch from the Sieve maintainer.
2786 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2787 was also set, the appendfile transport was unnecessarily scanning a
2788 directory of message files (e.g. for maildir delivery) to find the count
2789 of files (along with the size), even though it did not need this
2790 information. It now does the scan only if it needs to find either the
2791 size of the count of files.
2793 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2795 PH/52 Two bugs concerned with error handling when the smtp transport is
2798 (i) Exim was not creating retry information for temporary errors given
2799 for individual recipients after the DATA command when the smtp transport
2800 was used in LMTP mode. This meant that they could be retried too
2801 frequently, and not timed out correctly.
2803 (ii) Exim was setting the flag that allows error details to be returned
2804 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2805 recipients that were returned after the DATA command.
2807 PH/53 This is related to PH/52, but is more general: for any failing address,
2808 when detailed error information was permitted to be returned to the
2809 sender, but the error was temporary, then after the final timeout, only
2810 "retry timeout exceeded" was returned. Now it returns the full error as
2811 well as "retry timeout exceeded".
2813 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2814 do this, and (what is worse) MTAs that accept it.
2816 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2817 will now be deprecated.
2819 PH/56 New os.c-cygwin from the Cygwin maintainer.
2821 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2822 formats (previously only available in exim formats via -bpr, -bpru,
2823 and -bpra. Now also available in native and exiqgrep formats)
2825 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2826 with very large, slow to parse queues
2828 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2830 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2832 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2833 responses to authentication challenges, though it was showing the
2834 challenges; (ii) I've removed the CR characters from the debug output for
2837 PH/58 Allow for the insertion of a newline as well as a space when a string
2838 is turned into more than one encoded-word during RFC 2047 encoding. The
2839 Sieve code now uses this.
2841 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2842 data_4xx, lost_connection, tls_required.
2844 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2845 message was being sent as an SMTP response.
2847 PH/61 Add -l and -k options to exicyclog.
2849 PH/62 When verifying, if an address was redirected to one new address, so that
2850 verification continued, and the new address failed or deferred after
2851 having set something in $address_data, the value of $address_data was not
2852 passed back to the ACL. This was different to the case when no
2853 redirection occurred. The value is now passed back in both cases.
2855 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2856 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2857 use login_cap.h, so on its own it isn't the distinguishing feature. The
2858 new name refers directly to the setclassresources() function.
2860 PH/65 Added configuration files for NetBSD3.
2862 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2864 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2865 is preferred over IPv4.
2867 PH/68 The bounce_return_message and bounce_return_body options were not being
2868 honoured for bounces generated during the reception of non-SMTP messages.
2869 In particular, this applied to messages rejected by the ACL. This bug has
2870 been fixed. However, if bounce_return_message is true and bounce_return_
2871 body is false, the headers that are returned for a non-SMTP message
2872 include only those that have been read before the error was detected.
2873 (In the case of an ACL rejection, they have all been read.)
2875 PH/69 The HTML version of the specification is now built in a directory called
2876 spec_html instead of spec.html, because the latter looks like a path with
2877 a MIME-type, and this confuses some software.
2879 PH/70 Catch two compiler warnings in sieve.c.
2881 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2882 function verify_get_ident() calls ip_connect() to connect a socket, but
2883 if the "connect()" function timed out, ip_connect() used to close the
2884 socket. However, verify_get_ident() also closes the socket later, and in
2885 between Exim writes to the log, which may get opened at this point. When
2886 the socket was closed in ip_connect(), the log could get the same file
2887 descriptor number as the socket. This naturally causes chaos. The fix is
2888 not to close the socket in ip_connect(); the socket should be closed by
2889 the function that creates it. There was only one place in the code where
2890 this was missing, in the iplookup router, which I don't think anybody now
2891 uses, but I've fixed it anyway.
2893 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2894 well as to direct DNS lookups. Otherwise the handling of names in host
2895 lists is inconsistent and therefore confusing.
2901 PH/01 Two changes to the default runtime configuration:
2903 (1) Move the checks for relay_from_hosts and authenticated clients from
2904 after to before the (commented out) DNS black list checks.
2906 (2) Add control=submission to the relay_from_hosts and authenticated
2907 clients checks, on the grounds that messages accepted by these
2908 statements are most likely to be submissions.
2910 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2912 (1) Generate an error if the third argument for the ${prvs expansion is
2915 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2918 (3) Reset the variables that are obtained from the first argument of
2919 ${prvscheck and used in the second argument before leaving the code,
2920 because their memory is reclaimed, so using them afterwards may do
2923 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2924 one (it's much easier than Tom thought :-).
2926 (5) Because of (4), we can now allow for the use of $prvscheck_result
2927 inside the third argument.
2929 PH/03 For some reason, the default setting of PATH when running a command from
2930 a pipe transport was just "/usr/bin". I have changed it to
2933 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2934 anything to be listed in the output from -bV.
2936 PH/05 When a filter generated an autoreply, the entire To: header line was
2937 quoted in the delivery log line, like this:
2939 => >A.N.Other <ano@some.domain> <original@ddress> ...
2941 This has been changed so that it extracts the operative address. There
2942 may be more than one such address. If so, they are comma-separated, like
2945 => >ano@some.domain,ona@other.domain <original@ddress> ...
2947 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2948 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2949 not being looked up in the rDNS to get a host name, Exim was showing the
2950 IP address twice in Received: lines, even though the IP addresses were
2951 identical. For example:
2953 Received: from [1.2.3.4] (helo=[1.2.3.4])
2955 However, if the real host name was known, it was omitting the HELO data
2956 if it matched the actual IP address. This has been tidied up so that it
2957 doesn't show the same IP address twice.
2959 PH/07 When both +timestamp and +memory debugging was on, the value given by
2960 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2961 called by the debug printing, thereby overwriting the timestamp buffer.
2962 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2964 PH/08 When the original message was included in an autoreply transport, it
2965 always said "this is a copy of the message, including all the headers",
2966 even if body_only or headers_only was set. It now gives an appropriate
2969 PH/09 Applied a patch from the Sieve maintainer which:
2971 o fixes some comments
2972 o adds the (disabled) notify extension core
2973 o adds some debug output for the result of if/elsif tests
2974 o points to the current vacation draft in the documentation
2975 and documents the missing references header update
2979 o fixes a bug in processing the envelope test (when testing
2980 multiple envelope elements, the last element determinted the
2983 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2984 Electronic Mail") by including:
2986 Auto-submitted: auto-generated
2988 in the messages that it generates (bounce messages and others, such as
2989 warnings). In the case of bounce messages for non-SMTP mesages, there was
2990 also a typo: it was using "Auto_submitted" (underscore instead of
2991 hyphen). Since every message generated by Exim is necessarily in response
2992 to another message, thes have all been changed to:
2994 Auto-Submitted: auto-replied
2996 in accordance with these statements in the RFC:
2998 The auto-replied keyword:
3000 - SHOULD be used on messages sent in direct response to another
3001 message by an automatic process,
3003 - MUST NOT be used on manually-generated messages,
3005 - MAY be used on Delivery Status Notifications (DSNs) and Message
3006 Disposition Notifications (MDNs),
3008 - MUST NOT be used on messages generated by automatic or periodic
3009 processes, except for messages which are automatic responses to
3012 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3013 to the default Received: header definition.
3015 PH/12 Added log selector acl_warn_skipped (default on).
3017 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3018 variables because (a) they are in the wrong storage pool and (b) even if
3019 they were copied, it wouldn't work properly because of the caching.
3021 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3022 checking when decoding. Apparently there are clients that generate
3023 overlong encoded strings. Why am I not surprised?
3025 PH/15 If the first argument of "${if match_address" was not empty, but did not
3026 contain an "@" character, Exim crashed. Now it writes a panic log message
3027 and treats the condition as false.
3029 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3031 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3032 extension "envelope-auth". The code is finished and in agreement with
3033 other implementations, but there is no documentation so far and in fact,
3034 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3035 not changing the active code.
3037 Print executed "if" and "elsif" statements when debugging is used. This
3038 helps a great deal to understand what a filter does.
3040 Document more things not specified clearly in RFC3028. I had all this
3041 sorted out, when out of a sudden new issues came to my mind. Oops."
3043 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3046 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3047 canonical form (as documented). However, after a host name lookup from
3048 the IP address, check_host() was doing a simple string comparison with
3049 addresses acquired from the DNS when checking that the found name did
3050 have the original IP as one of its addresses. Since any found IPv6
3051 addresses are likely to be in abbreviated form, the comparison could
3052 fail. Luckily, there already exists a function for doing the comparison
3053 by converting both addresses to binary, so now that is used instead of
3054 the text comparison.
3056 PH/20 There was another similar case to PH/19, when a complete host name was
3057 given in a host list; looking up its IP address could give an abbreviated
3058 form, whereas the current host's name might or might not be abbreviated.
3059 The same fix has been applied.
3065 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3066 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3069 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3070 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3072 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3074 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3075 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3076 Also, for "Auto-Submitted", treat anything other than "no" as
3077 non-personal, in accordance with RFC 3834. (Previously it treated
3078 anything starting "auto-" as non-personal.)
3080 TF/01 The control=submission/name=... option had a problem with syntax
3081 errors if the name included a slash character. The /name= option
3082 now slurps the rest of the string, so it can include any characters
3083 but it must come last in the list of options (after /sender_retain
3086 PH/05 Some modifications to the interface to the fake nameserver for the new
3094 TK/01 Added the "success_on_redirect" address verification option. See
3095 NewStuff for rationale and an example.
3097 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3099 PH/02 Patch to exigrep to allow it to work on syslog lines.
3101 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3102 fread() to skip over the body file's header line, because in Cygwin the
3103 header line is locked and is inaccessible.
3105 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3106 co-exist for some time) to make it clear that it is the Exim ID that is
3107 referenced, not the Message-ID: header line.
3109 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3110 string_format() function, because snprintf() does not exist on all
3113 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3114 ${stat: expansion item.
3116 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3117 protocol synchronization error", to keep the pedants happy.
3119 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3120 well as for IRIX systems, when gcc is being used. See the host.c source
3123 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3125 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3128 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3129 transport and to the smtp transport in LMTP mode.
3131 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3133 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3134 run a filter in a subprocess. This could lead to confusion in subsequent
3135 lookups in the parent process. There should also be a search_tidyup() at
3136 the end of the subprocess.
3138 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3139 only if the host matched helo_try_verify_hosts, which caused the
3140 verification to occur when the EHLO/HELO command was issued. The ACL just
3141 tested the remembered result. Now, if a previous verification attempt has
3142 not happened, "verify = helo" does it there and then.
3144 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3146 TK/03 Fix log output including CR from clamd.
3148 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3149 exist provoked a memory error which could cause a segfault.
3151 PH/15 Installed PCRE 6.2
3153 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3155 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3156 of the problem. Specifically, suggested +O2 rather than +O1 for the
3159 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3161 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3162 record, the port was not correctly logged when the outgoing_port log
3163 selector was set (it logged the transort's default port).
3165 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3166 fallback_hosts, and "hosts" in the smtp transport.
3168 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3169 host errors such as "Connection refused".
3171 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3172 authentication with radiusclient 0.4.9:
3174 - Error returned from rc_read_config was caught wrongly
3175 - Username/password not passed on to radius server due to wrong length.
3177 The presumption is that some radiusclient API changes for 4.51/PH/17
3178 were not taken care of correctly. The code is still untested by me (my
3179 Linux distribution still has 0.3.2 of radiusclient), but it was
3180 contributed by a Radius user.
3182 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3183 expanding the "port" option of the smtp transport.
3185 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3186 while reading a MIME header. Thanks to Tom Hughes for a patch.
3188 PH/24 Include config.h inside local_scan.h so that configuration settings are
3191 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3192 that in an ACL for RCPT (for example), you can examine exactly what was
3195 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3196 commands, but it was not correctly comparing the address with the actual
3197 client host address. Thus, it would show the EHLO address in Received:
3198 header lines when this was not necessary.
3200 PH/27 Added the % operator to ${eval:}.
3202 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3203 it should be ignoring failures (because with -C, for example, it has lost
3204 privilege). It wasn't ignoring creation failures other than "already
3207 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3210 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3211 because an input line was too long, either on its own, or by virtue of
3212 too many continuations, the temporary file was not being removed, and the
3213 return code was incorrect.
3215 PH/31 Missing "BOOL" in function definition in filtertest.c.
3217 PH/32 Applied Sieve patches from the maintainer.
3219 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3221 PH/33 Added "verify = not_blind".
3223 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3224 Local/Makefile (with some defaults set). These are used in built scripts
3225 such as exicyclog, but they have never been used in the exim_install
3226 script (though there are many overriding facilities there). I have
3227 arranged that the exim_install script now takes note of these two
3230 PH/35 Installed configuration files for Dragonfly.
3232 PH/36 When a locally submitted message by a trusted user did not contain a
3233 From: header, and the sender address was obtained from -f or from an SMTP
3234 MAIL command, and the trusted user did not use -F to supply a sender
3235 name, $originator_name was incorrectly used when constructing a From:
3236 header. Furthermore, $originator_name was used for submission mode
3237 messages from external hosts without From: headers in a similar way,
3238 which is clearly wrong.
3240 PH/37 Added control=suppress_local_fixups.
3242 PH/38 When log_selector = +received_sender was set, and the addition of the
3243 sender made the log line's construction buffer exactly full, or one byte
3244 less than full, an overflow happened when the terminating "\n" was
3247 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3248 when the result of a list match is failure because a DNS lookup failed.
3250 PH/40 RM_COMMAND is now used in the building process.
3252 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3253 the "build-* directories that it finds.
3255 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3256 address in a domain literal was a prefix of an interface address.
3258 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3259 when verifying a sender address, unless rewrite_headers is false.
3261 PH/44 Wrote a long comment about why errors_to addresses are verified as
3262 recipients, not senders.
3264 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3265 the ratelimit ACL was added.
3267 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3269 PH/46 Added extra information about PostgreSQL errors to the error string.
3271 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3272 avoiding the need to install special zones in a real server. This is
3273 backwards compatible; if it can't find the fake resolver, it drops back.
3274 Thus, both old and new test suites can be run.
3276 TF/02 Added util/ratelimit.pl
3278 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3279 clock is set back in time.
3281 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3282 Candler <B.Candler@pobox.com>.
3284 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3285 OK for addresses that are the result of redirections.
3287 PH/48 A number of further additions for the benefit of the new test suite,
3288 including a fake gethostbyname() that interfaces to the fake DNS resolver
3291 TF/06 The fix for widen_domains has also been applied to qualify_single and
3292 search_parents which are the other dnslookup options that can cause
3295 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3298 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3300 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3301 service" if it encountered an MX record with an empty target hostname.
3302 The message is now "an MX or SRV record indicated no SMTP service".
3304 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3305 if the DNS of the sending site is misconfigured. This is quite a
3306 common situation. This change restores the behaviour of treating a
3307 helo verification defer as a failure.
3309 PH/51 If self=fail was set on a router, the bounce message did not include the
3310 actual error message.
3316 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3318 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3319 command itself ended in error, the underlying message about the transport
3320 filter timeout was being overwritten with the pipe command error. Now the
3321 underlying error message should be appended to the second error message.
3323 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3325 PH/02 Reduce the amount of output that "make" produces by default. Full output
3326 can still be requested.
3328 PH/03 The warning log line about a condition test deferring for a "warn" verb
3329 was being output only once per connection, rather than after each
3330 occurrence (because it was using the same function as for successful
3331 "warn" verbs). This seems wrong, so I have changed it.
3333 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3334 it should not have, which might have caused a crash in the right
3335 circumstances, but probably never did.
3337 PH/04 Installed a modified version of Tony Finch's patch to make submission
3338 mode fix the return path as well as the Sender: header line, and to
3339 add a /name= option so that you can make the user's friendly name appear
3342 TF/03 Added the control = fakedefer ACL modifier.
3344 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3345 Mark Lowes for thorough testing.
3347 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3349 TK/03 Merged latest SRS patch from Miles Wilton.
3351 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3352 with the definition in sysexits.h (which is #included earlier).
3353 Fortunately, Exim does not actually use EX_OK. The code used to try to
3354 preserve the sysexits.h value, by assumimg that macro definitions were
3355 scanned for macro replacements. I have been disabused of this notion,
3356 so now the code just undefines EX_OK before #including unistd.h.
3358 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3359 in the smtp transport. When a block could not be written in a single
3360 write() function, the timeout was being re-applied to each part-write.
3361 This seems wrong - if the receiver was accepting one byte at a time it
3362 would take for ever. The timeout is now adjusted when this happens. It
3363 doesn't have to be particularly precise.
3365 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3366 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3368 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3369 without a domain if the check to <postmaster@domain> fails.
3371 SC/01 Eximstats: added -xls and the ability to specify output files
3372 (patch written by Frank Heydlauf).
3374 SC/02 Eximstats: use FileHandles for outputing results.
3376 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3378 SC/04 Eximstats: fixed display of large numbers with -nvr option
3380 SC/05 Eximstats: fixed merging of reports with empty tables.
3382 SC/06 Eximstats: added the -include_original_destination flag
3384 SC/07 Eximstats: removed tabs and trailing whitespace.
3386 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3388 TK/06 MBOX spool code: Add real "From " MBOX separator line
3389 so the .eml file is really in mbox format (even though
3390 most programs do not really care). Patch from Alex Miller.
3392 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3393 The latter is generated from $received_to and is only set if the
3394 message has one envelope recipient. SA can use these headers,
3395 obviously out-of-the-box. Patch from Alex Miller.
3397 PH/08 The ${def test on a variable was returning false if the variable's
3398 value was "0", contrary to what the specification has always said!
3399 The result should be true unless the variable is empty.
3401 PH/09 The syntax error of a character other than { following "${if
3402 def:variable_name" (after optional whitespace) was not being diagnosed.
3403 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3404 accidental colon was present, for example, could give incorrect results.
3406 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3407 result is used (not including appendfile, where other changes are about
3410 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3411 This involved changing a lot of size variables from int to off_t. It
3412 should work with maildirs and everything.
3414 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3415 spamd dying while we are connected to it.
3417 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3420 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3421 function for BDB 4.3.
3423 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3425 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3426 the sources for PCRE have been re-arranged and more files are now
3429 PH/15 The code I had for printing potentially long long variables in PH/11
3430 above was not the best (it lost precision). The length of off_t variables
3431 is now inspected at build time, and an appropriate printing format (%ld
3432 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3433 to be "long long int" or "long int". This is needed for the internal
3434 formatting function string_vformat().
3436 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3437 the configuration file to be ":syslog", then the script "guesses" where
3438 the logs files are, rather than using the compiled in default. In our
3439 case the guess is not the same as the compiled default, so the script
3440 suddenly stopped working when I started to use syslog. The patch checks
3441 to see if log_file_path is "". If so, it attempts to read it from exim
3442 with no configuration file to get the compiled in version, before it
3443 falls back to the previous guessing code."
3445 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3446 implementing BATV in an Exim configuration. See NewStuff for the gory
3449 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3450 Makefile that are specific to HP-UX.
3452 PH/18 If the "use_postmaster" option was set for a recipient callout together
3453 with the "random" option, the postmaster address was used as the MAIL
3454 FROM address for the random test, but not for the subsequent recipient
3455 test. It is now used for both.
3457 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3458 patch removes a few documentation additions to RFC 3028, because the
3459 latest draft now contains them. It adds the new en;ascii-case comparator
3460 and a new error check for 8bit text in MIME parts. Comparator and
3461 require names are now matched exactly. I enabled the subaddress
3462 extension, but it is not well tested yet (read: it works for me)."
3464 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3465 rework some of the code of TK/09 above to avoid the hardwired use of
3466 "%lld" and "long long". Replaced the call to snprintf() with a call to
3469 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3470 records point to non-existent hosts", "retry timeout exceeded", and
3471 "retry time not reached for any host after a long failure period".
3473 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3474 experimental DomainKeys support:
3476 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3477 (2) On an error such as an illegally used "control", the wrong name for
3478 the control was given.
3480 These problems did NOT occur unless DomainKeys support was compiled.
3482 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3484 PH/24 Added ${if match_ip condition.
3486 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3487 databases so that it will be absolutely obvious if a crash occurs in the
3488 DB library. This is a regular occurrence (often caused by mis-matched
3491 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3492 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3493 picked up on a user's system that detects such things. There doesn't seem
3494 to be a gcc warning option for this - only an attribute that has to be
3495 put on the function's prototype. It seems that in Fedora Core 4 they have
3496 set this on a number of new functions. No doubt there will be more in due
3499 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3500 specify a transport. However, if an address that was verified by such a
3501 router was the subject of a callout, Exim crashed because it tried to
3502 read the rcpt_include_affixes from the non-existent transport. Now it
3503 just assumes that the setting of that option is false. This bug was
3504 introduced by 4.51/PH/31.
3506 PH/28 Changed -d+all to exclude +memory, because that information is very
3507 rarely of interest, but it makes the output a lot bigger. People tend to
3508 do -d+all out of habit.
3510 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3511 code in os-type was giving problems when libc.so lives in lib64, like on
3514 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3515 aren't the modern standard, and it seems that some systems' include files
3516 don't always have them. Exim was already checking for some of the newer
3517 ones like T_AAAA, and defining it itself. I've added checks for all the
3518 record types that Exim uses.
3520 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3521 not automatically generating a new one, as it is supposed to. This
3522 prevented TLS from working. If the file did exist, but contained invalid
3523 data, a new version was generated, as expected. It was only the case of a
3524 non-existent file that was broken.
3526 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3527 with a change in libdomainkeys > 0.64.
3529 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3530 from DNS. If the selector record carries the flag, it now has
3531 precedence over the domain-wide flag.
3533 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3535 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3536 the use of a port name that isn't defined in /etc/services) occurred, the
3537 message was deferred as in a normal delivery, and thus remained on the
3538 spool, instead of being failed because of the mua_wrapper setting. This
3539 is now fixed, and I tidied up some of the mua_wrapper messages at the
3542 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3543 the messages in a hash of arrays rather than using individual hashes.
3544 This is a bit cleaner and results in dramatic memory savings, albeit
3545 at a slight CPU cost.
3547 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3548 as requested by Marc Sherman.
3550 SC/10 Eximstats: added histograms for user specified patterns as requested
3553 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3555 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3556 fopen() in the content-scanning modules that did not already have it.
3562 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3563 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3565 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3567 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3569 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3570 to test Sieve filters that use "vacation".
3572 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3573 that changes the way the GnuTLS parameters are stored in the cache file.
3574 The new format can be generated externally. For backward compatibility,
3575 if the data in the cache doesn't make sense, Exim assumes it has read an
3576 old-format file, and it generates new data and writes a new file. This
3577 means that you can't go back to an older release without removing the
3580 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3581 work if there are any delivery delays because "one_time" forces the
3582 parent to be marked "delivered", so its unseen clone is never tried
3583 again. For this reason, Exim now forbids the simultaneous setting of
3586 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3587 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3588 entry if you want to know the details. The fix, however, neglected to
3589 consider the case when local delivery batching is involved. The test for
3590 "previously delivered" was not happening when checking to see if an
3591 address could be batched with a previous (undelivered) one; under
3592 certain circumstances this could lead to multiple deliveries to the same
3595 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3596 in its include files, and this causes problems building Exim.
3598 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3599 header_syntax) but Exim was just ignoring anything given after a slash.
3600 In particular, this caused confusion with an attempt to use "verify =
3601 reverse_host_lookup/defer_ok". An error is now given when options are
3602 supplied for verify items that do not have them. (Maybe reverse_host_
3603 lookup should have a defer_ok option, but that's a different point.)
3605 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3606 defined by RFC 821) to 2048, because there were problems with some AUTH
3607 commands, and RFC 1869 says the size should be increased for extended
3608 SMTP commands that take arguments.
3610 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3613 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3614 "unknown" error; now it says that the functionality isn't in the binary.
3616 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3617 an address' error message when a string expansion fails (syntax or
3618 whatever). Otherwise the password may appear in the log. Following change
3619 PH/42 below, there is no longer a chance of it appearing in a bounce
3622 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3624 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3625 of its addresses was ever tried. (Bugzilla bug #2.)
3627 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3628 the result incorrectly in the debug output. (It correctly added a newline
3629 to what was transported.)
3631 TF/01 Added $received_time.
3633 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3634 commented out examples of how to interface to a virus scanner and to
3635 SpamAssassin. Also added commented examples of av_scanner and
3636 spamd_address settings.
3638 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3639 and controls are allowed in which ACLs. There were a couple of minor
3640 errors. Some of the entries in the conditions table (which is a table of
3641 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3642 negation of where the condition IS allowed.
3644 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3646 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3647 header file does not have a version number, so I've had to invent a new
3648 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3649 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3650 radiusclient), but it was contributed by a Radius user.
3652 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3653 files or directories, for OpenSSL.
3655 PH/19 When an Exim process that is running as root has to create an Exim log
3656 file, it does so in a subprocess that runs as exim:exim so as to get the
3657 ownership right at creation (otherwise, other Exim processes might see
3658 the file with the wrong ownership). There was no test for failure of this
3659 fork() call, which would lead to the process getting stuck as it waited
3660 for a non-existent subprocess. Forks do occasionally fail when resources
3661 run out. I reviewed all the other calls to fork(); they all seem to check
3664 PH/20 When checking for unexpected SMTP input at connect time (before writing
3665 the banner), Exim was not dealing correctly with a non-positive return
3666 from the read() function. If the client had disconnected by this time,
3667 the result was a log entry for a synchronization error with an empty
3668 string after "input=" when read() returned zero. If read() returned -1
3669 (an event I could not check), uninitialized data bytes were printed.
3670 There were reports of junk text (parts of files, etc) appearing after
3673 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3675 PH/22 Added support for macro redefinition, and (re)definition in between
3676 driver and ACL definitions.
3678 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3679 forgetting to use the resulting value; it was using the unexpanded value.
3681 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3682 hadn't been configured. The fix is from Juergen Kreileder, who
3683 understands it better than I do:
3685 "Here's what I see happening with three configured cyrus_sasl
3686 authenticators configured (plain, login, cram-md5):
3688 On startup auth_cyrus_sasl_init() gets called for each of these.
3689 This means three calls to sasl_listmech() without a specified mech_list.
3690 => SASL tests which mechs of all available mechs actually work
3691 => three warnings about OTP not working
3692 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3694 With the patch, sasl_listmech() also gets called three times. But now
3695 SASL's mech_list option is set to the server_mech specified in the the
3696 authenticator. Or in other words, the answer from sasl_listmech()
3697 gets limited to just the mech you're testing for (which is different
3699 => the return list contains just 'plain' or 'login', 'cram-md5' or
3700 nothing depending on the value of ob->server_mech.
3702 I've just tested the patch: Authentication still works fine,
3703 unavailable mechs specified in the exim configuration are still
3704 caught, and the auth.log warnings about OTP are gone."
3706 PH/25 When debugging is enabled, the contents of the command line are added
3707 to the debugging output, even when log_selector=+arguments is not
3710 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3711 answer is "GNU", and only if the return is "GNU/something" is the answer
3714 PH/27 $acl_verify_message is now set immediately after the failure of a
3715 verification in an ACL, and so is available in subsequent modifiers. In
3716 particular, the message can be preserved by coding like this:
3718 warn !verify = sender
3719 set acl_m0 = $acl_verify_message
3721 Previously, $acl_verify_message was set only while expanding "message"
3722 and "log_message" when a very denied access.
3724 PH/28 Modified OS/os.c-Linux with
3726 -#ifndef OS_LOAD_AVERAGE
3727 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3729 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3730 nomenclature these days.)
3732 PH/29 Installed patch from the Sieve maintainer that adds the options
3733 sieve_useraddress and sieve_subaddress to the redirect router.
3735 PH/30 In these circumstances:
3736 . Two addresses routed to the same list of hosts;
3737 . First host does not offer TLS;
3738 . First host accepts first address;
3739 . First host gives temporary error to second address;
3740 . Second host offers TLS and a TLS session is established;
3741 . Second host accepts second address.
3742 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3743 and peerdn, if requested) that were in fact used only for the second
3746 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3747 attention to any local part prefix or suffix that was matched by the
3748 router that accepted the address. It now behaves in the same way as it
3749 does for delivery: the affixes are removed from the local part unless
3750 rcpt_include_affixes is set on the transport.
3752 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3753 timeout during the DATA phase of an incoming message.
3755 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3756 applied a patch sanctioned by the Sieve maintainer.
3758 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3759 the uid or gid is negative. A case of a negative gid caused this to be
3760 noticed. The fix allows for either to be negative.
3762 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3763 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3766 PH/36 The change PH/12 above was broken. Fixed it.
3768 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3769 the grounds that routing the same address twice would always produce the
3770 same answer. This might have been true once, but it is certainly no
3771 longer true now. Routing a child address may depend on the previous
3772 routing that produced that child. Some complicated redirection strategies
3773 went wrong when messages had multiple recipients, and made Exim's
3774 behaviour dependent on the order in which the addresses were given.
3776 I have moved the duplicate checking until after the routing is complete.
3777 Exim scans the addresses that are assigned to local and remote
3778 transports, and removes any duplicates. This means that more work will be
3779 done, as duplicates will always all be routed, but duplicates are
3780 presumably rare, so I don't expect this is of any significance.
3782 For deliveries to pipes, files, and autoreplies, the duplicate checking
3783 still happens during the routing process, since they are not going to be
3786 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3787 It corrects a timeout issue with spamd. This is Ian's comment: "The
3788 background is that sometimes spamd either never reads data from a
3789 connection it has accepted, or it never writes response data. The exiscan
3790 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3791 blindly assumes that writes won't block so it may never time out."
3793 PH/39 Allow G after quota size as well as K and M.
3795 PH/40 The value set for $authenticated_id in an authenticator may not contain
3796 binary zeroes or newlines because the value is written to log lines and
3797 to spool files. There was no check on this. Now the value is run through
3798 the string_printing() function so that such characters are converted to
3799 printable escape sequences.
3801 PH/41 $message_linecount is a new variable that contains the total number of
3802 lines in the message. Compare $body_linecount, which is the count for the
3805 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3806 bounce and delay warning messages, except in certain special cases, which
3809 (a) An SMTP error message from a remote host;
3810 (b) A message specified in a :fail: redirection;
3811 (c) A message specified in a "fail" command in a system filter;
3812 (d) A message specified in a FAIL return from the queryprogram router;
3813 (e) A message specified by the cannot_route_message router option.
3815 In these cases only, Exim does include the error details in bounce and
3816 warning messages. There are also a few cases where bland messages such
3817 as "unrouteable address" or "local delivery error" are given.
3819 PH/43 $value is now also set for the "else" part of a ${run expansion.
3821 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3822 being worked on, but at least Exim now implements the latest version to
3825 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3826 process to complete was treated as a delivery failure, a timeout while
3827 writing the message to the pipe was logged, but erroneously treated as a
3828 successful delivery. Such timeouts include transport filter timeouts. For
3829 consistency with the overall process timeout, these timeouts are now
3830 treated as errors, giving rise to delivery failures by default. However,
3831 there is now a new Boolean option for the pipe transport called
3832 timeout_defer, which, if set TRUE, converts the failures into defers for
3833 both kinds of timeout. A transport filter timeout is now identified in
3836 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3837 systems where "make" and "gmake" are different, calling "gmake" at top
3838 level broke things. I've arranged for the value of $(MAKE) to be passed
3839 from the Makefile to this script so that it can call the same version of
3843 A note about Exim versions 4.44 and 4.50
3844 ----------------------------------------
3846 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3847 changes of various kinds. As a consequence, a big documentation update was
3848 needed. This delayed the release for rather longer than seemed good, especially
3849 in the light of a couple of (minor) security issues. Therefore, the changes
3850 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3851 release. So 4.44 and 4.50 are in effect two different branches that both start
3854 I have left the 4.50 change log unchanged; it contains all the changes since
3855 4.43. The change log for 4.44 is below; many of its items are identical to
3856 those for 4.50. This seems to be the most sensible way to preserve the
3857 historical information.
3863 1. Minor wording change to the doc/README.SIEVE file.
3865 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3866 computation of the current number of files was incorrect.
3868 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3869 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3872 4. Give more explanation in the error message when the command for a transport
3873 filter fails to execute.
3875 5. There are several places where Exim runs a non-Exim command in a
3876 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3877 was being done only for the command run by the queryprogram router. It is
3878 now done for all such subprocesses. The other cases are: ${run, transport
3879 filters, and the commands run by the lmtp and pipe transports.
3881 6. Added CONFIGURE_GROUP build-time option.
3883 7. Some older OS have a limit of 256 on the maximum number of file
3884 descriptors. Exim was using setrlimit() to set 1000 as a large value
3885 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3886 systems. I've change it so that if it can't get 1000, it tries for 256.
3888 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3889 was an oversight, and furthermore, ever since the addition of extra
3890 controls (e.g. 4.43/32), the checks on when to allow different forms of
3891 "control" were broken. There should now be diagnostics for all cases when a
3892 control that does not make sense is encountered.
3894 9. Added the /retain_sender option to "control=submission".
3896 10. $recipients is now available in the predata ACL (oversight).
3898 11. Tidy the search cache before the fork to do a delivery from a message
3899 received from the command line. Otherwise the child will trigger a lookup
3900 failure and thereby defer the delivery if it tries to use (for example) a
3901 cached ldap connection that the parent has called unbind on.
3903 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3904 of $address_data from the recipient verification was clobbered by the
3905 sender verification.
3907 13. The value of address_data from a sender verification is now available in
3908 $sender_address_data in subsequent conditions in the ACL statement.
3910 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3912 15. Added a new option "connect=<time>" to callout options, to set a different
3915 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3916 was its contents. (It was OK if the option was not defined at all.)
3918 17. A "Completed" log line is now written for messages that are removed from
3919 the spool by the -Mrm option.
3921 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3922 information about exactly what failed.
3924 19. Added -dd to debug only the daemon process.
3926 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3927 handles timeouts, both on the server side and network timeouts. Renamed the
3928 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3930 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3931 It is now set to "smtps".
3933 22. $host_address is now set to the target address during the checking of
3934 ignore_target_hosts.
3936 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3937 being passed; this would have caused $sender_host_name to have been used if
3938 matching the list had actually called for a host name (not very likely,
3939 since this list is usually IP addresses). A host name is now passed as
3942 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3943 code that specifies a non-restarting handler (typically sigaction() in
3944 modern systems) in an attempt to fix a rare and obscure crash bug.
3946 25. Narrowed the window for a race in the daemon that could cause it to ignore
3947 SIGCHLD signals. This is not a major problem, because they are used only to
3948 wake it up if nothing else does.
3950 26. A malformed maildirsize file could cause Exim to calculate negative values
3951 for the mailbox size or file count. Odd effects could occur as a result.
3952 The maildirsize information is now recalculated if the size or filecount
3955 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3956 support for a long time. Removed HAVE_SYS_VFS_H.
3958 28. Installed the latest version of exipick from John Jetmore.
3960 29. In an address list, if the pattern was not a regular expression, an empty
3961 subject address (from a bounce message) matched only if the pattern was an
3962 empty string. Non-empty patterns were not even tested. This was the wrong
3963 because it is perfectly reasonable to use an empty address as part of a
3964 database query. An empty address is now tested by patterns that are
3965 lookups. However, all the other forms of pattern expect the subject to
3966 contain a local part and a domain, and therefore, for them, an empty
3967 address still always fails if the pattern is not itself empty.
3969 30. Exim went into a mad DNS loop when attempting to do a callout where the
3970 host was specified on an smtp transport, and looking it up yielded more
3971 than one IP address.
3973 31. Re-factored the code for checking spool and log partition space into a
3974 function that finds that data and another that does the check. The former
3975 is then used to implement four new variables: $spool_space, $log_space,
3976 $spool_inodes, and $log_inodes.
3978 32. The RFC2047 encoding function was originally intended for short strings
3979 such as real names; it was not keeping to the 75-character limit for
3980 encoded words that the RFC imposes. It now respects the limit, and
3981 generates multiple encoded words if necessary. To be on the safe side, I
3982 have increased the buffer size for the ${rfc2047: expansion operator from
3985 33. It is now permitted to omit both strings after an "if" condition; if the
3986 condition is true, the result is "true". As before, when the second string
3987 is omitted, a false condition yields an empty string. This makes it less
3988 cumbersome to write custom ACL and router conditions.
3990 34. Failure to deliver a bounce message always caused it to be frozen, even if
3991 there was an errors_to setting on the router. The errors_to setting is now
3994 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3995 canonical form (fully expanded) before being placed in
3996 $sender_host_address.
3998 36. The table in the code that translates DNS record types into text (T_A to
3999 "A" for instance) was missing entries for NS and CNAME. It is just possible
4000 that this could have caused confusion if both these types were looked up
4001 for the same domain, because the text type is used as part of Exim's
4002 per-process caching. But the chance of anyone hitting this buglet seems
4005 37. The dnsdb lookup has been extended in a number of ways.
4007 (1) There is a new type, "zns", which walks up the domain tree until it
4008 finds some nameserver records. It should be used with care.
4010 (2) There is a new type, "mxh", which is like "mx" except that it returns
4011 just the host names, not the priorities.
4013 (3) It is now possible to give a list of domains (or IP addresses) to be
4014 looked up. The behaviour when one of the lookups defers can be
4015 controlled by a keyword.
4017 (4) It is now possible to specify the separator character for use when
4018 multiple records are returned.
4020 38. The dnslists ACL condition has been extended: it is now possible to supply
4021 a list of IP addresses and/or domains to be looked up in a particular DNS
4024 39. Added log_selector=+queue_time_overall.
4026 40. When running the queue in the test harness, wait just a tad after forking a
4027 delivery process, to get repeatability of debugging output.
4029 41. Include certificate and key file names in error message when GnuTLS fails
4030 to set them up, because the GnuTLS error message doesn't include the name
4031 of the failing file when there is a problem reading it.
4033 42. Allow both -bf and -bF in the same test run.
4035 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4037 44. The "Exiscan patch" is now merged into the mainline Exim source.
4039 45. Sometimes the final signoff response after QUIT could fail to get
4040 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4041 < 0 before doing a fflush(). This bug looks as though it goes back to the
4042 introduction of TLS in release 3.20, but "sometimes" must have been rare
4043 because the tests only now provoked it.
4045 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4046 (this can affect the format of dates).
4048 47. exim_tidydb, when checking for the continued existence of a message for
4049 which it has found a message-specific retry record, was not finding
4050 messages that were in split spool directories. Consequently, it was
4051 deleting retry records that should have stayed in existence.
4053 48. Steve fixed some bugs in eximstats.
4055 49. The SPA authentication driver was not abandoning authentication and moving
4056 on to the next authenticator when an expansion was forced to fail,
4057 contradicting the general specification for all authenticators. Instead it
4058 was generating a temporary error. It now behaves as specified.
4060 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4061 (the order specifies the preference for clients). The order is now AES256,
4062 AES128, 3DES, ARCFOUR128.
4064 51. Small patch to Sieve code - explicitly set From: when generating an
4067 52. Exim crashed if a remote delivery caused a very long error message to be
4068 recorded - for instance if somebody sent an entire SpamAssassin report back
4069 as a large number of 550 error lines. This bug was coincidentally fixed by
4070 increasing the size of one of Exim's internal buffers (big_buffer) that
4071 happened as part of the Exiscan merge. However, to be on the safe side, I
4072 have made the code more robust (and fixed the comments that describe what
4075 53. Now that there can be additional text after "Completed" in log lines (if
4076 the queue_time_overall log selector is set), a one-byte patch to exigrep
4077 was needed to allow it to recognize "Completed" as not the last thing in
4080 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4081 patch that reportedly fixes this has been added. I am not expert enough to
4082 create a test for it. This is what the patch creator wrote:
4084 "I found a little strange behaviour of ldap code when working with
4085 Windows 2003 AD Domain, where users was placed in more than one
4086 Organization Units. When I tried to give exim partial DN, the exit code
4087 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4088 But simultaneously result of request was absolutely normal ldap result,
4089 so I produce this patch..."
4091 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4092 REFERENCE, so I have modified the code to exclude the patch when that macro
4095 55. Some experimental protocols are using DNS PTR records for new purposes. The
4096 keys for these records are domain names, not reversed IP addresses. The
4097 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4098 leaves it alone. Component reversal etc. now happens only for IP addresses.
4101 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4103 57. Double the size of the debug message buffer (to 2048) so that more of very
4104 long debug lines gets shown.
4106 58. The exicyclog utility now does better if the number of log files to keep
4107 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4109 59. Two changes related to the smtp_active_hostname option:
4111 (1) $smtp_active_hostname is now available as a variable.
4112 (2) The default for smtp_banner uses $smtp_active_hostname instead
4113 of $primary_hostname.
4115 60. The host_aton() function is supposed to be passed a string that is known
4116 to be a valid IP address. However, in the case of IPv6 addresses, it was
4117 not checking this. This is a hostage to fortune. Exim now panics and dies
4118 if the condition is not met. A case was found where this could be provoked
4119 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4120 components; fortuitously, this particular loophole had already been fixed
4121 by change 4.50/55 above.
4123 If there are any other similar loopholes, the new check in host_aton()
4124 itself should stop them being exploited. The report I received stated that
4125 data on the command line could provoke the exploit when Exim was running as
4126 exim, but did not say which command line option was involved. All I could
4127 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4128 running as the user.
4131 61. There was a buffer overflow vulnerability in the SPA authentication code
4132 (which came originally from the Samba project). I have added a test to the
4133 spa_base64_to_bits() function which I hope fixes it.
4136 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4137 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4139 63. The daemon start-up calls getloadavg() while still root for those OS that
4140 need the first call to be done as root, but it missed one case: when
4141 deliver_queue_load_max is set with deliver_drop_privilege. This is
4142 necessary for the benefit of the queue runner, because there is no re-exec
4143 when deliver_drop_privilege is set.
4145 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4146 This has been fixed.
4148 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4149 was in use, was not putting the data itself into the right store pool;
4150 consequently, it could be overwritten for a subsequent message in the same
4151 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4154 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4156 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4157 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4158 the function were treating the return as a boolean value, which happened to
4159 work because 0=false and not-0=true, but is not correct code.
4161 68. The host_aton() function was not handling scoped IPv6 addresses (those
4162 with, for example, "%eth0" on the end) correctly.
4164 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4165 negated items (that is, ~something) in unsigned ints. Some compilers
4166 apparently mutter when there is no cast.
4168 70. If an address verification called from an ACL failed, and did not produce a
4169 user-specific message (i.e. there was only a "system" message), nothing was
4170 put in $acl_verify_message. In this situation, it now puts the system
4173 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4174 session; change 4.31/43 added the unwanted input to the log line - except
4175 that it did not do this in the start of session case. It now does.
4177 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4178 This is wrong and can cause the other end to generate a synchronization
4179 error if it is another Exim or anything else that does the synchronization
4180 check. A QUIT command is no longer sent after a timeout.
4182 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4183 during host lookups.
4185 74. The defer_ok option of callout verification was not working if it was used
4186 when verifying addresses in header lines, that is, for this case:
4188 verify = header_sender/callout=defer_ok
4190 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4191 those file descriptors could be used for SMTP connections. If anything
4192 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4193 could be sent to the SMTP client, causing chaos. The daemon now opens
4194 stdin, stdout, and stderr to /dev/null when it puts itself into the
4197 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4198 log by default. The user can override this with suitable Perl magic.
4200 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4201 the log message when discard triggers, was not working for the DATA ACL or
4202 for the non-SMTP ACL.
4204 78. Error message wording change in sieve.c.
4206 79. If smtp_accept_max_per_host was set, the number of connections could be
4207 restricted to fewer than expected, because the daemon was trying to set up
4208 a new connection before checking whether the processes handling previous
4209 connections had finished. The check for completed processes is now done
4210 earlier. On busy systems, this bug wouldn't be noticed because something
4211 else would have woken the daemon, and it would have reaped the completed
4214 80. If a message was submitted locally by a user whose login name contained one
4215 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4216 It caused a spool format error. I have fixed the spool reading code. A
4217 related problem was that the "from" clause in the Received: line became
4218 illegal because of the space(s). It is now covered by ${quote_local_part.
4220 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4223 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4225 83. Patch from Sieve maintainer to fix unterminated string problem in
4226 "vacation" handling.
4228 84. Some minor changes to the Linux configuration files to help with other
4229 OS variants using glibc.
4231 85. One more patch for Sieve to update vacation handling to latest spec.
4234 ----------------------------------------------------
4235 See the note above about the 4.44 and 4.50 releases.
4236 ----------------------------------------------------
4242 1. Change 4.43/35 introduced a bug that caused file counts to be
4243 incorrectly computed when quota_filecount was set in an appendfile
4246 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4247 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4250 3. Give more explanation in the error message when the command for a transport
4251 filter fails to execute.
4253 4. There are several places where Exim runs a non-Exim command in a
4254 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4255 was being done only for the command run by the queryprogram router. It is
4256 now done for all such subprocesses. The other cases are: ${run, transport
4257 filters, and the commands run by the lmtp and pipe transports.
4259 5. Some older OS have a limit of 256 on the maximum number of file
4260 descriptors. Exim was using setrlimit() to set 1000 as a large value
4261 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4262 systems. I've change it so that if it can't get 1000, it tries for 256.
4264 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4265 was an oversight, and furthermore, ever since the addition of extra
4266 controls (e.g. 4.43/32), the checks on when to allow different forms of
4267 "control" were broken. There should now be diagnostics for all cases when a
4268 control that does not make sense is encountered.
4270 7. $recipients is now available in the predata ACL (oversight).
4272 8. Tidy the search cache before the fork to do a delivery from a message
4273 received from the command line. Otherwise the child will trigger a lookup
4274 failure and thereby defer the delivery if it tries to use (for example) a
4275 cached ldap connection that the parent has called unbind on.
4277 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4278 of $address_data from the recipient verification was clobbered by the
4279 sender verification.
4281 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4282 was its contents. (It was OK if the option was not defined at all.)
4284 11. A "Completed" log line is now written for messages that are removed from
4285 the spool by the -Mrm option.
4287 12. $host_address is now set to the target address during the checking of
4288 ignore_target_hosts.
4290 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4291 being passed; this would have caused $sender_host_name to have been used if
4292 matching the list had actually called for a host name (not very likely,
4293 since this list is usually IP addresses). A host name is now passed as
4296 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4297 code that specifies a non-restarting handler (typically sigaction() in
4298 modern systems) in an attempt to fix a rare and obscure crash bug.
4300 15. Narrowed the window for a race in the daemon that could cause it to ignore
4301 SIGCHLD signals. This is not a major problem, because they are used only to
4302 wake it up if nothing else does.
4304 16. A malformed maildirsize file could cause Exim to calculate negative values
4305 for the mailbox size or file count. Odd effects could occur as a result.
4306 The maildirsize information is now recalculated if the size or filecount
4309 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4310 support for a long time. Removed HAVE_SYS_VFS_H.
4312 18. Updated exipick to current release from John Jetmore.
4314 19. Allow an empty sender to be matched against a lookup in an address list.
4315 Previously the only cases considered were a regular expression, or an
4318 20. Exim went into a mad DNS lookup loop when doing a callout where the
4319 host was specified on the transport, if the DNS lookup yielded more than
4322 21. The RFC2047 encoding function was originally intended for short strings
4323 such as real names; it was not keeping to the 75-character limit for
4324 encoded words that the RFC imposes. It now respects the limit, and
4325 generates multiple encoded words if necessary. To be on the safe side, I
4326 have increased the buffer size for the ${rfc2047: expansion operator from
4329 22. Failure to deliver a bounce message always caused it to be frozen, even if
4330 there was an errors_to setting on the router. The errors_to setting is now
4333 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4334 canonical form (fully expanded) before being placed in
4335 $sender_host_address.
4337 24. Updated eximstats to version 1.33
4339 25. Include certificate and key file names in error message when GnuTLS fails
4340 to set them up, because the GnuTLS error message doesn't include the name
4341 of the failing file when there is a problem reading it.
4343 26. Expand error message when OpenSSL has problems setting up cert/key files.
4346 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4347 (this can affect the format of dates).
4349 28. exim_tidydb, when checking for the continued existence of a message for
4350 which it has found a message-specific retry record, was not finding
4351 messages that were in split spool directories. Consequently, it was
4352 deleting retry records that should have stayed in existence.
4354 29. eximstats updated to version 1.35
4355 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4356 1.35 - bugfix such that pie charts by volume are generated correctly
4358 30. The SPA authentication driver was not abandoning authentication and moving
4359 on to the next authenticator when an expansion was forced to fail,
4360 contradicting the general specification for all authenticators. Instead it
4361 was generating a temporary error. It now behaves as specified.
4363 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4364 (the order specifies the preference for clients). The order is now AES256,
4365 AES128, 3DES, ARCFOUR128.
4367 31. Small patch to Sieve code - explicitly set From: when generating an
4370 32. Exim crashed if a remote delivery caused a very long error message to be
4371 recorded - for instance if somebody sent an entire SpamAssassin report back
4372 as a large number of 550 error lines. This bug was coincidentally fixed by
4373 increasing the size of one of Exim's internal buffers (big_buffer) that
4374 happened as part of the Exiscan merge. However, to be on the safe side, I
4375 have made the code more robust (and fixed the comments that describe what
4378 33. Some experimental protocols are using DNS PTR records for new purposes. The
4379 keys for these records are domain names, not reversed IP addresses. The
4380 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4381 leaves it alone. Component reversal etc. now happens only for IP addresses.
4384 34. The host_aton() function is supposed to be passed a string that is known
4385 to be a valid IP address. However, in the case of IPv6 addresses, it was
4386 not checking this. This is a hostage to fortune. Exim now panics and dies
4387 if the condition is not met. A case was found where this could be provoked
4388 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4389 components; fortuitously, this particular loophole had already been fixed
4390 by change 4.50/55 or 4.44/33 above.
4392 If there are any other similar loopholes, the new check in host_aton()
4393 itself should stop them being exploited. The report I received stated that
4394 data on the command line could provoke the exploit when Exim was running as
4395 exim, but did not say which command line option was involved. All I could
4396 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4397 running as the user.
4400 35. There was a buffer overflow vulnerability in the SPA authentication code
4401 (which came originally from the Samba project). I have added a test to the
4402 spa_base64_to_bits() function which I hope fixes it.
4405 36. The daemon start-up calls getloadavg() while still root for those OS that
4406 need the first call to be done as root, but it missed one case: when
4407 deliver_queue_load_max is set with deliver_drop_privilege. This is
4408 necessary for the benefit of the queue runner, because there is no re-exec
4409 when deliver_drop_privilege is set.
4411 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4412 was in use, was not putting the data itself into the right store pool;
4413 consequently, it could be overwritten for a subsequent message in the same
4414 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4417 38. Sometimes the final signoff response after QUIT could fail to get
4418 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4419 < 0 before doing a fflush(). This bug looks as though it goes back to the
4420 introduction of TLS in release 3.20, but "sometimes" must have been rare
4421 because the tests only now provoked it.
4427 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4428 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4429 Now it returns an integer. A number of calls were still expecting a T/F
4430 return. Fortuitously, in all cases, the tests worked in OK situations,
4431 which is the norm. However, things would have gone wrong on any write
4432 failures on the smtp file descriptor. This function is used when sending
4433 messages over SMTP and also when doing verify callouts.
4435 2. When Exim is called to do synchronous delivery of a locally submitted
4436 message (the -odf or -odi options), it no longer closes stderr before doing
4439 3. Implemented the mua_wrapper option.
4441 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4443 5. Implemented the functions header_remove(), header_testname(),
4444 header_add_at_position(), and receive_remove_recipient(), and exported them
4447 6. If an ACL "warn" statement specified the addition of headers, Exim already
4448 inserted X-ACL-Warn: at the start if there was no header name. However, it
4449 was not making this test for the second and subsequent header lines if
4450 there were newlines in the string. This meant that an invalid header could
4451 be inserted if Exim was badly configured.
4453 7. Allow an ACL "warn" statement to add header lines at the start or after all
4454 the Received: headers, as well as at the end.
4456 8. Added the rcpt_4xx retry error code.
4458 9. Added postmaster_mailfrom=xxx to callout verification option.
4460 10. Added mailfrom=xxxx to the callout verification option, for verify=
4463 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4464 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4466 12. Inserted some casts to stop certain compilers warning when using pointer
4467 differences as field lengths or precisions in printf-type calls (mostly
4468 affecting debugging statements).
4470 13. Added optional readline() support for -be (dynamically loaded).
4472 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4473 same clock tick as a message's arrival, so that its received time was the
4474 same as the "first fail" time on the retry record, and that message
4475 remained on the queue past the ultimate address timeout, every queue runner
4476 would try a delivery (because it was past the ultimate address timeout) but
4477 after another failure, the ultimate address timeout, which should have then
4478 bounced the address, did not kick in. This was a "< instead of <=" error;
4479 in most cases the first failure would have been in the next clock tick
4480 after the received time, and all would be well.
4482 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4483 being recognized when the domain list was tested by the match_domain
4484 condition in an expansion string.
4486 16. Added the ${str2b64: operator.
4488 17. Exim was always calling setrlimit() to set a large limit for the number of
4489 processes, without checking whether the existing limit was already
4490 adequate. (It did check for the limit on file descriptors.) Furthermore,
4491 errors from getrlimit() and setrlimit() were being ignored. Now they are
4492 logged to the main and panic logs, but Exim does carry on, to try to do its
4493 job under whatever limits there are.
4495 18. Imported PCRE 5.0.
4497 19. Trivial typo in log message " temporarily refused connection" (the leading
4500 20. If the log selector return_path_on_delivery was set and an address was
4501 redirected to /dev/null, the delivery process crashed because it assumed
4502 that a return path would always be set for a "successful" delivery. In this
4503 case, the whole delivery is bypassed as an optimization, and therefore no
4506 21. Internal re-arrangement: the function for sending a challenge and reading
4507 a response while authentication was assuming a zero-terminated challenge
4508 string. It's now changed to take a pointer and a length, to allow for
4509 binary data in such strings.
4511 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4513 23. Exim was not respecting finduser_retries when seeking the login of the
4514 uid under which it was called; it was always trying 10 times. (The default
4515 setting of finduser_retries is zero.) Also, it was sleeping after the final
4516 failure, which is pointless.
4518 24. Implemented tls_on_connect_ports.
4520 25. Implemented acl_smtp_predata.
4522 26. If the domain in control=submission is set empty, Exim assumes that the
4523 authenticated id is a complete email address when it generates From: or
4524 Sender: header lines.
4526 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4527 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4528 chown and chgrp in /bin and hostname in /usr/bin.
4530 28. Exim was keeping the "process log" file open after each use, just as it
4531 does for the main log. This opens the possibility of it remaining open for
4532 long periods when the USR1 signal hits a daemon. Occasional processlog
4533 errors were reported, that could have been caused by this. Anyway, it seems
4534 much more sensible not to leave this file open at all, so that is what now
4537 29. The long-running daemon process does not normally write to the log once it
4538 has entered its main loop, and it closes the log before doing so. This is
4539 so that log files can straightforwardly be renamed and moved. However,
4540 there are a couple of unusual error situations where the daemon does write
4541 log entries, and I had neglected to close the log afterwards.
4543 30. The text of an SMTP error response that was received during a remote
4544 delivery was being truncated at 512 bytes. This is too short for some of
4545 the long messages that one sometimes sees. I've increased the limit to
4548 31. It is now possible to make retry rules that apply only when a message has a
4549 specific sender, in particular, an empty sender.
4551 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4552 it possible to be selective about when SMTP synchronization is enforced.
4554 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4556 32. Implemented hosts_connection_nolog.
4558 33. Added an ACL for QUIT.
4560 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4563 35. Added mailbox_size and mailbox_filecount to appendfile.
4565 36. Added control = no_multiline_responses to ACLs.
4567 37. There was a bug in the logic of the code that waits for the clock to tick
4568 in the case where the clock went backwards by a substantial amount such
4569 that the microsecond fraction of "now" was more than the microsecond
4570 fraction of "then" (but the whole seconds number was less).
4572 38. Added support for the libradius Radius client library this is found on
4573 FreeBSD (previously only the radiusclient library was supported).
4579 1. When certain lookups returned multiple values in the form name=value, the
4580 quoting of the values was not always being done properly. Specifically:
4581 (a) If the value started with a double quote, but contained no whitespace,
4583 (b) If the value contained whitespace other than a space character (i.e.
4584 tabs or newlines or carriage returns) it was not quoted.
4585 This fix has been applied to the mysql and pgsql lookups by writing a
4586 separate quoting function and calling it from the lookup code. The fix
4587 should probably also be applied to nisplus, ibase and oracle lookups, but
4588 since I cannot test any of those, I have not disturbed their existing code.
4590 2. A hit in the callout cache for a specific address caused a log line with no
4591 reason for rejecting RCPT. Now it says "Previous (cached) callout
4592 verification failure".
4594 3. There was an off-by-one bug in the queryprogram router. An over-long
4595 return line was truncated at 256 instead of 255 characters, thereby
4596 overflowing its buffer with the terminating zero. As well as fixing this, I
4597 have increased the buffer size to 1024 (and made a note to document this).
4599 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4600 when Exim is waiting for an SMTP response from a remote server, Exim
4601 restarts its select() call on the socket, thereby resetting its timeout.
4602 This is not a problem when such interrupts are rare. Somebody set up a cron
4603 job to run exiwhat every 2 minutes, which is less than the normal select()
4604 timeout (5 or 10 minutes). This meant that the select() timeout never
4605 kicked in because it was always reset. I have fixed this by comparing the
4606 time when an interrupt arrives with the time at the start of the first call
4607 to select(). If more time than the timeout has elapsed, the interrupt is
4608 treated as a timeout.
4610 5. Some internal re-factoring in preparation for the addition of Sieve
4611 extensions (by MH). In particular, the "personal" test is moved to a
4612 separate function, and given an option for scanning Cc: and Bcc: (which is
4613 not set for Exim filters).
4615 6. When Exim created an email address using the login of the caller as the
4616 local part (e.g. when creating a From: or Sender: header line), it was not
4617 quoting the local part when it contained special characters such as @.
4619 7. Installed new OpenBSD configuration files.
4621 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4622 try to make them clearer.
4624 9. Callout options, other than the timeout value, were being ignored when
4625 verifying sender addresses in header lines. For example, when using
4627 verify = header_sender/callout=no_cache
4629 the cache was (incorrectly) being used.
4631 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4632 only the Cygwin environment.
4634 11. When return_path_on_delivery was set as a log selector, if different remote
4635 addresses in the same message used different return paths and parallel
4636 remote delivery occurred, the wrong values would sometimes be logged.
4637 (Whenever a remote delivery process finished, the return path value from
4638 the most recently started remote delivery process was logged.)
4640 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4641 lines when AUTH and/or TLS are in use. This is the "received protocol"
4642 field. Exim used to use "asmtp" for authenticated SMTP, without any
4643 indication (in the protocol name) for TLS use. Now it follows the RFC and
4644 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4645 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4646 names appear in log lines as well as in Received: header lines.
4648 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4649 extensions, and comparison tests, and to fix some bugs.
4651 14. Changes to the "personal" filter test:
4653 (1) The test was buggy in that it was just doing the equivalent of
4654 "contains" tests on header lines. For example, if a user's address was
4655 anne@some.where, the "personal" test would incorrectly be true for
4657 To: susanne@some.where
4659 This test is now done by extracting each address from the header in turn,
4660 and checking the entire address. Other tests that are part of "personal"
4661 are now done using regular expressions (for example, to check local parts
4662 of addresses in From: header lines).
4664 (2) The list of non-personal local parts in From: addresses has been
4665 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4666 taken from the Sieve specification recommendations.
4668 (3) If the message contains any header line starting with "List-" it is
4669 treated as non-personal.
4671 (4) The test for "circular" in the Subject: header line has been removed
4672 because it now seems ill-conceived.
4674 15. Minor typos in src/EDITME comments corrected.
4676 16. Installed latest exipick from John Jetmore.
4678 17. If headers_add on a router specified a text string that was too long for
4679 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4680 of string_sprintf() is now avoided.
4682 18. $message_body_size was not set (it was always zero) when running the DATA
4683 ACL and the local_scan() function.
4685 19. For the "mail" command in an Exim filter, no default was being set for
4686 the once_repeat time, causing a random time value to be used if "once" was
4687 specified. (If the value happened to be <= 0, no repeat happened.) The
4688 default is now 0s, meaning "never repeat". The "vacation" command was OK
4689 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4690 (I found it when inspecting the code).
4692 20. There is now an overall timeout for performing a callout verification. It
4693 defaults to 4 times the callout timeout, which applies to individual SMTP
4694 commands during the callout. The overall timeout applies when there is more
4695 than one host that can be tried. The timeout is checked before trying the
4696 next host. This prevents very long delays if there are a large number of
4697 hosts and all are timing out (e.g. when the network connections are timing
4698 out). The value of the overall timeout can be changed by specifying an
4699 additional sub-option for "callout", called "maxwait". For example:
4701 verify = sender/callout=5s,maxwait=20s
4703 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4704 to the end before writing, but this should make it even safer).
4706 22. Exim was forgetting that it had advertised PIPELINING for the second and
4707 subsequent messages on an SMTP connection. It was also not resetting its
4708 memory on STARTTLS and an internal HELO.
4710 23. When Exim logs an SMTP synchronization error within a session, it now
4711 records whether PIPELINING has been advertised or not.
4713 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4714 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4715 rather than long int.
4717 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4719 26. Added the never_mail option to autoreply.
4725 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4726 crash if the getsockname() call failed; this can happen if a connection is
4727 closed very soon after it is established. The problem was simply in the
4728 order in which certain operations were done, causing Exim to try to write
4729 to the SMTP stream before it had set up the file descriptor. The bug has
4730 been fixed by making things happen in the correct order.
4736 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4737 before the connection was closed, thus losing the rejection response.
4739 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4740 some early Solaris releases, but causes trouble in current releases where
4741 socklen_t is defined.
4743 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4746 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4749 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4750 that releasing the top of it at the end releases what was used for sub-
4751 expansions (unless the block got too big). However, discard this block if
4752 the first thing is a variable or header, so that we can use its block when
4753 it is dynamic (useful for very large $message_headers, for example).
4755 6. Lookups now cache *every* query, not just the most recent. A new, separate
4756 store pool is used for this. It can be recovered when all lookup caches are
4757 flushed. Lookups now release memory at the end of their result strings.
4758 This has involved some general refactoring of the lookup sources.
4760 7. Some code has been added to the store_xxx() functions to reduce the amount
4761 of flapping under certain conditions.
4763 8. log_incoming_interface used to affect only the <= reception log lines. Now
4764 it causes the local interface and port to be added to several more SMTP log
4765 lines, for example "SMTP connection from", and rejection lines.
4767 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4769 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4771 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4772 could be overwritten at the end of the current message (or the start of a
4773 new message if it was set in a HELO ACL). The value is now preserved for
4774 the duration of the SMTP connection.
4776 12. If a transport had a headers_rewrite setting, and a matching header line
4777 contained an unqualified address, that address was qualified, even if it
4778 did not match any rewriting rules. The underlying bug was that the values
4779 of the flags that permit the existence of unqualified sender and recipient
4780 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4781 non-local messages, and by -bnq for local messages) were not being
4782 preserved with the message after it was received.
4784 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4785 "next input=" as part of the text comprising the host identity instead of
4786 the correct text. The code was using the same buffer for two different
4787 strings. However, depending on which order the printing function evaluated
4788 its arguments, the bug did not always show up. Under Linux, for example, my
4789 test suite worked just fine.
4791 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4792 doesn't work with some older versions of Perl. It has been changed to "my",
4793 which in any case is probably the better facility to use.
4795 15. A really picky compiler found some instances of statements for creating
4796 error messages that either had too many or two few arguments for the format
4799 16. The size of the buffer for calls to the DNS resolver has been increased
4800 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4801 for addresses that have a lot of PTR records. This alleviates a problem; it
4802 does not fully solve it.
4804 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4805 buffer now truncates the list and logs the incident, which is the same
4806 action as happens when Exim is looking up a host name and its aliases.
4807 Previously in this situation something unpredictable would happen;
4808 sometimes it was "internal error: store_reset failed".
4810 18. If a server dropped the connection unexpectedly when an Exim client was
4811 using GnuTLS and trying to read a response, the client delivery process
4812 crashed while trying to generate an error log message.
4814 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4815 string, for example:
4817 warn message = H1: something\nH2: something
4819 the text was added as a single header line from Exim's point of view
4820 though it ended up OK in the delivered message. However, searching for the
4821 second and subsequent header lines using $h_h2: did not work. This has been
4822 fixed. Similarly, if a system filter added multiple headers in this way,
4823 the routers could not see them.
4825 20. Expanded the error message when iplsearch is called with an invalid key to
4826 suggest using net-iplsearch in a host list.
4828 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4829 ACLs are no longer actually imposed (and a message to that effect is
4832 22. If a "gecos" field in a passwd entry contained escaped characters, in
4833 particular, if it contained a \" sequence, Exim got it wrong when building
4834 a From: or a Sender: header from that name. A second bug also caused
4835 incorrect handling when an unquoted " was present following a character
4836 that needed quoting.
4838 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4839 was not being matched caselessly.
4841 24. Arranged for all hyphens in the exim.8 source to be escaped with
4844 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4845 properly. Recipient callout cache records were still being keyed to include
4846 the sender, even when use_sender was set false. This led to far more
4847 callouts that were necessary. The sender is no longer included in the key
4848 when use_sender is false.
4850 26. Added "control = submission" modifier to ACLs.
4852 27. Added the ${base62d: operator to decode base 62 numbers.
4854 28. dnsdb lookups can now access SRV records.
4856 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4857 the configuration file.
4859 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4860 (-v) mode. This makes the output for a verbose queue run more intelligible.
4862 31. Added a use_postmaster feature to recipient callouts.
4864 32. Added the $body_zerocount variable, containing the number of binary zero
4865 bytes in the message body.
4867 33. The time of last modification of the "new" subdirectory is now used as the
4868 "mailbox time last read" when there is a quota error for a maildir
4871 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4873 35. Added +ignore_unknown as a special item in host lists.
4875 36. Code for decoding IPv6 addresses in host lists is now included, even if
4876 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4877 address was recognized as an IP address, but was then not correctly decoded
4878 into binary, causing unexpected and incorrect effects when compared with
4885 1. Very minor rewording of debugging text in manualroute to say "list of
4886 hosts" instead of "hostlist".
4888 2. If verify=header_syntax was set, and a header line with an unqualified
4889 address (no domain) and a large number of spaces between the end of the
4890 name and the colon was received, the reception process suffered a buffer
4891 overflow, and (when I tested it) crashed. This was caused by some obsolete
4892 code that should have been removed. The fix is to remove it!
4894 3. When running in the test harness, delay a bit after writing a bounce
4895 message to get a bit more predictability in the log output.
4897 4. Added a call to search_tidyup() just before forking a reception process. In
4898 theory, someone could use a lookup in the expansion of smtp_accept_max_
4899 per_host which, without the tidyup, could leave open a database connection.
4901 5. Added the variables $recipient_data and $sender_data which get set from a
4902 lookup success in an ACL "recipients" or "senders" condition, or a router
4903 "senders" option, similar to $domain_data and $local_part_data.
4905 6. Moved the writing of debug_print from before to after the "senders" test
4908 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4909 problems for message scanning, either using a data ACL, or using
4910 local_scan() because the Received: header was not generated till after they
4911 were called (in order to set the time as the time of reception completion).
4912 I have revised the way this works. The header is now generated after the
4913 body is received, but before the ACL or local_scan() are called. After they
4914 are run, the timestamp in the header is updated.
4920 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4921 before starting a queue runner without re-exec. This happened only when
4922 deliver_drop_privilege was set or when the Exim user was set to root. The
4923 effect of the bug was that timeouts during subsequent deliveries caused
4924 crashes instead of being properly handled. The handler is now left at its
4925 default (and expected) setting.
4927 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4928 message, again when deliver_drop_privilege is set or Exim is run as root.
4929 The bug described in (1) was not present in this case, but the tidying up
4930 of the other signals was missing. I have made the two cases consistent.
4932 3. The ignore_target_hosts setting on a manualroute router was being ignored
4933 for hosts that were looked up using the /MX notation.
4935 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4938 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4939 operated on the sender address. After changing the $sender_address to <>
4940 for the sender address verify, Exim was re-instated it as the original
4941 (before rewriting) address, but remembering that it had rewritten it, so it
4942 wasn't rewriting it again. This bug also had the effect of breaking the
4943 sender address verification caching when the sender address was rewritten.
4945 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4946 This has been changed so that if the ip literal address matches
4947 ignore_target_hosts, the router declines.
4949 7. Added expansion conditions match_domain, match_address, and match_local_
4950 part (NOT match_host).
4952 8. The placeholder for the Received: header didn't have a length field set.
4954 9. Added code to Exim itself and to exim_lock to test for a specific race
4955 condition that could lead to file corruption when using MBX delivery. The
4956 issue is with the lockfile that is created in /tmp. If this file is removed
4957 after a process has opened it but before that process has acquired a lock,
4958 there is the potential for a second process to recreate the file and also
4959 acquire a lock. This could lead to two Exim processes writing to the file
4960 at the same time. The added code performs the same test as UW imapd; it
4961 checks after acquiring the lock that its file descriptor still refers to
4962 the same named file.
4964 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4965 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4968 11. Added the smtp_active_hostname option. If used, this will typically be made
4969 to depend on the incoming interface address. Because $interface_address is
4970 not set up until the daemon has forked a reception process, error responses
4971 that can happen earlier (such as "too many connections") no longer contain
4974 12. If an expansion in a condition on a "warn" statement fails because a lookup
4975 defers, the "warn" statement is abandoned, and the next ACL statement is
4976 processed. Previously this caused the whole ACL to be aborted.
4978 13. Added the iplsearch lookup type.
4980 14. Added ident_timeout as a log selector.
4982 15. Added tls_certificate_verified as a log selector.
4984 16. Added a global option tls_require_ciphers (compare the smtp transport
4985 option of the same name). This controls incoming TLS connections.
4987 17. I finally figured out how to make tls_require_ciphers do a similar thing
4988 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4989 before starting the TLS session.
4991 18. Tabs are now shown as \t in -bP output.
4993 19. If the log selector return_path_on_delivery was set, Exim crashed when
4994 bouncing a message because it had too many Received: header lines.
4996 20. If two routers both had headers_remove settings, and the first one included
4997 a superfluous trailing colon, the final name in the first list and the
4998 first name in the second list were incorrectly joined into one item (with a
4999 colon in the middle).
5005 1. Added -C and -D options to the exinext utility, mainly to make it easier
5006 to include in the automated testing, but these could be helpful when
5007 multiple configurations are in use.
5009 2. The exinext utility was not formatting the output nicely when there was
5010 an alternate port involved in the retry record key, nor when there was a
5011 message id as well (for retries that were specific to a specific message
5012 and a specific host). It was also confused by IPv6 addresses, because of
5013 the additional colons they contain. I have fixed the IPv4 problem, and
5014 patched it up to do a reasonable job for IPv6.
5016 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5017 delivery, the log line now contains "pipelined" if PIPELINING was used.
5019 4. An SMTP transport process used to panic and die if the bind() call to set
5020 an explicit outgoing interface failed. This has been changed; it is now
5021 treated in the same way as a connect() failure.
5023 5. A reference to $sender_host_name in the part of a conditional expansion
5024 that was being skipped was still causing a DNS lookup. This no longer
5027 6. The def: expansion condition was not recognizing references to header lines
5028 that used bh_ and bheader_.
5030 7. Added the _cache feature to named lists.
5032 8. The code for checking quota_filecount in the appendfile transport was
5033 allowing one more file than it should have been.
5035 9. For compatibility with Sendmail, the command line option
5043 and sets the incoming protocol and host name (for trusted callers). The
5044 host name and its colon can be omitted when only the protocol is to be set.
5045 Note the Exim already has two private options, -pd and -ps, that refer to
5046 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5047 "s", but I don't think that's a major issue.
5049 10. A number of refactoring changes to the code, none of which should affect
5052 (a) The number of logging options was getting close to filling up the
5053 32-bit word that was used as a bit map. I have split them into two classes:
5054 those that are passed in the argument to log_write(), and those that are
5055 only ever tested independently outside of that function. These are now in
5056 separate 32-bit words, so there is plenty of room for expansion again.
5057 There is no change in the user interface or the logging behaviour.
5059 (b) When building, for example, log lines, the code previously used a
5060 macro that called string_cat() twice, in order to add two strings. This is
5061 not really sufficiently general. Furthermore, there was one instance where
5062 it was actually wrong because one of the argument was used twice, and in
5063 one call a function was used. (As it happened, calling the function twice
5064 did not affect the overall behaviour.) The macro has been replaced by a
5065 function that can join an arbitrary number of extra strings onto a growing
5068 (c) The code for expansion conditions now uses a table and a binary chop
5069 instead of a serial search (which was left over from when there were very
5070 few conditions). Also, it now recognizes conditions like "pam" even when
5071 the relevant support is not compiled in: a suitably worded error message is
5072 given if an attempt is made to use such a condition.
5074 11. Added ${time_interval:xxxxx}.
5076 12. A bug was causing one of the ddress fields not to be passed back correctly
5077 from remote delivery subprocesses. The field in question was not being
5078 subsequently used, so this caused to problems in practice.
5080 13. Added new log selectors queue_time and deliver_time.
5082 14. Might have fixed a bug in maildirsizefile handling that threw up
5083 "unexpected character" debug warnings, and recalculated the data
5084 unnecessarily. In any case, I expanded the warning message to give more
5087 15. Added the message "Restricted characters in address" to the statements in
5088 the default ACL that block characters like @ and % in local parts.
5090 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5091 Three changes have been made:
5093 (a) There was a serious bug; a negative response to MAIL caused the whole
5094 recipient domain to be cached as invalid, thereby blocking all messages
5095 to all local parts at the same domain, from all senders. This bug has
5096 been fixed. The domain is no longer cached after a negative response to
5097 MAIL if the sender used is not empty.
5099 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5102 (c) A new callout option, "use_sender" has been added for people who want
5103 the modified behaviour.
5109 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5112 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5113 indeed breaks things for older releases.
5115 3. Added additional logging to the case where there is a problem reading data
5116 from a filter that is running in a subprocess using a pipe, in order to
5117 try to track down a specific problem.
5119 4. Testing facility fudge: when running in the test harness and attempting
5120 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5121 getting "No route to host". Convert this to a timeout.
5123 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5126 6. Some OS don't have socklen_t but use size_t instead. This affects the
5127 fifth argument of getsockopt() amongst other things. This is now
5128 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5129 can be set for individual OS. I have set it for SunOS5, OSF1, and
5130 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5131 some earlier ones do not.
5133 7. Change 4.30/15 was not doing the test caselessly.
5135 8. The standard form for an IPv6 address literal was being rejected by address
5136 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5137 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5138 this, as well as the form without the "IPv6" on the front (but only when
5139 address literals are enabled, of course).
5141 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5143 10. Exim crashed if a message with an empty sender address specified by -f
5144 encountered a router with an errors_to setting. This could be provoked only
5145 by a command such as
5149 where an empty string was supplied; "<>" did not hit this bug.
5151 11. Installed PCRE release 4.5.
5153 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5154 remained set. It is now erased.
5156 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5157 times from message ids (which are base 36 rather than the normal 62).
5159 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5160 were being counted as actual protocol errors, and logged if the log
5161 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5162 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5163 and DATA following a set of rejected RCPTs do not count as protocol errors.
5164 In other words, Exim assumes they were pipelined, though this may not
5165 actually be the case. Of course, in all cases the client gets an
5166 appropriate error code.
5168 15. If a lookup fails in an ACL condition, a message about the failure may
5169 be available; it is used if testing the ACL cannot continue, because most
5170 such messages specify what the cause of the deferral is. However, some
5171 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5172 that caused an old message to be retained and used if a later statement
5173 caused a defer, replacing the real cause of the deferral.
5175 16. If an IP address had so many PTR records that the DNS lookup buffer
5176 was not large enough to hold them, Exim could crash while trying to process
5177 the truncated data. It now detects and logs this case.
5179 17. Further to 4.21/58, another change has been made: if (and only if) the
5180 first line of a message (the first header line) ends with CRLF, a bare LF
5181 in a subsequent header line has a space inserted after it, so as not to
5182 terminate the header.
5184 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5185 unnecessarily, used atoi() instead of strtol(), and didn't check the
5186 termination when getting file sizes from file names by regex.
5188 19. Completely re-implemented the support for maildirsize files, in the light
5189 of a number of problems with the previous contributed implementation
5190 (4.30/29). In particular:
5192 . If the quota is zero, the maildirsize file is maintained, but no quota is
5195 . If the maildir directory does not exist, it is created before any attempt
5196 to write a maildirsize file.
5198 . The quota value in the file is just a cache; if the quota is changed in
5199 the transport, the new value overrides.
5201 . A regular expression is available for excluding directories from the
5204 20. The autoreply transport checks the characters in options that define the
5205 message's headers; it allows continued headers, but it was checking with
5206 isspace() after an embedded newline instead of explicitly looking for a
5209 21. If all the "regular" hosts to which an address was routed had passed their
5210 expiry times, and had not reached their retry times, the address was
5211 bounced, even if fallback hosts were defined. Now Exim should go on to try
5214 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5215 equivalent code in the SMTP transport. Some hosts send humungous responses
5216 to HELO/EHLO, more than 1024 it seems.
5218 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5219 gives compiler warnings in some environments. I've now done it "properly",
5222 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5223 (because of problems with the built-in one) was declared to return uschar *
5224 instead of char *, causing compiler failure.
5226 25. Fixed a file descriptor leak when processing alias/forward files.
5228 26. Fixed a minor format string issue in dbfn.c.
5230 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5232 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5233 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5236 29. When a sender address is verified, it is cached, to save repeating the test
5237 when there is more than one recipient in a message. However, when the
5238 verification involves a callout, it is possible for different callout
5239 options to be set for different recipients. It is too complicated to keep
5240 track of this in the cache, so now Exim always runs a verification when a
5241 callout is required, relying on the callout cache for the optimization.
5242 The overhead is duplication of the address routing, but this should not be
5245 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5246 to be verified with callout=postmaster, and the main callout worked but the
5247 postmaster check failed, the verification correctly failed. However, if a
5248 subsequent RCPT command asked for sender verification *without* the
5249 postmaster check, incorrect caching caused this verification also to fail,
5252 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5253 it was not caching the DNS options (qualify_single, search_parents) that
5254 were used when the lookup failed. A subsequent lookup with different
5255 options therefore always gave the same answer, though there were cases
5256 where it should not have. (Example: a "domains = !$mx_any" option on a
5257 dnslookup router: the "domains" option is always processed without any
5258 widening, but the router might have qualify_single set.) Now Exim uses the
5259 cached value only when the same options are set.
5261 32. Added John Jetmore's "exipick" utility to the distribution.
5263 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5264 than a timeout (e.g. a certificate is required, and is not provided), an
5265 Exim server now closes the connection immediately. Previously it waited for
5266 the client to close - but if the client is SSL, it seems that they each
5267 wait for each other, leading to a delay before one of them times out.
5269 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5270 maintained 0.8.x compatibility because I don't think many are using it, and
5271 it is clearly obsolete.
5273 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5276 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5277 client certificate was expired. A simple patch fixes this, though I don't
5278 understand the full logic of why the verify callback is called multiple
5281 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5282 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5283 which causes problems with some clients (such as the Certicom SSL Plus
5284 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5285 disables the coutermeasure allowing Eudora to connect."
5287 38. Exim was not checking that a write() to a log file succeeded. This could
5288 lead to Bad Things if a log got too big, in particular if it hit a file
5289 size limit. Exim now panics and dies if it cannot write to a log file, just
5290 as it does if it cannot open a log file.
5292 39. Modified OS/Makefile-Linux so that it now contains
5294 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5296 The two -D definitions ensure that Exim is compiled with large file
5297 support, which makes it possible to handle log files that are bigger than
5300 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5301 instance) a domain was checked against a named list that involved a lookup,
5302 causing $domain_data to be set, then another domain was checked against the
5303 same list, then the first domain was re-checked, the value of $domain_data
5304 after the final check could be wrong. In particular, if the second check
5305 failed, it could be set empty. This bug probably also applied to
5308 41. The strip_trailing_dot option was not being applied to the address given
5309 with the -f command-line option.
5311 42. The code for reading a message's header from the spool was incrementing
5312 $received_count, but never initializing it. This meant that the value was
5313 incorrect (doubled) while delivering a message in the same process in which
5314 it was received. In the most common configuration of Exim, this never
5315 happens - a fresh exec is done - but it can happen when
5316 deliver_drop_privilege is set.
5318 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5319 it now includes up to 150 characters of the unexpected data in the log
5322 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5323 and building data strings. The size of both of these buffers was 10 000
5324 bytes - far larger than anybody would *ever* want, thought I. Needless to
5325 say, somebody hit the limit. I have increased the maximum line length to
5326 20 000 and the maximum data length of concatenated lines to 100 000. I have
5327 also fixed two bugs, because there was no checking on these buffers. Tsk,
5328 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5329 buffer is too small.
5331 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5332 lsearch lookups. Now it does.
5334 46. When parsing a route_list item in a manualroute router, a fixed-length
5335 buffer was used for the list of hosts. I made this 1024 bytes long,
5336 thinking that nobody would ever have a list of hosts that long. Wrong.
5337 Somebody had a whole pile of complicated expansion conditions, and the
5338 string was silently truncated, leading to an expansion error. It turns out
5339 that it is easier to change to an unlimited length (owing to other changes
5340 that have happened since this code was originally written) than to build
5341 structure for giving a limitation error. The length of the item that
5342 expands into the list of hosts is now unlimited.
5344 47. The lsearch lookup could not handle data where the length of text line was
5345 more than 4095 characters. Such lines were truncated, leading to shortened
5346 data being returned. It should now handle lines of any length.
5348 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5349 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5352 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5353 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5354 no longer happens in comment lines. A list of replacements is now placed
5355 at the head of all of the source files, except those whose only change is
5356 to replace PERL_COMMAND in the very first #! line.
5358 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5359 messages on the queue, with a bottom-up merge sort, using code contributed
5360 by Michael Haardt. This should make operations like -bp somewhat faster on
5361 large queues. It won't affect queue runners, except when queue_run_in_order
5364 51. Installed eximstats 1.31 in the distribution.
5366 52. Added support for SRV lookups to the dnslookup router.
5368 53. If an ACL referred to $message_body or $message_body_end, the value was not
5369 reset for any messages that followed in the same SMTP session.
5371 54. The store-handling optimization for building very long strings was not
5372 differentiating between the different store pools. I don't think this
5373 actually made any difference in practice, but I've tidied it.
5375 55. While running the routers to verify a sender address, $sender_address
5376 was still set to the sender address. This is wrong, because when routing to
5377 send a bounce to the sender, it would be empty. Therefore, I have changed
5378 it so that, while verifying a sender address, $sender_address is set to <>.
5379 (There is no change to what happens when verifying a recipient address.)
5381 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5382 target A or AAAA records (if not already returned) without resetting the
5383 qualify_single or search_parents options of the DNS resolver. These are
5384 inappropriate in this case because the targets of MX and SRV records must
5385 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5386 target that, when qualified, matched something in the local domain. These
5387 two options are now turned off when doing these lookups.
5389 57. It seems that at least some releases of Reiserfs (which does not have the
5390 concept of a fixed number of inodes) returns zero and not -1 for the
5391 number of available inodes. This interacted badly with check_spool_inodes,
5392 which assumed that -1 was the "no such thing" setting. What I have done is
5393 to check that the total number of inodes is greater than zero before doing
5394 the test of how many are available.
5396 58. When a "warn" ACL statement has a log_message modifier, the message is
5397 remembered, and not repeated. This is to avoid a lot of repetition when a
5398 message has many recipients that cause the same warning to be written.
5399 Howewer, Exim was preserving the list of already written lines for an
5400 entire SMTP session, which doesn't seem right. The memory is now reset if a
5401 new message is started.
5403 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5404 debugging output unless log_rewrite was also set.
5406 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5407 of (int)(handle) when we know that handle contains (void *)(-1).
5409 61. The Exim daemon panic-logs an error return when it closes the incoming
5410 connection. However "connection reset by peer" seems to be common, and
5411 isn't really an error worthy of noting specially, so that particular error
5414 62. When Exim is trying to find all the local interfaces, it used to panic and
5415 die if the ioctl to get the interface flags failed. However, it seems that
5416 on at least one OS (Solaris 9) it is possible to have an interface that is
5417 included in the list of interfaces, but for which you get a failure error
5418 for this call. This happens when the interface is not "plumbed" into a
5419 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5420 failure of the "get flags" call assumes that the interface is down.
5422 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5423 makes life easier for people who are doing arithmetic on fields extracted
5424 from dates, where you often get leading zeros that should not be
5425 interpreted as octal.
5427 64. Added qualify_domain to the redirect router, to override the global
5430 65. If a pathologically long header line contained very many addresses (the
5431 report of this problem mentioned 10 000) and each of them was rewritten,
5432 Exim could use up a very large amount of memory. (It kept on making new
5433 copies of the header line as it rewrote, and never released the old ones.)
5434 At the expense of a bit more processing, the header rewriting function has
5435 been changed so that it no longer eats memory in this way.
5437 66. The generation of the Received: header has been moved from the time that a
5438 message starts to be received, to the time that it finishes. The timestamp
5439 in the Received: header should now be very close to that of the <= log
5440 line. There are two side-effects of this change:
5442 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5443 logged header lines no longer include the local Received: line, because
5444 it has not yet been created. The same applies to a copy of the message
5445 that is returned to a non-SMTP sender when a message is rejected.
5447 (b) When a filter file is tested using -bf, no additional Received: header
5448 is added to the test message. After some thought, I decided that this
5451 This change does not affect the value of $received_for. It is still set
5452 after address rewriting, but before local_scan() is called.
5454 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5456 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5457 gave an unhelpful panic error message, and a defer error. I have managed to
5458 change this behaviour so that it now rejects any supplied certificate,
5459 which seems right, as the list of acceptable certificates is empty.
5461 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5462 gave an unhelpful defer error. I have not managed to make this reject any
5463 supplied certificates, but the error message it gives is "no certificate
5464 supplied", which is not helpful.
5466 70. exigrep's output now also includes lines that are not associated with any
5467 message, but which match the given pattern. Implemented by a patch from
5468 Martin Sluka, which also tidied up the Perl a bit.
5470 71. Recipient callout verification, like sender verification, was using <> in
5471 the MAIL FROM command. This isn't really the right thing, since the actual
5472 sender may affect whether the remote host accepts the recipient or not. I
5473 have changed it to use the actual sender in the callout; this means that
5474 the cache record is now keyed on a recipient/sender pair, not just the
5475 recipient address. There doesn't seem to be a real danger of callout loops,
5476 since a callout by the remote host to check the sender would use <>.
5477 [SEE ABOVE: changed after hitting problems.]
5479 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5480 temporary errors. However, in the case of such a code being given after
5481 the end of a data transmission (i.e. after ".") Exim was failing to write
5482 a retry record for the message. (Yes, there was some broken host that was
5483 actually sending 8xx at this point.)
5485 73. An unknown lookup type in a host list could cause Exim to panic-die when
5486 the list was checked. (An example that provoked this was putting <; in the
5487 middle of a list instead of at the start.) If this happened during a DATA
5488 ACL check, a -D file could be left lying around. This kind of configuration
5489 error no longer causes Exim to die; instead it causes a defer errror. The
5490 incident is still logged to the main and panic logs.
5492 74. Buglet left over from Exim 3 conversion. The message "too many messages
5493 in one connection" was written to the rejectlog but not the mainlog, except
5494 when address rewriting (yes!) was being logged.
5496 75. Added write_rejectlog option.
5498 76. When a system filter was run not as root (that is, when system_filter_user
5499 was set), the values of the $n variables were not being returned to the
5500 main process; thus, they were not subsequently available in the $sn
5503 77. Added +return_path_on_delivery log selector.
5505 78. A connection timeout was being treated differently from recipients deferred
5506 when testing hosts_max_try with a message that was older than the host's
5507 retry timeout. (The host should not be counted, thus allowing all hosts to
5508 be tried at least once before bouncing.) This may have been the cause of an
5509 occasionally reported bug whereby a message would remain on the queue
5510 longer than the retry timeout, but would be bounced if a delivery was
5511 forced. I say "may" because I never totally pinned down the problem;
5512 setting up timeout/retry tests is difficult. See also the next item.
5514 79. The ultimate address timeout was not being applied to errors that involved
5515 a combination of host plus message (for example, a timeout on a MAIL
5516 command). When an address resolved to a number of possible hosts, and they
5517 were not all tried for each delivery (e.g. because of hosts_max_try), a
5518 message could remain on the queue longer than the retry timeout.
5520 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5523 81. Fixed an obscure SMTP outgoing bug which required at least the following
5524 conditions: (a) there was another message waiting for the same server;
5525 (b) the server returned 5xx to all RCPT commands in the first message so
5526 that the message was not completed; (c) the server dropped the connection
5527 or gave a negative response to the RSET that Exim sends to abort the
5528 transaction. The observed case was a dropped connection after DATA that had
5529 been sent in pipelining mode. That is, the server had advertised PIPELINING
5530 but was not implementing it correctly. The effect of the bug was incorrect
5531 behaviour, such as trying another host, and this could lead to a crash.
5537 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5538 and daemon.c were passed as pointers to ints; they should have been
5539 pointers to socklen_t variables (which are typically unsigned ints).
5541 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5544 3. Fixed a really odd bug that affected only the testing scheme; patching a
5545 certain fixed string in the binary changed the value of another string that
5546 happened to be identical to the end of the original first string.
5548 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5549 name", it returns that address as the IP address. On some operating
5550 systems (e.g. Solaris), it also passes back the IP address string as the
5551 "host name". However, on others (e.g. Linux), it passes back an empty
5552 string. Exim wasn't checking for this, and was changing the host name to an
5553 empty string, assuming it had been canonicized.
5555 5. Although rare, it is permitted to have more than one PTR record for a given
5556 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5557 all the names associated with an address, because they do in Solaris.
5558 However, it seems that they do not in Linux for data that comes from the
5559 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5560 I found this out when I moved to a new Linux workstation and tried to run
5561 the Exim test suite.
5563 To get round this problem I have changed the code so that it now does its
5564 own call to the DNS to look up PTR records when searching for a host name.
5565 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5566 addresses that are only in /etc/hosts are still found.
5568 This behaviour is, however, controlled by an option called host_lookup_
5569 order, which defaults to "bydns:byaddr". If people want to use the other
5570 order, or indeed, just use one or the other means of lookup, they can
5571 specify it in this variable.
5573 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5574 some operating systems, this comes back from gethostbyaddr() as an empty
5575 string, and this is what Exim used to test for. However, it seems that in
5576 other systems, "." is yielded. Exim now tests for this case too.
5578 7. The values of check_spool_space and check_log_space are now held internally
5579 as a number of kilobytes instead of an absolute number of bytes. If a
5580 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5581 kilobyte. This means that much larger values can be stored.
5583 8. Exim monitor: an attempt to get the action menu when not actually pointing
5584 at a message produces an empty menu entitled "No message selected". This
5585 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5586 no entries in it ("Shell widget menu has zero width and/or height"). So I
5587 have added a single, blank menu entry in this case.
5589 9. Added ${quote_local_part.
5591 10. MIME decoding is now applied to the contents of Subject: header lines when
5594 11. Now that a reference to $sender_host_address automatically causes a reverse
5595 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5596 host lookup before query-style lookups in lists that might use this
5597 variable. This has therefore been abolished, and the "net-" prefix is no
5598 longer necessary for query-style lookups.
5600 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5601 have been a typo for LFLAGS, so it has been changed.
5603 13. The install script calls Exim with "-C /dev/null" in order to find the
5604 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5605 to be output. Howeve, since Exim outputs its version number before the
5606 error, it didn't break the script. It just looked ugly. I fixed this by
5607 always allowing "-C /dev/null" if the caller is root.
5609 14. Ignore overlarge ACL variable number when reading spool file - insurance
5610 against a later release with more variables having written the file.
5612 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5613 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5614 this, as well as the form without the "IPv6" on the front.
5616 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5617 OS/Makefile-Darwin file.
5619 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5620 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5622 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5623 variable. However, some OS use ints for inodes. I've added cast to long int
5624 to get rid of the compiler warning.
5626 19. I had forgotten to lock out "/../" in configuration file names when
5627 ALT_CONFIG_PREFIX was set.
5629 20. Routers used for verification do not need to specify transports. However,
5630 if such a router generated a host list, and callout was configured, Exim
5631 crashed, because it could not find a port number from the (non-existent)
5632 transport. It now assumes port 25 in this circumstance.
5634 21. Added the -t option to exigrep.
5636 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5637 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5638 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5639 them is set, LOOKUP_LSEARCH is forced.
5641 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5643 24. Added sender and host information to the "rejected by local_scan()" log
5644 line; previously there was no indication of these.
5646 25. Added .include_if_exists.
5648 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5649 Linux. It turns out that not all file systems support this. Apparently some
5650 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5651 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5652 not supported on the file descriptor, is now ignored when Exim is trying to
5653 sync a directory. This applies only to Linux.
5655 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5657 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5658 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5661 29. Added support for maildirsize files from supplied patch (modified a bit).
5663 30. The use of :fail: followed by an empty string could lead Exim to respond to
5664 sender verification failures with (e.g.):
5666 550 Verification failed for <xxx>
5667 550 Sender verify failed
5669 where the first response line was missing the '-' that indicates it is not
5670 the final line of the response.
5672 31. The loop for finding the name of the user that called Exim had a hardwired
5673 limit of 10; it now uses the value of finduser_retries, which is used for
5674 all other user lookups.
5676 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5679 33. Exim was neglecting to zero errno before one call of strtol() when
5680 expanding a string and expecting an integer value. On some systems this
5681 resulted in spurious "integer overflow" errors. Also, it was casting the
5682 result into an int without checking.
5684 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5685 did not work. The code looks as if it has *never* worked, though it appears
5686 to have been documented since at least releast 1.62. I have made it work.
5688 35. The "timeout_DNS" error in retry rules, also documented since at least
5689 1.62, also never worked. As it isn't clear exactly what this means, and
5690 clearly it isn't a major issue, I have abolished the feature by treating it
5691 as "timeout", and writing a warning to the main and panic logs.
5693 36. The display of retry rules for -brt wasn't always showing the error code
5696 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5697 timeout_connect_A, timeout_connect_MX.
5699 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5700 to the empty sender.
5702 39. The daemon was not analysing the content of -oX till after it had closed
5703 stderr and disconnected from the controlling terminal. This meant that any
5704 syntax errors were only noted on the panic log, and the return code from
5705 the command was 0. By re-arranging the code a little, I've made the
5706 decoding happen first, so such errors now appear on stderr, and the return
5707 code is 1. However, the actual setting up of the sockets still happens in
5708 the disconnected process, so errors there are still only recorded on the
5711 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5712 connections (as happens on some IP stacks) was logged at start up time as
5713 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5714 it from "IPv6 and IPv4", which means that two separate sockets are being
5717 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5718 says whether AF_INET or AF_INET6 was passed as an argument.
5720 42. Exiwhat output was messed up when time zones were included in log
5723 43. Exiwhat now gives more information about the daemon's listening ports,
5724 and whether -tls-on-connect was used.
5726 44. The "port" option of the smtp transport is now expanded.
5728 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5729 silently ignored. Now an error message is written to the main and panic
5732 46. There's a new ACL modifier called "logwrite" which writes to a log file
5733 as soon as it is encountered.
5735 47. Added $local_user_uid and $local_user_gid at routing time.
5737 48. Exim crashed when trying to verify a sender address that was being
5740 49. Exim was recognizing only a space character after ".include". It now also
5741 recognizes a tab character.
5743 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5744 extracting the relevant information from the specification. The man page no
5745 longer contains scrambled data for the -d option, and I've added a section
5746 at the front about calling Exim under different names.
5748 51. Added "extra_headers" argument to the "mail" command in filter files.
5750 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5753 53. Installed eximstats 1.29.
5755 54. Added transport_filter_timeout as a generic transport option.
5757 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5758 Cc: header lines. This was required by RFC 822, but it not required by RFC
5761 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5762 incoming messages that did not have them. Now it does so only if the
5763 message originates locally, that is, if there is no associated remote host
5764 address. When Resent- header lines are present, this applies to the Resent-
5765 lines rather than the non-Resent- lines.
5767 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5768 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5770 58. Messages for configuration errors now include the name of the main
5771 configuration file - useful now that there may be more than one file in a
5772 list (.included file names were always shown).
5774 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5775 for those rare installations that do not start the daemon as root or run it
5776 setuid root. I've cut out the call to initgroups() if the daemon is not
5779 60. The Exim user and group can now be bound into the binary as text strings
5780 that are looked up at the start of Exim's processing.
5782 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5784 62. Added $mailstore_basename variable.
5786 63. Installed patch to sieve.c from Michael Haardt.
5788 64. When Exim failed to open the panic log after failing to open the main log,
5789 the original message it was trying to log was written to stderr and debug
5790 output, but if they were not available (the usual case in production), it
5791 was lost. Now it is written to syslog before the two lines that record the
5792 failures to open the logs.
5794 65. Users' Exim filters run in subprocesses under the user's uid. It is
5795 possible for a "deliver" command or an alias in a "personal" command to
5796 provoke an address rewrite. If logging of address rewriting is configured,
5797 this fails because the process is not running as root or exim. There may be
5798 a better way of dealing with this, but for the moment (because 4.30 needs
5799 to be released), I have disabled address rewrite logging when running a
5800 filter in a non-root, non-exim process.
5806 1. The buildconfig auxiliary program wasn't quoting the value set for
5807 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5808 not defined. This bug was masked in 4.22 by the effect that was fixed in
5811 2. Some messages that were rejected after a message id was allocated were
5812 shown as "incomplete" by exigrep. It no longer does this for messages that
5813 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5815 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5816 have allow_domain_literals set, the ID did not get logged in the <= line.
5817 Domain literals are now always recognized in Message-ID: header lines.
5819 4. The first argument for a ${extract expansion item is the key name or field
5820 number. Leading and trailing spaces in this item were not being ignored,
5821 causing some misleading effects.
5823 5. When deliver_drop_privilege was set, single queue runner processes started
5824 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5825 same command in the process it spins off) were not dropping privilege.
5827 6. When the daemon running as "exim" started a queue runner, it always
5828 re-executed Exim in the spun-off process. This is a waste of effort when
5829 deliver_drop_privilege is set. The new process now just calls the
5830 queue-runner function directly.
5836 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5839 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5840 ignored. Though the use of -oP was forcing the writing of a pid file, it
5841 was always written to the default place.
5843 3. If the message "no IP address found for host xxxx" is generated during
5844 incoming verification, it is now followed by identification of the incoming
5845 connection (so you can more easily find what provoked it).
5847 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5849 5. Added some features to "harden" Exim a bit more against certain attacks:
5851 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5852 be put in Local/Makefile. This is like the never_users runtime option,
5853 but it cannot be overridden. The default setting is "root".
5855 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5856 prefix string with which any file named in a -C command line option
5859 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5860 is retained for -C and -D only if the caller of Exim is root. Without
5861 it, the exim user may also use -C and -D and retain privilege.
5863 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5864 command line option is disabled.
5866 6. Macro names set by the -D option must start with an upper case letter, just
5867 like macro names defined in the configuration file.
5869 7. Added "dereference=" facility to LDAP.
5871 8. Two instances of the typo "uknown" in the source files are fixed.
5873 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5874 the Configure-Makefile script screwed up while processing it.
5876 10. Incorporated PCRE 4.4.
5878 11. The SMTP synchronization check was not operating right at the start of an
5879 SMTP session. For example, it could not catch a HELO sent before the client
5880 waited for the greeting. There is now a check for outstanding input at the
5881 point when the greeting is written. Because of the duplex, asynchronous
5882 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5883 way, but not yet received, when the check is performed.
5885 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5886 on TCP/IP sockets, because this apparently causes some broken clients to
5889 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5890 unchanged) from the Cygwin maintainer.
5892 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5893 was supported instead of testing for mbx. Effectively a typo.
5895 15. The spa authenticator server code was not checking that the input it
5896 received was valid base64.
5898 16. The debug output line for the "set" modifier in ACLs was not showing the
5899 name of the variable that was being set.
5901 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5903 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5904 lookup on its own. Something else was needed to trigger the lookup. For
5905 example, a match in host_lookup or the need for a host name in a host list.
5906 Now, if $sender_host_name is referenced and the host name has not yet been
5907 looked up, a lookup is performed. If the lookup fails, the variable remains
5908 empty, and $host_lookup_failed is set to "1".
5910 19. Added "eqi" as a case-independent comparison operator.
5912 20. The saslauthd authentication condition could segfault if neither service
5913 nor realm was specified.
5915 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5916 error message that was logged was misleading, and incoming SMTP
5917 connections were dropped. The message is now more accurate, and temporary
5918 errors are given to SMTP connections.
5920 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5921 (except RSET) with a 421 error, until QUIT is received. However, it was
5922 failing to send a response to QUIT.
5924 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5925 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5926 ACL is now run after the helo_try_verify_hosts code.
5928 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5929 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5930 case-independent, so other case variants are also recognized). Apparently
5931 some systems use these upper case variants.
5933 25. If more than two messages were waiting for the same host, and a transport
5934 filter was specified for the transport, Exim sent two messages over the
5935 same TCP/IP connection, and then failed with "socket operation on non-
5936 socket" when it tried to send the third.
5938 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5940 27. The extern definition of crypt16() in expand.c was not being excluded when
5941 the OS had its own crypt16() function.
5943 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5944 as a preferred synonym for return_size_limit, both as an option and as an
5947 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5949 30. Changed the default configuration ACL to relax the local part checking rule
5950 for addresses that are not in any local domains. For these addresses,
5951 slashes and pipe symbols are allowed within local parts, but the sequence
5952 /../ is explicitly forbidden.
5954 31. SPA server authentication was not clearing the challenge buffer before
5957 32. log_message in a "warn" ACL statement was writing to the reject log as
5958 well as to the main log, which contradicts the documentation and doesn't
5959 seem right (because no rejection is happening). So I have stopped it.
5961 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5962 However, I am unable to do any testing of this.
5964 34. Fixed an infelicity in the appendfile transport. When checking directories
5965 for a mailbox, to see if any needed to be created, it was accidentally
5966 using path names with one or more superfluous leading slashes; tracing
5967 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5969 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5970 contents are added to the log line that is written for every discarded
5971 recipient. (Previously a log_message setting was ignored.)
5973 36. The ${quote: operator now quotes the string if it is empty.
5975 37. The install script runs exim in order to find its version number. If for
5976 some reason other than non-existence or emptiness, which it checks, it
5977 could not run './exim', it was installing it with an empty version number,
5978 i.e. as "exim-". This error state is now caught, and the installation is
5981 38. An argument was missing from the function that creates an error message
5982 when Exim fails to connect to the socket for saslauthd authentication.
5983 This could cause Exim to crash, or give a corrupted message.
5985 39. Added isip, isip4, and isip6 to ${if conditions.
5987 40. The ACL variables $acl_xx are now saved with the message, and can be
5988 accessed later in routers, transports, and filters.
5990 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5991 strings in the file are not string-expanded.
5993 42. If a MAIL command specified a SIZE value that was too large to fit into an
5994 int variable, the check against message_size_limit failed. Such values are
5995 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5996 day this will have to be increased, but I don't think I want to be around
5997 when emails are that large.
6004 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6005 iconv() is not standard in FreeBSD.
6007 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6008 IPv6 enabled. The observed symptom was a segmentation fault on return from
6009 the function os_common_find_running_interfaces() in src/os.c.
6011 3. In the check_special_case() function in daemon.c I had used "errno" as an
6012 argument name, which causes warnings on some systems. This was basically a
6013 typo, since it was named "eno" in the comments!
6015 4. The code that waits for the clock to tick (at a resolution of some fraction
6016 of a second) so as to ensure message-id uniqueness was always waiting for
6017 at least one whole tick, when it could have waited for less. [This is
6018 almost certainly not relevant at current processor speeds, where it is
6019 unlikely to ever wait at all. But we try to future-proof.]
6021 5. The function that sleeps for a time interval that includes fractions of a
6022 second contained a race. It did not block SIGALRM between setting the
6023 timer, and suspending (a couple of lines later). If the interval was short
6024 and the sigsuspend() was delayed until after it had expired, the suspension
6025 never ended. On busy systems this could lead to processes getting stuck for
6028 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6029 process, before it forks any delivery processes. The open lookup caching
6030 mechanism meant that the open file or database connection was passed into
6031 the delivery process. The problem was that delivery processes always tidy
6032 up cached lookup data. This could cause a problem for the next delivery
6033 process started by the queue runner, because the external queue runner
6034 process does not know about the closure. So the next delivery process
6035 still has data in the lookup cache. In the case of a file lookup, there was
6036 no problem because closing a file descriptor in a subprocess doesn't affect
6037 the parent. However, if the lookup was caching a connection to a database,
6038 the connection was closed, and the second delivery process was likely to
6039 see errors such as "PGSQL: query failed: server closed the connection
6040 unexpectedly". The problem has been fixed by closing all cached lookups
6041 in a queue runner before running a delivery process.
6043 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6044 seem to have the "const" qualifier which it has on other OS. I've
6047 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6048 *of the same type* (client or server) with the same public name that an
6049 error should be diagnosed.
6051 9. When Exim looked up a host name for an IP address, but failed to find the
6052 original IP address when looking up the host name (a safety check), it
6053 output the message "<ip address> does not match any IP for NULL", which was
6054 confusing, to say the least. The bug was that the host name should have
6055 appeared instead of "NULL".
6057 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6058 user that is built into the binary, and the -C or -D options is used, root
6059 privilege is dropped before the configuration file is read. In addition,
6060 logging is switched to stderr instead of the normal log files. If the
6061 configuration then re-defines the Exim user, the unprivileged environment
6062 is probably not what is expected, so Exim logs a panic warning message (but
6065 However, if deliver_drop_privilege is set, the unprivileged state may well
6066 be exactly what is intended, so the warning has been cut out in that case,
6067 and Exim is allowed to try to write to its normal log files.
6073 1. smtp_return_error_details was not giving details for temporary sender
6074 or receiver verification errors.
6076 2. Diagnose a configuration error if two authenticators have the same public
6079 3. Exim used not to create the message log file for a message until the first
6080 delivery attempt. This could be confusing when incoming messages were held
6081 for policy or load reasons. The message log file is now created at the time
6082 the message is received, and an initial "Received" line is written to it.
6084 4. The automatically generated man page for command line options had a minor
6085 bug that caused no ill effects; however, a more serious problem was that
6086 the procedure for building the man page automatically didn't always
6087 operate. Consequently, release 4.20 contains an out-of-date version. This
6088 shouldn't happen again.
6090 5. When building Exim with embedded Perl support, the script that builds the
6091 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6092 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6094 6. The freeze_tell option was not being used for messages that were frozen on
6095 arrival, either by an ACL or by local_scan().
6097 7. Added the smtp_incomplete_transaction log selector.
6099 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6100 was accepting AUTH without a new EHLO.
6102 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6103 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6106 10. Logging of TCP/IP connections (when configured) now happens in the main
6107 daemon process instead of the child process, so that the TCP/IP connection
6108 count is more accurate (but it can never be perfect).
6110 11. The use of "drop" in a nested ACL was not being handled correctly in the
6111 outer ACL. Now, if condition failure induced by the nested "drop" causes
6112 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6113 or "require"), the connection is dropped.
6115 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6116 that yield "discard" can now be used with an "accept" or a "discard" verb,
6117 but an error is generated for any others (because I can't see a useful way
6118 to define what should happen).
6120 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6121 are now processed in the same way as lines in the Exim configuration file.
6122 In particular, continuation lines are supported.
6124 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6126 15. Added -ti meaning -t -i.
6128 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6129 domains, and warn by logging if others are found.
6131 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6132 returned from GIFCONF when Exim is trying to find the list of interfaces on
6133 a host. The code in os.c has been modified to copy each ifreq to an aligned
6134 structure in all cases.
6136 Also, in some cases, the returned ifreq's were being copied to a 'struct
6137 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6138 means the last couple of bytes of an IPv6 address could be chopped if the
6139 ifreq contained only a normal sockaddr (14 bytes storage).
6141 18. Named domain lists were not supported in the hosts_treat_as_local option.
6142 An entry such as +xxxx was not recognized, and was treated as a literal
6145 19. Ensure that header lines added by a DATA ACL are included in the reject log
6146 if the ACL subsequently rejects the message.
6148 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6149 MD5 (which is deprecated).
6151 21. When testing a filter file using -bf, Exim was writing a message when it
6152 took the sender from a "From " line in the message, but it was not doing so
6153 when it took $return_path from a Return-Path: header line. It now does.
6155 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6156 with a valid header line field name (a series of printing characters
6157 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6159 23. Changed "disc" in the source to "disk" to conform to the documentation and
6160 the book and for uniformity.
6162 24. Ignore Sendmail's -Ooption=value command line item.
6164 25. When execve() failed while trying to run a command in a pipe transport,
6165 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6166 could be confused with a return value of 69 from the command itself. This
6167 has been changed to 127, the value the shell returns if it is asked to run
6168 a non-existent command. The wording for the related log line suggests a
6169 non-existent command as the problem.
6171 26. If received_header_text expands to an empty string, do not add a Received:
6172 header line to the message. (Well, it adds a token one on the spool, but
6173 marks it "old" so that it doesn't get used or transmitted.)
6175 27. Installed eximstats 1.28 (addition of -nt option).
6177 28. There was no check for failure on the call to getsockname() in the daemon
6178 code. This can fail if there is a shortage of resources on the system, with
6179 ENOMEM, for example. A temporary error is now given on failure.
6181 29. Contrary to the C standard, it seems that in some environments, the
6182 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6183 program. Exim now does this explicitly; it affects the formatting of
6184 timestamps using strftime().
6186 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6187 complaints. I've now initialized all the variables, to avoid this.
6189 32. Header lines added by a system filter were not being "seen" during
6190 transport-time rewrites.
6192 33. The info_callback() function passed to OpenSSL is set up with type void
6193 (*)(SSL *, int, int), as described somewhere. However, when calling the
6194 function (actually a macro) that sets it up, the type void(*)() is
6195 expected. I've put in a cast to prevent warnings from picky compilers.
6197 34. If a DNS black list lookup found a CNAME record, but there were no A
6198 records associated with the domain it pointed at, Exim crashed.
6200 35. If a DNS black list lookup returned more than one A record, Exim ignored
6201 all but the first. It now scans all returned addresses if a particular IP
6202 value is being sought. In this situation, the contents of the
6203 $dnslist_value variable are a list of all the addresses, separated by a
6206 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6207 to accept a host name and all its aliases if the forward lookup for any of
6208 them yielded the IP address of the incoming connection. Now it accepts only
6209 those names whose forward lookup yields the correct IP address. Any other
6210 names are discarded. This closes a loophole whereby a rogue DNS
6211 administrator could create reverse DNS records to break through a
6212 wildcarded host restriction in an ACL.
6214 37. If a user filter or a system filter that ran in a subprocess used any of
6215 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6216 the wrong values were passed to the pipe command ($thisaddress had the
6217 value of $0, $0 had the value of $1, etc). This bug was introduced by
6218 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6220 38. Improved the line breaking for long SMTP error messages from ACLs.
6221 Previously, if there was no break point between 40 and 75 characters, Exim
6222 left the rest of the message alone. Two changes have been made: (a) I've
6223 reduced the minimum length to 35 characters; (b) if it can't find a break
6224 point between 35 and 75 characters, it looks ahead and uses the first one
6225 that it finds. This may give the occasional overlong line, but at least the
6226 remaining text gets split now.
6228 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6229 file descriptors might be low, and that setting 1000 would always raise it.
6230 It turns out that in some environments, the limit is already over 1000 and
6231 that lowering it causes trouble. So now Exim takes care not to decrease it.
6233 40. When delivering a message, the value of $return_path is set to $sender_
6234 address at the start of routing (routers may change the value). By an
6235 oversight, this default was not being set up when an address was tested by
6236 -bt or -bv, which affected the outcome if any router or filter referred to
6239 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6240 log only when "message" or "log_message" are set. However, if one of the
6241 conditions was an address verification, or a call to a nested ACL, the
6242 messages generated by the underlying test were being passed through. This
6243 no longer happens. The underlying message is available in $acl_verify_
6244 message for both "message" and "log_message" expansions, so it can be
6245 passed through if needed.
6247 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6248 new expansion $bh_ to give the encoded byte string without charset
6249 translation. Translation happens only if iconv() is available; HAVE_ICONV
6250 indicates this at build time. HEADERS_CHARSET gives the charset to
6251 translate to; headers_charset can change it in the configuration, and
6252 "headers charset" can change it in an individual filter file.
6254 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6255 that creates RFC 2047 encoded "words" labels them as that charset instead
6256 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6257 expansion operator; (ii) when Exim creates a From: line for a local
6258 message; (iii) when a header line is rewritten to include a "phrase" part.
6260 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6261 buggy, causing it to skip the first lines of messages whose message ID
6262 ended in 'D'. This would not have bitten before Exim release 4.14, because
6263 message IDs were unlikely to end in 'D' before then. The effect was to have
6264 incorrect size information for certain domains.
6266 45. #include "config.h" was missing at the start of the crypt16.c module. This
6267 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6270 46. If there was a timeout during a "random" callout check, Exim treated it as
6271 a failure of the random address, and carried on sending RSET and the real
6272 address. If the delay was just some slowness somewhere, the response to the
6273 original RCPT would be taken as a response to RSET and so on, causing
6274 mayhem of various kinds.
6276 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6277 when I implemented it. It didn't allow for the fact that some option values
6278 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6279 the right test for positive values.
6281 48. Domain names in DNS records are case-independent. Exim always looks them up
6282 in lower case. Some resolvers return domain names in exactly the case they
6283 appear in the zone file, that is, they may contain uppercase letters. Not
6284 all resolvers do this - some return always lower case. Exim was treating a
6285 change of case by a resolver as a change of domain, similar to a widening
6286 of a domain abbreviation. This triggered its re-routing code and so it was
6287 trying to route what was effectively the same domain again. This normally
6288 caused routing to fail (because the router wouldn't handle the domain
6289 twice). Now Exim checks for this case specially, and just changes the
6290 casing of the domain that it ultimately uses when it transmits the message
6293 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6296 50. If a filter generated a file delivery with a non-absolute name (possible if
6297 no home directory exists for the router), the forbid_file option was not
6300 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6301 the existing equality matching.
6303 52. Exim was using ints instead of ino_t variables in some places where it was
6304 dealing with inode numbers.
6306 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6307 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6308 called TMPDIR, and if it finds it is different, it changes its value.
6310 54. The smtp_printf() function is now made available to local_scan() so
6311 additional output lines can be written before returning. There is also an
6312 smtp_fflush() function to enable the detection of a dropped connection.
6313 The variables smtp_input and smtp_batched_input are exported to
6316 55. Changed the default runtime configuration: the message "Unknown user"
6317 has been removed from the ACL, and instead placed on the localuser router,
6318 using the cannot_route_message feature. This means that any verification
6319 failures that generate their own messages won't get overridden. Similarly,
6320 the "Unrouteable address" message that was in the ACL for unverifiable
6321 relay addresses has also been removed.
6323 56. Added hosts_avoid_esmtp to the smtp transport.
6325 57. The exicyclog script was not checking for the esoteric option
6326 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6327 will work only if exicyclog is run under the appropriate euid.
6329 58. Following a discussion on the list, the rules by which Exim recognises line
6330 endings on incoming messages have been changed. The -dropcr and drop_cr
6331 options are now no-ops, retained only for backwards compatibility. The
6332 following line terminators are recognized: LF CRLF CR. However, special
6333 processing applies to CR:
6335 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6336 nor a local message in the state where . is a terminator.
6338 (ii) If a bare CR is encountered in a header line, an extra space is added
6339 after the line terminator so as not to end the header. The reasoning
6340 behind this is that bare CRs in header lines are most likely either
6341 to be mistakes, or people trying to play silly games.
6343 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6344 was being incorrectly given as 18 bytes larger than it should have been.
6345 This is a VOB (very old bug).
6347 60. This may never have affected anything current, but just in case it has:
6348 When the local host is found other than at the start of a list of hosts,
6349 the local host, those with the same MX, and any that follow, are discarded.
6350 When the list in question was part of a longer list of hosts, the following
6351 hosts (not currently being processed) were also being discarded. This no
6352 longer happens. I'm not sure if this situation could ever has previously
6355 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6358 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6359 header. This is something that is recommended in a new Internet Draft, and
6360 is something that is documented as being done by Sendmail. There are two
6361 possible values. For messages generated by the autoreply transport, Exim
6364 Auto-Submitted: auto-replied
6366 whereas for all other generated messages (e.g. bounces) it adds
6368 Auto-Submitted: auto-generated
6370 63. The "personal" condition in filters now includes a test for the
6371 Auto-Submitted: header. If it contains the string "auto-" the message it
6372 not considered personal.
6374 64. Added rcpt_include_affixes as a generic transport option.
6376 65. Added queue_only_override (default true).
6378 66. Added the syslog_duplication option.
6380 67. If what should have been the first header line of a message consisted of
6381 a space followed by a colon, Exim was mis-interpreting it as a header line.
6382 It isn't of course - it is syntactically invalid and should therefore be
6383 treated as the start of the message body. The misbehaviour could have
6384 caused a number of strange effects, including loss of data in subsequent
6385 header lines, and spool format errors.
6387 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6388 client host had authenticated. This control can now be exercised by an ACL
6389 for more flexibility.
6391 69. By default, callouts do not happen when testing with -bh. There is now a
6392 variant, -bhc, which does actually run the callout code, including
6393 consulting and updating the callout cache.
6395 70. Added support for saslauthd authentication, courtesy of Alexander
6398 71. If statvfs() failed on the spool or log directories while checking their
6399 size for availability, Exim confusingly gave the error "space shortage".
6400 Furthermore, in debugging mode it crashed with a floating point exception.
6401 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6402 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6403 serious problem, Exim now writes to the main and panic logs when this
6404 happens, with details of the failure. It then refuses to accept the
6405 incoming message, giving the message "spool directory problem" or "log
6406 directory problem" with a 421 code for SMTP messages.
6408 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6409 0, 1, and 2 exist, because some OS complain for execs without them (see
6410 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6411 descriptors. However, the code omitted to check that the open succeeded,
6412 causing mysterious errors if for some reason the permissions on /dev/null
6413 got screwed. Now Exim writes a message to the main and panic logs, and
6414 bombs out if it can't open /dev/null.
6416 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6417 interact so that it is all more flexible. It is supposed to remain
6418 backwards compatible. Also added extra_local_interfaces.
6420 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6421 to bomb out with an assertion failure - to the client this appears as a
6422 connection drop. This problem occurs in the part of the code that was taken
6423 from the Samba project. Fortunately, the assertion is in a very simple
6424 function, so I have fixed this by reproducing the function inline in the
6425 one place where it is called, and arranging for authentication to fail
6426 instead of killing the process with assert().
6428 75. The SPA client code was not working when the server requested OEM rather
6429 than Unicode encoding.
6431 76. Added code to make require_files with a specific uid setting more usable in
6432 the case where statting the file as root fails - usually a non-root-mounted
6433 NFS file system. When this happens and the failure is EACCES, Exim now
6434 forks a subprocess and does the per-uid checking as the relevant uid.
6436 77. Added process_log_path.
6438 78. If log_file_path was not explicitly set, a setting of check_log_space or
6439 check_log_inodes was ignored.
6441 79. If a space check for the spool or log partitions fails, the incident is now
6442 logged. Of course, in the latter case the data may get lost...
6444 80. Added the %p formatting code to string_format() so that it can be used to
6445 print addresses in debug_print(). Adjusted all the address printing in the
6446 debugging in store.c to use %p rather than %d.
6448 81. There was a concern that a line of code in smtp_in.c could overflow a
6449 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6450 initially expressed, the concern was not well-founded, because trailing
6451 spaces are removed early. However, if the trailing spaces were followed by
6452 a NULL, they did not get removed, so the overflow was possible. Two fixes
6455 (a) I re-wrote the offending code in a cleaner fashion.
6456 (b) If an incoming SMTP command contains a NULL character, it is rejected
6459 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6460 runs initgroups(), so that if the Exim user is in any additional groups,
6461 they will be used during message reception.
6467 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob