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.
162 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
165 JH/02 The smtp transport option "multi_domain" is now expanded.
167 JH/03 The smtp transport now requests PRDR by default, if the server offers
170 JH/04 Certificate name checking on server certificates, when exim is a client,
171 is now done by default. The transport option tls_verify_cert_hostnames
172 can be used to disable this per-host. The build option
173 EXPERIMENTAL_CERTNAMES is withdrawn.
175 JH/05 The value of the tls_verify_certificates smtp transport and main options
176 default to the word "system" to access the system default CA bundle.
177 For GnuTLS, only version 3.0.20 or later.
179 JH/06 Verification of the server certificate for a TLS connection is now tried
180 (but not required) by default. The verification status is now logged by
181 default, for both outbound TLS and client-certificate supplying inbound
184 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
187 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
188 Status Notification (bounce) messages are now MIME format per RFC 3464.
189 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
190 under the control of the dsn_advertise_hosts option, and routers may
191 have a dsn_lasthop option.
193 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
194 default, modifiable by a malware= option. The list separator for
195 the options can now be changed in the usual way. Bug 68.
197 JH/10 The smtp_receive_timeout main option is now expanded before use.
199 JH/11 The incoming_interface log option now also enables logging of the
200 local interface on delivery outgoing connections.
202 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
203 if the interface and destination host and port all match.
205 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
208 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
209 Patch from Andrew Lewis.
211 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
212 now supports optional time-restrictions, weighting, and priority
213 modifiers per server. Patch originally by <rommer@active.by>.
215 JH/16 The spamd_address main option now supports a mixed list of local
216 and remote servers. Remote servers can be IPv6 addresses, and
217 specify a port-range.
219 JH/17 Bug 68: The spamd_address main option now supports an optional
220 timeout value per server.
222 JH/18 Bug 1581: Router and transport options headers_add/remove can
223 now have the list separator specified.
225 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
228 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
231 JH/21 Support for the A6 type of dns record is withdrawn.
233 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
234 rather than the verbs used.
236 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
237 from 255 to 1024 chars.
239 JH/24 Verification callouts now attempt to use TLS by default.
241 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
242 are generic router options now. The defaults didn't change.
244 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
245 Original patch from Alexander Shikoff, worked over by JH.
247 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
248 files if ZCAT_COMMAND is not executable.
250 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
252 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
254 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
255 Normally benign, it bites when the pair was led to by a CNAME;
256 modern usage is to not canoicalize the domain to a CNAME target
257 (and we were inconsistent anyway for A-only vs AAAA+A).
259 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
261 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
262 when evaluating $sender_host_dnssec.
264 JH/31 Check the HELO verification lookup for DNSSEC, adding new
265 $sender_helo_dnssec variable.
267 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
269 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
271 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
273 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
274 documented as working, but never had. Support all but $spam_report.
276 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
277 added for tls authenticator.
282 TL/01 When running the test suite, the README says that variables such as
283 no_msglog_check are global and can be placed anywhere in a specific
284 test's script, however it was observed that placement needed to be near
285 the beginning for it to behave that way. Changed the runtest perl
286 script to read through the entire script once to detect and set these
287 variables, reset to the beginning of the script, and then run through
288 the script parsing/test process like normal.
290 TL/02 The BSD's have an arc4random API. One of the functions to induce
291 adding randomness was arc4random_stir(), but it has been removed in
292 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
293 function when detected.
295 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
296 cause callback expansion.
298 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
299 syntax errors in an expansion can be treated as a string instead of
300 logging or causing an error, due to the internal use of bool_lax
301 instead of bool when processing it.
303 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
304 server certificates when making smtp deliveries.
306 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
308 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
310 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
312 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
313 Merged patch from Sebastian Wiedenroth.
315 JH/05 Fix results-pipe from transport process. Several recipients, combined
316 with certificate use, exposed issues where response data items split
317 over buffer boundaries were not parsed properly. This eventually
318 resulted in duplicates being sent. This issue only became common enough
319 to notice due to the introduction of conection certificate information,
320 the item size being so much larger. Found and fixed by Wolfgang Breyha.
322 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
323 size buffer was used, resulting in syntax errors when an expansion
326 JH/07 Add support for directories of certificates when compiled with a GnuTLS
327 version 3.3.6 or later.
329 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
330 is EXPERIMENTAL_EVENT, the main-configuration and transport options
331 both become "event_action", the variables become $event_name, $event_data
332 and $event_defer_errno. There is a new variable $verify_mode, usable in
333 routers, transports and related events. The tls:cert event is now also
334 raised for inbound connections, if the main configuration event_action
337 TL/06 In test suite, disable OCSP for old versions of openssl which contained
338 early OCSP support, but no stapling (appears to be less than 1.0.0).
340 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
341 server certificate names available under the smtp transport option
342 "tls_verify_cert_hostname" now do not permit multi-component wildcard
345 JH/10 Time-related extraction expansions from certificates now use the main
346 option "timezone" setting for output formatting, and are consistent
347 between OpenSSL and GnuTLS compilations. Bug 1541.
349 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
350 encoded parameter in the incoming message. Bug 1558.
352 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
353 include certificate info, eximon was claiming there were spoolfile
356 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
358 JH/14 Log delivery-related information more consistently, using the sequence
359 "H=<name> [<ip>]" wherever possible.
361 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
362 are problematic for Debian distribution, omit them from the release
365 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
367 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
370 JH/17 Fix a null-indirection in certextract expansions when a nondefault
371 output list separator was used.
376 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
377 checkers that were complaining about end of non-void function with no
380 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
381 This was a regression intruduced in 4.83 by another bugfix.
383 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
385 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
386 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
392 TF/01 Correctly close the server side of TLS when forking for delivery.
394 When a message was received over SMTP with TLS, Exim failed to clear up
395 the incoming connection properly after forking off the child process to
396 deliver the message. In some situations the subsequent outgoing
397 delivery connection happened to have the same fd number as the incoming
398 connection previously had. Exim would try to use TLS and fail, logging
399 a "Bad file descriptor" error.
401 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
402 utilities have not been installed.
404 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
405 temporary space as the ACL may create new global variables.
407 TL/01 LDAP support uses per connection or global context settings, depending
408 upon the detected version of the libraries at build time.
410 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
411 to extract and use the src ip:port in logging and expansions as if it
412 were a direct connection from the outside internet. PPv2 support was
413 updated based on HAProxy spec change in May 2014.
415 JH/02 Add ${listextract {number}{list}{success}{fail}}.
417 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
418 Properly escape header and check for NULL return.
420 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
423 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
425 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
426 characters in header names, implemented as a verify condition.
427 Contributed by Michael Fischer v. Mollard.
429 TL/05 Rename SPF condition results err_perm and err_temp to standardized
430 results permerror and temperror. Previous values are deprecated but
431 still accepted. In a future release, err_perm and err_temp will be
432 completely removed, which will be a backward incompatibility if the
433 ACL tests for either of these two old results. Patch contributed by
434 user bes-internal on the mailing list.
436 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
438 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
439 selectors, in both main and reject logs.
441 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
444 JH/07 Add malware type "sock" for talking to simple daemon.
446 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
448 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
449 routers/transports under cutthrough routing.
451 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
452 numbers. Touch up "bool" conditional to keep the same definition.
454 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
456 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
458 JH/12 Expand items in router/transport headers_add or headers_remove lists
459 individually rather than the list as a whole. Bug 1452.
461 Required for reasonable handling of multiple headers_ options when
462 they may be empty; requires that headers_remove items with embedded
463 colons must have them doubled (or the list-separator changed).
465 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
466 view the policy declared in the DMARC record. Currently, $dmarc_status
467 is a combined value of both the record presence and the result of the
470 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
472 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
473 dnslookup router and the smtp transport (applying to the forward
476 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
477 of ldap servers used for a specific lookup. Patch provided by Heiko
480 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
481 New variable $lookup_dnssec_authenticated for observability.
483 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
484 Patch submitted by Lars Timman.
486 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
488 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
489 Requires trusted mode and valid format message id, aborts otherwise.
490 Patch contributed by Heiko Schlichting.
492 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
493 certextract with support for various fields. Bug 1358.
495 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
496 is requested by default, modifiable by smtp transport option
499 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
500 operate on certificate variables to give certificate fingerprints
501 Also new ${sha256:cert_variable}.
503 JH/23 The PRDR feature is moved from being Experimental into the mainline.
505 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
506 Christian Aistleitner.
508 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
510 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
511 file. Patch from Wolfgang Breyha.
513 JH/25 Expand the coverage of the delivery $host and $host_address to
514 client authenticators run in verify callout. Bug 1476.
516 JH/26 Port service names are now accepted for tls_on_connect_ports, to
517 align with daemon_smtp_ports. Bug 72.
519 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
520 support and error reporting did not work properly.
522 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
523 and is readable. Patch from Andrew Colin Kissa.
525 TL/14 Enhance documentation of ${run expansion and how it parses the
526 commandline after expansion, particularly in the case when an
527 unquoted variable expansion results in an empty value.
529 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
531 PP/02 Fix internal collision of T_APL on systems which support RFC3123
532 by renaming away from it. Addresses GH issue 15, reported by
535 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
537 TL/15 SECURITY: prevent double expansion in math comparison functions
538 (can expand unsanitized data). Not remotely exploitable.
545 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
547 PP/02 Make -n do something, by making it not do something.
548 When combined with -bP, the name of an option is not output.
550 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
553 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
554 $sender_host_name and config options to manage this, and basic check
557 PP/05 DSCP support for outbound connections and control modifier for inbound.
559 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
560 (Only plugin which currently uses this is kerberos4, which nobody should
561 be using, but we should make it available and other future plugins might
562 conceivably use it, even though it would break NAT; stuff *should* be
563 using channel bindings instead).
565 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
566 name; added for Sendmail compatibility; requires admin caller.
567 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
568 just ignore it); requires trusted caller.
569 Also parse but ignore: -Ac -Am -X<logfile>
572 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
574 TL/02 Add +smtp_confirmation as a default logging option.
576 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
577 Patch by Magnus Holmgren from 2007-02-20.
579 TL/04 Bugzilla 1281 - Spec typo.
580 Bugzilla 1283 - Spec typo.
581 Bugzilla 1290 - Spec grammar fixes.
583 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
585 TL/06 Add Experimental DMARC support using libopendmarc libraries.
587 TL/07 Fix an out of order global option causing a segfault. Reported to dev
588 mailing list by by Dmitry Isaikin.
590 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
592 JH/02 Support "G" suffix to numbers in ${if comparisons.
594 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
596 NM/01 Bugzilla 1197 - Spec typo
597 Bugzilla 1196 - Spec examples corrections
599 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
601 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
602 gnutls_enable_pkcs11, but renamed to more accurately indicate its
605 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
606 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
608 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
609 "acl {{name}{arg}...}", and optional args on acl condition
612 JH/05 Permit multiple router/transport headers_add/remove lines.
614 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
616 JH/07 Avoid using a waiting database for a single-message-only transport.
617 Performance patch from Paul Fisher. Bugzilla 1262.
619 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
622 JH/09 Add $headers_added variable, with content from use of ACL modifier
623 add_header (but not yet added to the message). Bugzilla 199.
625 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
626 Pulled from Bugzilla 817 by Wolfgang Breyha.
628 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
630 (nb: this is the same fix as in Exim 4.80.1)
632 JH/11 Add A= logging on delivery lines, and a client_set_id option on
635 JH/12 Add optional authenticated_sender logging to A= and a log_selector
638 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
640 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
641 advertise SMTP AUTH mechanism to us, instead of a generic
642 protocol violation error. Also, make Exim more robust to bad
643 data from the Dovecot auth socket.
645 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
647 When a queue runner is handling a message, Exim first routes the
648 recipient addresses, during which it prunes them based on the retry
649 hints database. After that it attempts to deliver the message to
650 any remaining recipients. It then updates the hints database using
653 So if a recipient address works intermittently, it can get repeatedly
654 deferred at routing time. The retry hints record remains fresh so the
655 address never reaches the final cutoff time.
657 This is a fairly common occurrence when a user is bumping up against
658 their storage quota. Exim had some logic in its local delivery code
659 to deal with this. However it did not apply to per-recipient defers
660 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
662 This change adds a proper retry rule check during routing so that the
663 final cutoff time is checked against the message's age. We only do
664 this check if there is an address retry record and there is not a
665 domain retry record; this implies that previous attempts to handle
666 the address had the retry_use_local_parts option turned on. We use
667 this as an approximation for the destination being like a local
668 delivery, as in LMTP.
670 I suspect this new check makes the old local delivery cutoff check
671 redundant, but I have not verified this so I left the code in place.
673 TF/02 Correct gecos expansion when From: is a prefix of the username.
675 Test 0254 submits a message to Exim with the header
679 When I ran the test suite under the user fanf2, Exim expanded
680 the header to contain my full name, whereas it should have added
681 a Resent-Sender: header. It erroneously treats any prefix of the
682 username as equal to the username.
684 This change corrects that bug.
686 GF/01 DCC debug and logging tidyup
687 Error conditions log to paniclog rather than rejectlog.
688 Debug lines prefixed by "DCC: " to remove any ambiguity.
690 TF/03 Avoid unnecessary rebuilds of lookup-related code.
692 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
693 Bug spotted by Jeremy Harris; was flawed since initial commit.
694 Would have resulted in OCSP responses post-SNI triggering an Exim
695 NULL dereference and crash.
697 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
699 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
700 Bug detection, analysis and fix by Samuel Thibault.
701 Bugzilla 1331, Debian bug #698092.
703 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
705 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
706 Server implementation by Todd Lyons, client by JH.
707 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
708 config variable "prdr_enable" controls whether the server
709 advertises the facility. If the client requests PRDR a new
710 acl_data_smtp_prdr ACL is called once for each recipient, after
711 the body content is received and before the acl_smtp_data ACL.
712 The client is controlled by bolth of: a hosts_try_prdr option
713 on the smtp transport, and the server advertisement.
714 Default client logging of deliveries and rejections involving
715 PRDR are flagged with the string "PRDR".
717 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
718 fclose(). Diagnosis by Todd Lyons.
720 PP/17 Update configure.default to handle IPv6 localhost better.
721 Patch by Alain Williams (plus minor tweaks).
724 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
725 This is now consistent with GnuTLS, and is now documented: the
726 previous undocumented portable approach to treating the option as
727 unset was to force an expansion failure. That still works, and
728 an empty string is now equivalent.
730 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
731 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
732 not performing validation itself.
734 PP/20 Added force_command boolean option to pipe transport.
735 Patch from Nick Koston, of cPanel Inc.
737 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
740 TF/04 Added udpsend ACL modifer and hexquote expansion operator
742 PP/21 Fix eximon continuous updating with timestamped log-files.
743 Broken in a format-string cleanup in 4.80, missed when I repaired the
744 other false fix of the same issue.
745 Report and fix from Heiko Schlichting.
748 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
749 Report from Prashanth Katuri.
751 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
752 It's SecureTransport, so affects any MacOS clients which use the
753 system-integrated TLS libraries, including email clients.
755 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
756 using a MIME ACL for non-SMTP local injection.
757 Report and assistance in diagnosis by Warren Baker.
759 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
761 JH/16 Fix comparisons for 64b. Bugzilla 1385.
763 TL/09 Add expansion variable $authenticated_fail_id to keep track of
764 last id that failed so it may be referenced in subsequent ACL's.
766 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
769 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
770 ldap library initialization, allowing self-signed CA's to be
771 used. Also properly sets require_cert option later in code by
772 using NULL (global ldap config) instead of ldap handle (per
773 session). Bug diagnosis and testing by alxgomz.
775 TL/12 Enhanced documentation in the ratelimit.pl script provided in
776 the src/util/ subdirectory.
778 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
779 renamed to Transport Post Delivery Action by Jeremy Harris, as
782 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
783 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
784 redis_servers = needs to be configured which will be used by the redis
785 lookup. Patch from Warren Baker, of The Packet Hub.
787 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
789 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
790 hostname or reverse DNS when processing a host list. Used suggestions
791 from multiple comments on this bug.
793 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
795 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
796 Missed a few lines, added it to make the runtest require no keyboard
799 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
800 contains upper case chars. Make router use caseful_local_part.
802 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
803 support when GnuTLS has been built with p11-kit.
809 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
811 This, or similar/improved, will also be change PP/11 of 4.82.
817 PP/01 Handle short writes when writing local log-files.
818 In practice, only affects FreeBSD (8 onwards).
819 Bugzilla 1053, with thanks to Dmitry Isaikin.
821 NM/01 Bugzilla 949 - Documentation tweak
823 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
826 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
828 PP/02 Implemented gsasl authenticator.
830 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
832 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
833 `pkg-config foo` for cflags/libs.
835 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
836 with rest of GSASL and with heimdal_gssapi.
838 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
839 `pkg-config foo` for cflags/libs for the TLS implementation.
841 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
842 properties get this fed in as external SSF. A number of robustness
843 and debugging improvements to the cyrus_sasl authenticator.
845 PP/08 cyrus_sasl server now expands the server_realm option.
847 PP/09 Bugzilla 1214 - Log authentication information in reject log.
848 Patch by Jeremy Harris.
850 PP/10 Added dbmjz lookup type.
852 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
854 PP/12 MAIL args handles TAB as well as SP, for better interop with
855 non-compliant senders.
856 Analysis and variant patch by Todd Lyons.
858 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
859 Bug report from Lars Müller <lars@samba.org> (via SUSE),
860 Patch from Dirk Mueller <dmueller@suse.com>
862 PP/13 tls_peerdn now print-escaped for spool files.
863 Observed some $tls_peerdn in wild which contained \n, which resulted
864 in spool file corruption.
866 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
867 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
868 or write after TLS renegotiation, which otherwise led to messages
871 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
872 as a tracking header (ie: a signed header comes before the signature).
873 Patch from Wolfgang Breyha.
875 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
876 comma-sep list; embedded commas doubled.
878 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
880 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
882 Report and patch from Dmitry Banschikov.
884 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
885 Removed SSL_clear() after SSL_new() which led to protocol negotiation
886 failures. We appear to now support TLS1.1+ with Exim.
888 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
889 lets Exim select keys and certificates based upon TLS SNI from client.
890 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
891 before an outbound SMTP session. New log_selector, +tls_sni.
893 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
894 NULL dereference. Report and patch from Alun Jones.
896 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
897 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
898 Not seeing resolver debug output on NetBSD, but suspect this is a
899 resolver implementation change.
901 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
902 Left warnings. Added "eximon gdb" invocation mode.
904 PP/21 Defaulting "accept_8bitmime" to true, not false.
906 PP/22 Added -bw for inetd wait mode support.
908 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
909 locate the relevant includes and libraries. Made this the default.
911 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
912 Bugzilla 1246, report and most of solution from Tomasz Kusy.
914 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
915 This may cause build issues on older platforms.
917 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
918 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
919 gnutls_require_mac & gnutls_require_protocols (no longer supported).
920 Added SNI support via GnuTLS too.
921 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
923 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
925 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
926 Applied second patch from Janne, implementing suggestion to default
927 multiple-strings-in-record handling to match SPF spec.
929 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
931 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
932 read-only, out of scope).
933 Patch from Wolfgang Breyha, report from Stuart Northfield.
935 PP/29 Fix three issues highlighted by clang analyser static analysis.
936 Only crash-plausible issue would require the Cambridge-specific
937 iplookup router and a misconfiguration.
938 Report from Marcin Mirosław.
940 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
942 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
943 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
944 As part of this, removing so much warning spew let me fix some minor
945 real issues in debug logging.
947 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
948 assignment on my part. Fixed.
950 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
951 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
952 Janne Snabb (who went above and beyond: thank you).
954 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
955 string otherwise requires a connection and a bunch more work and it's
956 relatively easy to get wrong. Should also expose TLS library linkage
959 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
960 64-bit ${eval} (JH/03).
962 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
963 GNU libc to support some of the 64-bit stuff, should not lead to
964 conflicts. Defined before os.h is pulled in, so if a given platform
965 needs to override this, it can.
967 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
968 protection layer was required, which is not implemented.
969 Bugzilla 1254, patch from Wolfgang Breyha.
971 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
972 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
973 tls_dhparam take prime identifiers. Also unbreak combination of
974 OpenSSL+DH_params+TLSSNI.
976 PP/39 Disable SSLv2 by default in OpenSSL support.
982 PP/01 Solaris build fix for Oracle's LDAP libraries.
983 Bugzilla 1109, patch from Stephen Usher.
985 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
987 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
990 TF/02 Fix a couple more cases where we did not log the error message
991 when unlink() failed. See also change 4.74-TF/03.
993 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
994 lock up or crash if it happened to be inside a call to libc when it
995 got a SIGUSR1 from exiwhat.
997 The SIGUSR1 handler appends the current process status to the process
998 log which is later printed by exiwhat. It used to use the general
999 purpose logging code to do this, but several functions it calls are
1000 not safe for signals.
1002 The new output code in the SIGUSR1 handler is specific to the process
1003 log, and simple enough that it's easy to inspect for signal safety.
1004 Removing some special cases also simplifies the general logging code.
1005 Removing the spurious timestamps from the process log simplifies
1008 TF/04 Improved ratelimit ACL condition.
1010 The /noupdate option has been deprecated in favour of /readonly which
1011 has clearer semantics. The /leaky, /strict, and /readonly update modes
1012 are mutually exclusive. The update mode is no longer included in the
1013 database key; it just determines when the database is updated. (This
1014 means that when you upgrde Exim will forget old rate measurements.)
1016 Exim now checks that the per_* options are used with an update mode that
1017 makes sense for the current ACL. For example, when Exim is processing a
1018 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1019 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1020 must specify per_mail/readonly. If you omit the update mode it defaults to
1021 /leaky where that makes sense (as before) or /readonly where required.
1023 The /noupdate option is now undocumented but still supported for
1024 backwards compatibility. It is equivalent to /readonly except that in
1025 ACLs where /readonly is required you may specify /leaky/noupdate or
1026 /strict/noupdate which are treated the same as /readonly.
1028 A useful new feature is the /count= option. This is a generalization
1029 of the per_byte option, so that you can measure the throughput of other
1030 aggregate values. For example, the per_byte option is now equivalent
1031 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1033 The per_rcpt option has been generalized using the /count= mechanism
1034 (though it's more complicated than the per_byte equivalence). When it is
1035 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1036 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1037 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1038 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1039 Note that using per_rcpt with a non-readonly update mode in more than
1040 one ACL will cause the recipients to be double-counted. (The per_mail
1041 and per_byte options don't have this problem.)
1043 The handling of very low rates has changed slightly. If the computed rate
1044 is less than the event's count (usually one) then this event is the first
1045 after a long gap. In this case the rate is set to the same as this event's
1046 count, so that the first message of a spam run is counted properly.
1048 The major new feature is a mechanism for counting the rate of unique
1049 events. The new per_addr option counts the number of different
1050 recipients that someone has sent messages to in the last time period. It
1051 behaves like per_rcpt if all the recipient addresses are different, but
1052 duplicate recipient addresses do not increase the measured rate. Like
1053 the /count= option this is a general mechanism, so the per_addr option
1054 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1055 example, measure the rate that a client uses different sender addresses
1056 with the options per_mail/unique=$sender_address. There are further
1057 details in the main documentation.
1059 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1061 TF/06 Removed a few PCRE remnants.
1063 TF/07 Automatically extract Exim's version number from tags in the git
1064 repository when doing development or release builds.
1066 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1067 Bugzilla 879. Patch from Paul Fisher.
1069 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1070 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1073 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1075 PP/05 Variable $av_failed, true if the AV scanner deferred.
1076 Bugzilla 1078. Patch from John Horne.
1078 PP/06 Stop make process more reliably on build failure.
1079 Bugzilla 1087. Patch from Heiko Schlittermann.
1081 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1082 Bugzilla 1089. Patch from Heiko Schlittermann.
1084 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1085 Bugzilla 1131. Patch from Holger Weiß.
1087 PP/09 Handle IPv6 addresses with SPF.
1088 Bugzilla 860. Patch from Wolfgang Breyha.
1090 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1092 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1095 PP/11 match_* no longer expand right-hand-side by default.
1096 New compile-time build option, EXPAND_LISTMATCH_RHS.
1097 New expansion conditions, "inlist", "inlisti".
1099 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1101 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1103 PP/14 fix log_write() format string regression from TF/03.
1104 Bugzilla 1152. Patch from Dmitry Isaikin.
1110 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1112 PP/02 Harmonised TLS library version reporting; only show if debugging.
1113 Layout now matches that introduced for other libraries in 4.74 PP/03.
1115 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1117 PP/04 New "dns_use_edns0" global option.
1119 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1122 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1123 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1125 TK/01 Updated PolarSSL code to 0.14.2.
1126 Bugzilla 1097. Patch from Andreas Metzler.
1128 PP/07 Catch divide-by-zero in ${eval:...}.
1129 Fixes bugzilla 1102.
1131 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1134 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1135 format-string attack -- SECURITY: remote arbitrary code execution.
1137 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1138 time unintentionally subject to list matching rules, letting the header
1139 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1140 arbitrary string expansion). This allowed for information disclosure.
1142 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1143 INT_MIN/-1 -- value coerced to INT_MAX.
1149 NM/01 Workround for PCRE version dependency in version reporting
1152 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1153 This fixes portability to compilers other than gcc, notably
1154 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1156 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1157 makefiles for portability to HP-UX and POSIX correctness.
1159 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1160 Also via indented variable definition in the Makefile.
1161 (Debugging by Oliver Heesakkers).
1163 PP/02 Restore caching of spamd results with expanded spamd_address.
1164 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1166 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1167 Improves build reliability. Fix from: Frank Elsner
1169 NM/02 Fix wide character breakage in the rfc2047 coding
1170 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1172 NM/03 Allow underscore in dnslist lookups
1173 Fixes bug 1026. Patch from Graeme Fowler
1175 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1176 Code patches from Adam Ciarcinski of NetBSD.
1178 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1181 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1182 is logged, to avoid truncation. Patch from John Horne.
1184 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1185 Patch from Jakob Hirsch.
1187 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1188 SQL string expansion failure details.
1189 Patch from Andrey Oktyabrski.
1191 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1192 Patch from Simon Arlott.
1194 PP/09 New lookups functionality failed to compile on old gcc which rejects
1195 extern declarations in function scope.
1196 Patch from Oliver Fleischmann
1198 PP/10 Use sig_atomic_t for flags set from signal handlers.
1199 Check getgroups() return and improve debugging.
1200 Fixed developed for diagnosis in bug 927 (which turned out to be
1203 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1204 Patch from Mark Zealey.
1206 PP/12 Bugzilla 1056: Improved spamd server selection.
1207 Patch from Mark Zealey.
1209 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1210 Based on patch from Heiko Schlittermann.
1212 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1213 Patch from Uwe Doering, sign-off by Michael Haardt.
1215 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1216 variable declaration deep within a block. Bug and patch from
1219 PP/15 lookups-Makefile IRIX compatibilty coercion.
1221 PP/16 Make DISABLE_DKIM build knob functional.
1223 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1224 Patch by Simon Arlott
1226 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1227 variable argument macros. Our copy now differs from upstream.
1233 TF/01 Failure to get a lock on a hints database can have serious
1234 consequences so log it to the panic log.
1236 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1237 controlled using the smtp_confirmation log selector.
1239 TF/03 Include the error message when we fail to unlink a spool file.
1241 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1242 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1243 for maintaining out-of-tree patches for some time.
1245 PP/01 Bugzilla 139: Documentation and portability issues.
1246 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1247 Handle per-OS dynamic-module compilation flags.
1249 PP/02 Let /dev/null have normal permissions.
1250 The 4.73 fixes were a little too stringent and complained about the
1251 permissions on /dev/null. Exempt it from some checks.
1252 Reported by Andreas M. Kirchwitz.
1254 PP/03 Report version information for many libraries, including
1255 Exim version information for dynamically loaded libraries. Created
1256 version.h, now support a version extension string for distributors
1257 who patch heavily. Dynamic module ABI change.
1259 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1260 privilege escalation vulnerability whereby the Exim run-time user
1261 can cause root to append content of the attacker's choosing to
1264 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1267 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1268 If dropping privileges for untrusted macros, we disabled normal logging
1269 on the basis that it would fail; for the Exim run-time user, this is not
1270 the case, and it resulted in successful deliveries going unlogged.
1271 Fixed. Reported by Andreas Metzler.
1277 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1278 only prepend for the Resent-* case. Fixes regression introduced in
1279 Exim 4.70 by NM/22 for Bugzilla 607.
1281 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1282 increasing numbers of administrators be bitten by this.
1284 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1286 PP/03 Bugzilla 994: added openssl_options main configuration option.
1288 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1290 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1292 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1294 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1295 without a peer certificate, leading to a segfault because of an
1296 assumption that peers always have certificates. Be a little more
1297 paranoid. Problem reported by Martin Tscholak.
1299 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1300 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1301 NB: ClamAV planning to remove STREAM in "middle of 2010".
1302 CL also introduces -bmalware, various -d+acl logging additions and
1303 more caution in buffer sizes.
1305 PP/09 Implemented reverse_ip expansion operator.
1307 PP/10 Bugzilla 937: provide a "debug" ACL control.
1309 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1311 PP/12 Bugzilla 973: Implement --version.
1313 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1315 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1317 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1319 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1320 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1321 ignore trailing whitespace.
1323 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1325 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1328 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1329 Notification from Dr Andrew Aitchison.
1331 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1332 ExtendedDetectionInfo response format.
1333 Notification from John Horne.
1335 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1338 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1339 XSL and documented dependency on system catalogs, with examples of how
1342 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1345 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1346 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1347 configuration file which is writeable by the Exim user or group.
1349 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1350 of configuration files to cover files specified with the -C option if
1351 they are going to be used with root privileges, not just the default
1354 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1355 option (effectively making it always true).
1357 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1358 files to be used while preserving root privileges.
1360 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1361 that rogue child processes cannot use them.
1363 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1364 run-time user, instead of root.
1366 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1367 Exim run-time user without dropping privileges.
1369 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1370 result string, instead of calling string_vformat() twice with the same
1373 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1374 for other users. Others should always drop root privileges if they use
1375 -C on the command line, even for a whitelisted configure file.
1377 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1379 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1385 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1386 $data_path, and $header_path variables; fixed documentation bugs and
1389 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1390 exipick to access non-standard spools, including the "frozen" queue
1393 NM/01 Bugzilla 965: Support mysql stored procedures.
1394 Patch from Alain Williams
1396 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1398 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1399 Patch from Andreas Metzler
1401 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1402 Patch from Kirill Miazine
1404 NM/05 Bugzilla 671: Added umask to procmail example.
1406 JJ/03 installed exipick 20100323.0, fixing doc bug
1408 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1409 directory. Notification and patch from Dan Rosenberg.
1411 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1413 TK/02 Improve log output when DKIM signing operation fails.
1415 MH/01 Treat the transport option dkim_domain as a colon separated
1416 list, not as a single string, and sign the message with each element,
1417 omitting multiple occurences of the same signer.
1419 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1420 Bugzilla 985, 986. Patch by Simon Arlott
1422 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1423 Patch by Simon Arlott
1425 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1426 MBX locking. Notification from Dan Rosenberg.
1432 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1434 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1436 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1438 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1440 NM/04 Bugzilla 915: Flag broken perl installation during build.
1446 TK/01 Added patch by Johannes Berg that expands the main option
1447 "spamd_address" if it starts with a dollar sign.
1449 TK/02 Write list of recipients to X-Envelope-Sender header when building
1450 the mbox-format spool file for content scanning (suggested by Jakob
1453 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1454 (http://www.dcc-servers.net/) support via dccifd. Activated by
1455 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1457 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1458 by Mark Daniel Reidel <mr@df.eu>.
1460 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1461 When building exim an external PCRE library is now needed -
1462 PCRE is a system library on the majority of modern systems.
1463 See entry on PCRE_LIBS in EDITME file.
1465 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1466 conversation. Added nologin parameter to request.
1467 Patch contributed by Kirill Miazine.
1469 TF/01 Do not log submission mode rewrites if they do not change the address.
1471 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1473 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1474 log files in place. Contributed by Roberto Lima.
1476 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1478 TF/03 Bugzilla 615: When checking the local_parts router precondition
1479 after a local_part_suffix or local_part_prefix option, Exim now
1480 does not use the address's named list lookup cache, since this
1481 contains cached lookups for the whole local part.
1483 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1484 Robert Millan. Documentation is in experimental-spec.txt.
1486 TF/04 Bugzilla 668: Fix parallel build (make -j).
1488 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1490 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1491 Patch provided by Jan Srzednicki.
1493 TF/05 Leading white space used to be stripped from $spam_report which
1494 wrecked the formatting. Now it is preserved.
1496 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1497 that they are available at delivery time.
1499 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1501 TF/08 TLS error reporting now respects the incoming_interface and
1502 incoming_port log selectors.
1504 TF/09 Produce a more useful error message if an SMTP transport's hosts
1505 setting expands to an empty string.
1507 NM/06 Bugzilla 744: EXPN did not work under TLS.
1508 Patch provided by Phil Pennock.
1510 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1511 Patch provided by Richard Godbee.
1513 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1514 acl_smtp_notquit, added index entry.
1516 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1517 Patch provided by Eugene Bujak.
1519 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1520 accept(). Patch provided by Maxim Dounin.
1522 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1523 Patch provided by Phil Pennock.
1525 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1527 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1528 Patch provided by Brad "anomie" Jorsch.
1530 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1531 Patch provided by Dean Brooks.
1533 TK/05 Add native DKIM support (does not depend on external libraries).
1535 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1536 Patch provided by Graeme Fowler.
1538 NM/16 Bugzilla 851: Documentation example syntax fix.
1540 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1542 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1545 NM/19 Bugzilla 745: TLS version reporting.
1546 Patch provided by Phil Pennock.
1548 NM/20 Bugzilla 167: bool: condition support.
1549 Patch provided by Phil Pennock.
1551 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1552 clients. Patch provided by Phil Pennock.
1554 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1555 Patch provided by Brad "anomie" Jorsch.
1557 NM/23 Bugzilla 687: Fix misparses in eximstats.
1558 Patch provided by Heiko Schlittermann.
1560 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1561 Patch provided by Heiko Schlittermann.
1563 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1564 plus update to original patch.
1566 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1568 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1569 Patch provided by David Brownlee.
1571 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1573 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1575 NM/30 Bugzilla 888: TLS documentation bugfixes.
1577 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1579 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1580 Unlike the original bugzilla I have changed all shell scripts in src tree.
1582 NM/33 Bugzilla 898: Transport filter timeout fix.
1583 Patch by Todd Rinaldo.
1585 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1586 Patch by Serge Demonchaux.
1588 NM/35 Bugzilla 39: Base64 decode bug fixes.
1589 Patch by Jakob Hirsch.
1591 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1593 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1595 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1597 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1603 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1604 ALT-N's libdkim that I have put here:
1605 http://duncanthrax.net/exim-experimental/
1607 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1608 were called 'true' and it seems that C99 defines that as a reserved
1609 keyword to be used with 'bool' variable types. That means you could
1610 not include C99-style headers which use bools without triggering
1611 build errors in sieve.c.
1613 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1614 as mailq or other aliases. Changed the --help handling significantly
1615 to do whats expected. exim_usage() emits usage/help information.
1617 SC/01 Added the -bylocaldomain option to eximstats.
1619 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1621 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1623 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1629 PH/01 Another patch from the Sieve maintainer.
1631 PH/02 When an IPv6 address is converted to a string for single-key lookup
1632 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1633 dots are used instead of colons so that keys in lsearch files need not
1634 contain colons. This was done some time before quoting was made available
1635 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1636 (notated using the quote facility) so as to distinguish them from IPv4
1637 keys. This meant that lookups for IP addresses in host lists did not work
1638 for iplsearch lookups.
1640 This has been fixed by arranging for IPv6 addresses to be expressed with
1641 colons if the lookup type is iplsearch. This is not incompatible, because
1642 previously such lookups could never work.
1644 The situation is now rather anomolous, since one *can* have colons in
1645 ordinary lsearch keys. However, making the change in all cases is
1646 incompatible and would probably break a number of configurations.
1648 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1651 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1652 conversion specification without a maximum field width, thereby enabling
1653 a rogue spamd server to cause a buffer overflow. While nobody in their
1654 right mind would setup Exim to query an untrusted spamd server, an
1655 attacker that gains access to a server running spamd could potentially
1656 exploit this vulnerability to run arbitrary code as the Exim user.
1658 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1659 $primary_hostname instead of what libspf2 thinks the hosts name is.
1661 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1662 a directory entry by the name of the lookup key. Previously, if a
1663 symlink pointed to a non-existing file or a file in a directory that
1664 Exim lacked permissions to read, a lookup for a key matching that
1665 symlink would fail. Now it is enough that a matching directory entry
1666 exists, symlink or not. (Bugzilla 503.)
1668 PH/03 The body_linecount and body_zerocount variables are now exported in the
1671 PH/04 Added the $dnslist_matched variable.
1673 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1674 This means they are set thereafter only if the connection becomes
1677 PH/06 Added the client_condition to authenticators so that some can be skipped
1678 by clients under certain conditions.
1680 PH/07 The error message for a badly-placed control=no_multiline_responses left
1681 "_responses" off the end of the name.
1683 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1685 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1686 (without spaces) instead of just copying the configuration text.
1688 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1690 PH/11 Added $max_received_linelength.
1692 PH/12 Added +ignore_defer and +include_defer to host lists.
1694 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1695 way in which PCRE > 7.0 is built.
1697 PH/14 Implemented queue_only_load_latch.
1699 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1700 MAIL command. The effect was to mangle the value on 64-bit systems.
1702 PH/16 Another patch from the Sieve maintainer.
1704 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1706 PH/18 If a system quota error occurred while trying to create the file for
1707 a maildir delivery, the message "Mailbox is full" was not appended to the
1708 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1709 only to a quota excession during the actual writing of the file.
1711 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1712 characters?) which causes problems in log lines. The DN values are now
1713 passed through string_printing() before being added to log lines.
1715 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1716 and InterBase are left for another time.)
1718 PH/21 Added message_body_newlines option.
1720 PH/22 Guard against possible overflow in moan_check_errorcopy().
1722 PH/23 POSIX allows open() to be a macro; guard against that.
1724 PH/24 If the recipient of an error message contained an @ in the local part
1725 (suitably quoted, of course), incorrect values were put in $domain and
1726 $local_part during the evaluation of errors_copy.
1732 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1733 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1736 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1737 issue a MAIL command.
1739 PH/02 In an ACL statement such as
1741 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1743 if a client was not listed at all, or was listed with a value other than
1744 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1745 the condition was not true (as it should be), so access was not denied.
1746 The bug was that the ! inversion was incorrectly passed on to the second
1747 item. This has been fixed.
1749 PH/03 Added additional dnslists conditions == and =& which are different from
1750 = and & when the dns lookup returns more than one IP address.
1752 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1753 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1755 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1756 FSYNC, which compiles an option called disable_fsync that allows for
1757 bypassing fsync(). The documentation is heavily laced with warnings.
1759 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1761 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1762 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1763 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1764 including adding "make clean"; (3) Added -fPIC when compiling the test
1765 dynamically loaded module, to get rid of a warning.
1767 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1768 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1769 = 0s. The bug is otherwise harmless.
1771 PH/07 There was a bug in the dovecot authenticator such that the value of
1772 $auth1 could be overwritten, and so not correctly preserved, after a
1773 successful authentication. This usually meant that the value preserved by
1774 the server_setid option was incorrect.
1776 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1778 PH/09 Installed PCRE release 7.0.
1780 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1781 run for batched SMTP input. It is now run at the start of every message
1782 in the batch. While fixing this I discovered that the process information
1783 (output by running exiwhat) was not always getting set for -bs and -bS
1784 input. This is fixed, and it now also says "batched" for BSMTP.
1786 PH/11 Added control=no_pipelining.
1788 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1789 patch, slightly modified), and move the expansion of helo_data till after
1790 the connection is made in the smtp transport (so it can use these
1793 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1795 PH/14 Added log_selector = +pid.
1797 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1799 PH/16 Add ${if forany and ${if forall.
1801 PH/17 Added dsn_from option to vary the From: line in DSNs.
1803 PH/18 Flush SMTP output before performing a callout, unless control =
1804 no_callout_flush is set.
1806 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1807 was true (the default) a successful delivery failed to delete the retry
1808 item, thus causing premature timeout of the address. The bug is now
1811 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1813 PH/21 Long custom messages for fakedefer and fakereject are now split up
1814 into multiline reponses in the same way that messages for "deny" and
1815 other ACL rejections are.
1817 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1818 with slight modification.
1820 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1821 draft, changing the syntax and factoring some duplicate code".
1823 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1824 for deliveries of the second and subsequent messages over the same SMTP
1827 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1828 ${reduce, with only minor "tidies".
1830 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1832 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1833 expansion side effects.
1835 PH/27 When a message times out after an over-quota error from an Exim-imposed
1836 quota, the bounce message says "mailbox is full". This message was not
1837 being given when it was a system quota that was exceeded. It now should
1840 MH/03 Made $recipients available in local_scan(). local_scan() already has
1841 better access to the recipient list through recipients_list[], but
1842 $recipients can be useful in postmaster-provided expansion strings.
1844 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1845 in the case of a MAIL command with additional options following the
1846 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1847 were accidentally chopped off.
1849 PH/29 SMTP synchronization checks are implemented when a command is read -
1850 there is a check that no more input is waiting when there shouldn't be
1851 any. However, for some commands, a delay in an ACL can mean that it is
1852 some time before the response is written. In this time, more input might
1853 arrive, invalidly. So now there are extra checks after an ACL has run for
1854 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1855 pipelining has not been advertised.
1857 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1859 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1860 returned in the SMTP response when smtp_return_error_details was false.
1861 This has been fixed.
1863 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1864 instead of the C I/O that was originally supplied, because problems were
1865 reported on Solaris.
1867 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1868 Exim which did not show up earlier: it was assuming that a call to
1869 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1870 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1871 was a macro that became an assignment, so it seemed to work. This has
1872 changed to a proper function call with a void return, hence the compile
1873 error. Exim's code has been fixed.
1875 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1878 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1880 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1882 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1883 because it turns out to be too big or there is a timeout) there is no
1884 "Completed" line in the log. When some messages of this type were
1885 selected by exigrep, they were listed as "not completed". Others were
1886 picked up by some special patterns. I have improved the selection
1887 criteria to be more general.
1889 PH/38 The host_find_failed option in the manualroute router can now be set
1890 to "ignore", to completely ignore a host whose IP address cannot be
1891 found. If all hosts are ignored, the behaviour is controlled by the new
1892 host_all_ignored option.
1894 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1895 homing or because of multiple MX records with /mx) generated more than
1896 one IP address, and the following item turned out to be the local host,
1897 all the secondary addresses of the first item were incorrectly removed
1898 from the list, along with the local host and any following hosts (which
1899 is what is supposed to happen).
1901 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1902 whoever called Exim into the -H file. In the case of the daemon it was
1903 behaving confusingly. When first started, it used values for whoever
1904 started the daemon, but after a SIGHUP it used the Exim user (because it
1905 calls itself on a restart). I have changed the code so that it now always
1908 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1909 message are rejected with the same error (e.g. no authentication or bad
1910 sender address), and a DATA command is nevertheless sent (as can happen
1911 with PIPELINING or a stupid MUA), the error message that was given to the
1912 RCPT commands is included in the rejection of the DATA command. This is
1913 intended to be helpful for MUAs that show only the final error to their
1916 PH/42 Another patch from the Sieve maintainer.
1918 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1919 Eximstats - Fixed some broken HTML links and added missing column headers
1921 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1922 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1924 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1926 PH/43 Yet another patch from the Sieve maintainer.
1928 PH/44 I found a way to check for a TCP/IP connection going away before sending
1929 the response to the final '.' that terminates a message, but only in the
1930 case where the client has not sent further data following the '.'
1931 (unfortunately, this is allowed). However, in many cases there won't be
1932 any further data because there won't be any more messages to send. A call
1933 to select() can be used: if it shows that the input is "ready", there is
1934 either input waiting, or the socket has been closed. An attempt to read
1935 the next input character can distinguish the two cases. Previously, Exim
1936 would have sent an OK response which the client would never have see.
1937 This could lead to message repetition. This fix should cure that, at
1938 least in a lot of common cases.
1940 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1941 advertised in response to EHLO.
1947 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1948 fixed by 4.65/MH/01 (is this a record?) are fixed:
1950 (i) An empty string was always treated as zero by the numeric comparison
1951 operators. This behaviour has been restored.
1953 (ii) It is documented that the numeric comparison operators always treat
1954 their arguments as decimal numbers. This was broken in that numbers
1955 starting with 0 were being interpreted as octal.
1957 While fixing these problems I realized that there was another issue that
1958 hadn't been noticed. Values of message_size_limit (both the global option
1959 and the transport option) were treated as octal if they started with 0.
1960 The documentation was vague. These values are now always treated as
1961 decimal, and I will make that clear in the documentation.
1967 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1968 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1971 MH/01 Don't check that the operands of numeric comparison operators are
1972 integers when their expansion is in "skipping" mode (fixes bug
1973 introduced by 4.64-PH/07).
1975 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1976 child addresses, Exim now panics and dies. Previously, because the count
1977 is held in a short int, deliveries were likely to be lost. As such a
1978 large number of recipients for a single message is ridiculous
1979 (performance will be very, very poor), I have chosen to impose a limit
1980 rather than extend the field.
1986 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1987 leftover -K file (the existence of which was triggered by #402).
1988 While we were at it, introduced process PID as part of the -K
1989 filename. This should rule out race conditions when creating
1992 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1993 processing considerably. Previous code took too long for large mails,
1994 triggering a timeout which in turn triggers #401.
1996 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1997 in the DK code in transports.c. sendfile() is not really portable,
1998 hence the _LINUX specificness.
2000 TF/01 In the add_headers option to the mail command in an Exim filter,
2001 there was a bug that Exim would claim a syntax error in any
2002 header after the first one which had an odd number of characters
2005 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2006 callout verification, Exim cached a "reject" for the entire domain. This
2007 is correct for most verifications, but it is not correct for a recipient
2008 verification with use_sender or use_postmaster set, because in that case
2009 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2010 case of MAIL FROM:<> rejection from other early rejections (e.g.
2011 rejection of HELO). When verifying a recipient using a non-null MAIL
2012 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2013 Whatever the result of the callout, the value of the domain cache is
2014 left unchanged (for any other kind of callout, getting as far as trying
2015 RCPT means that the domain itself is ok).
2017 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2020 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2021 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2022 session, the daemon ignores ECONNECT errors and logs others; it now
2023 ignores EPIPE as well.
2025 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2026 (quoted-printable decoding).
2028 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2029 later the small subsequent patch to fix an introduced bug.
2031 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2033 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2035 PH/08 An error is now given if message_size_limit is specified negative.
2037 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2038 to be given (somewhat) arbitrary names.
2040 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2043 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2044 miscellaneous code fixes
2046 PH/10 Added the log_reject_target ACL modifier to specify where to log
2049 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2050 hostname. This is wrong, because it relates to the incoming message (and
2051 probably the interface on which it is arriving) and not to the outgoing
2052 callout (which could be using a different interface). This has been
2053 changed to use the value of the helo_data option from the smtp transport
2054 instead - this is what is used when a message is actually being sent. If
2055 there is no remote transport (possible with a router that sets up host
2056 addresses), $smtp_active_hostname is used.
2058 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2059 tweaks were necessary in order to get it to work (see also 21 below):
2060 (a) The code assumed that strncpy() returns a negative number on buffer
2061 overflow, which isn't the case. Replaced with Exim's string_format()
2063 (b) There were several signed/unsigned issues. I just did the minimum
2064 hacking in of casts. There is scope for a larger refactoring.
2065 (c) The code used strcasecmp() which is not a standard C function.
2066 Replaced with Exim's strcmpic() function.
2067 (d) The code set only $1; it now sets $auth1 as well.
2068 (e) A simple test gave the error "authentication client didn't specify
2069 service in request". It would seem that Dovecot has changed its
2070 interface. Fortunately there's a specification; I followed it and
2071 changed what the client sends and it appears to be working now.
2073 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2076 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2077 address A is aliased to B and C, where B exists and C does not. Without
2078 -v the output is "A verified" because verification stops after a
2079 successful redirection if more than one address is generated. However,
2080 with -v the child addresses are also verified. Exim was outputting "A
2081 failed to verify" and then showing the successful verification for C,
2082 with its parentage. It now outputs "B failed to verify", showing B's
2083 parentage before showing the successful verification of C.
2085 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2086 look up a TXT record in a specific list after matching in a combined
2089 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2090 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2091 they consult the DNS. I had assumed they would set it the way they
2092 wanted; and indeed my experiments on Linux seem to show that in some
2093 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2094 To be on the safe side, however, I have now made the interface to
2095 host_find_byname() similar to host_find_bydns(), with an argument
2096 containing the DNS resolver options. The host_find_byname() function now
2097 sets these options at its start, just as host_find_bydns() does. The smtp
2098 transport options dns_qualify_single and dns_search_parents are passed to
2099 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2100 of host_find_byname() use the default settings of RES_DEFNAMES
2101 (qualify_single) but not RES_DNSRCH (search_parents).
2103 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2104 spool_read_header() do less string testing, by means of a preliminary
2105 switch on the second character of optional "-foo" lines. (This is
2106 overdue, caused by the large number of possibilities that now exist.
2107 Originally there were few.) While I was there, I also converted the
2108 str(n)cmp tests so they don't re-test the leading "-" and the first
2109 character, in the hope this might squeeze out yet more improvement.
2111 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2112 flag allowing group syntax was set by the header_syntax check but not
2113 turned off, possible causing trouble later; (2) The flag was not being
2114 set at all for the header_verify test, causing "group"-style headers to
2115 be rejected. I have now set it in this case, and also caused header_
2116 verify to ignore an empty address taken from a group. While doing this, I
2117 came across some other cases where the code for allowing group syntax
2118 while scanning a header line wasn't quite right (mostly, not resetting
2119 the flag correctly in the right place). These bugs could have caused
2120 trouble for malformed header lines. I hope it is now all correct.
2122 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2123 with the "reply" argument non-NULL. The code, however (which originally
2124 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2125 but it didn't always do it. This confused somebody who was copying the
2126 code for some other use. I have removed all the tests.
2128 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2129 feature that was used to support insecure browsers during the U.S. crypto
2130 embargo. It requires special client support, and Exim is probably the
2131 only MTA that supported it -- and would never use it because real RSA is
2132 always available. This code has been removed, because it had the bad
2133 effect of slowing Exim down by computing (never used) parameters for the
2134 RSA_EXPORT functionality.
2136 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2137 authenticator to fail if there's a tab character in the incoming data
2138 (there should never be unless someone is messing about, as it's supposed
2139 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2140 if the connection is using TLS or if the remote IP is the same as the
2141 local IP, and the "valid-client-cert option" if a client certificate has
2144 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2145 authenticators. This can be used for authorization after authentication
2146 succeeds. (In the case of plaintext, it servers for both authentication
2149 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2150 if any retry times were supplied.
2152 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2153 connection, where there is no client IP address to check. In this
2154 situation, the verify now always succeeds.
2156 PH/25 Applied John Jetmore's -Mset patch.
2158 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2160 PH/27 In a string expansion for a processed (not raw) header when multiple
2161 headers of the same name were present, leading whitespace was being
2162 removed from all of them, but trailing whitespace was being removed only
2163 from the last one. Now trailing whitespace is removed from each header
2164 before concatenation. Completely empty headers in a concatenation (as
2165 before) are ignored.
2167 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2168 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2170 PH/29 [Removed. This was a change that I later backed out, and forgot to
2171 correct the ChangeLog entry (that I had efficiently created) before
2172 committing the later change.]
2174 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2175 address errors (4xx response to RCPT) over the same connection as other
2176 messages routed to the same hosts. Such deliveries are always "forced",
2177 so retry times are not inspected. This resulted in far too many retries
2178 for the affected addresses. The effect occurred only when there were more
2179 hosts than the hosts_max_try setting in the smtp transport when it had
2180 the 4xx errors. Those hosts that it had tried were not added to the list
2181 of hosts for which the message was waiting, so if all were tried, there
2182 was no problem. Two fixes have been applied:
2184 (i) If there are any address or message errors in an SMTP delivery, none
2185 of the hosts (tried or untried) are now added to the list of hosts
2186 for which the message is waiting, so the message should not be a
2187 candidate for sending over the same connection that was used for a
2188 successful delivery of some other message. This seems entirely
2189 reasonable: after all the message is NOT "waiting for some host".
2190 This is so "obvious" that I'm not sure why it wasn't done
2191 previously. Hope I haven't missed anything, but it can't do any
2192 harm, as the worst effect is to miss an optimization.
2194 (ii) If, despite (i), such a delivery is accidentally attempted, the
2195 routing retry time is respected, so at least it doesn't keep
2196 hammering the server.
2198 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2199 in ${readsocket because some servers need this prod.
2201 PH/32 Added some extra debug output when updating a wait-xxx database.
2203 PH/33 The hint "could be header name not terminated by colon", which has been
2204 given for certain expansion errors for a long time, was not being given
2205 for the ${if def:h_colon_omitted{... case.
2207 PH/34 The spec says: "With one important exception, whenever a domain list is
2208 being scanned, $domain contains the subject domain." There was at least
2209 one case where this was not true.
2211 PH/35 The error "getsockname() failed: connection reset by peer" was being
2212 written to the panic log as well as the main log, but it isn't really
2213 panic-worthy as it just means the connection died rather early on. I have
2214 removed the panic log writing for the ECONNRESET error when getsockname()
2217 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2218 runs only) independently of the message's sender address. This meant
2219 that, if the 4xx error was in fact related to the sender, a different
2220 message to the same recipient with a different sender could confuse
2221 things. In particualar, this can happen when sending to a greylisting
2222 server, but other circumstances could also provoke similar problems.
2223 I have changed the default so that the retry time for these errors is now
2224 based a combination of the sender and recipient addresses. This change
2225 can be overridden by setting address_retry_include_sender=false in the
2228 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2229 remote server are returned as part of bounce messages. This was not
2230 happening for LMTP over a pipe (the lmtp transport), but now it is the
2231 same for both kinds of LMTP.
2233 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2234 in header lines that were in fact CNAMEs. This is no longer the case.
2236 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2237 and queue runs started by the daemon processed all messages. This has
2238 been fixed so that -R and -S can now usefully be given with -q<time>.
2240 PH/40 Import PCRE release 6.7 (fixes some bugs).
2242 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2244 PH/42 Give an error if -q is specified more than once.
2246 PH/43 Renamed the variables $interface_address and $interface_port as
2247 $received_ip_address and $received_port, to make it clear that these
2248 values apply to message reception, and not to the outgoing interface when
2249 a message is delivered. (The old names remain recognized, of course.)
2251 PH/44 There was no timeout on the connect() call when using a Unix domain
2252 socket in the ${readsocket expansion. There now is.
2254 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2255 be meaningful with "accept".
2257 SC/01 Eximstats V1.43
2258 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2260 SC/02 Eximstats V1.44
2261 Use a glob alias rather than an array ref in the generated
2262 parser. This improves both readability and performance.
2264 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2265 Collect SpamAssassin and rejection statistics.
2266 Don't display local sender or destination tables unless
2267 there is data to show.
2268 Added average volumes into the top table text output.
2270 SC/04 Eximstats V1.46
2271 Collect data on the number of addresses (recipients)
2272 as well as the number of messages.
2274 SC/05 Eximstats V1.47
2275 Added 'Message too big' to the list of mail rejection
2276 reasons (thanks to Marco Gaiarin).
2278 SC/06 Eximstats V1.48
2279 Mainlog lines which have GMT offsets and are too short to
2280 have a flag are now skipped.
2282 SC/07 Eximstats V1.49 (Alain Williams)
2283 Added the -emptyok flag.
2285 SC/08 Eximstats V1.50
2286 Fixes for obtaining the IP address from reject messages.
2288 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2289 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2290 whitesspace changes from 4.64-PH/27
2292 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2295 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2296 are found, allow negative numbers in numeric criteria)
2298 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2300 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2301 to match changes made in 4.64-PH/43
2303 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2305 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2306 "domain" is actually an IP address, and force a failure. This locks out
2307 those revolvers/nameservers that support "A-for-A" lookups, in
2308 contravention of the specifications.
2310 PH/48 When a host name was looked up from an IP address, and the subsequent
2311 forward lookup of the name timed out, the host name was left in
2312 $sender_host_name, contrary to the specification.
2314 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2315 restricted to single-key lookups, Exim was not diagnosing an error if
2316 * or *@ was used with a query-style lookup.
2318 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2320 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2321 long ago, but noone interested enough thought of it. Let's just say that
2322 the "1.1" means that there are some new functions that weren't there at
2323 some point in the past.
2325 PH/51 Error processing for expansion failure of helo_data from an smtp
2326 transport during callout processing was broken.
2328 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2329 tested/used via the -bh/-bhc/-bs options.
2331 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2332 bug, fixed in subsequent PCRE releases).
2334 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2335 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2337 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2343 SC/01 Use a glob alias rather than an array ref in eximstats generated
2344 parser. This improves both readability and performance.
2346 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2347 Don't display local sender or destination tables in eximstats unless
2348 there is data to show.
2349 Added average volumes into the eximstats top table text output.
2351 SC/03 Collect data on the number of addresses (recipients) as well
2352 as the number of messages in eximstats.
2354 TF/01 Correct an error in the documentation for the redirect router. Exim
2355 does (usually) call initgroups() when daemonizing.
2357 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2358 with consistent privilege compared to when running as a daemon.
2360 TF/03 Note in the spec that $authenticated_id is not set for local
2361 submissions from trusted users.
2363 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2364 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2366 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2367 by adding some example configuration directives to the default
2368 configuration file. A little bit of work is required to uncomment the
2369 directives and define how usernames and passwords are checked, but
2370 there is now a framework to start from.
2372 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2373 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2374 without this. I don't know how relevant this is to other LDAP libraries.
2376 PH/02 Add the verb name to the "unknown ACL verb" error.
2378 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2380 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2382 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2383 directory not expanded when it should be if an expanded home directory
2384 was set for the address (which is overridden by the transport).
2386 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2389 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2390 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2391 because it is too late at that time, and has no effect.
2393 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2394 security issue with \' (bugzilla #107). I could not use the
2395 PQescapeStringConn() function, because it needs a PGconn value as one of
2398 PH/08 When testing addresses using -bt, indicate those final addresses that
2399 are duplicates that would not cause an additional delivery. At least one
2400 person was confused, thinking that -bt output corresponded to deliveries.
2401 (Suppressing duplicates isn't a good idea as you lose the information
2402 about possibly different redirections that led to the duplicates.)
2404 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2405 systems where poll() doesn't work, in particular OS X.
2407 PH/10 Added more information to debugging output for retry time not reached.
2409 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2410 operations in malware.c.
2412 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2415 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2416 syslog_duplication set false, log lines that would normally be written
2417 both the the main log and to the reject log were not written to syslog at
2420 PH/14 In the default configuration, change the use of "message" in ACL warn
2421 statements to "add_header".
2423 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2424 not followed by a command (e.g. "seen endif").
2426 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2427 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2430 PH/17 Added extra conditions to the default value of delay_warning_condition
2434 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2435 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2436 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2439 The Auto-Submitted: and various List- headers are standardised, whereas I
2440 don't think Precedence: ever was.
2442 PH/18 Refactored debugging code in route_finduser() to show more information,
2443 in particular, the error code if getpwnam() issues one.
2445 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2446 This is apparently needed in addition to the PH/07 change above to avoid
2447 any possible encoding problems.
2449 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2450 but not after initializing Perl.
2452 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2453 output them only if debugging. By default they are written stderr,
2454 apparently, which is not desirable.
2456 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2459 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2462 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2464 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2465 authenticated or an ident call has been made. Suppress the default
2466 values for $authenticated_id and $authenticated_sender (but permit -oMai
2467 and -oMas) when testing with -bh.
2469 PH/24 Re-jigged the order of the tests in the default configuration so that the
2470 tests for valid domains and recipients precede the DNS black list and CSA
2471 tests, on the grounds that those ones are more expensive.
2473 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2474 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2475 This bug exists in every version of Exim that I still have, right back to
2478 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2479 However, an attempt to turn on case-sensitivity in a regex key by
2480 including (?-i) didn't work because the subject string was already
2481 lowercased, and the effects were non-intuitive. It turns out that a
2482 one-line patch can be used to allow (?-i) to work as expected.
2488 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2489 other effects) broke the use of negated acl sub-conditions.
2491 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2494 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2495 "Deny" causes Exim to reject the incoming connection with a 554 error.
2496 Unfortunately, if there is a major crisis, such as a disk failure,
2497 tcp-wrappers gives "deny", whereas what one would like would be some
2498 kind of temporary error. A kludge has been added to help with this.
2499 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2500 554 error is used if errno is still zero or contains ENOENT (which occurs
2501 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2504 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2506 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2507 errors. Otherwise a message that provokes a temporary error (when other
2508 messages do not) can cause a whole host to time out.
2510 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2511 addresses were routed directly to files or pipes from a redirect router.
2512 File deliveries just didn't batch; pipe deliveries might have suffered
2515 PH/06 A failure to get a lock for a hints database would erroneously always say
2516 "Failed to get write lock", even when it was really a read lock.
2518 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2519 of 0600. This has been changed to use the value of the lockfile_mode
2520 option (which defaults to 0600).
2522 PH/08 Applied small patch from the Sieve maintainer.
2524 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2525 folder from quota calculations, a direct delivery into this folder messed
2526 up the contents of the maildirsize file. This was because the regex was
2527 used only to exclude .Trash (or whatever) when the size of the mailbox
2528 was calculated. There was no check that a delivery was happening into an
2529 excluded directory. This bug has been fixed by ignoring all quota
2530 processing for deliveries into excluded directories.
2532 PH/10 Added the maildirfolder_create_regex option to appendfile.
2538 PH/01 The code for finding all the local interface addresses on a FreeBSD
2539 system running IPv6 was broken. This may well have applied to all BSD
2540 systems, as well as to others that have similar system calls. The broken
2541 code found IPv4 interfaces correctly, but gave incorrect values for the
2542 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2543 that it would not match correctly against @[] and not recognize the IPv6
2546 PH/02 The ipliteral router was not recognizing addresses of the form user@
2547 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2549 PH/03 Added disable_ipv6.
2551 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2552 decoded form, because it is most often used to construct To: headers
2553 lines in autoreplies, and the decoded form may well be syntactically
2554 invalid. However, $reply_address has leading white space removed, and all
2555 newlines turned into spaces so that the autoreply transport does not
2558 PH/05 If group was specified without a user on a router, and no group or user
2559 was specified on a transport, the group from the router was ignored.
2561 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2562 for visible compile-time settings that can be used to change these
2563 numbers, for those that want even more. Backwards compatibility with old
2564 spool files has been maintained. However, going back to a previous Exim
2565 release will lost any variables that are in spool files.
2567 PH/07 Two small changes when running in the test harness: increase delay when
2568 passing a TCP/IP connection to a new process, in case the original
2569 process has to generate a bounce, and remove special handling of
2570 127.0.0.2 (sic), which is no longer necessary.
2572 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2573 be the same on different OS.
2575 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2578 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2579 whether --show-vars was specified or not
2581 JJ/02 exipick: Added support for new ACL variable spool format introduced
2584 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2585 syntactically invalid From: or Reply-to: line, and a filter used this to
2586 generate an autoreply, and therefore failed to obtain an address for the
2587 autoreply, Exim could try to deliver to a non-existent relative file
2588 name, causing unrelated and misleading errors. What now happens is that
2589 it logs this as a hard delivery error, but does not attempt to create a
2592 PH/11 The exinext utility has a -C option for testing purposes, but although
2593 the given file was scanned by exinext itself; it wasn't being passed on
2594 when Exim was called.
2596 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2597 an end-of-file indication when reading a command response.
2599 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2600 compiled. In many other places in Exim, IPv6 addresses are always
2601 recognized, so I have changed this. It also means that IPv4 domain
2602 literals of the form [IPV4:n.n.n.n] are now always recognized.
2604 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2605 used if the router is not running as root, for example, when verifying at
2606 ACL time, or when using -bh. The debugging output from this situation was
2607 non-existent - all you got was a failure to exec. I have made two
2610 (a) Failures to set uid/gid, the current directory, or a process leader
2611 in a subprocess such as that created by queryprogram now generate
2612 suitable debugging ouput when -d is set.
2614 (b) The queryprogram router detects when it is not running as root,
2615 outputs suitable debugging information if -d is set, and then runs
2616 the subprocess without attempting to change uid/gid.
2618 PH/15 Minor change to Makefile for building test_host (undocumented testing
2621 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2622 additional section of a DNS packet that returns MX or SRV records.
2623 Instead, it always explicitly searches for A/AAAA records. This avoids
2624 major problems that occur when a DNS server includes only records of one
2625 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2626 fixed another bug: if SRV records were looked up and the corresponding
2627 address records were *not* found in the additional section, the port
2628 values from the SRV records were lost.
2630 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2631 using the correct key (the original address) when searching the retry
2632 rules in order to find which one to use for generating the retry hint.
2634 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2635 adding the default one. Similarly, if it contains a Reply-To: header, the
2636 errors_reply_to option, if set, is not used.
2638 PH/19 When calculating a retry time, Exim used to measure the "time since
2639 failure" by looking at the "first failed" field in the retry record. Now
2640 it does not use this if it is later than than the arrival time of the
2641 message. Instead it uses the arrival time. This makes for better
2642 behaviour in cases where some deliveries succeed, thus re-setting the
2643 "first failed" field. An example is a quota failure for a huge message
2644 when small messages continue to be delivered. Without this change, the
2645 "time since failure" will always be short, possible causing more frequent
2646 delivery attempts for the huge message than are intended.
2647 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2649 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2650 $1, $2, $3) because the numerical variables can be reset during some
2651 expansion items (e.g. "match"), thereby losing the authentication data.
2653 PH/21 Make -bV show the size of off_t variables so that the test suite can
2654 decide whether to run tests for quotas > 2G.
2656 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2657 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2658 filecount value is greater than 2G or if a quota value is greater than 2G
2659 on a system where the size of off_t is not greater than 4, a panic error
2662 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2663 never match. The debug and -bh output now contains an explicit error
2664 message indicating a malformed IPv4 address or mask.
2666 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2667 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2668 PH/23 above applies.
2670 PH/25 Do not write to syslog when running in the test harness. The only
2671 occasion when this arises is a failure to open the main or panic logs
2672 (for which there is an explicit test).
2674 PH/26 Added the /no_tell option to "control=freeze".
2676 PH/27 If a host name lookup failed very early in a connection, for example, if
2677 the IP address matched host_lookup and the reverse lookup yielded a name
2678 that did not have a forward lookup, an error message of the form "no IP
2679 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2680 could be logged. Now it outputs the IP address instead of "NULL".
2682 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2683 allows the sender and the authenticated sender to be set when
2684 submitting a message from within Exim. Since child_open_exim() is
2685 documented for local_scan(), the new function should be too.
2687 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2688 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2689 results in an empty string is now treated as unset.
2691 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2693 PH/31 Added sender_verify_fail logging option.
2695 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2696 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2697 tidied the source and removed it altogether.
2699 PH/33 When a queue run was abandoned because the load average was too high, a
2700 log line was always written; now it is written only if the queue_run log
2701 selector is set. In addition, the log line for abandonment now contains
2702 information about the queue run such as the pid. This is always present
2703 in "start" and "stop" lines but was omitted from the "abandon" line.
2705 PH/34 Omit spaces between a header name and the colon in the error message that
2706 is given when verify = headers_syntax fails (if there are lots of them,
2707 the message gets confusing).
2709 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2710 names, as there are now some PTR records that contain slashes. This check
2711 is only to protect against broken name servers that fall over on strange
2712 characters, so the fact that it applies to all lookups doesn't matter.
2714 PH/36 Now that the new test suite is complete, we can remove some of the
2715 special code in Exim that was needed for the old test suite. For example,
2716 sorting DNS records because real resolvers return them in an arbitrary
2717 order. The new test suite's fake resolver always returns records in the
2720 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2721 bounces) except when queue_only is set, to avoid logging races between
2722 the different processes.
2724 PH/38 Panic-die if .include specifies a non-absolute path.
2726 PH/39 A tweak to the "H" retry rule from its user.
2728 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2729 a label. They prevented compilation on older perls.
2731 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2732 a warning to be raised on newish perls.
2734 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2735 on queue. Changes to match documented behaviour of showing count of
2736 messages matching specified criteria.
2738 PH/40 Changed the default ident timeout from 30s to 5s.
2740 PH/41 Added support for the use of login_cap features, on those BSD systems
2741 that have them, for controlling the resources used by pipe deliveries.
2743 PH/42 The content-scanning code uses fopen() to create files in which to put
2744 message data. Previously it was not paying any attention to the mode of
2745 the files. Exim runs with umask(0) because the rest of the code creates
2746 files with open(), and sets the required mode explicitly. Thus, these
2747 files were ending up world-writeable. This was not a big issue, because,
2748 being within the spool directory, they were not world-accessible. I have
2749 created a function called modefopen, which takes an additional mode
2750 argument. It sets umask(777), creates the file, chmods it to the required
2751 mode, then resets the umask. All the relevant calls to fopen() in the
2752 content scanning code have been changed to use this function.
2754 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2755 to 24 hours. This avoids potential overflow problems when processing G
2756 and H retry rules. I suspect nobody ever tinkers with this value.
2758 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2760 PH/45 When the plaintext authenticator is running as a client, the server's
2761 challenges are checked to ensure they are valid base64 strings. By
2762 default, the authentication attempt is cancelled if an invalid string is
2763 received. Setting client_ignore_invalid_base64 true ignores these errors.
2764 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2765 they are received. Thus, the responses can be made to depend on the
2766 challenges. If an invalid string is ignored, an empty string is placed in
2769 PH/46 Messages that are created by the autoreply transport now contains a
2770 References: header, in accordance with RFCs 2822 and 3834.
2772 PH/47 Added authenticated_sender_force to the smtp transport.
2774 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2776 PH/49 Installed latest patch from the Sieve maintainer.
2778 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2779 was also set, the appendfile transport was unnecessarily scanning a
2780 directory of message files (e.g. for maildir delivery) to find the count
2781 of files (along with the size), even though it did not need this
2782 information. It now does the scan only if it needs to find either the
2783 size of the count of files.
2785 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2787 PH/52 Two bugs concerned with error handling when the smtp transport is
2790 (i) Exim was not creating retry information for temporary errors given
2791 for individual recipients after the DATA command when the smtp transport
2792 was used in LMTP mode. This meant that they could be retried too
2793 frequently, and not timed out correctly.
2795 (ii) Exim was setting the flag that allows error details to be returned
2796 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2797 recipients that were returned after the DATA command.
2799 PH/53 This is related to PH/52, but is more general: for any failing address,
2800 when detailed error information was permitted to be returned to the
2801 sender, but the error was temporary, then after the final timeout, only
2802 "retry timeout exceeded" was returned. Now it returns the full error as
2803 well as "retry timeout exceeded".
2805 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2806 do this, and (what is worse) MTAs that accept it.
2808 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2809 will now be deprecated.
2811 PH/56 New os.c-cygwin from the Cygwin maintainer.
2813 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2814 formats (previously only available in exim formats via -bpr, -bpru,
2815 and -bpra. Now also available in native and exiqgrep formats)
2817 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2818 with very large, slow to parse queues
2820 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2822 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2824 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2825 responses to authentication challenges, though it was showing the
2826 challenges; (ii) I've removed the CR characters from the debug output for
2829 PH/58 Allow for the insertion of a newline as well as a space when a string
2830 is turned into more than one encoded-word during RFC 2047 encoding. The
2831 Sieve code now uses this.
2833 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2834 data_4xx, lost_connection, tls_required.
2836 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2837 message was being sent as an SMTP response.
2839 PH/61 Add -l and -k options to exicyclog.
2841 PH/62 When verifying, if an address was redirected to one new address, so that
2842 verification continued, and the new address failed or deferred after
2843 having set something in $address_data, the value of $address_data was not
2844 passed back to the ACL. This was different to the case when no
2845 redirection occurred. The value is now passed back in both cases.
2847 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2848 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2849 use login_cap.h, so on its own it isn't the distinguishing feature. The
2850 new name refers directly to the setclassresources() function.
2852 PH/65 Added configuration files for NetBSD3.
2854 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2856 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2857 is preferred over IPv4.
2859 PH/68 The bounce_return_message and bounce_return_body options were not being
2860 honoured for bounces generated during the reception of non-SMTP messages.
2861 In particular, this applied to messages rejected by the ACL. This bug has
2862 been fixed. However, if bounce_return_message is true and bounce_return_
2863 body is false, the headers that are returned for a non-SMTP message
2864 include only those that have been read before the error was detected.
2865 (In the case of an ACL rejection, they have all been read.)
2867 PH/69 The HTML version of the specification is now built in a directory called
2868 spec_html instead of spec.html, because the latter looks like a path with
2869 a MIME-type, and this confuses some software.
2871 PH/70 Catch two compiler warnings in sieve.c.
2873 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2874 function verify_get_ident() calls ip_connect() to connect a socket, but
2875 if the "connect()" function timed out, ip_connect() used to close the
2876 socket. However, verify_get_ident() also closes the socket later, and in
2877 between Exim writes to the log, which may get opened at this point. When
2878 the socket was closed in ip_connect(), the log could get the same file
2879 descriptor number as the socket. This naturally causes chaos. The fix is
2880 not to close the socket in ip_connect(); the socket should be closed by
2881 the function that creates it. There was only one place in the code where
2882 this was missing, in the iplookup router, which I don't think anybody now
2883 uses, but I've fixed it anyway.
2885 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2886 well as to direct DNS lookups. Otherwise the handling of names in host
2887 lists is inconsistent and therefore confusing.
2893 PH/01 Two changes to the default runtime configuration:
2895 (1) Move the checks for relay_from_hosts and authenticated clients from
2896 after to before the (commented out) DNS black list checks.
2898 (2) Add control=submission to the relay_from_hosts and authenticated
2899 clients checks, on the grounds that messages accepted by these
2900 statements are most likely to be submissions.
2902 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2904 (1) Generate an error if the third argument for the ${prvs expansion is
2907 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2910 (3) Reset the variables that are obtained from the first argument of
2911 ${prvscheck and used in the second argument before leaving the code,
2912 because their memory is reclaimed, so using them afterwards may do
2915 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2916 one (it's much easier than Tom thought :-).
2918 (5) Because of (4), we can now allow for the use of $prvscheck_result
2919 inside the third argument.
2921 PH/03 For some reason, the default setting of PATH when running a command from
2922 a pipe transport was just "/usr/bin". I have changed it to
2925 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2926 anything to be listed in the output from -bV.
2928 PH/05 When a filter generated an autoreply, the entire To: header line was
2929 quoted in the delivery log line, like this:
2931 => >A.N.Other <ano@some.domain> <original@ddress> ...
2933 This has been changed so that it extracts the operative address. There
2934 may be more than one such address. If so, they are comma-separated, like
2937 => >ano@some.domain,ona@other.domain <original@ddress> ...
2939 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2940 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2941 not being looked up in the rDNS to get a host name, Exim was showing the
2942 IP address twice in Received: lines, even though the IP addresses were
2943 identical. For example:
2945 Received: from [1.2.3.4] (helo=[1.2.3.4])
2947 However, if the real host name was known, it was omitting the HELO data
2948 if it matched the actual IP address. This has been tidied up so that it
2949 doesn't show the same IP address twice.
2951 PH/07 When both +timestamp and +memory debugging was on, the value given by
2952 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2953 called by the debug printing, thereby overwriting the timestamp buffer.
2954 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2956 PH/08 When the original message was included in an autoreply transport, it
2957 always said "this is a copy of the message, including all the headers",
2958 even if body_only or headers_only was set. It now gives an appropriate
2961 PH/09 Applied a patch from the Sieve maintainer which:
2963 o fixes some comments
2964 o adds the (disabled) notify extension core
2965 o adds some debug output for the result of if/elsif tests
2966 o points to the current vacation draft in the documentation
2967 and documents the missing references header update
2971 o fixes a bug in processing the envelope test (when testing
2972 multiple envelope elements, the last element determinted the
2975 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2976 Electronic Mail") by including:
2978 Auto-submitted: auto-generated
2980 in the messages that it generates (bounce messages and others, such as
2981 warnings). In the case of bounce messages for non-SMTP mesages, there was
2982 also a typo: it was using "Auto_submitted" (underscore instead of
2983 hyphen). Since every message generated by Exim is necessarily in response
2984 to another message, thes have all been changed to:
2986 Auto-Submitted: auto-replied
2988 in accordance with these statements in the RFC:
2990 The auto-replied keyword:
2992 - SHOULD be used on messages sent in direct response to another
2993 message by an automatic process,
2995 - MUST NOT be used on manually-generated messages,
2997 - MAY be used on Delivery Status Notifications (DSNs) and Message
2998 Disposition Notifications (MDNs),
3000 - MUST NOT be used on messages generated by automatic or periodic
3001 processes, except for messages which are automatic responses to
3004 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3005 to the default Received: header definition.
3007 PH/12 Added log selector acl_warn_skipped (default on).
3009 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3010 variables because (a) they are in the wrong storage pool and (b) even if
3011 they were copied, it wouldn't work properly because of the caching.
3013 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3014 checking when decoding. Apparently there are clients that generate
3015 overlong encoded strings. Why am I not surprised?
3017 PH/15 If the first argument of "${if match_address" was not empty, but did not
3018 contain an "@" character, Exim crashed. Now it writes a panic log message
3019 and treats the condition as false.
3021 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3023 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3024 extension "envelope-auth". The code is finished and in agreement with
3025 other implementations, but there is no documentation so far and in fact,
3026 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3027 not changing the active code.
3029 Print executed "if" and "elsif" statements when debugging is used. This
3030 helps a great deal to understand what a filter does.
3032 Document more things not specified clearly in RFC3028. I had all this
3033 sorted out, when out of a sudden new issues came to my mind. Oops."
3035 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3038 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3039 canonical form (as documented). However, after a host name lookup from
3040 the IP address, check_host() was doing a simple string comparison with
3041 addresses acquired from the DNS when checking that the found name did
3042 have the original IP as one of its addresses. Since any found IPv6
3043 addresses are likely to be in abbreviated form, the comparison could
3044 fail. Luckily, there already exists a function for doing the comparison
3045 by converting both addresses to binary, so now that is used instead of
3046 the text comparison.
3048 PH/20 There was another similar case to PH/19, when a complete host name was
3049 given in a host list; looking up its IP address could give an abbreviated
3050 form, whereas the current host's name might or might not be abbreviated.
3051 The same fix has been applied.
3057 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3058 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3061 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3062 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3064 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3066 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3067 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3068 Also, for "Auto-Submitted", treat anything other than "no" as
3069 non-personal, in accordance with RFC 3834. (Previously it treated
3070 anything starting "auto-" as non-personal.)
3072 TF/01 The control=submission/name=... option had a problem with syntax
3073 errors if the name included a slash character. The /name= option
3074 now slurps the rest of the string, so it can include any characters
3075 but it must come last in the list of options (after /sender_retain
3078 PH/05 Some modifications to the interface to the fake nameserver for the new
3086 TK/01 Added the "success_on_redirect" address verification option. See
3087 NewStuff for rationale and an example.
3089 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3091 PH/02 Patch to exigrep to allow it to work on syslog lines.
3093 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3094 fread() to skip over the body file's header line, because in Cygwin the
3095 header line is locked and is inaccessible.
3097 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3098 co-exist for some time) to make it clear that it is the Exim ID that is
3099 referenced, not the Message-ID: header line.
3101 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3102 string_format() function, because snprintf() does not exist on all
3105 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3106 ${stat: expansion item.
3108 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3109 protocol synchronization error", to keep the pedants happy.
3111 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3112 well as for IRIX systems, when gcc is being used. See the host.c source
3115 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3117 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3120 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3121 transport and to the smtp transport in LMTP mode.
3123 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3125 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3126 run a filter in a subprocess. This could lead to confusion in subsequent
3127 lookups in the parent process. There should also be a search_tidyup() at
3128 the end of the subprocess.
3130 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3131 only if the host matched helo_try_verify_hosts, which caused the
3132 verification to occur when the EHLO/HELO command was issued. The ACL just
3133 tested the remembered result. Now, if a previous verification attempt has
3134 not happened, "verify = helo" does it there and then.
3136 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3138 TK/03 Fix log output including CR from clamd.
3140 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3141 exist provoked a memory error which could cause a segfault.
3143 PH/15 Installed PCRE 6.2
3145 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3147 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3148 of the problem. Specifically, suggested +O2 rather than +O1 for the
3151 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3153 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3154 record, the port was not correctly logged when the outgoing_port log
3155 selector was set (it logged the transort's default port).
3157 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3158 fallback_hosts, and "hosts" in the smtp transport.
3160 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3161 host errors such as "Connection refused".
3163 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3164 authentication with radiusclient 0.4.9:
3166 - Error returned from rc_read_config was caught wrongly
3167 - Username/password not passed on to radius server due to wrong length.
3169 The presumption is that some radiusclient API changes for 4.51/PH/17
3170 were not taken care of correctly. The code is still untested by me (my
3171 Linux distribution still has 0.3.2 of radiusclient), but it was
3172 contributed by a Radius user.
3174 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3175 expanding the "port" option of the smtp transport.
3177 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3178 while reading a MIME header. Thanks to Tom Hughes for a patch.
3180 PH/24 Include config.h inside local_scan.h so that configuration settings are
3183 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3184 that in an ACL for RCPT (for example), you can examine exactly what was
3187 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3188 commands, but it was not correctly comparing the address with the actual
3189 client host address. Thus, it would show the EHLO address in Received:
3190 header lines when this was not necessary.
3192 PH/27 Added the % operator to ${eval:}.
3194 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3195 it should be ignoring failures (because with -C, for example, it has lost
3196 privilege). It wasn't ignoring creation failures other than "already
3199 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3202 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3203 because an input line was too long, either on its own, or by virtue of
3204 too many continuations, the temporary file was not being removed, and the
3205 return code was incorrect.
3207 PH/31 Missing "BOOL" in function definition in filtertest.c.
3209 PH/32 Applied Sieve patches from the maintainer.
3211 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3213 PH/33 Added "verify = not_blind".
3215 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3216 Local/Makefile (with some defaults set). These are used in built scripts
3217 such as exicyclog, but they have never been used in the exim_install
3218 script (though there are many overriding facilities there). I have
3219 arranged that the exim_install script now takes note of these two
3222 PH/35 Installed configuration files for Dragonfly.
3224 PH/36 When a locally submitted message by a trusted user did not contain a
3225 From: header, and the sender address was obtained from -f or from an SMTP
3226 MAIL command, and the trusted user did not use -F to supply a sender
3227 name, $originator_name was incorrectly used when constructing a From:
3228 header. Furthermore, $originator_name was used for submission mode
3229 messages from external hosts without From: headers in a similar way,
3230 which is clearly wrong.
3232 PH/37 Added control=suppress_local_fixups.
3234 PH/38 When log_selector = +received_sender was set, and the addition of the
3235 sender made the log line's construction buffer exactly full, or one byte
3236 less than full, an overflow happened when the terminating "\n" was
3239 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3240 when the result of a list match is failure because a DNS lookup failed.
3242 PH/40 RM_COMMAND is now used in the building process.
3244 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3245 the "build-* directories that it finds.
3247 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3248 address in a domain literal was a prefix of an interface address.
3250 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3251 when verifying a sender address, unless rewrite_headers is false.
3253 PH/44 Wrote a long comment about why errors_to addresses are verified as
3254 recipients, not senders.
3256 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3257 the ratelimit ACL was added.
3259 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3261 PH/46 Added extra information about PostgreSQL errors to the error string.
3263 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3264 avoiding the need to install special zones in a real server. This is
3265 backwards compatible; if it can't find the fake resolver, it drops back.
3266 Thus, both old and new test suites can be run.
3268 TF/02 Added util/ratelimit.pl
3270 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3271 clock is set back in time.
3273 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3274 Candler <B.Candler@pobox.com>.
3276 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3277 OK for addresses that are the result of redirections.
3279 PH/48 A number of further additions for the benefit of the new test suite,
3280 including a fake gethostbyname() that interfaces to the fake DNS resolver
3283 TF/06 The fix for widen_domains has also been applied to qualify_single and
3284 search_parents which are the other dnslookup options that can cause
3287 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3290 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3292 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3293 service" if it encountered an MX record with an empty target hostname.
3294 The message is now "an MX or SRV record indicated no SMTP service".
3296 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3297 if the DNS of the sending site is misconfigured. This is quite a
3298 common situation. This change restores the behaviour of treating a
3299 helo verification defer as a failure.
3301 PH/51 If self=fail was set on a router, the bounce message did not include the
3302 actual error message.
3308 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3310 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3311 command itself ended in error, the underlying message about the transport
3312 filter timeout was being overwritten with the pipe command error. Now the
3313 underlying error message should be appended to the second error message.
3315 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3317 PH/02 Reduce the amount of output that "make" produces by default. Full output
3318 can still be requested.
3320 PH/03 The warning log line about a condition test deferring for a "warn" verb
3321 was being output only once per connection, rather than after each
3322 occurrence (because it was using the same function as for successful
3323 "warn" verbs). This seems wrong, so I have changed it.
3325 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3326 it should not have, which might have caused a crash in the right
3327 circumstances, but probably never did.
3329 PH/04 Installed a modified version of Tony Finch's patch to make submission
3330 mode fix the return path as well as the Sender: header line, and to
3331 add a /name= option so that you can make the user's friendly name appear
3334 TF/03 Added the control = fakedefer ACL modifier.
3336 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3337 Mark Lowes for thorough testing.
3339 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3341 TK/03 Merged latest SRS patch from Miles Wilton.
3343 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3344 with the definition in sysexits.h (which is #included earlier).
3345 Fortunately, Exim does not actually use EX_OK. The code used to try to
3346 preserve the sysexits.h value, by assumimg that macro definitions were
3347 scanned for macro replacements. I have been disabused of this notion,
3348 so now the code just undefines EX_OK before #including unistd.h.
3350 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3351 in the smtp transport. When a block could not be written in a single
3352 write() function, the timeout was being re-applied to each part-write.
3353 This seems wrong - if the receiver was accepting one byte at a time it
3354 would take for ever. The timeout is now adjusted when this happens. It
3355 doesn't have to be particularly precise.
3357 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3358 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3360 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3361 without a domain if the check to <postmaster@domain> fails.
3363 SC/01 Eximstats: added -xls and the ability to specify output files
3364 (patch written by Frank Heydlauf).
3366 SC/02 Eximstats: use FileHandles for outputing results.
3368 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3370 SC/04 Eximstats: fixed display of large numbers with -nvr option
3372 SC/05 Eximstats: fixed merging of reports with empty tables.
3374 SC/06 Eximstats: added the -include_original_destination flag
3376 SC/07 Eximstats: removed tabs and trailing whitespace.
3378 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3380 TK/06 MBOX spool code: Add real "From " MBOX separator line
3381 so the .eml file is really in mbox format (even though
3382 most programs do not really care). Patch from Alex Miller.
3384 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3385 The latter is generated from $received_to and is only set if the
3386 message has one envelope recipient. SA can use these headers,
3387 obviously out-of-the-box. Patch from Alex Miller.
3389 PH/08 The ${def test on a variable was returning false if the variable's
3390 value was "0", contrary to what the specification has always said!
3391 The result should be true unless the variable is empty.
3393 PH/09 The syntax error of a character other than { following "${if
3394 def:variable_name" (after optional whitespace) was not being diagnosed.
3395 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3396 accidental colon was present, for example, could give incorrect results.
3398 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3399 result is used (not including appendfile, where other changes are about
3402 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3403 This involved changing a lot of size variables from int to off_t. It
3404 should work with maildirs and everything.
3406 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3407 spamd dying while we are connected to it.
3409 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3412 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3413 function for BDB 4.3.
3415 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3417 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3418 the sources for PCRE have been re-arranged and more files are now
3421 PH/15 The code I had for printing potentially long long variables in PH/11
3422 above was not the best (it lost precision). The length of off_t variables
3423 is now inspected at build time, and an appropriate printing format (%ld
3424 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3425 to be "long long int" or "long int". This is needed for the internal
3426 formatting function string_vformat().
3428 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3429 the configuration file to be ":syslog", then the script "guesses" where
3430 the logs files are, rather than using the compiled in default. In our
3431 case the guess is not the same as the compiled default, so the script
3432 suddenly stopped working when I started to use syslog. The patch checks
3433 to see if log_file_path is "". If so, it attempts to read it from exim
3434 with no configuration file to get the compiled in version, before it
3435 falls back to the previous guessing code."
3437 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3438 implementing BATV in an Exim configuration. See NewStuff for the gory
3441 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3442 Makefile that are specific to HP-UX.
3444 PH/18 If the "use_postmaster" option was set for a recipient callout together
3445 with the "random" option, the postmaster address was used as the MAIL
3446 FROM address for the random test, but not for the subsequent recipient
3447 test. It is now used for both.
3449 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3450 patch removes a few documentation additions to RFC 3028, because the
3451 latest draft now contains them. It adds the new en;ascii-case comparator
3452 and a new error check for 8bit text in MIME parts. Comparator and
3453 require names are now matched exactly. I enabled the subaddress
3454 extension, but it is not well tested yet (read: it works for me)."
3456 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3457 rework some of the code of TK/09 above to avoid the hardwired use of
3458 "%lld" and "long long". Replaced the call to snprintf() with a call to
3461 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3462 records point to non-existent hosts", "retry timeout exceeded", and
3463 "retry time not reached for any host after a long failure period".
3465 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3466 experimental DomainKeys support:
3468 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3469 (2) On an error such as an illegally used "control", the wrong name for
3470 the control was given.
3472 These problems did NOT occur unless DomainKeys support was compiled.
3474 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3476 PH/24 Added ${if match_ip condition.
3478 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3479 databases so that it will be absolutely obvious if a crash occurs in the
3480 DB library. This is a regular occurrence (often caused by mis-matched
3483 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3484 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3485 picked up on a user's system that detects such things. There doesn't seem
3486 to be a gcc warning option for this - only an attribute that has to be
3487 put on the function's prototype. It seems that in Fedora Core 4 they have
3488 set this on a number of new functions. No doubt there will be more in due
3491 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3492 specify a transport. However, if an address that was verified by such a
3493 router was the subject of a callout, Exim crashed because it tried to
3494 read the rcpt_include_affixes from the non-existent transport. Now it
3495 just assumes that the setting of that option is false. This bug was
3496 introduced by 4.51/PH/31.
3498 PH/28 Changed -d+all to exclude +memory, because that information is very
3499 rarely of interest, but it makes the output a lot bigger. People tend to
3500 do -d+all out of habit.
3502 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3503 code in os-type was giving problems when libc.so lives in lib64, like on
3506 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3507 aren't the modern standard, and it seems that some systems' include files
3508 don't always have them. Exim was already checking for some of the newer
3509 ones like T_AAAA, and defining it itself. I've added checks for all the
3510 record types that Exim uses.
3512 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3513 not automatically generating a new one, as it is supposed to. This
3514 prevented TLS from working. If the file did exist, but contained invalid
3515 data, a new version was generated, as expected. It was only the case of a
3516 non-existent file that was broken.
3518 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3519 with a change in libdomainkeys > 0.64.
3521 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3522 from DNS. If the selector record carries the flag, it now has
3523 precedence over the domain-wide flag.
3525 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3527 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3528 the use of a port name that isn't defined in /etc/services) occurred, the
3529 message was deferred as in a normal delivery, and thus remained on the
3530 spool, instead of being failed because of the mua_wrapper setting. This
3531 is now fixed, and I tidied up some of the mua_wrapper messages at the
3534 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3535 the messages in a hash of arrays rather than using individual hashes.
3536 This is a bit cleaner and results in dramatic memory savings, albeit
3537 at a slight CPU cost.
3539 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3540 as requested by Marc Sherman.
3542 SC/10 Eximstats: added histograms for user specified patterns as requested
3545 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3547 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3548 fopen() in the content-scanning modules that did not already have it.
3554 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3555 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3557 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3559 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3561 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3562 to test Sieve filters that use "vacation".
3564 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3565 that changes the way the GnuTLS parameters are stored in the cache file.
3566 The new format can be generated externally. For backward compatibility,
3567 if the data in the cache doesn't make sense, Exim assumes it has read an
3568 old-format file, and it generates new data and writes a new file. This
3569 means that you can't go back to an older release without removing the
3572 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3573 work if there are any delivery delays because "one_time" forces the
3574 parent to be marked "delivered", so its unseen clone is never tried
3575 again. For this reason, Exim now forbids the simultaneous setting of
3578 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3579 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3580 entry if you want to know the details. The fix, however, neglected to
3581 consider the case when local delivery batching is involved. The test for
3582 "previously delivered" was not happening when checking to see if an
3583 address could be batched with a previous (undelivered) one; under
3584 certain circumstances this could lead to multiple deliveries to the same
3587 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3588 in its include files, and this causes problems building Exim.
3590 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3591 header_syntax) but Exim was just ignoring anything given after a slash.
3592 In particular, this caused confusion with an attempt to use "verify =
3593 reverse_host_lookup/defer_ok". An error is now given when options are
3594 supplied for verify items that do not have them. (Maybe reverse_host_
3595 lookup should have a defer_ok option, but that's a different point.)
3597 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3598 defined by RFC 821) to 2048, because there were problems with some AUTH
3599 commands, and RFC 1869 says the size should be increased for extended
3600 SMTP commands that take arguments.
3602 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3605 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3606 "unknown" error; now it says that the functionality isn't in the binary.
3608 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3609 an address' error message when a string expansion fails (syntax or
3610 whatever). Otherwise the password may appear in the log. Following change
3611 PH/42 below, there is no longer a chance of it appearing in a bounce
3614 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3616 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3617 of its addresses was ever tried. (Bugzilla bug #2.)
3619 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3620 the result incorrectly in the debug output. (It correctly added a newline
3621 to what was transported.)
3623 TF/01 Added $received_time.
3625 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3626 commented out examples of how to interface to a virus scanner and to
3627 SpamAssassin. Also added commented examples of av_scanner and
3628 spamd_address settings.
3630 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3631 and controls are allowed in which ACLs. There were a couple of minor
3632 errors. Some of the entries in the conditions table (which is a table of
3633 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3634 negation of where the condition IS allowed.
3636 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3638 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3639 header file does not have a version number, so I've had to invent a new
3640 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3641 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3642 radiusclient), but it was contributed by a Radius user.
3644 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3645 files or directories, for OpenSSL.
3647 PH/19 When an Exim process that is running as root has to create an Exim log
3648 file, it does so in a subprocess that runs as exim:exim so as to get the
3649 ownership right at creation (otherwise, other Exim processes might see
3650 the file with the wrong ownership). There was no test for failure of this
3651 fork() call, which would lead to the process getting stuck as it waited
3652 for a non-existent subprocess. Forks do occasionally fail when resources
3653 run out. I reviewed all the other calls to fork(); they all seem to check
3656 PH/20 When checking for unexpected SMTP input at connect time (before writing
3657 the banner), Exim was not dealing correctly with a non-positive return
3658 from the read() function. If the client had disconnected by this time,
3659 the result was a log entry for a synchronization error with an empty
3660 string after "input=" when read() returned zero. If read() returned -1
3661 (an event I could not check), uninitialized data bytes were printed.
3662 There were reports of junk text (parts of files, etc) appearing after
3665 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3667 PH/22 Added support for macro redefinition, and (re)definition in between
3668 driver and ACL definitions.
3670 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3671 forgetting to use the resulting value; it was using the unexpanded value.
3673 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3674 hadn't been configured. The fix is from Juergen Kreileder, who
3675 understands it better than I do:
3677 "Here's what I see happening with three configured cyrus_sasl
3678 authenticators configured (plain, login, cram-md5):
3680 On startup auth_cyrus_sasl_init() gets called for each of these.
3681 This means three calls to sasl_listmech() without a specified mech_list.
3682 => SASL tests which mechs of all available mechs actually work
3683 => three warnings about OTP not working
3684 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3686 With the patch, sasl_listmech() also gets called three times. But now
3687 SASL's mech_list option is set to the server_mech specified in the the
3688 authenticator. Or in other words, the answer from sasl_listmech()
3689 gets limited to just the mech you're testing for (which is different
3691 => the return list contains just 'plain' or 'login', 'cram-md5' or
3692 nothing depending on the value of ob->server_mech.
3694 I've just tested the patch: Authentication still works fine,
3695 unavailable mechs specified in the exim configuration are still
3696 caught, and the auth.log warnings about OTP are gone."
3698 PH/25 When debugging is enabled, the contents of the command line are added
3699 to the debugging output, even when log_selector=+arguments is not
3702 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3703 answer is "GNU", and only if the return is "GNU/something" is the answer
3706 PH/27 $acl_verify_message is now set immediately after the failure of a
3707 verification in an ACL, and so is available in subsequent modifiers. In
3708 particular, the message can be preserved by coding like this:
3710 warn !verify = sender
3711 set acl_m0 = $acl_verify_message
3713 Previously, $acl_verify_message was set only while expanding "message"
3714 and "log_message" when a very denied access.
3716 PH/28 Modified OS/os.c-Linux with
3718 -#ifndef OS_LOAD_AVERAGE
3719 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3721 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3722 nomenclature these days.)
3724 PH/29 Installed patch from the Sieve maintainer that adds the options
3725 sieve_useraddress and sieve_subaddress to the redirect router.
3727 PH/30 In these circumstances:
3728 . Two addresses routed to the same list of hosts;
3729 . First host does not offer TLS;
3730 . First host accepts first address;
3731 . First host gives temporary error to second address;
3732 . Second host offers TLS and a TLS session is established;
3733 . Second host accepts second address.
3734 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3735 and peerdn, if requested) that were in fact used only for the second
3738 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3739 attention to any local part prefix or suffix that was matched by the
3740 router that accepted the address. It now behaves in the same way as it
3741 does for delivery: the affixes are removed from the local part unless
3742 rcpt_include_affixes is set on the transport.
3744 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3745 timeout during the DATA phase of an incoming message.
3747 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3748 applied a patch sanctioned by the Sieve maintainer.
3750 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3751 the uid or gid is negative. A case of a negative gid caused this to be
3752 noticed. The fix allows for either to be negative.
3754 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3755 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3758 PH/36 The change PH/12 above was broken. Fixed it.
3760 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3761 the grounds that routing the same address twice would always produce the
3762 same answer. This might have been true once, but it is certainly no
3763 longer true now. Routing a child address may depend on the previous
3764 routing that produced that child. Some complicated redirection strategies
3765 went wrong when messages had multiple recipients, and made Exim's
3766 behaviour dependent on the order in which the addresses were given.
3768 I have moved the duplicate checking until after the routing is complete.
3769 Exim scans the addresses that are assigned to local and remote
3770 transports, and removes any duplicates. This means that more work will be
3771 done, as duplicates will always all be routed, but duplicates are
3772 presumably rare, so I don't expect this is of any significance.
3774 For deliveries to pipes, files, and autoreplies, the duplicate checking
3775 still happens during the routing process, since they are not going to be
3778 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3779 It corrects a timeout issue with spamd. This is Ian's comment: "The
3780 background is that sometimes spamd either never reads data from a
3781 connection it has accepted, or it never writes response data. The exiscan
3782 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3783 blindly assumes that writes won't block so it may never time out."
3785 PH/39 Allow G after quota size as well as K and M.
3787 PH/40 The value set for $authenticated_id in an authenticator may not contain
3788 binary zeroes or newlines because the value is written to log lines and
3789 to spool files. There was no check on this. Now the value is run through
3790 the string_printing() function so that such characters are converted to
3791 printable escape sequences.
3793 PH/41 $message_linecount is a new variable that contains the total number of
3794 lines in the message. Compare $body_linecount, which is the count for the
3797 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3798 bounce and delay warning messages, except in certain special cases, which
3801 (a) An SMTP error message from a remote host;
3802 (b) A message specified in a :fail: redirection;
3803 (c) A message specified in a "fail" command in a system filter;
3804 (d) A message specified in a FAIL return from the queryprogram router;
3805 (e) A message specified by the cannot_route_message router option.
3807 In these cases only, Exim does include the error details in bounce and
3808 warning messages. There are also a few cases where bland messages such
3809 as "unrouteable address" or "local delivery error" are given.
3811 PH/43 $value is now also set for the "else" part of a ${run expansion.
3813 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3814 being worked on, but at least Exim now implements the latest version to
3817 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3818 process to complete was treated as a delivery failure, a timeout while
3819 writing the message to the pipe was logged, but erroneously treated as a
3820 successful delivery. Such timeouts include transport filter timeouts. For
3821 consistency with the overall process timeout, these timeouts are now
3822 treated as errors, giving rise to delivery failures by default. However,
3823 there is now a new Boolean option for the pipe transport called
3824 timeout_defer, which, if set TRUE, converts the failures into defers for
3825 both kinds of timeout. A transport filter timeout is now identified in
3828 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3829 systems where "make" and "gmake" are different, calling "gmake" at top
3830 level broke things. I've arranged for the value of $(MAKE) to be passed
3831 from the Makefile to this script so that it can call the same version of
3835 A note about Exim versions 4.44 and 4.50
3836 ----------------------------------------
3838 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3839 changes of various kinds. As a consequence, a big documentation update was
3840 needed. This delayed the release for rather longer than seemed good, especially
3841 in the light of a couple of (minor) security issues. Therefore, the changes
3842 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3843 release. So 4.44 and 4.50 are in effect two different branches that both start
3846 I have left the 4.50 change log unchanged; it contains all the changes since
3847 4.43. The change log for 4.44 is below; many of its items are identical to
3848 those for 4.50. This seems to be the most sensible way to preserve the
3849 historical information.
3855 1. Minor wording change to the doc/README.SIEVE file.
3857 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3858 computation of the current number of files was incorrect.
3860 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3861 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3864 4. Give more explanation in the error message when the command for a transport
3865 filter fails to execute.
3867 5. There are several places where Exim runs a non-Exim command in a
3868 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3869 was being done only for the command run by the queryprogram router. It is
3870 now done for all such subprocesses. The other cases are: ${run, transport
3871 filters, and the commands run by the lmtp and pipe transports.
3873 6. Added CONFIGURE_GROUP build-time option.
3875 7. Some older OS have a limit of 256 on the maximum number of file
3876 descriptors. Exim was using setrlimit() to set 1000 as a large value
3877 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3878 systems. I've change it so that if it can't get 1000, it tries for 256.
3880 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3881 was an oversight, and furthermore, ever since the addition of extra
3882 controls (e.g. 4.43/32), the checks on when to allow different forms of
3883 "control" were broken. There should now be diagnostics for all cases when a
3884 control that does not make sense is encountered.
3886 9. Added the /retain_sender option to "control=submission".
3888 10. $recipients is now available in the predata ACL (oversight).
3890 11. Tidy the search cache before the fork to do a delivery from a message
3891 received from the command line. Otherwise the child will trigger a lookup
3892 failure and thereby defer the delivery if it tries to use (for example) a
3893 cached ldap connection that the parent has called unbind on.
3895 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3896 of $address_data from the recipient verification was clobbered by the
3897 sender verification.
3899 13. The value of address_data from a sender verification is now available in
3900 $sender_address_data in subsequent conditions in the ACL statement.
3902 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3904 15. Added a new option "connect=<time>" to callout options, to set a different
3907 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3908 was its contents. (It was OK if the option was not defined at all.)
3910 17. A "Completed" log line is now written for messages that are removed from
3911 the spool by the -Mrm option.
3913 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3914 information about exactly what failed.
3916 19. Added -dd to debug only the daemon process.
3918 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3919 handles timeouts, both on the server side and network timeouts. Renamed the
3920 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3922 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3923 It is now set to "smtps".
3925 22. $host_address is now set to the target address during the checking of
3926 ignore_target_hosts.
3928 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3929 being passed; this would have caused $sender_host_name to have been used if
3930 matching the list had actually called for a host name (not very likely,
3931 since this list is usually IP addresses). A host name is now passed as
3934 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3935 code that specifies a non-restarting handler (typically sigaction() in
3936 modern systems) in an attempt to fix a rare and obscure crash bug.
3938 25. Narrowed the window for a race in the daemon that could cause it to ignore
3939 SIGCHLD signals. This is not a major problem, because they are used only to
3940 wake it up if nothing else does.
3942 26. A malformed maildirsize file could cause Exim to calculate negative values
3943 for the mailbox size or file count. Odd effects could occur as a result.
3944 The maildirsize information is now recalculated if the size or filecount
3947 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3948 support for a long time. Removed HAVE_SYS_VFS_H.
3950 28. Installed the latest version of exipick from John Jetmore.
3952 29. In an address list, if the pattern was not a regular expression, an empty
3953 subject address (from a bounce message) matched only if the pattern was an
3954 empty string. Non-empty patterns were not even tested. This was the wrong
3955 because it is perfectly reasonable to use an empty address as part of a
3956 database query. An empty address is now tested by patterns that are
3957 lookups. However, all the other forms of pattern expect the subject to
3958 contain a local part and a domain, and therefore, for them, an empty
3959 address still always fails if the pattern is not itself empty.
3961 30. Exim went into a mad DNS loop when attempting to do a callout where the
3962 host was specified on an smtp transport, and looking it up yielded more
3963 than one IP address.
3965 31. Re-factored the code for checking spool and log partition space into a
3966 function that finds that data and another that does the check. The former
3967 is then used to implement four new variables: $spool_space, $log_space,
3968 $spool_inodes, and $log_inodes.
3970 32. The RFC2047 encoding function was originally intended for short strings
3971 such as real names; it was not keeping to the 75-character limit for
3972 encoded words that the RFC imposes. It now respects the limit, and
3973 generates multiple encoded words if necessary. To be on the safe side, I
3974 have increased the buffer size for the ${rfc2047: expansion operator from
3977 33. It is now permitted to omit both strings after an "if" condition; if the
3978 condition is true, the result is "true". As before, when the second string
3979 is omitted, a false condition yields an empty string. This makes it less
3980 cumbersome to write custom ACL and router conditions.
3982 34. Failure to deliver a bounce message always caused it to be frozen, even if
3983 there was an errors_to setting on the router. The errors_to setting is now
3986 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3987 canonical form (fully expanded) before being placed in
3988 $sender_host_address.
3990 36. The table in the code that translates DNS record types into text (T_A to
3991 "A" for instance) was missing entries for NS and CNAME. It is just possible
3992 that this could have caused confusion if both these types were looked up
3993 for the same domain, because the text type is used as part of Exim's
3994 per-process caching. But the chance of anyone hitting this buglet seems
3997 37. The dnsdb lookup has been extended in a number of ways.
3999 (1) There is a new type, "zns", which walks up the domain tree until it
4000 finds some nameserver records. It should be used with care.
4002 (2) There is a new type, "mxh", which is like "mx" except that it returns
4003 just the host names, not the priorities.
4005 (3) It is now possible to give a list of domains (or IP addresses) to be
4006 looked up. The behaviour when one of the lookups defers can be
4007 controlled by a keyword.
4009 (4) It is now possible to specify the separator character for use when
4010 multiple records are returned.
4012 38. The dnslists ACL condition has been extended: it is now possible to supply
4013 a list of IP addresses and/or domains to be looked up in a particular DNS
4016 39. Added log_selector=+queue_time_overall.
4018 40. When running the queue in the test harness, wait just a tad after forking a
4019 delivery process, to get repeatability of debugging output.
4021 41. Include certificate and key file names in error message when GnuTLS fails
4022 to set them up, because the GnuTLS error message doesn't include the name
4023 of the failing file when there is a problem reading it.
4025 42. Allow both -bf and -bF in the same test run.
4027 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4029 44. The "Exiscan patch" is now merged into the mainline Exim source.
4031 45. Sometimes the final signoff response after QUIT could fail to get
4032 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4033 < 0 before doing a fflush(). This bug looks as though it goes back to the
4034 introduction of TLS in release 3.20, but "sometimes" must have been rare
4035 because the tests only now provoked it.
4037 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4038 (this can affect the format of dates).
4040 47. exim_tidydb, when checking for the continued existence of a message for
4041 which it has found a message-specific retry record, was not finding
4042 messages that were in split spool directories. Consequently, it was
4043 deleting retry records that should have stayed in existence.
4045 48. Steve fixed some bugs in eximstats.
4047 49. The SPA authentication driver was not abandoning authentication and moving
4048 on to the next authenticator when an expansion was forced to fail,
4049 contradicting the general specification for all authenticators. Instead it
4050 was generating a temporary error. It now behaves as specified.
4052 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4053 (the order specifies the preference for clients). The order is now AES256,
4054 AES128, 3DES, ARCFOUR128.
4056 51. Small patch to Sieve code - explicitly set From: when generating an
4059 52. Exim crashed if a remote delivery caused a very long error message to be
4060 recorded - for instance if somebody sent an entire SpamAssassin report back
4061 as a large number of 550 error lines. This bug was coincidentally fixed by
4062 increasing the size of one of Exim's internal buffers (big_buffer) that
4063 happened as part of the Exiscan merge. However, to be on the safe side, I
4064 have made the code more robust (and fixed the comments that describe what
4067 53. Now that there can be additional text after "Completed" in log lines (if
4068 the queue_time_overall log selector is set), a one-byte patch to exigrep
4069 was needed to allow it to recognize "Completed" as not the last thing in
4072 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4073 patch that reportedly fixes this has been added. I am not expert enough to
4074 create a test for it. This is what the patch creator wrote:
4076 "I found a little strange behaviour of ldap code when working with
4077 Windows 2003 AD Domain, where users was placed in more than one
4078 Organization Units. When I tried to give exim partial DN, the exit code
4079 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4080 But simultaneously result of request was absolutely normal ldap result,
4081 so I produce this patch..."
4083 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4084 REFERENCE, so I have modified the code to exclude the patch when that macro
4087 55. Some experimental protocols are using DNS PTR records for new purposes. The
4088 keys for these records are domain names, not reversed IP addresses. The
4089 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4090 leaves it alone. Component reversal etc. now happens only for IP addresses.
4093 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4095 57. Double the size of the debug message buffer (to 2048) so that more of very
4096 long debug lines gets shown.
4098 58. The exicyclog utility now does better if the number of log files to keep
4099 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4101 59. Two changes related to the smtp_active_hostname option:
4103 (1) $smtp_active_hostname is now available as a variable.
4104 (2) The default for smtp_banner uses $smtp_active_hostname instead
4105 of $primary_hostname.
4107 60. The host_aton() function is supposed to be passed a string that is known
4108 to be a valid IP address. However, in the case of IPv6 addresses, it was
4109 not checking this. This is a hostage to fortune. Exim now panics and dies
4110 if the condition is not met. A case was found where this could be provoked
4111 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4112 components; fortuitously, this particular loophole had already been fixed
4113 by change 4.50/55 above.
4115 If there are any other similar loopholes, the new check in host_aton()
4116 itself should stop them being exploited. The report I received stated that
4117 data on the command line could provoke the exploit when Exim was running as
4118 exim, but did not say which command line option was involved. All I could
4119 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4120 running as the user.
4123 61. There was a buffer overflow vulnerability in the SPA authentication code
4124 (which came originally from the Samba project). I have added a test to the
4125 spa_base64_to_bits() function which I hope fixes it.
4128 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4129 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4131 63. The daemon start-up calls getloadavg() while still root for those OS that
4132 need the first call to be done as root, but it missed one case: when
4133 deliver_queue_load_max is set with deliver_drop_privilege. This is
4134 necessary for the benefit of the queue runner, because there is no re-exec
4135 when deliver_drop_privilege is set.
4137 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4138 This has been fixed.
4140 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4141 was in use, was not putting the data itself into the right store pool;
4142 consequently, it could be overwritten for a subsequent message in the same
4143 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4146 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4148 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4149 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4150 the function were treating the return as a boolean value, which happened to
4151 work because 0=false and not-0=true, but is not correct code.
4153 68. The host_aton() function was not handling scoped IPv6 addresses (those
4154 with, for example, "%eth0" on the end) correctly.
4156 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4157 negated items (that is, ~something) in unsigned ints. Some compilers
4158 apparently mutter when there is no cast.
4160 70. If an address verification called from an ACL failed, and did not produce a
4161 user-specific message (i.e. there was only a "system" message), nothing was
4162 put in $acl_verify_message. In this situation, it now puts the system
4165 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4166 session; change 4.31/43 added the unwanted input to the log line - except
4167 that it did not do this in the start of session case. It now does.
4169 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4170 This is wrong and can cause the other end to generate a synchronization
4171 error if it is another Exim or anything else that does the synchronization
4172 check. A QUIT command is no longer sent after a timeout.
4174 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4175 during host lookups.
4177 74. The defer_ok option of callout verification was not working if it was used
4178 when verifying addresses in header lines, that is, for this case:
4180 verify = header_sender/callout=defer_ok
4182 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4183 those file descriptors could be used for SMTP connections. If anything
4184 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4185 could be sent to the SMTP client, causing chaos. The daemon now opens
4186 stdin, stdout, and stderr to /dev/null when it puts itself into the
4189 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4190 log by default. The user can override this with suitable Perl magic.
4192 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4193 the log message when discard triggers, was not working for the DATA ACL or
4194 for the non-SMTP ACL.
4196 78. Error message wording change in sieve.c.
4198 79. If smtp_accept_max_per_host was set, the number of connections could be
4199 restricted to fewer than expected, because the daemon was trying to set up
4200 a new connection before checking whether the processes handling previous
4201 connections had finished. The check for completed processes is now done
4202 earlier. On busy systems, this bug wouldn't be noticed because something
4203 else would have woken the daemon, and it would have reaped the completed
4206 80. If a message was submitted locally by a user whose login name contained one
4207 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4208 It caused a spool format error. I have fixed the spool reading code. A
4209 related problem was that the "from" clause in the Received: line became
4210 illegal because of the space(s). It is now covered by ${quote_local_part.
4212 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4215 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4217 83. Patch from Sieve maintainer to fix unterminated string problem in
4218 "vacation" handling.
4220 84. Some minor changes to the Linux configuration files to help with other
4221 OS variants using glibc.
4223 85. One more patch for Sieve to update vacation handling to latest spec.
4226 ----------------------------------------------------
4227 See the note above about the 4.44 and 4.50 releases.
4228 ----------------------------------------------------
4234 1. Change 4.43/35 introduced a bug that caused file counts to be
4235 incorrectly computed when quota_filecount was set in an appendfile
4238 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4239 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4242 3. Give more explanation in the error message when the command for a transport
4243 filter fails to execute.
4245 4. There are several places where Exim runs a non-Exim command in a
4246 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4247 was being done only for the command run by the queryprogram router. It is
4248 now done for all such subprocesses. The other cases are: ${run, transport
4249 filters, and the commands run by the lmtp and pipe transports.
4251 5. Some older OS have a limit of 256 on the maximum number of file
4252 descriptors. Exim was using setrlimit() to set 1000 as a large value
4253 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4254 systems. I've change it so that if it can't get 1000, it tries for 256.
4256 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4257 was an oversight, and furthermore, ever since the addition of extra
4258 controls (e.g. 4.43/32), the checks on when to allow different forms of
4259 "control" were broken. There should now be diagnostics for all cases when a
4260 control that does not make sense is encountered.
4262 7. $recipients is now available in the predata ACL (oversight).
4264 8. Tidy the search cache before the fork to do a delivery from a message
4265 received from the command line. Otherwise the child will trigger a lookup
4266 failure and thereby defer the delivery if it tries to use (for example) a
4267 cached ldap connection that the parent has called unbind on.
4269 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4270 of $address_data from the recipient verification was clobbered by the
4271 sender verification.
4273 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4274 was its contents. (It was OK if the option was not defined at all.)
4276 11. A "Completed" log line is now written for messages that are removed from
4277 the spool by the -Mrm option.
4279 12. $host_address is now set to the target address during the checking of
4280 ignore_target_hosts.
4282 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4283 being passed; this would have caused $sender_host_name to have been used if
4284 matching the list had actually called for a host name (not very likely,
4285 since this list is usually IP addresses). A host name is now passed as
4288 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4289 code that specifies a non-restarting handler (typically sigaction() in
4290 modern systems) in an attempt to fix a rare and obscure crash bug.
4292 15. Narrowed the window for a race in the daemon that could cause it to ignore
4293 SIGCHLD signals. This is not a major problem, because they are used only to
4294 wake it up if nothing else does.
4296 16. A malformed maildirsize file could cause Exim to calculate negative values
4297 for the mailbox size or file count. Odd effects could occur as a result.
4298 The maildirsize information is now recalculated if the size or filecount
4301 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4302 support for a long time. Removed HAVE_SYS_VFS_H.
4304 18. Updated exipick to current release from John Jetmore.
4306 19. Allow an empty sender to be matched against a lookup in an address list.
4307 Previously the only cases considered were a regular expression, or an
4310 20. Exim went into a mad DNS lookup loop when doing a callout where the
4311 host was specified on the transport, if the DNS lookup yielded more than
4314 21. The RFC2047 encoding function was originally intended for short strings
4315 such as real names; it was not keeping to the 75-character limit for
4316 encoded words that the RFC imposes. It now respects the limit, and
4317 generates multiple encoded words if necessary. To be on the safe side, I
4318 have increased the buffer size for the ${rfc2047: expansion operator from
4321 22. Failure to deliver a bounce message always caused it to be frozen, even if
4322 there was an errors_to setting on the router. The errors_to setting is now
4325 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4326 canonical form (fully expanded) before being placed in
4327 $sender_host_address.
4329 24. Updated eximstats to version 1.33
4331 25. Include certificate and key file names in error message when GnuTLS fails
4332 to set them up, because the GnuTLS error message doesn't include the name
4333 of the failing file when there is a problem reading it.
4335 26. Expand error message when OpenSSL has problems setting up cert/key files.
4338 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4339 (this can affect the format of dates).
4341 28. exim_tidydb, when checking for the continued existence of a message for
4342 which it has found a message-specific retry record, was not finding
4343 messages that were in split spool directories. Consequently, it was
4344 deleting retry records that should have stayed in existence.
4346 29. eximstats updated to version 1.35
4347 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4348 1.35 - bugfix such that pie charts by volume are generated correctly
4350 30. The SPA authentication driver was not abandoning authentication and moving
4351 on to the next authenticator when an expansion was forced to fail,
4352 contradicting the general specification for all authenticators. Instead it
4353 was generating a temporary error. It now behaves as specified.
4355 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4356 (the order specifies the preference for clients). The order is now AES256,
4357 AES128, 3DES, ARCFOUR128.
4359 31. Small patch to Sieve code - explicitly set From: when generating an
4362 32. Exim crashed if a remote delivery caused a very long error message to be
4363 recorded - for instance if somebody sent an entire SpamAssassin report back
4364 as a large number of 550 error lines. This bug was coincidentally fixed by
4365 increasing the size of one of Exim's internal buffers (big_buffer) that
4366 happened as part of the Exiscan merge. However, to be on the safe side, I
4367 have made the code more robust (and fixed the comments that describe what
4370 33. Some experimental protocols are using DNS PTR records for new purposes. The
4371 keys for these records are domain names, not reversed IP addresses. The
4372 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4373 leaves it alone. Component reversal etc. now happens only for IP addresses.
4376 34. The host_aton() function is supposed to be passed a string that is known
4377 to be a valid IP address. However, in the case of IPv6 addresses, it was
4378 not checking this. This is a hostage to fortune. Exim now panics and dies
4379 if the condition is not met. A case was found where this could be provoked
4380 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4381 components; fortuitously, this particular loophole had already been fixed
4382 by change 4.50/55 or 4.44/33 above.
4384 If there are any other similar loopholes, the new check in host_aton()
4385 itself should stop them being exploited. The report I received stated that
4386 data on the command line could provoke the exploit when Exim was running as
4387 exim, but did not say which command line option was involved. All I could
4388 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4389 running as the user.
4392 35. There was a buffer overflow vulnerability in the SPA authentication code
4393 (which came originally from the Samba project). I have added a test to the
4394 spa_base64_to_bits() function which I hope fixes it.
4397 36. The daemon start-up calls getloadavg() while still root for those OS that
4398 need the first call to be done as root, but it missed one case: when
4399 deliver_queue_load_max is set with deliver_drop_privilege. This is
4400 necessary for the benefit of the queue runner, because there is no re-exec
4401 when deliver_drop_privilege is set.
4403 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4404 was in use, was not putting the data itself into the right store pool;
4405 consequently, it could be overwritten for a subsequent message in the same
4406 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4409 38. Sometimes the final signoff response after QUIT could fail to get
4410 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4411 < 0 before doing a fflush(). This bug looks as though it goes back to the
4412 introduction of TLS in release 3.20, but "sometimes" must have been rare
4413 because the tests only now provoked it.
4419 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4420 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4421 Now it returns an integer. A number of calls were still expecting a T/F
4422 return. Fortuitously, in all cases, the tests worked in OK situations,
4423 which is the norm. However, things would have gone wrong on any write
4424 failures on the smtp file descriptor. This function is used when sending
4425 messages over SMTP and also when doing verify callouts.
4427 2. When Exim is called to do synchronous delivery of a locally submitted
4428 message (the -odf or -odi options), it no longer closes stderr before doing
4431 3. Implemented the mua_wrapper option.
4433 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4435 5. Implemented the functions header_remove(), header_testname(),
4436 header_add_at_position(), and receive_remove_recipient(), and exported them
4439 6. If an ACL "warn" statement specified the addition of headers, Exim already
4440 inserted X-ACL-Warn: at the start if there was no header name. However, it
4441 was not making this test for the second and subsequent header lines if
4442 there were newlines in the string. This meant that an invalid header could
4443 be inserted if Exim was badly configured.
4445 7. Allow an ACL "warn" statement to add header lines at the start or after all
4446 the Received: headers, as well as at the end.
4448 8. Added the rcpt_4xx retry error code.
4450 9. Added postmaster_mailfrom=xxx to callout verification option.
4452 10. Added mailfrom=xxxx to the callout verification option, for verify=
4455 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4456 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4458 12. Inserted some casts to stop certain compilers warning when using pointer
4459 differences as field lengths or precisions in printf-type calls (mostly
4460 affecting debugging statements).
4462 13. Added optional readline() support for -be (dynamically loaded).
4464 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4465 same clock tick as a message's arrival, so that its received time was the
4466 same as the "first fail" time on the retry record, and that message
4467 remained on the queue past the ultimate address timeout, every queue runner
4468 would try a delivery (because it was past the ultimate address timeout) but
4469 after another failure, the ultimate address timeout, which should have then
4470 bounced the address, did not kick in. This was a "< instead of <=" error;
4471 in most cases the first failure would have been in the next clock tick
4472 after the received time, and all would be well.
4474 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4475 being recognized when the domain list was tested by the match_domain
4476 condition in an expansion string.
4478 16. Added the ${str2b64: operator.
4480 17. Exim was always calling setrlimit() to set a large limit for the number of
4481 processes, without checking whether the existing limit was already
4482 adequate. (It did check for the limit on file descriptors.) Furthermore,
4483 errors from getrlimit() and setrlimit() were being ignored. Now they are
4484 logged to the main and panic logs, but Exim does carry on, to try to do its
4485 job under whatever limits there are.
4487 18. Imported PCRE 5.0.
4489 19. Trivial typo in log message " temporarily refused connection" (the leading
4492 20. If the log selector return_path_on_delivery was set and an address was
4493 redirected to /dev/null, the delivery process crashed because it assumed
4494 that a return path would always be set for a "successful" delivery. In this
4495 case, the whole delivery is bypassed as an optimization, and therefore no
4498 21. Internal re-arrangement: the function for sending a challenge and reading
4499 a response while authentication was assuming a zero-terminated challenge
4500 string. It's now changed to take a pointer and a length, to allow for
4501 binary data in such strings.
4503 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4505 23. Exim was not respecting finduser_retries when seeking the login of the
4506 uid under which it was called; it was always trying 10 times. (The default
4507 setting of finduser_retries is zero.) Also, it was sleeping after the final
4508 failure, which is pointless.
4510 24. Implemented tls_on_connect_ports.
4512 25. Implemented acl_smtp_predata.
4514 26. If the domain in control=submission is set empty, Exim assumes that the
4515 authenticated id is a complete email address when it generates From: or
4516 Sender: header lines.
4518 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4519 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4520 chown and chgrp in /bin and hostname in /usr/bin.
4522 28. Exim was keeping the "process log" file open after each use, just as it
4523 does for the main log. This opens the possibility of it remaining open for
4524 long periods when the USR1 signal hits a daemon. Occasional processlog
4525 errors were reported, that could have been caused by this. Anyway, it seems
4526 much more sensible not to leave this file open at all, so that is what now
4529 29. The long-running daemon process does not normally write to the log once it
4530 has entered its main loop, and it closes the log before doing so. This is
4531 so that log files can straightforwardly be renamed and moved. However,
4532 there are a couple of unusual error situations where the daemon does write
4533 log entries, and I had neglected to close the log afterwards.
4535 30. The text of an SMTP error response that was received during a remote
4536 delivery was being truncated at 512 bytes. This is too short for some of
4537 the long messages that one sometimes sees. I've increased the limit to
4540 31. It is now possible to make retry rules that apply only when a message has a
4541 specific sender, in particular, an empty sender.
4543 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4544 it possible to be selective about when SMTP synchronization is enforced.
4546 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4548 32. Implemented hosts_connection_nolog.
4550 33. Added an ACL for QUIT.
4552 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4555 35. Added mailbox_size and mailbox_filecount to appendfile.
4557 36. Added control = no_multiline_responses to ACLs.
4559 37. There was a bug in the logic of the code that waits for the clock to tick
4560 in the case where the clock went backwards by a substantial amount such
4561 that the microsecond fraction of "now" was more than the microsecond
4562 fraction of "then" (but the whole seconds number was less).
4564 38. Added support for the libradius Radius client library this is found on
4565 FreeBSD (previously only the radiusclient library was supported).
4571 1. When certain lookups returned multiple values in the form name=value, the
4572 quoting of the values was not always being done properly. Specifically:
4573 (a) If the value started with a double quote, but contained no whitespace,
4575 (b) If the value contained whitespace other than a space character (i.e.
4576 tabs or newlines or carriage returns) it was not quoted.
4577 This fix has been applied to the mysql and pgsql lookups by writing a
4578 separate quoting function and calling it from the lookup code. The fix
4579 should probably also be applied to nisplus, ibase and oracle lookups, but
4580 since I cannot test any of those, I have not disturbed their existing code.
4582 2. A hit in the callout cache for a specific address caused a log line with no
4583 reason for rejecting RCPT. Now it says "Previous (cached) callout
4584 verification failure".
4586 3. There was an off-by-one bug in the queryprogram router. An over-long
4587 return line was truncated at 256 instead of 255 characters, thereby
4588 overflowing its buffer with the terminating zero. As well as fixing this, I
4589 have increased the buffer size to 1024 (and made a note to document this).
4591 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4592 when Exim is waiting for an SMTP response from a remote server, Exim
4593 restarts its select() call on the socket, thereby resetting its timeout.
4594 This is not a problem when such interrupts are rare. Somebody set up a cron
4595 job to run exiwhat every 2 minutes, which is less than the normal select()
4596 timeout (5 or 10 minutes). This meant that the select() timeout never
4597 kicked in because it was always reset. I have fixed this by comparing the
4598 time when an interrupt arrives with the time at the start of the first call
4599 to select(). If more time than the timeout has elapsed, the interrupt is
4600 treated as a timeout.
4602 5. Some internal re-factoring in preparation for the addition of Sieve
4603 extensions (by MH). In particular, the "personal" test is moved to a
4604 separate function, and given an option for scanning Cc: and Bcc: (which is
4605 not set for Exim filters).
4607 6. When Exim created an email address using the login of the caller as the
4608 local part (e.g. when creating a From: or Sender: header line), it was not
4609 quoting the local part when it contained special characters such as @.
4611 7. Installed new OpenBSD configuration files.
4613 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4614 try to make them clearer.
4616 9. Callout options, other than the timeout value, were being ignored when
4617 verifying sender addresses in header lines. For example, when using
4619 verify = header_sender/callout=no_cache
4621 the cache was (incorrectly) being used.
4623 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4624 only the Cygwin environment.
4626 11. When return_path_on_delivery was set as a log selector, if different remote
4627 addresses in the same message used different return paths and parallel
4628 remote delivery occurred, the wrong values would sometimes be logged.
4629 (Whenever a remote delivery process finished, the return path value from
4630 the most recently started remote delivery process was logged.)
4632 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4633 lines when AUTH and/or TLS are in use. This is the "received protocol"
4634 field. Exim used to use "asmtp" for authenticated SMTP, without any
4635 indication (in the protocol name) for TLS use. Now it follows the RFC and
4636 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4637 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4638 names appear in log lines as well as in Received: header lines.
4640 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4641 extensions, and comparison tests, and to fix some bugs.
4643 14. Changes to the "personal" filter test:
4645 (1) The test was buggy in that it was just doing the equivalent of
4646 "contains" tests on header lines. For example, if a user's address was
4647 anne@some.where, the "personal" test would incorrectly be true for
4649 To: susanne@some.where
4651 This test is now done by extracting each address from the header in turn,
4652 and checking the entire address. Other tests that are part of "personal"
4653 are now done using regular expressions (for example, to check local parts
4654 of addresses in From: header lines).
4656 (2) The list of non-personal local parts in From: addresses has been
4657 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4658 taken from the Sieve specification recommendations.
4660 (3) If the message contains any header line starting with "List-" it is
4661 treated as non-personal.
4663 (4) The test for "circular" in the Subject: header line has been removed
4664 because it now seems ill-conceived.
4666 15. Minor typos in src/EDITME comments corrected.
4668 16. Installed latest exipick from John Jetmore.
4670 17. If headers_add on a router specified a text string that was too long for
4671 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4672 of string_sprintf() is now avoided.
4674 18. $message_body_size was not set (it was always zero) when running the DATA
4675 ACL and the local_scan() function.
4677 19. For the "mail" command in an Exim filter, no default was being set for
4678 the once_repeat time, causing a random time value to be used if "once" was
4679 specified. (If the value happened to be <= 0, no repeat happened.) The
4680 default is now 0s, meaning "never repeat". The "vacation" command was OK
4681 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4682 (I found it when inspecting the code).
4684 20. There is now an overall timeout for performing a callout verification. It
4685 defaults to 4 times the callout timeout, which applies to individual SMTP
4686 commands during the callout. The overall timeout applies when there is more
4687 than one host that can be tried. The timeout is checked before trying the
4688 next host. This prevents very long delays if there are a large number of
4689 hosts and all are timing out (e.g. when the network connections are timing
4690 out). The value of the overall timeout can be changed by specifying an
4691 additional sub-option for "callout", called "maxwait". For example:
4693 verify = sender/callout=5s,maxwait=20s
4695 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4696 to the end before writing, but this should make it even safer).
4698 22. Exim was forgetting that it had advertised PIPELINING for the second and
4699 subsequent messages on an SMTP connection. It was also not resetting its
4700 memory on STARTTLS and an internal HELO.
4702 23. When Exim logs an SMTP synchronization error within a session, it now
4703 records whether PIPELINING has been advertised or not.
4705 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4706 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4707 rather than long int.
4709 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4711 26. Added the never_mail option to autoreply.
4717 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4718 crash if the getsockname() call failed; this can happen if a connection is
4719 closed very soon after it is established. The problem was simply in the
4720 order in which certain operations were done, causing Exim to try to write
4721 to the SMTP stream before it had set up the file descriptor. The bug has
4722 been fixed by making things happen in the correct order.
4728 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4729 before the connection was closed, thus losing the rejection response.
4731 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4732 some early Solaris releases, but causes trouble in current releases where
4733 socklen_t is defined.
4735 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4738 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4741 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4742 that releasing the top of it at the end releases what was used for sub-
4743 expansions (unless the block got too big). However, discard this block if
4744 the first thing is a variable or header, so that we can use its block when
4745 it is dynamic (useful for very large $message_headers, for example).
4747 6. Lookups now cache *every* query, not just the most recent. A new, separate
4748 store pool is used for this. It can be recovered when all lookup caches are
4749 flushed. Lookups now release memory at the end of their result strings.
4750 This has involved some general refactoring of the lookup sources.
4752 7. Some code has been added to the store_xxx() functions to reduce the amount
4753 of flapping under certain conditions.
4755 8. log_incoming_interface used to affect only the <= reception log lines. Now
4756 it causes the local interface and port to be added to several more SMTP log
4757 lines, for example "SMTP connection from", and rejection lines.
4759 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4761 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4763 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4764 could be overwritten at the end of the current message (or the start of a
4765 new message if it was set in a HELO ACL). The value is now preserved for
4766 the duration of the SMTP connection.
4768 12. If a transport had a headers_rewrite setting, and a matching header line
4769 contained an unqualified address, that address was qualified, even if it
4770 did not match any rewriting rules. The underlying bug was that the values
4771 of the flags that permit the existence of unqualified sender and recipient
4772 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4773 non-local messages, and by -bnq for local messages) were not being
4774 preserved with the message after it was received.
4776 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4777 "next input=" as part of the text comprising the host identity instead of
4778 the correct text. The code was using the same buffer for two different
4779 strings. However, depending on which order the printing function evaluated
4780 its arguments, the bug did not always show up. Under Linux, for example, my
4781 test suite worked just fine.
4783 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4784 doesn't work with some older versions of Perl. It has been changed to "my",
4785 which in any case is probably the better facility to use.
4787 15. A really picky compiler found some instances of statements for creating
4788 error messages that either had too many or two few arguments for the format
4791 16. The size of the buffer for calls to the DNS resolver has been increased
4792 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4793 for addresses that have a lot of PTR records. This alleviates a problem; it
4794 does not fully solve it.
4796 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4797 buffer now truncates the list and logs the incident, which is the same
4798 action as happens when Exim is looking up a host name and its aliases.
4799 Previously in this situation something unpredictable would happen;
4800 sometimes it was "internal error: store_reset failed".
4802 18. If a server dropped the connection unexpectedly when an Exim client was
4803 using GnuTLS and trying to read a response, the client delivery process
4804 crashed while trying to generate an error log message.
4806 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4807 string, for example:
4809 warn message = H1: something\nH2: something
4811 the text was added as a single header line from Exim's point of view
4812 though it ended up OK in the delivered message. However, searching for the
4813 second and subsequent header lines using $h_h2: did not work. This has been
4814 fixed. Similarly, if a system filter added multiple headers in this way,
4815 the routers could not see them.
4817 20. Expanded the error message when iplsearch is called with an invalid key to
4818 suggest using net-iplsearch in a host list.
4820 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4821 ACLs are no longer actually imposed (and a message to that effect is
4824 22. If a "gecos" field in a passwd entry contained escaped characters, in
4825 particular, if it contained a \" sequence, Exim got it wrong when building
4826 a From: or a Sender: header from that name. A second bug also caused
4827 incorrect handling when an unquoted " was present following a character
4828 that needed quoting.
4830 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4831 was not being matched caselessly.
4833 24. Arranged for all hyphens in the exim.8 source to be escaped with
4836 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4837 properly. Recipient callout cache records were still being keyed to include
4838 the sender, even when use_sender was set false. This led to far more
4839 callouts that were necessary. The sender is no longer included in the key
4840 when use_sender is false.
4842 26. Added "control = submission" modifier to ACLs.
4844 27. Added the ${base62d: operator to decode base 62 numbers.
4846 28. dnsdb lookups can now access SRV records.
4848 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4849 the configuration file.
4851 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4852 (-v) mode. This makes the output for a verbose queue run more intelligible.
4854 31. Added a use_postmaster feature to recipient callouts.
4856 32. Added the $body_zerocount variable, containing the number of binary zero
4857 bytes in the message body.
4859 33. The time of last modification of the "new" subdirectory is now used as the
4860 "mailbox time last read" when there is a quota error for a maildir
4863 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4865 35. Added +ignore_unknown as a special item in host lists.
4867 36. Code for decoding IPv6 addresses in host lists is now included, even if
4868 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4869 address was recognized as an IP address, but was then not correctly decoded
4870 into binary, causing unexpected and incorrect effects when compared with
4877 1. Very minor rewording of debugging text in manualroute to say "list of
4878 hosts" instead of "hostlist".
4880 2. If verify=header_syntax was set, and a header line with an unqualified
4881 address (no domain) and a large number of spaces between the end of the
4882 name and the colon was received, the reception process suffered a buffer
4883 overflow, and (when I tested it) crashed. This was caused by some obsolete
4884 code that should have been removed. The fix is to remove it!
4886 3. When running in the test harness, delay a bit after writing a bounce
4887 message to get a bit more predictability in the log output.
4889 4. Added a call to search_tidyup() just before forking a reception process. In
4890 theory, someone could use a lookup in the expansion of smtp_accept_max_
4891 per_host which, without the tidyup, could leave open a database connection.
4893 5. Added the variables $recipient_data and $sender_data which get set from a
4894 lookup success in an ACL "recipients" or "senders" condition, or a router
4895 "senders" option, similar to $domain_data and $local_part_data.
4897 6. Moved the writing of debug_print from before to after the "senders" test
4900 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4901 problems for message scanning, either using a data ACL, or using
4902 local_scan() because the Received: header was not generated till after they
4903 were called (in order to set the time as the time of reception completion).
4904 I have revised the way this works. The header is now generated after the
4905 body is received, but before the ACL or local_scan() are called. After they
4906 are run, the timestamp in the header is updated.
4912 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4913 before starting a queue runner without re-exec. This happened only when
4914 deliver_drop_privilege was set or when the Exim user was set to root. The
4915 effect of the bug was that timeouts during subsequent deliveries caused
4916 crashes instead of being properly handled. The handler is now left at its
4917 default (and expected) setting.
4919 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4920 message, again when deliver_drop_privilege is set or Exim is run as root.
4921 The bug described in (1) was not present in this case, but the tidying up
4922 of the other signals was missing. I have made the two cases consistent.
4924 3. The ignore_target_hosts setting on a manualroute router was being ignored
4925 for hosts that were looked up using the /MX notation.
4927 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4930 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4931 operated on the sender address. After changing the $sender_address to <>
4932 for the sender address verify, Exim was re-instated it as the original
4933 (before rewriting) address, but remembering that it had rewritten it, so it
4934 wasn't rewriting it again. This bug also had the effect of breaking the
4935 sender address verification caching when the sender address was rewritten.
4937 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4938 This has been changed so that if the ip literal address matches
4939 ignore_target_hosts, the router declines.
4941 7. Added expansion conditions match_domain, match_address, and match_local_
4942 part (NOT match_host).
4944 8. The placeholder for the Received: header didn't have a length field set.
4946 9. Added code to Exim itself and to exim_lock to test for a specific race
4947 condition that could lead to file corruption when using MBX delivery. The
4948 issue is with the lockfile that is created in /tmp. If this file is removed
4949 after a process has opened it but before that process has acquired a lock,
4950 there is the potential for a second process to recreate the file and also
4951 acquire a lock. This could lead to two Exim processes writing to the file
4952 at the same time. The added code performs the same test as UW imapd; it
4953 checks after acquiring the lock that its file descriptor still refers to
4954 the same named file.
4956 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4957 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4960 11. Added the smtp_active_hostname option. If used, this will typically be made
4961 to depend on the incoming interface address. Because $interface_address is
4962 not set up until the daemon has forked a reception process, error responses
4963 that can happen earlier (such as "too many connections") no longer contain
4966 12. If an expansion in a condition on a "warn" statement fails because a lookup
4967 defers, the "warn" statement is abandoned, and the next ACL statement is
4968 processed. Previously this caused the whole ACL to be aborted.
4970 13. Added the iplsearch lookup type.
4972 14. Added ident_timeout as a log selector.
4974 15. Added tls_certificate_verified as a log selector.
4976 16. Added a global option tls_require_ciphers (compare the smtp transport
4977 option of the same name). This controls incoming TLS connections.
4979 17. I finally figured out how to make tls_require_ciphers do a similar thing
4980 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4981 before starting the TLS session.
4983 18. Tabs are now shown as \t in -bP output.
4985 19. If the log selector return_path_on_delivery was set, Exim crashed when
4986 bouncing a message because it had too many Received: header lines.
4988 20. If two routers both had headers_remove settings, and the first one included
4989 a superfluous trailing colon, the final name in the first list and the
4990 first name in the second list were incorrectly joined into one item (with a
4991 colon in the middle).
4997 1. Added -C and -D options to the exinext utility, mainly to make it easier
4998 to include in the automated testing, but these could be helpful when
4999 multiple configurations are in use.
5001 2. The exinext utility was not formatting the output nicely when there was
5002 an alternate port involved in the retry record key, nor when there was a
5003 message id as well (for retries that were specific to a specific message
5004 and a specific host). It was also confused by IPv6 addresses, because of
5005 the additional colons they contain. I have fixed the IPv4 problem, and
5006 patched it up to do a reasonable job for IPv6.
5008 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5009 delivery, the log line now contains "pipelined" if PIPELINING was used.
5011 4. An SMTP transport process used to panic and die if the bind() call to set
5012 an explicit outgoing interface failed. This has been changed; it is now
5013 treated in the same way as a connect() failure.
5015 5. A reference to $sender_host_name in the part of a conditional expansion
5016 that was being skipped was still causing a DNS lookup. This no longer
5019 6. The def: expansion condition was not recognizing references to header lines
5020 that used bh_ and bheader_.
5022 7. Added the _cache feature to named lists.
5024 8. The code for checking quota_filecount in the appendfile transport was
5025 allowing one more file than it should have been.
5027 9. For compatibility with Sendmail, the command line option
5035 and sets the incoming protocol and host name (for trusted callers). The
5036 host name and its colon can be omitted when only the protocol is to be set.
5037 Note the Exim already has two private options, -pd and -ps, that refer to
5038 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5039 "s", but I don't think that's a major issue.
5041 10. A number of refactoring changes to the code, none of which should affect
5044 (a) The number of logging options was getting close to filling up the
5045 32-bit word that was used as a bit map. I have split them into two classes:
5046 those that are passed in the argument to log_write(), and those that are
5047 only ever tested independently outside of that function. These are now in
5048 separate 32-bit words, so there is plenty of room for expansion again.
5049 There is no change in the user interface or the logging behaviour.
5051 (b) When building, for example, log lines, the code previously used a
5052 macro that called string_cat() twice, in order to add two strings. This is
5053 not really sufficiently general. Furthermore, there was one instance where
5054 it was actually wrong because one of the argument was used twice, and in
5055 one call a function was used. (As it happened, calling the function twice
5056 did not affect the overall behaviour.) The macro has been replaced by a
5057 function that can join an arbitrary number of extra strings onto a growing
5060 (c) The code for expansion conditions now uses a table and a binary chop
5061 instead of a serial search (which was left over from when there were very
5062 few conditions). Also, it now recognizes conditions like "pam" even when
5063 the relevant support is not compiled in: a suitably worded error message is
5064 given if an attempt is made to use such a condition.
5066 11. Added ${time_interval:xxxxx}.
5068 12. A bug was causing one of the ddress fields not to be passed back correctly
5069 from remote delivery subprocesses. The field in question was not being
5070 subsequently used, so this caused to problems in practice.
5072 13. Added new log selectors queue_time and deliver_time.
5074 14. Might have fixed a bug in maildirsizefile handling that threw up
5075 "unexpected character" debug warnings, and recalculated the data
5076 unnecessarily. In any case, I expanded the warning message to give more
5079 15. Added the message "Restricted characters in address" to the statements in
5080 the default ACL that block characters like @ and % in local parts.
5082 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5083 Three changes have been made:
5085 (a) There was a serious bug; a negative response to MAIL caused the whole
5086 recipient domain to be cached as invalid, thereby blocking all messages
5087 to all local parts at the same domain, from all senders. This bug has
5088 been fixed. The domain is no longer cached after a negative response to
5089 MAIL if the sender used is not empty.
5091 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5094 (c) A new callout option, "use_sender" has been added for people who want
5095 the modified behaviour.
5101 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5104 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5105 indeed breaks things for older releases.
5107 3. Added additional logging to the case where there is a problem reading data
5108 from a filter that is running in a subprocess using a pipe, in order to
5109 try to track down a specific problem.
5111 4. Testing facility fudge: when running in the test harness and attempting
5112 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5113 getting "No route to host". Convert this to a timeout.
5115 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5118 6. Some OS don't have socklen_t but use size_t instead. This affects the
5119 fifth argument of getsockopt() amongst other things. This is now
5120 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5121 can be set for individual OS. I have set it for SunOS5, OSF1, and
5122 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5123 some earlier ones do not.
5125 7. Change 4.30/15 was not doing the test caselessly.
5127 8. The standard form for an IPv6 address literal was being rejected by address
5128 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5129 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5130 this, as well as the form without the "IPv6" on the front (but only when
5131 address literals are enabled, of course).
5133 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5135 10. Exim crashed if a message with an empty sender address specified by -f
5136 encountered a router with an errors_to setting. This could be provoked only
5137 by a command such as
5141 where an empty string was supplied; "<>" did not hit this bug.
5143 11. Installed PCRE release 4.5.
5145 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5146 remained set. It is now erased.
5148 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5149 times from message ids (which are base 36 rather than the normal 62).
5151 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5152 were being counted as actual protocol errors, and logged if the log
5153 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5154 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5155 and DATA following a set of rejected RCPTs do not count as protocol errors.
5156 In other words, Exim assumes they were pipelined, though this may not
5157 actually be the case. Of course, in all cases the client gets an
5158 appropriate error code.
5160 15. If a lookup fails in an ACL condition, a message about the failure may
5161 be available; it is used if testing the ACL cannot continue, because most
5162 such messages specify what the cause of the deferral is. However, some
5163 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5164 that caused an old message to be retained and used if a later statement
5165 caused a defer, replacing the real cause of the deferral.
5167 16. If an IP address had so many PTR records that the DNS lookup buffer
5168 was not large enough to hold them, Exim could crash while trying to process
5169 the truncated data. It now detects and logs this case.
5171 17. Further to 4.21/58, another change has been made: if (and only if) the
5172 first line of a message (the first header line) ends with CRLF, a bare LF
5173 in a subsequent header line has a space inserted after it, so as not to
5174 terminate the header.
5176 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5177 unnecessarily, used atoi() instead of strtol(), and didn't check the
5178 termination when getting file sizes from file names by regex.
5180 19. Completely re-implemented the support for maildirsize files, in the light
5181 of a number of problems with the previous contributed implementation
5182 (4.30/29). In particular:
5184 . If the quota is zero, the maildirsize file is maintained, but no quota is
5187 . If the maildir directory does not exist, it is created before any attempt
5188 to write a maildirsize file.
5190 . The quota value in the file is just a cache; if the quota is changed in
5191 the transport, the new value overrides.
5193 . A regular expression is available for excluding directories from the
5196 20. The autoreply transport checks the characters in options that define the
5197 message's headers; it allows continued headers, but it was checking with
5198 isspace() after an embedded newline instead of explicitly looking for a
5201 21. If all the "regular" hosts to which an address was routed had passed their
5202 expiry times, and had not reached their retry times, the address was
5203 bounced, even if fallback hosts were defined. Now Exim should go on to try
5206 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5207 equivalent code in the SMTP transport. Some hosts send humungous responses
5208 to HELO/EHLO, more than 1024 it seems.
5210 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5211 gives compiler warnings in some environments. I've now done it "properly",
5214 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5215 (because of problems with the built-in one) was declared to return uschar *
5216 instead of char *, causing compiler failure.
5218 25. Fixed a file descriptor leak when processing alias/forward files.
5220 26. Fixed a minor format string issue in dbfn.c.
5222 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5224 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5225 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5228 29. When a sender address is verified, it is cached, to save repeating the test
5229 when there is more than one recipient in a message. However, when the
5230 verification involves a callout, it is possible for different callout
5231 options to be set for different recipients. It is too complicated to keep
5232 track of this in the cache, so now Exim always runs a verification when a
5233 callout is required, relying on the callout cache for the optimization.
5234 The overhead is duplication of the address routing, but this should not be
5237 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5238 to be verified with callout=postmaster, and the main callout worked but the
5239 postmaster check failed, the verification correctly failed. However, if a
5240 subsequent RCPT command asked for sender verification *without* the
5241 postmaster check, incorrect caching caused this verification also to fail,
5244 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5245 it was not caching the DNS options (qualify_single, search_parents) that
5246 were used when the lookup failed. A subsequent lookup with different
5247 options therefore always gave the same answer, though there were cases
5248 where it should not have. (Example: a "domains = !$mx_any" option on a
5249 dnslookup router: the "domains" option is always processed without any
5250 widening, but the router might have qualify_single set.) Now Exim uses the
5251 cached value only when the same options are set.
5253 32. Added John Jetmore's "exipick" utility to the distribution.
5255 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5256 than a timeout (e.g. a certificate is required, and is not provided), an
5257 Exim server now closes the connection immediately. Previously it waited for
5258 the client to close - but if the client is SSL, it seems that they each
5259 wait for each other, leading to a delay before one of them times out.
5261 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5262 maintained 0.8.x compatibility because I don't think many are using it, and
5263 it is clearly obsolete.
5265 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5268 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5269 client certificate was expired. A simple patch fixes this, though I don't
5270 understand the full logic of why the verify callback is called multiple
5273 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5274 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5275 which causes problems with some clients (such as the Certicom SSL Plus
5276 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5277 disables the coutermeasure allowing Eudora to connect."
5279 38. Exim was not checking that a write() to a log file succeeded. This could
5280 lead to Bad Things if a log got too big, in particular if it hit a file
5281 size limit. Exim now panics and dies if it cannot write to a log file, just
5282 as it does if it cannot open a log file.
5284 39. Modified OS/Makefile-Linux so that it now contains
5286 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5288 The two -D definitions ensure that Exim is compiled with large file
5289 support, which makes it possible to handle log files that are bigger than
5292 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5293 instance) a domain was checked against a named list that involved a lookup,
5294 causing $domain_data to be set, then another domain was checked against the
5295 same list, then the first domain was re-checked, the value of $domain_data
5296 after the final check could be wrong. In particular, if the second check
5297 failed, it could be set empty. This bug probably also applied to
5300 41. The strip_trailing_dot option was not being applied to the address given
5301 with the -f command-line option.
5303 42. The code for reading a message's header from the spool was incrementing
5304 $received_count, but never initializing it. This meant that the value was
5305 incorrect (doubled) while delivering a message in the same process in which
5306 it was received. In the most common configuration of Exim, this never
5307 happens - a fresh exec is done - but it can happen when
5308 deliver_drop_privilege is set.
5310 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5311 it now includes up to 150 characters of the unexpected data in the log
5314 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5315 and building data strings. The size of both of these buffers was 10 000
5316 bytes - far larger than anybody would *ever* want, thought I. Needless to
5317 say, somebody hit the limit. I have increased the maximum line length to
5318 20 000 and the maximum data length of concatenated lines to 100 000. I have
5319 also fixed two bugs, because there was no checking on these buffers. Tsk,
5320 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5321 buffer is too small.
5323 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5324 lsearch lookups. Now it does.
5326 46. When parsing a route_list item in a manualroute router, a fixed-length
5327 buffer was used for the list of hosts. I made this 1024 bytes long,
5328 thinking that nobody would ever have a list of hosts that long. Wrong.
5329 Somebody had a whole pile of complicated expansion conditions, and the
5330 string was silently truncated, leading to an expansion error. It turns out
5331 that it is easier to change to an unlimited length (owing to other changes
5332 that have happened since this code was originally written) than to build
5333 structure for giving a limitation error. The length of the item that
5334 expands into the list of hosts is now unlimited.
5336 47. The lsearch lookup could not handle data where the length of text line was
5337 more than 4095 characters. Such lines were truncated, leading to shortened
5338 data being returned. It should now handle lines of any length.
5340 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5341 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5344 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5345 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5346 no longer happens in comment lines. A list of replacements is now placed
5347 at the head of all of the source files, except those whose only change is
5348 to replace PERL_COMMAND in the very first #! line.
5350 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5351 messages on the queue, with a bottom-up merge sort, using code contributed
5352 by Michael Haardt. This should make operations like -bp somewhat faster on
5353 large queues. It won't affect queue runners, except when queue_run_in_order
5356 51. Installed eximstats 1.31 in the distribution.
5358 52. Added support for SRV lookups to the dnslookup router.
5360 53. If an ACL referred to $message_body or $message_body_end, the value was not
5361 reset for any messages that followed in the same SMTP session.
5363 54. The store-handling optimization for building very long strings was not
5364 differentiating between the different store pools. I don't think this
5365 actually made any difference in practice, but I've tidied it.
5367 55. While running the routers to verify a sender address, $sender_address
5368 was still set to the sender address. This is wrong, because when routing to
5369 send a bounce to the sender, it would be empty. Therefore, I have changed
5370 it so that, while verifying a sender address, $sender_address is set to <>.
5371 (There is no change to what happens when verifying a recipient address.)
5373 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5374 target A or AAAA records (if not already returned) without resetting the
5375 qualify_single or search_parents options of the DNS resolver. These are
5376 inappropriate in this case because the targets of MX and SRV records must
5377 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5378 target that, when qualified, matched something in the local domain. These
5379 two options are now turned off when doing these lookups.
5381 57. It seems that at least some releases of Reiserfs (which does not have the
5382 concept of a fixed number of inodes) returns zero and not -1 for the
5383 number of available inodes. This interacted badly with check_spool_inodes,
5384 which assumed that -1 was the "no such thing" setting. What I have done is
5385 to check that the total number of inodes is greater than zero before doing
5386 the test of how many are available.
5388 58. When a "warn" ACL statement has a log_message modifier, the message is
5389 remembered, and not repeated. This is to avoid a lot of repetition when a
5390 message has many recipients that cause the same warning to be written.
5391 Howewer, Exim was preserving the list of already written lines for an
5392 entire SMTP session, which doesn't seem right. The memory is now reset if a
5393 new message is started.
5395 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5396 debugging output unless log_rewrite was also set.
5398 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5399 of (int)(handle) when we know that handle contains (void *)(-1).
5401 61. The Exim daemon panic-logs an error return when it closes the incoming
5402 connection. However "connection reset by peer" seems to be common, and
5403 isn't really an error worthy of noting specially, so that particular error
5406 62. When Exim is trying to find all the local interfaces, it used to panic and
5407 die if the ioctl to get the interface flags failed. However, it seems that
5408 on at least one OS (Solaris 9) it is possible to have an interface that is
5409 included in the list of interfaces, but for which you get a failure error
5410 for this call. This happens when the interface is not "plumbed" into a
5411 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5412 failure of the "get flags" call assumes that the interface is down.
5414 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5415 makes life easier for people who are doing arithmetic on fields extracted
5416 from dates, where you often get leading zeros that should not be
5417 interpreted as octal.
5419 64. Added qualify_domain to the redirect router, to override the global
5422 65. If a pathologically long header line contained very many addresses (the
5423 report of this problem mentioned 10 000) and each of them was rewritten,
5424 Exim could use up a very large amount of memory. (It kept on making new
5425 copies of the header line as it rewrote, and never released the old ones.)
5426 At the expense of a bit more processing, the header rewriting function has
5427 been changed so that it no longer eats memory in this way.
5429 66. The generation of the Received: header has been moved from the time that a
5430 message starts to be received, to the time that it finishes. The timestamp
5431 in the Received: header should now be very close to that of the <= log
5432 line. There are two side-effects of this change:
5434 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5435 logged header lines no longer include the local Received: line, because
5436 it has not yet been created. The same applies to a copy of the message
5437 that is returned to a non-SMTP sender when a message is rejected.
5439 (b) When a filter file is tested using -bf, no additional Received: header
5440 is added to the test message. After some thought, I decided that this
5443 This change does not affect the value of $received_for. It is still set
5444 after address rewriting, but before local_scan() is called.
5446 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5448 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5449 gave an unhelpful panic error message, and a defer error. I have managed to
5450 change this behaviour so that it now rejects any supplied certificate,
5451 which seems right, as the list of acceptable certificates is empty.
5453 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5454 gave an unhelpful defer error. I have not managed to make this reject any
5455 supplied certificates, but the error message it gives is "no certificate
5456 supplied", which is not helpful.
5458 70. exigrep's output now also includes lines that are not associated with any
5459 message, but which match the given pattern. Implemented by a patch from
5460 Martin Sluka, which also tidied up the Perl a bit.
5462 71. Recipient callout verification, like sender verification, was using <> in
5463 the MAIL FROM command. This isn't really the right thing, since the actual
5464 sender may affect whether the remote host accepts the recipient or not. I
5465 have changed it to use the actual sender in the callout; this means that
5466 the cache record is now keyed on a recipient/sender pair, not just the
5467 recipient address. There doesn't seem to be a real danger of callout loops,
5468 since a callout by the remote host to check the sender would use <>.
5469 [SEE ABOVE: changed after hitting problems.]
5471 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5472 temporary errors. However, in the case of such a code being given after
5473 the end of a data transmission (i.e. after ".") Exim was failing to write
5474 a retry record for the message. (Yes, there was some broken host that was
5475 actually sending 8xx at this point.)
5477 73. An unknown lookup type in a host list could cause Exim to panic-die when
5478 the list was checked. (An example that provoked this was putting <; in the
5479 middle of a list instead of at the start.) If this happened during a DATA
5480 ACL check, a -D file could be left lying around. This kind of configuration
5481 error no longer causes Exim to die; instead it causes a defer errror. The
5482 incident is still logged to the main and panic logs.
5484 74. Buglet left over from Exim 3 conversion. The message "too many messages
5485 in one connection" was written to the rejectlog but not the mainlog, except
5486 when address rewriting (yes!) was being logged.
5488 75. Added write_rejectlog option.
5490 76. When a system filter was run not as root (that is, when system_filter_user
5491 was set), the values of the $n variables were not being returned to the
5492 main process; thus, they were not subsequently available in the $sn
5495 77. Added +return_path_on_delivery log selector.
5497 78. A connection timeout was being treated differently from recipients deferred
5498 when testing hosts_max_try with a message that was older than the host's
5499 retry timeout. (The host should not be counted, thus allowing all hosts to
5500 be tried at least once before bouncing.) This may have been the cause of an
5501 occasionally reported bug whereby a message would remain on the queue
5502 longer than the retry timeout, but would be bounced if a delivery was
5503 forced. I say "may" because I never totally pinned down the problem;
5504 setting up timeout/retry tests is difficult. See also the next item.
5506 79. The ultimate address timeout was not being applied to errors that involved
5507 a combination of host plus message (for example, a timeout on a MAIL
5508 command). When an address resolved to a number of possible hosts, and they
5509 were not all tried for each delivery (e.g. because of hosts_max_try), a
5510 message could remain on the queue longer than the retry timeout.
5512 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5515 81. Fixed an obscure SMTP outgoing bug which required at least the following
5516 conditions: (a) there was another message waiting for the same server;
5517 (b) the server returned 5xx to all RCPT commands in the first message so
5518 that the message was not completed; (c) the server dropped the connection
5519 or gave a negative response to the RSET that Exim sends to abort the
5520 transaction. The observed case was a dropped connection after DATA that had
5521 been sent in pipelining mode. That is, the server had advertised PIPELINING
5522 but was not implementing it correctly. The effect of the bug was incorrect
5523 behaviour, such as trying another host, and this could lead to a crash.
5529 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5530 and daemon.c were passed as pointers to ints; they should have been
5531 pointers to socklen_t variables (which are typically unsigned ints).
5533 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5536 3. Fixed a really odd bug that affected only the testing scheme; patching a
5537 certain fixed string in the binary changed the value of another string that
5538 happened to be identical to the end of the original first string.
5540 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5541 name", it returns that address as the IP address. On some operating
5542 systems (e.g. Solaris), it also passes back the IP address string as the
5543 "host name". However, on others (e.g. Linux), it passes back an empty
5544 string. Exim wasn't checking for this, and was changing the host name to an
5545 empty string, assuming it had been canonicized.
5547 5. Although rare, it is permitted to have more than one PTR record for a given
5548 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5549 all the names associated with an address, because they do in Solaris.
5550 However, it seems that they do not in Linux for data that comes from the
5551 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5552 I found this out when I moved to a new Linux workstation and tried to run
5553 the Exim test suite.
5555 To get round this problem I have changed the code so that it now does its
5556 own call to the DNS to look up PTR records when searching for a host name.
5557 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5558 addresses that are only in /etc/hosts are still found.
5560 This behaviour is, however, controlled by an option called host_lookup_
5561 order, which defaults to "bydns:byaddr". If people want to use the other
5562 order, or indeed, just use one or the other means of lookup, they can
5563 specify it in this variable.
5565 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5566 some operating systems, this comes back from gethostbyaddr() as an empty
5567 string, and this is what Exim used to test for. However, it seems that in
5568 other systems, "." is yielded. Exim now tests for this case too.
5570 7. The values of check_spool_space and check_log_space are now held internally
5571 as a number of kilobytes instead of an absolute number of bytes. If a
5572 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5573 kilobyte. This means that much larger values can be stored.
5575 8. Exim monitor: an attempt to get the action menu when not actually pointing
5576 at a message produces an empty menu entitled "No message selected". This
5577 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5578 no entries in it ("Shell widget menu has zero width and/or height"). So I
5579 have added a single, blank menu entry in this case.
5581 9. Added ${quote_local_part.
5583 10. MIME decoding is now applied to the contents of Subject: header lines when
5586 11. Now that a reference to $sender_host_address automatically causes a reverse
5587 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5588 host lookup before query-style lookups in lists that might use this
5589 variable. This has therefore been abolished, and the "net-" prefix is no
5590 longer necessary for query-style lookups.
5592 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5593 have been a typo for LFLAGS, so it has been changed.
5595 13. The install script calls Exim with "-C /dev/null" in order to find the
5596 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5597 to be output. Howeve, since Exim outputs its version number before the
5598 error, it didn't break the script. It just looked ugly. I fixed this by
5599 always allowing "-C /dev/null" if the caller is root.
5601 14. Ignore overlarge ACL variable number when reading spool file - insurance
5602 against a later release with more variables having written the file.
5604 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5605 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5606 this, as well as the form without the "IPv6" on the front.
5608 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5609 OS/Makefile-Darwin file.
5611 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5612 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5614 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5615 variable. However, some OS use ints for inodes. I've added cast to long int
5616 to get rid of the compiler warning.
5618 19. I had forgotten to lock out "/../" in configuration file names when
5619 ALT_CONFIG_PREFIX was set.
5621 20. Routers used for verification do not need to specify transports. However,
5622 if such a router generated a host list, and callout was configured, Exim
5623 crashed, because it could not find a port number from the (non-existent)
5624 transport. It now assumes port 25 in this circumstance.
5626 21. Added the -t option to exigrep.
5628 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5629 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5630 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5631 them is set, LOOKUP_LSEARCH is forced.
5633 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5635 24. Added sender and host information to the "rejected by local_scan()" log
5636 line; previously there was no indication of these.
5638 25. Added .include_if_exists.
5640 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5641 Linux. It turns out that not all file systems support this. Apparently some
5642 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5643 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5644 not supported on the file descriptor, is now ignored when Exim is trying to
5645 sync a directory. This applies only to Linux.
5647 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5649 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5650 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5653 29. Added support for maildirsize files from supplied patch (modified a bit).
5655 30. The use of :fail: followed by an empty string could lead Exim to respond to
5656 sender verification failures with (e.g.):
5658 550 Verification failed for <xxx>
5659 550 Sender verify failed
5661 where the first response line was missing the '-' that indicates it is not
5662 the final line of the response.
5664 31. The loop for finding the name of the user that called Exim had a hardwired
5665 limit of 10; it now uses the value of finduser_retries, which is used for
5666 all other user lookups.
5668 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5671 33. Exim was neglecting to zero errno before one call of strtol() when
5672 expanding a string and expecting an integer value. On some systems this
5673 resulted in spurious "integer overflow" errors. Also, it was casting the
5674 result into an int without checking.
5676 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5677 did not work. The code looks as if it has *never* worked, though it appears
5678 to have been documented since at least releast 1.62. I have made it work.
5680 35. The "timeout_DNS" error in retry rules, also documented since at least
5681 1.62, also never worked. As it isn't clear exactly what this means, and
5682 clearly it isn't a major issue, I have abolished the feature by treating it
5683 as "timeout", and writing a warning to the main and panic logs.
5685 36. The display of retry rules for -brt wasn't always showing the error code
5688 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5689 timeout_connect_A, timeout_connect_MX.
5691 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5692 to the empty sender.
5694 39. The daemon was not analysing the content of -oX till after it had closed
5695 stderr and disconnected from the controlling terminal. This meant that any
5696 syntax errors were only noted on the panic log, and the return code from
5697 the command was 0. By re-arranging the code a little, I've made the
5698 decoding happen first, so such errors now appear on stderr, and the return
5699 code is 1. However, the actual setting up of the sockets still happens in
5700 the disconnected process, so errors there are still only recorded on the
5703 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5704 connections (as happens on some IP stacks) was logged at start up time as
5705 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5706 it from "IPv6 and IPv4", which means that two separate sockets are being
5709 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5710 says whether AF_INET or AF_INET6 was passed as an argument.
5712 42. Exiwhat output was messed up when time zones were included in log
5715 43. Exiwhat now gives more information about the daemon's listening ports,
5716 and whether -tls-on-connect was used.
5718 44. The "port" option of the smtp transport is now expanded.
5720 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5721 silently ignored. Now an error message is written to the main and panic
5724 46. There's a new ACL modifier called "logwrite" which writes to a log file
5725 as soon as it is encountered.
5727 47. Added $local_user_uid and $local_user_gid at routing time.
5729 48. Exim crashed when trying to verify a sender address that was being
5732 49. Exim was recognizing only a space character after ".include". It now also
5733 recognizes a tab character.
5735 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5736 extracting the relevant information from the specification. The man page no
5737 longer contains scrambled data for the -d option, and I've added a section
5738 at the front about calling Exim under different names.
5740 51. Added "extra_headers" argument to the "mail" command in filter files.
5742 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5745 53. Installed eximstats 1.29.
5747 54. Added transport_filter_timeout as a generic transport option.
5749 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5750 Cc: header lines. This was required by RFC 822, but it not required by RFC
5753 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5754 incoming messages that did not have them. Now it does so only if the
5755 message originates locally, that is, if there is no associated remote host
5756 address. When Resent- header lines are present, this applies to the Resent-
5757 lines rather than the non-Resent- lines.
5759 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5760 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5762 58. Messages for configuration errors now include the name of the main
5763 configuration file - useful now that there may be more than one file in a
5764 list (.included file names were always shown).
5766 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5767 for those rare installations that do not start the daemon as root or run it
5768 setuid root. I've cut out the call to initgroups() if the daemon is not
5771 60. The Exim user and group can now be bound into the binary as text strings
5772 that are looked up at the start of Exim's processing.
5774 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5776 62. Added $mailstore_basename variable.
5778 63. Installed patch to sieve.c from Michael Haardt.
5780 64. When Exim failed to open the panic log after failing to open the main log,
5781 the original message it was trying to log was written to stderr and debug
5782 output, but if they were not available (the usual case in production), it
5783 was lost. Now it is written to syslog before the two lines that record the
5784 failures to open the logs.
5786 65. Users' Exim filters run in subprocesses under the user's uid. It is
5787 possible for a "deliver" command or an alias in a "personal" command to
5788 provoke an address rewrite. If logging of address rewriting is configured,
5789 this fails because the process is not running as root or exim. There may be
5790 a better way of dealing with this, but for the moment (because 4.30 needs
5791 to be released), I have disabled address rewrite logging when running a
5792 filter in a non-root, non-exim process.
5798 1. The buildconfig auxiliary program wasn't quoting the value set for
5799 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5800 not defined. This bug was masked in 4.22 by the effect that was fixed in
5803 2. Some messages that were rejected after a message id was allocated were
5804 shown as "incomplete" by exigrep. It no longer does this for messages that
5805 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5807 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5808 have allow_domain_literals set, the ID did not get logged in the <= line.
5809 Domain literals are now always recognized in Message-ID: header lines.
5811 4. The first argument for a ${extract expansion item is the key name or field
5812 number. Leading and trailing spaces in this item were not being ignored,
5813 causing some misleading effects.
5815 5. When deliver_drop_privilege was set, single queue runner processes started
5816 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5817 same command in the process it spins off) were not dropping privilege.
5819 6. When the daemon running as "exim" started a queue runner, it always
5820 re-executed Exim in the spun-off process. This is a waste of effort when
5821 deliver_drop_privilege is set. The new process now just calls the
5822 queue-runner function directly.
5828 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5831 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5832 ignored. Though the use of -oP was forcing the writing of a pid file, it
5833 was always written to the default place.
5835 3. If the message "no IP address found for host xxxx" is generated during
5836 incoming verification, it is now followed by identification of the incoming
5837 connection (so you can more easily find what provoked it).
5839 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5841 5. Added some features to "harden" Exim a bit more against certain attacks:
5843 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5844 be put in Local/Makefile. This is like the never_users runtime option,
5845 but it cannot be overridden. The default setting is "root".
5847 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5848 prefix string with which any file named in a -C command line option
5851 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5852 is retained for -C and -D only if the caller of Exim is root. Without
5853 it, the exim user may also use -C and -D and retain privilege.
5855 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5856 command line option is disabled.
5858 6. Macro names set by the -D option must start with an upper case letter, just
5859 like macro names defined in the configuration file.
5861 7. Added "dereference=" facility to LDAP.
5863 8. Two instances of the typo "uknown" in the source files are fixed.
5865 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5866 the Configure-Makefile script screwed up while processing it.
5868 10. Incorporated PCRE 4.4.
5870 11. The SMTP synchronization check was not operating right at the start of an
5871 SMTP session. For example, it could not catch a HELO sent before the client
5872 waited for the greeting. There is now a check for outstanding input at the
5873 point when the greeting is written. Because of the duplex, asynchronous
5874 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5875 way, but not yet received, when the check is performed.
5877 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5878 on TCP/IP sockets, because this apparently causes some broken clients to
5881 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5882 unchanged) from the Cygwin maintainer.
5884 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5885 was supported instead of testing for mbx. Effectively a typo.
5887 15. The spa authenticator server code was not checking that the input it
5888 received was valid base64.
5890 16. The debug output line for the "set" modifier in ACLs was not showing the
5891 name of the variable that was being set.
5893 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5895 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5896 lookup on its own. Something else was needed to trigger the lookup. For
5897 example, a match in host_lookup or the need for a host name in a host list.
5898 Now, if $sender_host_name is referenced and the host name has not yet been
5899 looked up, a lookup is performed. If the lookup fails, the variable remains
5900 empty, and $host_lookup_failed is set to "1".
5902 19. Added "eqi" as a case-independent comparison operator.
5904 20. The saslauthd authentication condition could segfault if neither service
5905 nor realm was specified.
5907 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5908 error message that was logged was misleading, and incoming SMTP
5909 connections were dropped. The message is now more accurate, and temporary
5910 errors are given to SMTP connections.
5912 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5913 (except RSET) with a 421 error, until QUIT is received. However, it was
5914 failing to send a response to QUIT.
5916 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5917 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5918 ACL is now run after the helo_try_verify_hosts code.
5920 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5921 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5922 case-independent, so other case variants are also recognized). Apparently
5923 some systems use these upper case variants.
5925 25. If more than two messages were waiting for the same host, and a transport
5926 filter was specified for the transport, Exim sent two messages over the
5927 same TCP/IP connection, and then failed with "socket operation on non-
5928 socket" when it tried to send the third.
5930 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5932 27. The extern definition of crypt16() in expand.c was not being excluded when
5933 the OS had its own crypt16() function.
5935 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5936 as a preferred synonym for return_size_limit, both as an option and as an
5939 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5941 30. Changed the default configuration ACL to relax the local part checking rule
5942 for addresses that are not in any local domains. For these addresses,
5943 slashes and pipe symbols are allowed within local parts, but the sequence
5944 /../ is explicitly forbidden.
5946 31. SPA server authentication was not clearing the challenge buffer before
5949 32. log_message in a "warn" ACL statement was writing to the reject log as
5950 well as to the main log, which contradicts the documentation and doesn't
5951 seem right (because no rejection is happening). So I have stopped it.
5953 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5954 However, I am unable to do any testing of this.
5956 34. Fixed an infelicity in the appendfile transport. When checking directories
5957 for a mailbox, to see if any needed to be created, it was accidentally
5958 using path names with one or more superfluous leading slashes; tracing
5959 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5961 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5962 contents are added to the log line that is written for every discarded
5963 recipient. (Previously a log_message setting was ignored.)
5965 36. The ${quote: operator now quotes the string if it is empty.
5967 37. The install script runs exim in order to find its version number. If for
5968 some reason other than non-existence or emptiness, which it checks, it
5969 could not run './exim', it was installing it with an empty version number,
5970 i.e. as "exim-". This error state is now caught, and the installation is
5973 38. An argument was missing from the function that creates an error message
5974 when Exim fails to connect to the socket for saslauthd authentication.
5975 This could cause Exim to crash, or give a corrupted message.
5977 39. Added isip, isip4, and isip6 to ${if conditions.
5979 40. The ACL variables $acl_xx are now saved with the message, and can be
5980 accessed later in routers, transports, and filters.
5982 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5983 strings in the file are not string-expanded.
5985 42. If a MAIL command specified a SIZE value that was too large to fit into an
5986 int variable, the check against message_size_limit failed. Such values are
5987 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5988 day this will have to be increased, but I don't think I want to be around
5989 when emails are that large.
5996 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5997 iconv() is not standard in FreeBSD.
5999 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6000 IPv6 enabled. The observed symptom was a segmentation fault on return from
6001 the function os_common_find_running_interfaces() in src/os.c.
6003 3. In the check_special_case() function in daemon.c I had used "errno" as an
6004 argument name, which causes warnings on some systems. This was basically a
6005 typo, since it was named "eno" in the comments!
6007 4. The code that waits for the clock to tick (at a resolution of some fraction
6008 of a second) so as to ensure message-id uniqueness was always waiting for
6009 at least one whole tick, when it could have waited for less. [This is
6010 almost certainly not relevant at current processor speeds, where it is
6011 unlikely to ever wait at all. But we try to future-proof.]
6013 5. The function that sleeps for a time interval that includes fractions of a
6014 second contained a race. It did not block SIGALRM between setting the
6015 timer, and suspending (a couple of lines later). If the interval was short
6016 and the sigsuspend() was delayed until after it had expired, the suspension
6017 never ended. On busy systems this could lead to processes getting stuck for
6020 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6021 process, before it forks any delivery processes. The open lookup caching
6022 mechanism meant that the open file or database connection was passed into
6023 the delivery process. The problem was that delivery processes always tidy
6024 up cached lookup data. This could cause a problem for the next delivery
6025 process started by the queue runner, because the external queue runner
6026 process does not know about the closure. So the next delivery process
6027 still has data in the lookup cache. In the case of a file lookup, there was
6028 no problem because closing a file descriptor in a subprocess doesn't affect
6029 the parent. However, if the lookup was caching a connection to a database,
6030 the connection was closed, and the second delivery process was likely to
6031 see errors such as "PGSQL: query failed: server closed the connection
6032 unexpectedly". The problem has been fixed by closing all cached lookups
6033 in a queue runner before running a delivery process.
6035 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6036 seem to have the "const" qualifier which it has on other OS. I've
6039 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6040 *of the same type* (client or server) with the same public name that an
6041 error should be diagnosed.
6043 9. When Exim looked up a host name for an IP address, but failed to find the
6044 original IP address when looking up the host name (a safety check), it
6045 output the message "<ip address> does not match any IP for NULL", which was
6046 confusing, to say the least. The bug was that the host name should have
6047 appeared instead of "NULL".
6049 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6050 user that is built into the binary, and the -C or -D options is used, root
6051 privilege is dropped before the configuration file is read. In addition,
6052 logging is switched to stderr instead of the normal log files. If the
6053 configuration then re-defines the Exim user, the unprivileged environment
6054 is probably not what is expected, so Exim logs a panic warning message (but
6057 However, if deliver_drop_privilege is set, the unprivileged state may well
6058 be exactly what is intended, so the warning has been cut out in that case,
6059 and Exim is allowed to try to write to its normal log files.
6065 1. smtp_return_error_details was not giving details for temporary sender
6066 or receiver verification errors.
6068 2. Diagnose a configuration error if two authenticators have the same public
6071 3. Exim used not to create the message log file for a message until the first
6072 delivery attempt. This could be confusing when incoming messages were held
6073 for policy or load reasons. The message log file is now created at the time
6074 the message is received, and an initial "Received" line is written to it.
6076 4. The automatically generated man page for command line options had a minor
6077 bug that caused no ill effects; however, a more serious problem was that
6078 the procedure for building the man page automatically didn't always
6079 operate. Consequently, release 4.20 contains an out-of-date version. This
6080 shouldn't happen again.
6082 5. When building Exim with embedded Perl support, the script that builds the
6083 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6084 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6086 6. The freeze_tell option was not being used for messages that were frozen on
6087 arrival, either by an ACL or by local_scan().
6089 7. Added the smtp_incomplete_transaction log selector.
6091 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6092 was accepting AUTH without a new EHLO.
6094 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6095 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6098 10. Logging of TCP/IP connections (when configured) now happens in the main
6099 daemon process instead of the child process, so that the TCP/IP connection
6100 count is more accurate (but it can never be perfect).
6102 11. The use of "drop" in a nested ACL was not being handled correctly in the
6103 outer ACL. Now, if condition failure induced by the nested "drop" causes
6104 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6105 or "require"), the connection is dropped.
6107 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6108 that yield "discard" can now be used with an "accept" or a "discard" verb,
6109 but an error is generated for any others (because I can't see a useful way
6110 to define what should happen).
6112 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6113 are now processed in the same way as lines in the Exim configuration file.
6114 In particular, continuation lines are supported.
6116 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6118 15. Added -ti meaning -t -i.
6120 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6121 domains, and warn by logging if others are found.
6123 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6124 returned from GIFCONF when Exim is trying to find the list of interfaces on
6125 a host. The code in os.c has been modified to copy each ifreq to an aligned
6126 structure in all cases.
6128 Also, in some cases, the returned ifreq's were being copied to a 'struct
6129 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6130 means the last couple of bytes of an IPv6 address could be chopped if the
6131 ifreq contained only a normal sockaddr (14 bytes storage).
6133 18. Named domain lists were not supported in the hosts_treat_as_local option.
6134 An entry such as +xxxx was not recognized, and was treated as a literal
6137 19. Ensure that header lines added by a DATA ACL are included in the reject log
6138 if the ACL subsequently rejects the message.
6140 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6141 MD5 (which is deprecated).
6143 21. When testing a filter file using -bf, Exim was writing a message when it
6144 took the sender from a "From " line in the message, but it was not doing so
6145 when it took $return_path from a Return-Path: header line. It now does.
6147 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6148 with a valid header line field name (a series of printing characters
6149 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6151 23. Changed "disc" in the source to "disk" to conform to the documentation and
6152 the book and for uniformity.
6154 24. Ignore Sendmail's -Ooption=value command line item.
6156 25. When execve() failed while trying to run a command in a pipe transport,
6157 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6158 could be confused with a return value of 69 from the command itself. This
6159 has been changed to 127, the value the shell returns if it is asked to run
6160 a non-existent command. The wording for the related log line suggests a
6161 non-existent command as the problem.
6163 26. If received_header_text expands to an empty string, do not add a Received:
6164 header line to the message. (Well, it adds a token one on the spool, but
6165 marks it "old" so that it doesn't get used or transmitted.)
6167 27. Installed eximstats 1.28 (addition of -nt option).
6169 28. There was no check for failure on the call to getsockname() in the daemon
6170 code. This can fail if there is a shortage of resources on the system, with
6171 ENOMEM, for example. A temporary error is now given on failure.
6173 29. Contrary to the C standard, it seems that in some environments, the
6174 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6175 program. Exim now does this explicitly; it affects the formatting of
6176 timestamps using strftime().
6178 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6179 complaints. I've now initialized all the variables, to avoid this.
6181 32. Header lines added by a system filter were not being "seen" during
6182 transport-time rewrites.
6184 33. The info_callback() function passed to OpenSSL is set up with type void
6185 (*)(SSL *, int, int), as described somewhere. However, when calling the
6186 function (actually a macro) that sets it up, the type void(*)() is
6187 expected. I've put in a cast to prevent warnings from picky compilers.
6189 34. If a DNS black list lookup found a CNAME record, but there were no A
6190 records associated with the domain it pointed at, Exim crashed.
6192 35. If a DNS black list lookup returned more than one A record, Exim ignored
6193 all but the first. It now scans all returned addresses if a particular IP
6194 value is being sought. In this situation, the contents of the
6195 $dnslist_value variable are a list of all the addresses, separated by a
6198 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6199 to accept a host name and all its aliases if the forward lookup for any of
6200 them yielded the IP address of the incoming connection. Now it accepts only
6201 those names whose forward lookup yields the correct IP address. Any other
6202 names are discarded. This closes a loophole whereby a rogue DNS
6203 administrator could create reverse DNS records to break through a
6204 wildcarded host restriction in an ACL.
6206 37. If a user filter or a system filter that ran in a subprocess used any of
6207 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6208 the wrong values were passed to the pipe command ($thisaddress had the
6209 value of $0, $0 had the value of $1, etc). This bug was introduced by
6210 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6212 38. Improved the line breaking for long SMTP error messages from ACLs.
6213 Previously, if there was no break point between 40 and 75 characters, Exim
6214 left the rest of the message alone. Two changes have been made: (a) I've
6215 reduced the minimum length to 35 characters; (b) if it can't find a break
6216 point between 35 and 75 characters, it looks ahead and uses the first one
6217 that it finds. This may give the occasional overlong line, but at least the
6218 remaining text gets split now.
6220 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6221 file descriptors might be low, and that setting 1000 would always raise it.
6222 It turns out that in some environments, the limit is already over 1000 and
6223 that lowering it causes trouble. So now Exim takes care not to decrease it.
6225 40. When delivering a message, the value of $return_path is set to $sender_
6226 address at the start of routing (routers may change the value). By an
6227 oversight, this default was not being set up when an address was tested by
6228 -bt or -bv, which affected the outcome if any router or filter referred to
6231 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6232 log only when "message" or "log_message" are set. However, if one of the
6233 conditions was an address verification, or a call to a nested ACL, the
6234 messages generated by the underlying test were being passed through. This
6235 no longer happens. The underlying message is available in $acl_verify_
6236 message for both "message" and "log_message" expansions, so it can be
6237 passed through if needed.
6239 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6240 new expansion $bh_ to give the encoded byte string without charset
6241 translation. Translation happens only if iconv() is available; HAVE_ICONV
6242 indicates this at build time. HEADERS_CHARSET gives the charset to
6243 translate to; headers_charset can change it in the configuration, and
6244 "headers charset" can change it in an individual filter file.
6246 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6247 that creates RFC 2047 encoded "words" labels them as that charset instead
6248 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6249 expansion operator; (ii) when Exim creates a From: line for a local
6250 message; (iii) when a header line is rewritten to include a "phrase" part.
6252 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6253 buggy, causing it to skip the first lines of messages whose message ID
6254 ended in 'D'. This would not have bitten before Exim release 4.14, because
6255 message IDs were unlikely to end in 'D' before then. The effect was to have
6256 incorrect size information for certain domains.
6258 45. #include "config.h" was missing at the start of the crypt16.c module. This
6259 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6262 46. If there was a timeout during a "random" callout check, Exim treated it as
6263 a failure of the random address, and carried on sending RSET and the real
6264 address. If the delay was just some slowness somewhere, the response to the
6265 original RCPT would be taken as a response to RSET and so on, causing
6266 mayhem of various kinds.
6268 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6269 when I implemented it. It didn't allow for the fact that some option values
6270 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6271 the right test for positive values.
6273 48. Domain names in DNS records are case-independent. Exim always looks them up
6274 in lower case. Some resolvers return domain names in exactly the case they
6275 appear in the zone file, that is, they may contain uppercase letters. Not
6276 all resolvers do this - some return always lower case. Exim was treating a
6277 change of case by a resolver as a change of domain, similar to a widening
6278 of a domain abbreviation. This triggered its re-routing code and so it was
6279 trying to route what was effectively the same domain again. This normally
6280 caused routing to fail (because the router wouldn't handle the domain
6281 twice). Now Exim checks for this case specially, and just changes the
6282 casing of the domain that it ultimately uses when it transmits the message
6285 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6288 50. If a filter generated a file delivery with a non-absolute name (possible if
6289 no home directory exists for the router), the forbid_file option was not
6292 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6293 the existing equality matching.
6295 52. Exim was using ints instead of ino_t variables in some places where it was
6296 dealing with inode numbers.
6298 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6299 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6300 called TMPDIR, and if it finds it is different, it changes its value.
6302 54. The smtp_printf() function is now made available to local_scan() so
6303 additional output lines can be written before returning. There is also an
6304 smtp_fflush() function to enable the detection of a dropped connection.
6305 The variables smtp_input and smtp_batched_input are exported to
6308 55. Changed the default runtime configuration: the message "Unknown user"
6309 has been removed from the ACL, and instead placed on the localuser router,
6310 using the cannot_route_message feature. This means that any verification
6311 failures that generate their own messages won't get overridden. Similarly,
6312 the "Unrouteable address" message that was in the ACL for unverifiable
6313 relay addresses has also been removed.
6315 56. Added hosts_avoid_esmtp to the smtp transport.
6317 57. The exicyclog script was not checking for the esoteric option
6318 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6319 will work only if exicyclog is run under the appropriate euid.
6321 58. Following a discussion on the list, the rules by which Exim recognises line
6322 endings on incoming messages have been changed. The -dropcr and drop_cr
6323 options are now no-ops, retained only for backwards compatibility. The
6324 following line terminators are recognized: LF CRLF CR. However, special
6325 processing applies to CR:
6327 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6328 nor a local message in the state where . is a terminator.
6330 (ii) If a bare CR is encountered in a header line, an extra space is added
6331 after the line terminator so as not to end the header. The reasoning
6332 behind this is that bare CRs in header lines are most likely either
6333 to be mistakes, or people trying to play silly games.
6335 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6336 was being incorrectly given as 18 bytes larger than it should have been.
6337 This is a VOB (very old bug).
6339 60. This may never have affected anything current, but just in case it has:
6340 When the local host is found other than at the start of a list of hosts,
6341 the local host, those with the same MX, and any that follow, are discarded.
6342 When the list in question was part of a longer list of hosts, the following
6343 hosts (not currently being processed) were also being discarded. This no
6344 longer happens. I'm not sure if this situation could ever has previously
6347 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6350 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6351 header. This is something that is recommended in a new Internet Draft, and
6352 is something that is documented as being done by Sendmail. There are two
6353 possible values. For messages generated by the autoreply transport, Exim
6356 Auto-Submitted: auto-replied
6358 whereas for all other generated messages (e.g. bounces) it adds
6360 Auto-Submitted: auto-generated
6362 63. The "personal" condition in filters now includes a test for the
6363 Auto-Submitted: header. If it contains the string "auto-" the message it
6364 not considered personal.
6366 64. Added rcpt_include_affixes as a generic transport option.
6368 65. Added queue_only_override (default true).
6370 66. Added the syslog_duplication option.
6372 67. If what should have been the first header line of a message consisted of
6373 a space followed by a colon, Exim was mis-interpreting it as a header line.
6374 It isn't of course - it is syntactically invalid and should therefore be
6375 treated as the start of the message body. The misbehaviour could have
6376 caused a number of strange effects, including loss of data in subsequent
6377 header lines, and spool format errors.
6379 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6380 client host had authenticated. This control can now be exercised by an ACL
6381 for more flexibility.
6383 69. By default, callouts do not happen when testing with -bh. There is now a
6384 variant, -bhc, which does actually run the callout code, including
6385 consulting and updating the callout cache.
6387 70. Added support for saslauthd authentication, courtesy of Alexander
6390 71. If statvfs() failed on the spool or log directories while checking their
6391 size for availability, Exim confusingly gave the error "space shortage".
6392 Furthermore, in debugging mode it crashed with a floating point exception.
6393 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6394 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6395 serious problem, Exim now writes to the main and panic logs when this
6396 happens, with details of the failure. It then refuses to accept the
6397 incoming message, giving the message "spool directory problem" or "log
6398 directory problem" with a 421 code for SMTP messages.
6400 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6401 0, 1, and 2 exist, because some OS complain for execs without them (see
6402 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6403 descriptors. However, the code omitted to check that the open succeeded,
6404 causing mysterious errors if for some reason the permissions on /dev/null
6405 got screwed. Now Exim writes a message to the main and panic logs, and
6406 bombs out if it can't open /dev/null.
6408 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6409 interact so that it is all more flexible. It is supposed to remain
6410 backwards compatible. Also added extra_local_interfaces.
6412 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6413 to bomb out with an assertion failure - to the client this appears as a
6414 connection drop. This problem occurs in the part of the code that was taken
6415 from the Samba project. Fortunately, the assertion is in a very simple
6416 function, so I have fixed this by reproducing the function inline in the
6417 one place where it is called, and arranging for authentication to fail
6418 instead of killing the process with assert().
6420 75. The SPA client code was not working when the server requested OEM rather
6421 than Unicode encoding.
6423 76. Added code to make require_files with a specific uid setting more usable in
6424 the case where statting the file as root fails - usually a non-root-mounted
6425 NFS file system. When this happens and the failure is EACCES, Exim now
6426 forks a subprocess and does the per-uid checking as the relevant uid.
6428 77. Added process_log_path.
6430 78. If log_file_path was not explicitly set, a setting of check_log_space or
6431 check_log_inodes was ignored.
6433 79. If a space check for the spool or log partitions fails, the incident is now
6434 logged. Of course, in the latter case the data may get lost...
6436 80. Added the %p formatting code to string_format() so that it can be used to
6437 print addresses in debug_print(). Adjusted all the address printing in the
6438 debugging in store.c to use %p rather than %d.
6440 81. There was a concern that a line of code in smtp_in.c could overflow a
6441 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6442 initially expressed, the concern was not well-founded, because trailing
6443 spaces are removed early. However, if the trailing spaces were followed by
6444 a NULL, they did not get removed, so the overflow was possible. Two fixes
6447 (a) I re-wrote the offending code in a cleaner fashion.
6448 (b) If an incoming SMTP command contains a NULL character, it is rejected
6451 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6452 runs initgroups(), so that if the Exim user is in any additional groups,
6453 they will be used during message reception.
6459 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob