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.
150 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
153 JH/02 The smtp transport option "multi_domain" is now expanded.
155 JH/03 The smtp transport now requests PRDR by default, if the server offers
158 JH/04 Certificate name checking on server certificates, when exim is a client,
159 is now done by default. The transport option tls_verify_cert_hostnames
160 can be used to disable this per-host. The build option
161 EXPERIMENTAL_CERTNAMES is withdrawn.
163 JH/05 The value of the tls_verify_certificates smtp transport and main options
164 default to the word "system" to access the system default CA bundle.
165 For GnuTLS, only version 3.0.20 or later.
167 JH/06 Verification of the server certificate for a TLS connection is now tried
168 (but not required) by default. The verification status is now logged by
169 default, for both outbound TLS and client-certificate supplying inbound
172 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
175 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
176 Status Notification (bounce) messages are now MIME format per RFC 3464.
177 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
178 under the control of the dsn_advertise_hosts option, and routers may
179 have a dsn_lasthop option.
181 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
182 default, modifiable by a malware= option. The list separator for
183 the options can now be changed in the usual way. Bug 68.
185 JH/10 The smtp_receive_timeout main option is now expanded before use.
187 JH/11 The incoming_interface log option now also enables logging of the
188 local interface on delivery outgoing connections.
190 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
191 if the interface and destination host and port all match.
193 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
196 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
197 Patch from Andrew Lewis.
199 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
200 now supports optional time-restrictions, weighting, and priority
201 modifiers per server. Patch originally by <rommer@active.by>.
203 JH/16 The spamd_address main option now supports a mixed list of local
204 and remote servers. Remote servers can be IPv6 addresses, and
205 specify a port-range.
207 JH/17 Bug 68: The spamd_address main option now supports an optional
208 timeout value per server.
210 JH/18 Bug 1581: Router and transport options headers_add/remove can
211 now have the list separator specified.
213 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
216 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
219 JH/21 Support for the A6 type of dns record is withdrawn.
221 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
222 rather than the verbs used.
224 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
225 from 255 to 1024 chars.
227 JH/24 Verification callouts now attempt to use TLS by default.
229 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
230 are generic router options now. The defaults didn't change.
232 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
233 Original patch from Alexander Shikoff, worked over by JH.
235 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
236 files if ZCAT_COMMAND is not executable.
238 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
240 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
242 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
243 Normally benign, it bites when the pair was led to by a CNAME;
244 modern usage is to not canoicalize the domain to a CNAME target
245 (and we were inconsistent anyway for A-only vs AAAA+A).
247 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
249 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
250 when evaluating $sender_host_dnssec.
252 JH/31 Check the HELO verification lookup for DNSSEC, adding new
253 $sender_helo_dnssec variable.
255 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
257 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
259 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
261 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
262 documented as working, but never had. Support all but $spam_report.
264 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
265 added for tls authenticator.
270 TL/01 When running the test suite, the README says that variables such as
271 no_msglog_check are global and can be placed anywhere in a specific
272 test's script, however it was observed that placement needed to be near
273 the beginning for it to behave that way. Changed the runtest perl
274 script to read through the entire script once to detect and set these
275 variables, reset to the beginning of the script, and then run through
276 the script parsing/test process like normal.
278 TL/02 The BSD's have an arc4random API. One of the functions to induce
279 adding randomness was arc4random_stir(), but it has been removed in
280 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
281 function when detected.
283 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
284 cause callback expansion.
286 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
287 syntax errors in an expansion can be treated as a string instead of
288 logging or causing an error, due to the internal use of bool_lax
289 instead of bool when processing it.
291 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
292 server certificates when making smtp deliveries.
294 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
296 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
298 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
300 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
301 Merged patch from Sebastian Wiedenroth.
303 JH/05 Fix results-pipe from transport process. Several recipients, combined
304 with certificate use, exposed issues where response data items split
305 over buffer boundaries were not parsed properly. This eventually
306 resulted in duplicates being sent. This issue only became common enough
307 to notice due to the introduction of conection certificate information,
308 the item size being so much larger. Found and fixed by Wolfgang Breyha.
310 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
311 size buffer was used, resulting in syntax errors when an expansion
314 JH/07 Add support for directories of certificates when compiled with a GnuTLS
315 version 3.3.6 or later.
317 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
318 is EXPERIMENTAL_EVENT, the main-configuration and transport options
319 both become "event_action", the variables become $event_name, $event_data
320 and $event_defer_errno. There is a new variable $verify_mode, usable in
321 routers, transports and related events. The tls:cert event is now also
322 raised for inbound connections, if the main configuration event_action
325 TL/06 In test suite, disable OCSP for old versions of openssl which contained
326 early OCSP support, but no stapling (appears to be less than 1.0.0).
328 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
329 server certificate names available under the smtp transport option
330 "tls_verify_cert_hostname" now do not permit multi-component wildcard
333 JH/10 Time-related extraction expansions from certificates now use the main
334 option "timezone" setting for output formatting, and are consistent
335 between OpenSSL and GnuTLS compilations. Bug 1541.
337 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
338 encoded parameter in the incoming message. Bug 1558.
340 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
341 include certificate info, eximon was claiming there were spoolfile
344 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
346 JH/14 Log delivery-related information more consistently, using the sequence
347 "H=<name> [<ip>]" wherever possible.
349 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
350 are problematic for Debian distribution, omit them from the release
353 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
355 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
358 JH/17 Fix a null-indirection in certextract expansions when a nondefault
359 output list separator was used.
364 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
365 checkers that were complaining about end of non-void function with no
368 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
369 This was a regression intruduced in 4.83 by another bugfix.
371 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
373 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
374 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
380 TF/01 Correctly close the server side of TLS when forking for delivery.
382 When a message was received over SMTP with TLS, Exim failed to clear up
383 the incoming connection properly after forking off the child process to
384 deliver the message. In some situations the subsequent outgoing
385 delivery connection happened to have the same fd number as the incoming
386 connection previously had. Exim would try to use TLS and fail, logging
387 a "Bad file descriptor" error.
389 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
390 utilities have not been installed.
392 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
393 temporary space as the ACL may create new global variables.
395 TL/01 LDAP support uses per connection or global context settings, depending
396 upon the detected version of the libraries at build time.
398 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
399 to extract and use the src ip:port in logging and expansions as if it
400 were a direct connection from the outside internet. PPv2 support was
401 updated based on HAProxy spec change in May 2014.
403 JH/02 Add ${listextract {number}{list}{success}{fail}}.
405 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
406 Properly escape header and check for NULL return.
408 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
411 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
413 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
414 characters in header names, implemented as a verify condition.
415 Contributed by Michael Fischer v. Mollard.
417 TL/05 Rename SPF condition results err_perm and err_temp to standardized
418 results permerror and temperror. Previous values are deprecated but
419 still accepted. In a future release, err_perm and err_temp will be
420 completely removed, which will be a backward incompatibility if the
421 ACL tests for either of these two old results. Patch contributed by
422 user bes-internal on the mailing list.
424 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
426 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
427 selectors, in both main and reject logs.
429 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
432 JH/07 Add malware type "sock" for talking to simple daemon.
434 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
436 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
437 routers/transports under cutthrough routing.
439 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
440 numbers. Touch up "bool" conditional to keep the same definition.
442 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
444 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
446 JH/12 Expand items in router/transport headers_add or headers_remove lists
447 individually rather than the list as a whole. Bug 1452.
449 Required for reasonable handling of multiple headers_ options when
450 they may be empty; requires that headers_remove items with embedded
451 colons must have them doubled (or the list-separator changed).
453 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
454 view the policy declared in the DMARC record. Currently, $dmarc_status
455 is a combined value of both the record presence and the result of the
458 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
460 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
461 dnslookup router and the smtp transport (applying to the forward
464 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
465 of ldap servers used for a specific lookup. Patch provided by Heiko
468 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
469 New variable $lookup_dnssec_authenticated for observability.
471 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
472 Patch submitted by Lars Timman.
474 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
476 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
477 Requires trusted mode and valid format message id, aborts otherwise.
478 Patch contributed by Heiko Schlichting.
480 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
481 certextract with support for various fields. Bug 1358.
483 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
484 is requested by default, modifiable by smtp transport option
487 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
488 operate on certificate variables to give certificate fingerprints
489 Also new ${sha256:cert_variable}.
491 JH/23 The PRDR feature is moved from being Experimental into the mainline.
493 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
494 Christian Aistleitner.
496 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
498 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
499 file. Patch from Wolfgang Breyha.
501 JH/25 Expand the coverage of the delivery $host and $host_address to
502 client authenticators run in verify callout. Bug 1476.
504 JH/26 Port service names are now accepted for tls_on_connect_ports, to
505 align with daemon_smtp_ports. Bug 72.
507 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
508 support and error reporting did not work properly.
510 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
511 and is readable. Patch from Andrew Colin Kissa.
513 TL/14 Enhance documentation of ${run expansion and how it parses the
514 commandline after expansion, particularly in the case when an
515 unquoted variable expansion results in an empty value.
517 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
519 PP/02 Fix internal collision of T_APL on systems which support RFC3123
520 by renaming away from it. Addresses GH issue 15, reported by
523 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
525 TL/15 SECURITY: prevent double expansion in math comparison functions
526 (can expand unsanitized data). Not remotely exploitable.
533 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
535 PP/02 Make -n do something, by making it not do something.
536 When combined with -bP, the name of an option is not output.
538 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
541 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
542 $sender_host_name and config options to manage this, and basic check
545 PP/05 DSCP support for outbound connections and control modifier for inbound.
547 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
548 (Only plugin which currently uses this is kerberos4, which nobody should
549 be using, but we should make it available and other future plugins might
550 conceivably use it, even though it would break NAT; stuff *should* be
551 using channel bindings instead).
553 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
554 name; added for Sendmail compatibility; requires admin caller.
555 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
556 just ignore it); requires trusted caller.
557 Also parse but ignore: -Ac -Am -X<logfile>
560 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
562 TL/02 Add +smtp_confirmation as a default logging option.
564 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
565 Patch by Magnus Holmgren from 2007-02-20.
567 TL/04 Bugzilla 1281 - Spec typo.
568 Bugzilla 1283 - Spec typo.
569 Bugzilla 1290 - Spec grammar fixes.
571 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
573 TL/06 Add Experimental DMARC support using libopendmarc libraries.
575 TL/07 Fix an out of order global option causing a segfault. Reported to dev
576 mailing list by by Dmitry Isaikin.
578 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
580 JH/02 Support "G" suffix to numbers in ${if comparisons.
582 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
584 NM/01 Bugzilla 1197 - Spec typo
585 Bugzilla 1196 - Spec examples corrections
587 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
589 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
590 gnutls_enable_pkcs11, but renamed to more accurately indicate its
593 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
594 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
596 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
597 "acl {{name}{arg}...}", and optional args on acl condition
600 JH/05 Permit multiple router/transport headers_add/remove lines.
602 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
604 JH/07 Avoid using a waiting database for a single-message-only transport.
605 Performance patch from Paul Fisher. Bugzilla 1262.
607 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
610 JH/09 Add $headers_added variable, with content from use of ACL modifier
611 add_header (but not yet added to the message). Bugzilla 199.
613 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
614 Pulled from Bugzilla 817 by Wolfgang Breyha.
616 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
618 (nb: this is the same fix as in Exim 4.80.1)
620 JH/11 Add A= logging on delivery lines, and a client_set_id option on
623 JH/12 Add optional authenticated_sender logging to A= and a log_selector
626 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
628 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
629 advertise SMTP AUTH mechanism to us, instead of a generic
630 protocol violation error. Also, make Exim more robust to bad
631 data from the Dovecot auth socket.
633 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
635 When a queue runner is handling a message, Exim first routes the
636 recipient addresses, during which it prunes them based on the retry
637 hints database. After that it attempts to deliver the message to
638 any remaining recipients. It then updates the hints database using
641 So if a recipient address works intermittently, it can get repeatedly
642 deferred at routing time. The retry hints record remains fresh so the
643 address never reaches the final cutoff time.
645 This is a fairly common occurrence when a user is bumping up against
646 their storage quota. Exim had some logic in its local delivery code
647 to deal with this. However it did not apply to per-recipient defers
648 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
650 This change adds a proper retry rule check during routing so that the
651 final cutoff time is checked against the message's age. We only do
652 this check if there is an address retry record and there is not a
653 domain retry record; this implies that previous attempts to handle
654 the address had the retry_use_local_parts option turned on. We use
655 this as an approximation for the destination being like a local
656 delivery, as in LMTP.
658 I suspect this new check makes the old local delivery cutoff check
659 redundant, but I have not verified this so I left the code in place.
661 TF/02 Correct gecos expansion when From: is a prefix of the username.
663 Test 0254 submits a message to Exim with the header
667 When I ran the test suite under the user fanf2, Exim expanded
668 the header to contain my full name, whereas it should have added
669 a Resent-Sender: header. It erroneously treats any prefix of the
670 username as equal to the username.
672 This change corrects that bug.
674 GF/01 DCC debug and logging tidyup
675 Error conditions log to paniclog rather than rejectlog.
676 Debug lines prefixed by "DCC: " to remove any ambiguity.
678 TF/03 Avoid unnecessary rebuilds of lookup-related code.
680 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
681 Bug spotted by Jeremy Harris; was flawed since initial commit.
682 Would have resulted in OCSP responses post-SNI triggering an Exim
683 NULL dereference and crash.
685 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
687 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
688 Bug detection, analysis and fix by Samuel Thibault.
689 Bugzilla 1331, Debian bug #698092.
691 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
693 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
694 Server implementation by Todd Lyons, client by JH.
695 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
696 config variable "prdr_enable" controls whether the server
697 advertises the facility. If the client requests PRDR a new
698 acl_data_smtp_prdr ACL is called once for each recipient, after
699 the body content is received and before the acl_smtp_data ACL.
700 The client is controlled by bolth of: a hosts_try_prdr option
701 on the smtp transport, and the server advertisement.
702 Default client logging of deliveries and rejections involving
703 PRDR are flagged with the string "PRDR".
705 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
706 fclose(). Diagnosis by Todd Lyons.
708 PP/17 Update configure.default to handle IPv6 localhost better.
709 Patch by Alain Williams (plus minor tweaks).
712 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
713 This is now consistent with GnuTLS, and is now documented: the
714 previous undocumented portable approach to treating the option as
715 unset was to force an expansion failure. That still works, and
716 an empty string is now equivalent.
718 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
719 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
720 not performing validation itself.
722 PP/20 Added force_command boolean option to pipe transport.
723 Patch from Nick Koston, of cPanel Inc.
725 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
728 TF/04 Added udpsend ACL modifer and hexquote expansion operator
730 PP/21 Fix eximon continuous updating with timestamped log-files.
731 Broken in a format-string cleanup in 4.80, missed when I repaired the
732 other false fix of the same issue.
733 Report and fix from Heiko Schlichting.
736 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
737 Report from Prashanth Katuri.
739 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
740 It's SecureTransport, so affects any MacOS clients which use the
741 system-integrated TLS libraries, including email clients.
743 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
744 using a MIME ACL for non-SMTP local injection.
745 Report and assistance in diagnosis by Warren Baker.
747 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
749 JH/16 Fix comparisons for 64b. Bugzilla 1385.
751 TL/09 Add expansion variable $authenticated_fail_id to keep track of
752 last id that failed so it may be referenced in subsequent ACL's.
754 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
757 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
758 ldap library initialization, allowing self-signed CA's to be
759 used. Also properly sets require_cert option later in code by
760 using NULL (global ldap config) instead of ldap handle (per
761 session). Bug diagnosis and testing by alxgomz.
763 TL/12 Enhanced documentation in the ratelimit.pl script provided in
764 the src/util/ subdirectory.
766 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
767 renamed to Transport Post Delivery Action by Jeremy Harris, as
770 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
771 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
772 redis_servers = needs to be configured which will be used by the redis
773 lookup. Patch from Warren Baker, of The Packet Hub.
775 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
777 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
778 hostname or reverse DNS when processing a host list. Used suggestions
779 from multiple comments on this bug.
781 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
783 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
784 Missed a few lines, added it to make the runtest require no keyboard
787 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
788 contains upper case chars. Make router use caseful_local_part.
790 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
791 support when GnuTLS has been built with p11-kit.
797 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
799 This, or similar/improved, will also be change PP/11 of 4.82.
805 PP/01 Handle short writes when writing local log-files.
806 In practice, only affects FreeBSD (8 onwards).
807 Bugzilla 1053, with thanks to Dmitry Isaikin.
809 NM/01 Bugzilla 949 - Documentation tweak
811 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
814 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
816 PP/02 Implemented gsasl authenticator.
818 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
820 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
821 `pkg-config foo` for cflags/libs.
823 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
824 with rest of GSASL and with heimdal_gssapi.
826 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
827 `pkg-config foo` for cflags/libs for the TLS implementation.
829 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
830 properties get this fed in as external SSF. A number of robustness
831 and debugging improvements to the cyrus_sasl authenticator.
833 PP/08 cyrus_sasl server now expands the server_realm option.
835 PP/09 Bugzilla 1214 - Log authentication information in reject log.
836 Patch by Jeremy Harris.
838 PP/10 Added dbmjz lookup type.
840 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
842 PP/12 MAIL args handles TAB as well as SP, for better interop with
843 non-compliant senders.
844 Analysis and variant patch by Todd Lyons.
846 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
847 Bug report from Lars Müller <lars@samba.org> (via SUSE),
848 Patch from Dirk Mueller <dmueller@suse.com>
850 PP/13 tls_peerdn now print-escaped for spool files.
851 Observed some $tls_peerdn in wild which contained \n, which resulted
852 in spool file corruption.
854 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
855 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
856 or write after TLS renegotiation, which otherwise led to messages
859 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
860 as a tracking header (ie: a signed header comes before the signature).
861 Patch from Wolfgang Breyha.
863 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
864 comma-sep list; embedded commas doubled.
866 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
868 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
870 Report and patch from Dmitry Banschikov.
872 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
873 Removed SSL_clear() after SSL_new() which led to protocol negotiation
874 failures. We appear to now support TLS1.1+ with Exim.
876 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
877 lets Exim select keys and certificates based upon TLS SNI from client.
878 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
879 before an outbound SMTP session. New log_selector, +tls_sni.
881 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
882 NULL dereference. Report and patch from Alun Jones.
884 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
885 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
886 Not seeing resolver debug output on NetBSD, but suspect this is a
887 resolver implementation change.
889 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
890 Left warnings. Added "eximon gdb" invocation mode.
892 PP/21 Defaulting "accept_8bitmime" to true, not false.
894 PP/22 Added -bw for inetd wait mode support.
896 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
897 locate the relevant includes and libraries. Made this the default.
899 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
900 Bugzilla 1246, report and most of solution from Tomasz Kusy.
902 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
903 This may cause build issues on older platforms.
905 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
906 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
907 gnutls_require_mac & gnutls_require_protocols (no longer supported).
908 Added SNI support via GnuTLS too.
909 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
911 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
913 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
914 Applied second patch from Janne, implementing suggestion to default
915 multiple-strings-in-record handling to match SPF spec.
917 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
919 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
920 read-only, out of scope).
921 Patch from Wolfgang Breyha, report from Stuart Northfield.
923 PP/29 Fix three issues highlighted by clang analyser static analysis.
924 Only crash-plausible issue would require the Cambridge-specific
925 iplookup router and a misconfiguration.
926 Report from Marcin Mirosław.
928 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
930 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
931 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
932 As part of this, removing so much warning spew let me fix some minor
933 real issues in debug logging.
935 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
936 assignment on my part. Fixed.
938 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
939 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
940 Janne Snabb (who went above and beyond: thank you).
942 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
943 string otherwise requires a connection and a bunch more work and it's
944 relatively easy to get wrong. Should also expose TLS library linkage
947 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
948 64-bit ${eval} (JH/03).
950 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
951 GNU libc to support some of the 64-bit stuff, should not lead to
952 conflicts. Defined before os.h is pulled in, so if a given platform
953 needs to override this, it can.
955 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
956 protection layer was required, which is not implemented.
957 Bugzilla 1254, patch from Wolfgang Breyha.
959 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
960 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
961 tls_dhparam take prime identifiers. Also unbreak combination of
962 OpenSSL+DH_params+TLSSNI.
964 PP/39 Disable SSLv2 by default in OpenSSL support.
970 PP/01 Solaris build fix for Oracle's LDAP libraries.
971 Bugzilla 1109, patch from Stephen Usher.
973 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
975 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
978 TF/02 Fix a couple more cases where we did not log the error message
979 when unlink() failed. See also change 4.74-TF/03.
981 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
982 lock up or crash if it happened to be inside a call to libc when it
983 got a SIGUSR1 from exiwhat.
985 The SIGUSR1 handler appends the current process status to the process
986 log which is later printed by exiwhat. It used to use the general
987 purpose logging code to do this, but several functions it calls are
988 not safe for signals.
990 The new output code in the SIGUSR1 handler is specific to the process
991 log, and simple enough that it's easy to inspect for signal safety.
992 Removing some special cases also simplifies the general logging code.
993 Removing the spurious timestamps from the process log simplifies
996 TF/04 Improved ratelimit ACL condition.
998 The /noupdate option has been deprecated in favour of /readonly which
999 has clearer semantics. The /leaky, /strict, and /readonly update modes
1000 are mutually exclusive. The update mode is no longer included in the
1001 database key; it just determines when the database is updated. (This
1002 means that when you upgrde Exim will forget old rate measurements.)
1004 Exim now checks that the per_* options are used with an update mode that
1005 makes sense for the current ACL. For example, when Exim is processing a
1006 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1007 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1008 must specify per_mail/readonly. If you omit the update mode it defaults to
1009 /leaky where that makes sense (as before) or /readonly where required.
1011 The /noupdate option is now undocumented but still supported for
1012 backwards compatibility. It is equivalent to /readonly except that in
1013 ACLs where /readonly is required you may specify /leaky/noupdate or
1014 /strict/noupdate which are treated the same as /readonly.
1016 A useful new feature is the /count= option. This is a generalization
1017 of the per_byte option, so that you can measure the throughput of other
1018 aggregate values. For example, the per_byte option is now equivalent
1019 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1021 The per_rcpt option has been generalized using the /count= mechanism
1022 (though it's more complicated than the per_byte equivalence). When it is
1023 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1024 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1025 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1026 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1027 Note that using per_rcpt with a non-readonly update mode in more than
1028 one ACL will cause the recipients to be double-counted. (The per_mail
1029 and per_byte options don't have this problem.)
1031 The handling of very low rates has changed slightly. If the computed rate
1032 is less than the event's count (usually one) then this event is the first
1033 after a long gap. In this case the rate is set to the same as this event's
1034 count, so that the first message of a spam run is counted properly.
1036 The major new feature is a mechanism for counting the rate of unique
1037 events. The new per_addr option counts the number of different
1038 recipients that someone has sent messages to in the last time period. It
1039 behaves like per_rcpt if all the recipient addresses are different, but
1040 duplicate recipient addresses do not increase the measured rate. Like
1041 the /count= option this is a general mechanism, so the per_addr option
1042 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1043 example, measure the rate that a client uses different sender addresses
1044 with the options per_mail/unique=$sender_address. There are further
1045 details in the main documentation.
1047 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1049 TF/06 Removed a few PCRE remnants.
1051 TF/07 Automatically extract Exim's version number from tags in the git
1052 repository when doing development or release builds.
1054 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1055 Bugzilla 879. Patch from Paul Fisher.
1057 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1058 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1061 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1063 PP/05 Variable $av_failed, true if the AV scanner deferred.
1064 Bugzilla 1078. Patch from John Horne.
1066 PP/06 Stop make process more reliably on build failure.
1067 Bugzilla 1087. Patch from Heiko Schlittermann.
1069 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1070 Bugzilla 1089. Patch from Heiko Schlittermann.
1072 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1073 Bugzilla 1131. Patch from Holger Weiß.
1075 PP/09 Handle IPv6 addresses with SPF.
1076 Bugzilla 860. Patch from Wolfgang Breyha.
1078 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1080 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1083 PP/11 match_* no longer expand right-hand-side by default.
1084 New compile-time build option, EXPAND_LISTMATCH_RHS.
1085 New expansion conditions, "inlist", "inlisti".
1087 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1089 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1091 PP/14 fix log_write() format string regression from TF/03.
1092 Bugzilla 1152. Patch from Dmitry Isaikin.
1098 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1100 PP/02 Harmonised TLS library version reporting; only show if debugging.
1101 Layout now matches that introduced for other libraries in 4.74 PP/03.
1103 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1105 PP/04 New "dns_use_edns0" global option.
1107 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1110 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1111 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1113 TK/01 Updated PolarSSL code to 0.14.2.
1114 Bugzilla 1097. Patch from Andreas Metzler.
1116 PP/07 Catch divide-by-zero in ${eval:...}.
1117 Fixes bugzilla 1102.
1119 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1122 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1123 format-string attack -- SECURITY: remote arbitrary code execution.
1125 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1126 time unintentionally subject to list matching rules, letting the header
1127 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1128 arbitrary string expansion). This allowed for information disclosure.
1130 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1131 INT_MIN/-1 -- value coerced to INT_MAX.
1137 NM/01 Workround for PCRE version dependency in version reporting
1140 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1141 This fixes portability to compilers other than gcc, notably
1142 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1144 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1145 makefiles for portability to HP-UX and POSIX correctness.
1147 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1148 Also via indented variable definition in the Makefile.
1149 (Debugging by Oliver Heesakkers).
1151 PP/02 Restore caching of spamd results with expanded spamd_address.
1152 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1154 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1155 Improves build reliability. Fix from: Frank Elsner
1157 NM/02 Fix wide character breakage in the rfc2047 coding
1158 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1160 NM/03 Allow underscore in dnslist lookups
1161 Fixes bug 1026. Patch from Graeme Fowler
1163 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1164 Code patches from Adam Ciarcinski of NetBSD.
1166 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1169 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1170 is logged, to avoid truncation. Patch from John Horne.
1172 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1173 Patch from Jakob Hirsch.
1175 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1176 SQL string expansion failure details.
1177 Patch from Andrey Oktyabrski.
1179 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1180 Patch from Simon Arlott.
1182 PP/09 New lookups functionality failed to compile on old gcc which rejects
1183 extern declarations in function scope.
1184 Patch from Oliver Fleischmann
1186 PP/10 Use sig_atomic_t for flags set from signal handlers.
1187 Check getgroups() return and improve debugging.
1188 Fixed developed for diagnosis in bug 927 (which turned out to be
1191 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1192 Patch from Mark Zealey.
1194 PP/12 Bugzilla 1056: Improved spamd server selection.
1195 Patch from Mark Zealey.
1197 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1198 Based on patch from Heiko Schlittermann.
1200 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1201 Patch from Uwe Doering, sign-off by Michael Haardt.
1203 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1204 variable declaration deep within a block. Bug and patch from
1207 PP/15 lookups-Makefile IRIX compatibilty coercion.
1209 PP/16 Make DISABLE_DKIM build knob functional.
1211 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1212 Patch by Simon Arlott
1214 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1215 variable argument macros. Our copy now differs from upstream.
1221 TF/01 Failure to get a lock on a hints database can have serious
1222 consequences so log it to the panic log.
1224 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1225 controlled using the smtp_confirmation log selector.
1227 TF/03 Include the error message when we fail to unlink a spool file.
1229 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1230 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1231 for maintaining out-of-tree patches for some time.
1233 PP/01 Bugzilla 139: Documentation and portability issues.
1234 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1235 Handle per-OS dynamic-module compilation flags.
1237 PP/02 Let /dev/null have normal permissions.
1238 The 4.73 fixes were a little too stringent and complained about the
1239 permissions on /dev/null. Exempt it from some checks.
1240 Reported by Andreas M. Kirchwitz.
1242 PP/03 Report version information for many libraries, including
1243 Exim version information for dynamically loaded libraries. Created
1244 version.h, now support a version extension string for distributors
1245 who patch heavily. Dynamic module ABI change.
1247 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1248 privilege escalation vulnerability whereby the Exim run-time user
1249 can cause root to append content of the attacker's choosing to
1252 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1255 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1256 If dropping privileges for untrusted macros, we disabled normal logging
1257 on the basis that it would fail; for the Exim run-time user, this is not
1258 the case, and it resulted in successful deliveries going unlogged.
1259 Fixed. Reported by Andreas Metzler.
1265 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1266 only prepend for the Resent-* case. Fixes regression introduced in
1267 Exim 4.70 by NM/22 for Bugzilla 607.
1269 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1270 increasing numbers of administrators be bitten by this.
1272 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1274 PP/03 Bugzilla 994: added openssl_options main configuration option.
1276 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1278 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1280 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1282 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1283 without a peer certificate, leading to a segfault because of an
1284 assumption that peers always have certificates. Be a little more
1285 paranoid. Problem reported by Martin Tscholak.
1287 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1288 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1289 NB: ClamAV planning to remove STREAM in "middle of 2010".
1290 CL also introduces -bmalware, various -d+acl logging additions and
1291 more caution in buffer sizes.
1293 PP/09 Implemented reverse_ip expansion operator.
1295 PP/10 Bugzilla 937: provide a "debug" ACL control.
1297 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1299 PP/12 Bugzilla 973: Implement --version.
1301 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1303 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1305 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1307 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1308 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1309 ignore trailing whitespace.
1311 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1313 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1316 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1317 Notification from Dr Andrew Aitchison.
1319 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1320 ExtendedDetectionInfo response format.
1321 Notification from John Horne.
1323 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1326 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1327 XSL and documented dependency on system catalogs, with examples of how
1330 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1333 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1334 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1335 configuration file which is writeable by the Exim user or group.
1337 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1338 of configuration files to cover files specified with the -C option if
1339 they are going to be used with root privileges, not just the default
1342 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1343 option (effectively making it always true).
1345 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1346 files to be used while preserving root privileges.
1348 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1349 that rogue child processes cannot use them.
1351 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1352 run-time user, instead of root.
1354 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1355 Exim run-time user without dropping privileges.
1357 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1358 result string, instead of calling string_vformat() twice with the same
1361 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1362 for other users. Others should always drop root privileges if they use
1363 -C on the command line, even for a whitelisted configure file.
1365 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1367 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1373 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1374 $data_path, and $header_path variables; fixed documentation bugs and
1377 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1378 exipick to access non-standard spools, including the "frozen" queue
1381 NM/01 Bugzilla 965: Support mysql stored procedures.
1382 Patch from Alain Williams
1384 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1386 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1387 Patch from Andreas Metzler
1389 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1390 Patch from Kirill Miazine
1392 NM/05 Bugzilla 671: Added umask to procmail example.
1394 JJ/03 installed exipick 20100323.0, fixing doc bug
1396 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1397 directory. Notification and patch from Dan Rosenberg.
1399 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1401 TK/02 Improve log output when DKIM signing operation fails.
1403 MH/01 Treat the transport option dkim_domain as a colon separated
1404 list, not as a single string, and sign the message with each element,
1405 omitting multiple occurences of the same signer.
1407 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1408 Bugzilla 985, 986. Patch by Simon Arlott
1410 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1411 Patch by Simon Arlott
1413 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1414 MBX locking. Notification from Dan Rosenberg.
1420 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1422 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1424 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1426 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1428 NM/04 Bugzilla 915: Flag broken perl installation during build.
1434 TK/01 Added patch by Johannes Berg that expands the main option
1435 "spamd_address" if it starts with a dollar sign.
1437 TK/02 Write list of recipients to X-Envelope-Sender header when building
1438 the mbox-format spool file for content scanning (suggested by Jakob
1441 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1442 (http://www.dcc-servers.net/) support via dccifd. Activated by
1443 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1445 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1446 by Mark Daniel Reidel <mr@df.eu>.
1448 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1449 When building exim an external PCRE library is now needed -
1450 PCRE is a system library on the majority of modern systems.
1451 See entry on PCRE_LIBS in EDITME file.
1453 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1454 conversation. Added nologin parameter to request.
1455 Patch contributed by Kirill Miazine.
1457 TF/01 Do not log submission mode rewrites if they do not change the address.
1459 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1461 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1462 log files in place. Contributed by Roberto Lima.
1464 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1466 TF/03 Bugzilla 615: When checking the local_parts router precondition
1467 after a local_part_suffix or local_part_prefix option, Exim now
1468 does not use the address's named list lookup cache, since this
1469 contains cached lookups for the whole local part.
1471 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1472 Robert Millan. Documentation is in experimental-spec.txt.
1474 TF/04 Bugzilla 668: Fix parallel build (make -j).
1476 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1478 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1479 Patch provided by Jan Srzednicki.
1481 TF/05 Leading white space used to be stripped from $spam_report which
1482 wrecked the formatting. Now it is preserved.
1484 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1485 that they are available at delivery time.
1487 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1489 TF/08 TLS error reporting now respects the incoming_interface and
1490 incoming_port log selectors.
1492 TF/09 Produce a more useful error message if an SMTP transport's hosts
1493 setting expands to an empty string.
1495 NM/06 Bugzilla 744: EXPN did not work under TLS.
1496 Patch provided by Phil Pennock.
1498 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1499 Patch provided by Richard Godbee.
1501 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1502 acl_smtp_notquit, added index entry.
1504 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1505 Patch provided by Eugene Bujak.
1507 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1508 accept(). Patch provided by Maxim Dounin.
1510 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1511 Patch provided by Phil Pennock.
1513 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1515 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1516 Patch provided by Brad "anomie" Jorsch.
1518 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1519 Patch provided by Dean Brooks.
1521 TK/05 Add native DKIM support (does not depend on external libraries).
1523 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1524 Patch provided by Graeme Fowler.
1526 NM/16 Bugzilla 851: Documentation example syntax fix.
1528 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1530 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1533 NM/19 Bugzilla 745: TLS version reporting.
1534 Patch provided by Phil Pennock.
1536 NM/20 Bugzilla 167: bool: condition support.
1537 Patch provided by Phil Pennock.
1539 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1540 clients. Patch provided by Phil Pennock.
1542 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1543 Patch provided by Brad "anomie" Jorsch.
1545 NM/23 Bugzilla 687: Fix misparses in eximstats.
1546 Patch provided by Heiko Schlittermann.
1548 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1549 Patch provided by Heiko Schlittermann.
1551 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1552 plus update to original patch.
1554 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1556 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1557 Patch provided by David Brownlee.
1559 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1561 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1563 NM/30 Bugzilla 888: TLS documentation bugfixes.
1565 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1567 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1568 Unlike the original bugzilla I have changed all shell scripts in src tree.
1570 NM/33 Bugzilla 898: Transport filter timeout fix.
1571 Patch by Todd Rinaldo.
1573 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1574 Patch by Serge Demonchaux.
1576 NM/35 Bugzilla 39: Base64 decode bug fixes.
1577 Patch by Jakob Hirsch.
1579 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1581 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1583 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1585 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1591 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1592 ALT-N's libdkim that I have put here:
1593 http://duncanthrax.net/exim-experimental/
1595 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1596 were called 'true' and it seems that C99 defines that as a reserved
1597 keyword to be used with 'bool' variable types. That means you could
1598 not include C99-style headers which use bools without triggering
1599 build errors in sieve.c.
1601 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1602 as mailq or other aliases. Changed the --help handling significantly
1603 to do whats expected. exim_usage() emits usage/help information.
1605 SC/01 Added the -bylocaldomain option to eximstats.
1607 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1609 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1611 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1617 PH/01 Another patch from the Sieve maintainer.
1619 PH/02 When an IPv6 address is converted to a string for single-key lookup
1620 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1621 dots are used instead of colons so that keys in lsearch files need not
1622 contain colons. This was done some time before quoting was made available
1623 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1624 (notated using the quote facility) so as to distinguish them from IPv4
1625 keys. This meant that lookups for IP addresses in host lists did not work
1626 for iplsearch lookups.
1628 This has been fixed by arranging for IPv6 addresses to be expressed with
1629 colons if the lookup type is iplsearch. This is not incompatible, because
1630 previously such lookups could never work.
1632 The situation is now rather anomolous, since one *can* have colons in
1633 ordinary lsearch keys. However, making the change in all cases is
1634 incompatible and would probably break a number of configurations.
1636 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1639 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1640 conversion specification without a maximum field width, thereby enabling
1641 a rogue spamd server to cause a buffer overflow. While nobody in their
1642 right mind would setup Exim to query an untrusted spamd server, an
1643 attacker that gains access to a server running spamd could potentially
1644 exploit this vulnerability to run arbitrary code as the Exim user.
1646 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1647 $primary_hostname instead of what libspf2 thinks the hosts name is.
1649 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1650 a directory entry by the name of the lookup key. Previously, if a
1651 symlink pointed to a non-existing file or a file in a directory that
1652 Exim lacked permissions to read, a lookup for a key matching that
1653 symlink would fail. Now it is enough that a matching directory entry
1654 exists, symlink or not. (Bugzilla 503.)
1656 PH/03 The body_linecount and body_zerocount variables are now exported in the
1659 PH/04 Added the $dnslist_matched variable.
1661 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1662 This means they are set thereafter only if the connection becomes
1665 PH/06 Added the client_condition to authenticators so that some can be skipped
1666 by clients under certain conditions.
1668 PH/07 The error message for a badly-placed control=no_multiline_responses left
1669 "_responses" off the end of the name.
1671 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1673 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1674 (without spaces) instead of just copying the configuration text.
1676 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1678 PH/11 Added $max_received_linelength.
1680 PH/12 Added +ignore_defer and +include_defer to host lists.
1682 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1683 way in which PCRE > 7.0 is built.
1685 PH/14 Implemented queue_only_load_latch.
1687 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1688 MAIL command. The effect was to mangle the value on 64-bit systems.
1690 PH/16 Another patch from the Sieve maintainer.
1692 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1694 PH/18 If a system quota error occurred while trying to create the file for
1695 a maildir delivery, the message "Mailbox is full" was not appended to the
1696 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1697 only to a quota excession during the actual writing of the file.
1699 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1700 characters?) which causes problems in log lines. The DN values are now
1701 passed through string_printing() before being added to log lines.
1703 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1704 and InterBase are left for another time.)
1706 PH/21 Added message_body_newlines option.
1708 PH/22 Guard against possible overflow in moan_check_errorcopy().
1710 PH/23 POSIX allows open() to be a macro; guard against that.
1712 PH/24 If the recipient of an error message contained an @ in the local part
1713 (suitably quoted, of course), incorrect values were put in $domain and
1714 $local_part during the evaluation of errors_copy.
1720 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1721 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1724 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1725 issue a MAIL command.
1727 PH/02 In an ACL statement such as
1729 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1731 if a client was not listed at all, or was listed with a value other than
1732 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1733 the condition was not true (as it should be), so access was not denied.
1734 The bug was that the ! inversion was incorrectly passed on to the second
1735 item. This has been fixed.
1737 PH/03 Added additional dnslists conditions == and =& which are different from
1738 = and & when the dns lookup returns more than one IP address.
1740 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1741 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1743 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1744 FSYNC, which compiles an option called disable_fsync that allows for
1745 bypassing fsync(). The documentation is heavily laced with warnings.
1747 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1749 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1750 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1751 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1752 including adding "make clean"; (3) Added -fPIC when compiling the test
1753 dynamically loaded module, to get rid of a warning.
1755 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1756 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1757 = 0s. The bug is otherwise harmless.
1759 PH/07 There was a bug in the dovecot authenticator such that the value of
1760 $auth1 could be overwritten, and so not correctly preserved, after a
1761 successful authentication. This usually meant that the value preserved by
1762 the server_setid option was incorrect.
1764 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1766 PH/09 Installed PCRE release 7.0.
1768 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1769 run for batched SMTP input. It is now run at the start of every message
1770 in the batch. While fixing this I discovered that the process information
1771 (output by running exiwhat) was not always getting set for -bs and -bS
1772 input. This is fixed, and it now also says "batched" for BSMTP.
1774 PH/11 Added control=no_pipelining.
1776 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1777 patch, slightly modified), and move the expansion of helo_data till after
1778 the connection is made in the smtp transport (so it can use these
1781 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1783 PH/14 Added log_selector = +pid.
1785 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1787 PH/16 Add ${if forany and ${if forall.
1789 PH/17 Added dsn_from option to vary the From: line in DSNs.
1791 PH/18 Flush SMTP output before performing a callout, unless control =
1792 no_callout_flush is set.
1794 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1795 was true (the default) a successful delivery failed to delete the retry
1796 item, thus causing premature timeout of the address. The bug is now
1799 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1801 PH/21 Long custom messages for fakedefer and fakereject are now split up
1802 into multiline reponses in the same way that messages for "deny" and
1803 other ACL rejections are.
1805 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1806 with slight modification.
1808 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1809 draft, changing the syntax and factoring some duplicate code".
1811 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1812 for deliveries of the second and subsequent messages over the same SMTP
1815 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1816 ${reduce, with only minor "tidies".
1818 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1820 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1821 expansion side effects.
1823 PH/27 When a message times out after an over-quota error from an Exim-imposed
1824 quota, the bounce message says "mailbox is full". This message was not
1825 being given when it was a system quota that was exceeded. It now should
1828 MH/03 Made $recipients available in local_scan(). local_scan() already has
1829 better access to the recipient list through recipients_list[], but
1830 $recipients can be useful in postmaster-provided expansion strings.
1832 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1833 in the case of a MAIL command with additional options following the
1834 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1835 were accidentally chopped off.
1837 PH/29 SMTP synchronization checks are implemented when a command is read -
1838 there is a check that no more input is waiting when there shouldn't be
1839 any. However, for some commands, a delay in an ACL can mean that it is
1840 some time before the response is written. In this time, more input might
1841 arrive, invalidly. So now there are extra checks after an ACL has run for
1842 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1843 pipelining has not been advertised.
1845 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1847 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1848 returned in the SMTP response when smtp_return_error_details was false.
1849 This has been fixed.
1851 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1852 instead of the C I/O that was originally supplied, because problems were
1853 reported on Solaris.
1855 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1856 Exim which did not show up earlier: it was assuming that a call to
1857 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1858 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1859 was a macro that became an assignment, so it seemed to work. This has
1860 changed to a proper function call with a void return, hence the compile
1861 error. Exim's code has been fixed.
1863 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1866 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1868 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1870 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1871 because it turns out to be too big or there is a timeout) there is no
1872 "Completed" line in the log. When some messages of this type were
1873 selected by exigrep, they were listed as "not completed". Others were
1874 picked up by some special patterns. I have improved the selection
1875 criteria to be more general.
1877 PH/38 The host_find_failed option in the manualroute router can now be set
1878 to "ignore", to completely ignore a host whose IP address cannot be
1879 found. If all hosts are ignored, the behaviour is controlled by the new
1880 host_all_ignored option.
1882 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1883 homing or because of multiple MX records with /mx) generated more than
1884 one IP address, and the following item turned out to be the local host,
1885 all the secondary addresses of the first item were incorrectly removed
1886 from the list, along with the local host and any following hosts (which
1887 is what is supposed to happen).
1889 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1890 whoever called Exim into the -H file. In the case of the daemon it was
1891 behaving confusingly. When first started, it used values for whoever
1892 started the daemon, but after a SIGHUP it used the Exim user (because it
1893 calls itself on a restart). I have changed the code so that it now always
1896 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1897 message are rejected with the same error (e.g. no authentication or bad
1898 sender address), and a DATA command is nevertheless sent (as can happen
1899 with PIPELINING or a stupid MUA), the error message that was given to the
1900 RCPT commands is included in the rejection of the DATA command. This is
1901 intended to be helpful for MUAs that show only the final error to their
1904 PH/42 Another patch from the Sieve maintainer.
1906 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1907 Eximstats - Fixed some broken HTML links and added missing column headers
1909 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1910 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1912 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1914 PH/43 Yet another patch from the Sieve maintainer.
1916 PH/44 I found a way to check for a TCP/IP connection going away before sending
1917 the response to the final '.' that terminates a message, but only in the
1918 case where the client has not sent further data following the '.'
1919 (unfortunately, this is allowed). However, in many cases there won't be
1920 any further data because there won't be any more messages to send. A call
1921 to select() can be used: if it shows that the input is "ready", there is
1922 either input waiting, or the socket has been closed. An attempt to read
1923 the next input character can distinguish the two cases. Previously, Exim
1924 would have sent an OK response which the client would never have see.
1925 This could lead to message repetition. This fix should cure that, at
1926 least in a lot of common cases.
1928 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1929 advertised in response to EHLO.
1935 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1936 fixed by 4.65/MH/01 (is this a record?) are fixed:
1938 (i) An empty string was always treated as zero by the numeric comparison
1939 operators. This behaviour has been restored.
1941 (ii) It is documented that the numeric comparison operators always treat
1942 their arguments as decimal numbers. This was broken in that numbers
1943 starting with 0 were being interpreted as octal.
1945 While fixing these problems I realized that there was another issue that
1946 hadn't been noticed. Values of message_size_limit (both the global option
1947 and the transport option) were treated as octal if they started with 0.
1948 The documentation was vague. These values are now always treated as
1949 decimal, and I will make that clear in the documentation.
1955 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1956 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1959 MH/01 Don't check that the operands of numeric comparison operators are
1960 integers when their expansion is in "skipping" mode (fixes bug
1961 introduced by 4.64-PH/07).
1963 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1964 child addresses, Exim now panics and dies. Previously, because the count
1965 is held in a short int, deliveries were likely to be lost. As such a
1966 large number of recipients for a single message is ridiculous
1967 (performance will be very, very poor), I have chosen to impose a limit
1968 rather than extend the field.
1974 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1975 leftover -K file (the existence of which was triggered by #402).
1976 While we were at it, introduced process PID as part of the -K
1977 filename. This should rule out race conditions when creating
1980 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1981 processing considerably. Previous code took too long for large mails,
1982 triggering a timeout which in turn triggers #401.
1984 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1985 in the DK code in transports.c. sendfile() is not really portable,
1986 hence the _LINUX specificness.
1988 TF/01 In the add_headers option to the mail command in an Exim filter,
1989 there was a bug that Exim would claim a syntax error in any
1990 header after the first one which had an odd number of characters
1993 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1994 callout verification, Exim cached a "reject" for the entire domain. This
1995 is correct for most verifications, but it is not correct for a recipient
1996 verification with use_sender or use_postmaster set, because in that case
1997 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1998 case of MAIL FROM:<> rejection from other early rejections (e.g.
1999 rejection of HELO). When verifying a recipient using a non-null MAIL
2000 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2001 Whatever the result of the callout, the value of the domain cache is
2002 left unchanged (for any other kind of callout, getting as far as trying
2003 RCPT means that the domain itself is ok).
2005 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2008 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2009 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2010 session, the daemon ignores ECONNECT errors and logs others; it now
2011 ignores EPIPE as well.
2013 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2014 (quoted-printable decoding).
2016 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2017 later the small subsequent patch to fix an introduced bug.
2019 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2021 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2023 PH/08 An error is now given if message_size_limit is specified negative.
2025 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2026 to be given (somewhat) arbitrary names.
2028 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2031 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2032 miscellaneous code fixes
2034 PH/10 Added the log_reject_target ACL modifier to specify where to log
2037 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2038 hostname. This is wrong, because it relates to the incoming message (and
2039 probably the interface on which it is arriving) and not to the outgoing
2040 callout (which could be using a different interface). This has been
2041 changed to use the value of the helo_data option from the smtp transport
2042 instead - this is what is used when a message is actually being sent. If
2043 there is no remote transport (possible with a router that sets up host
2044 addresses), $smtp_active_hostname is used.
2046 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2047 tweaks were necessary in order to get it to work (see also 21 below):
2048 (a) The code assumed that strncpy() returns a negative number on buffer
2049 overflow, which isn't the case. Replaced with Exim's string_format()
2051 (b) There were several signed/unsigned issues. I just did the minimum
2052 hacking in of casts. There is scope for a larger refactoring.
2053 (c) The code used strcasecmp() which is not a standard C function.
2054 Replaced with Exim's strcmpic() function.
2055 (d) The code set only $1; it now sets $auth1 as well.
2056 (e) A simple test gave the error "authentication client didn't specify
2057 service in request". It would seem that Dovecot has changed its
2058 interface. Fortunately there's a specification; I followed it and
2059 changed what the client sends and it appears to be working now.
2061 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2064 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2065 address A is aliased to B and C, where B exists and C does not. Without
2066 -v the output is "A verified" because verification stops after a
2067 successful redirection if more than one address is generated. However,
2068 with -v the child addresses are also verified. Exim was outputting "A
2069 failed to verify" and then showing the successful verification for C,
2070 with its parentage. It now outputs "B failed to verify", showing B's
2071 parentage before showing the successful verification of C.
2073 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2074 look up a TXT record in a specific list after matching in a combined
2077 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2078 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2079 they consult the DNS. I had assumed they would set it the way they
2080 wanted; and indeed my experiments on Linux seem to show that in some
2081 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2082 To be on the safe side, however, I have now made the interface to
2083 host_find_byname() similar to host_find_bydns(), with an argument
2084 containing the DNS resolver options. The host_find_byname() function now
2085 sets these options at its start, just as host_find_bydns() does. The smtp
2086 transport options dns_qualify_single and dns_search_parents are passed to
2087 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2088 of host_find_byname() use the default settings of RES_DEFNAMES
2089 (qualify_single) but not RES_DNSRCH (search_parents).
2091 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2092 spool_read_header() do less string testing, by means of a preliminary
2093 switch on the second character of optional "-foo" lines. (This is
2094 overdue, caused by the large number of possibilities that now exist.
2095 Originally there were few.) While I was there, I also converted the
2096 str(n)cmp tests so they don't re-test the leading "-" and the first
2097 character, in the hope this might squeeze out yet more improvement.
2099 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2100 flag allowing group syntax was set by the header_syntax check but not
2101 turned off, possible causing trouble later; (2) The flag was not being
2102 set at all for the header_verify test, causing "group"-style headers to
2103 be rejected. I have now set it in this case, and also caused header_
2104 verify to ignore an empty address taken from a group. While doing this, I
2105 came across some other cases where the code for allowing group syntax
2106 while scanning a header line wasn't quite right (mostly, not resetting
2107 the flag correctly in the right place). These bugs could have caused
2108 trouble for malformed header lines. I hope it is now all correct.
2110 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2111 with the "reply" argument non-NULL. The code, however (which originally
2112 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2113 but it didn't always do it. This confused somebody who was copying the
2114 code for some other use. I have removed all the tests.
2116 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2117 feature that was used to support insecure browsers during the U.S. crypto
2118 embargo. It requires special client support, and Exim is probably the
2119 only MTA that supported it -- and would never use it because real RSA is
2120 always available. This code has been removed, because it had the bad
2121 effect of slowing Exim down by computing (never used) parameters for the
2122 RSA_EXPORT functionality.
2124 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2125 authenticator to fail if there's a tab character in the incoming data
2126 (there should never be unless someone is messing about, as it's supposed
2127 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2128 if the connection is using TLS or if the remote IP is the same as the
2129 local IP, and the "valid-client-cert option" if a client certificate has
2132 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2133 authenticators. This can be used for authorization after authentication
2134 succeeds. (In the case of plaintext, it servers for both authentication
2137 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2138 if any retry times were supplied.
2140 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2141 connection, where there is no client IP address to check. In this
2142 situation, the verify now always succeeds.
2144 PH/25 Applied John Jetmore's -Mset patch.
2146 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2148 PH/27 In a string expansion for a processed (not raw) header when multiple
2149 headers of the same name were present, leading whitespace was being
2150 removed from all of them, but trailing whitespace was being removed only
2151 from the last one. Now trailing whitespace is removed from each header
2152 before concatenation. Completely empty headers in a concatenation (as
2153 before) are ignored.
2155 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2156 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2158 PH/29 [Removed. This was a change that I later backed out, and forgot to
2159 correct the ChangeLog entry (that I had efficiently created) before
2160 committing the later change.]
2162 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2163 address errors (4xx response to RCPT) over the same connection as other
2164 messages routed to the same hosts. Such deliveries are always "forced",
2165 so retry times are not inspected. This resulted in far too many retries
2166 for the affected addresses. The effect occurred only when there were more
2167 hosts than the hosts_max_try setting in the smtp transport when it had
2168 the 4xx errors. Those hosts that it had tried were not added to the list
2169 of hosts for which the message was waiting, so if all were tried, there
2170 was no problem. Two fixes have been applied:
2172 (i) If there are any address or message errors in an SMTP delivery, none
2173 of the hosts (tried or untried) are now added to the list of hosts
2174 for which the message is waiting, so the message should not be a
2175 candidate for sending over the same connection that was used for a
2176 successful delivery of some other message. This seems entirely
2177 reasonable: after all the message is NOT "waiting for some host".
2178 This is so "obvious" that I'm not sure why it wasn't done
2179 previously. Hope I haven't missed anything, but it can't do any
2180 harm, as the worst effect is to miss an optimization.
2182 (ii) If, despite (i), such a delivery is accidentally attempted, the
2183 routing retry time is respected, so at least it doesn't keep
2184 hammering the server.
2186 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2187 in ${readsocket because some servers need this prod.
2189 PH/32 Added some extra debug output when updating a wait-xxx database.
2191 PH/33 The hint "could be header name not terminated by colon", which has been
2192 given for certain expansion errors for a long time, was not being given
2193 for the ${if def:h_colon_omitted{... case.
2195 PH/34 The spec says: "With one important exception, whenever a domain list is
2196 being scanned, $domain contains the subject domain." There was at least
2197 one case where this was not true.
2199 PH/35 The error "getsockname() failed: connection reset by peer" was being
2200 written to the panic log as well as the main log, but it isn't really
2201 panic-worthy as it just means the connection died rather early on. I have
2202 removed the panic log writing for the ECONNRESET error when getsockname()
2205 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2206 runs only) independently of the message's sender address. This meant
2207 that, if the 4xx error was in fact related to the sender, a different
2208 message to the same recipient with a different sender could confuse
2209 things. In particualar, this can happen when sending to a greylisting
2210 server, but other circumstances could also provoke similar problems.
2211 I have changed the default so that the retry time for these errors is now
2212 based a combination of the sender and recipient addresses. This change
2213 can be overridden by setting address_retry_include_sender=false in the
2216 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2217 remote server are returned as part of bounce messages. This was not
2218 happening for LMTP over a pipe (the lmtp transport), but now it is the
2219 same for both kinds of LMTP.
2221 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2222 in header lines that were in fact CNAMEs. This is no longer the case.
2224 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2225 and queue runs started by the daemon processed all messages. This has
2226 been fixed so that -R and -S can now usefully be given with -q<time>.
2228 PH/40 Import PCRE release 6.7 (fixes some bugs).
2230 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2232 PH/42 Give an error if -q is specified more than once.
2234 PH/43 Renamed the variables $interface_address and $interface_port as
2235 $received_ip_address and $received_port, to make it clear that these
2236 values apply to message reception, and not to the outgoing interface when
2237 a message is delivered. (The old names remain recognized, of course.)
2239 PH/44 There was no timeout on the connect() call when using a Unix domain
2240 socket in the ${readsocket expansion. There now is.
2242 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2243 be meaningful with "accept".
2245 SC/01 Eximstats V1.43
2246 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2248 SC/02 Eximstats V1.44
2249 Use a glob alias rather than an array ref in the generated
2250 parser. This improves both readability and performance.
2252 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2253 Collect SpamAssassin and rejection statistics.
2254 Don't display local sender or destination tables unless
2255 there is data to show.
2256 Added average volumes into the top table text output.
2258 SC/04 Eximstats V1.46
2259 Collect data on the number of addresses (recipients)
2260 as well as the number of messages.
2262 SC/05 Eximstats V1.47
2263 Added 'Message too big' to the list of mail rejection
2264 reasons (thanks to Marco Gaiarin).
2266 SC/06 Eximstats V1.48
2267 Mainlog lines which have GMT offsets and are too short to
2268 have a flag are now skipped.
2270 SC/07 Eximstats V1.49 (Alain Williams)
2271 Added the -emptyok flag.
2273 SC/08 Eximstats V1.50
2274 Fixes for obtaining the IP address from reject messages.
2276 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2277 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2278 whitesspace changes from 4.64-PH/27
2280 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2283 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2284 are found, allow negative numbers in numeric criteria)
2286 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2288 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2289 to match changes made in 4.64-PH/43
2291 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2293 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2294 "domain" is actually an IP address, and force a failure. This locks out
2295 those revolvers/nameservers that support "A-for-A" lookups, in
2296 contravention of the specifications.
2298 PH/48 When a host name was looked up from an IP address, and the subsequent
2299 forward lookup of the name timed out, the host name was left in
2300 $sender_host_name, contrary to the specification.
2302 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2303 restricted to single-key lookups, Exim was not diagnosing an error if
2304 * or *@ was used with a query-style lookup.
2306 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2308 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2309 long ago, but noone interested enough thought of it. Let's just say that
2310 the "1.1" means that there are some new functions that weren't there at
2311 some point in the past.
2313 PH/51 Error processing for expansion failure of helo_data from an smtp
2314 transport during callout processing was broken.
2316 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2317 tested/used via the -bh/-bhc/-bs options.
2319 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2320 bug, fixed in subsequent PCRE releases).
2322 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2323 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2325 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2331 SC/01 Use a glob alias rather than an array ref in eximstats generated
2332 parser. This improves both readability and performance.
2334 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2335 Don't display local sender or destination tables in eximstats unless
2336 there is data to show.
2337 Added average volumes into the eximstats top table text output.
2339 SC/03 Collect data on the number of addresses (recipients) as well
2340 as the number of messages in eximstats.
2342 TF/01 Correct an error in the documentation for the redirect router. Exim
2343 does (usually) call initgroups() when daemonizing.
2345 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2346 with consistent privilege compared to when running as a daemon.
2348 TF/03 Note in the spec that $authenticated_id is not set for local
2349 submissions from trusted users.
2351 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2352 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2354 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2355 by adding some example configuration directives to the default
2356 configuration file. A little bit of work is required to uncomment the
2357 directives and define how usernames and passwords are checked, but
2358 there is now a framework to start from.
2360 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2361 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2362 without this. I don't know how relevant this is to other LDAP libraries.
2364 PH/02 Add the verb name to the "unknown ACL verb" error.
2366 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2368 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2370 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2371 directory not expanded when it should be if an expanded home directory
2372 was set for the address (which is overridden by the transport).
2374 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2377 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2378 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2379 because it is too late at that time, and has no effect.
2381 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2382 security issue with \' (bugzilla #107). I could not use the
2383 PQescapeStringConn() function, because it needs a PGconn value as one of
2386 PH/08 When testing addresses using -bt, indicate those final addresses that
2387 are duplicates that would not cause an additional delivery. At least one
2388 person was confused, thinking that -bt output corresponded to deliveries.
2389 (Suppressing duplicates isn't a good idea as you lose the information
2390 about possibly different redirections that led to the duplicates.)
2392 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2393 systems where poll() doesn't work, in particular OS X.
2395 PH/10 Added more information to debugging output for retry time not reached.
2397 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2398 operations in malware.c.
2400 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2403 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2404 syslog_duplication set false, log lines that would normally be written
2405 both the the main log and to the reject log were not written to syslog at
2408 PH/14 In the default configuration, change the use of "message" in ACL warn
2409 statements to "add_header".
2411 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2412 not followed by a command (e.g. "seen endif").
2414 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2415 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2418 PH/17 Added extra conditions to the default value of delay_warning_condition
2422 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2423 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2424 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2427 The Auto-Submitted: and various List- headers are standardised, whereas I
2428 don't think Precedence: ever was.
2430 PH/18 Refactored debugging code in route_finduser() to show more information,
2431 in particular, the error code if getpwnam() issues one.
2433 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2434 This is apparently needed in addition to the PH/07 change above to avoid
2435 any possible encoding problems.
2437 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2438 but not after initializing Perl.
2440 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2441 output them only if debugging. By default they are written stderr,
2442 apparently, which is not desirable.
2444 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2447 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2450 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2452 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2453 authenticated or an ident call has been made. Suppress the default
2454 values for $authenticated_id and $authenticated_sender (but permit -oMai
2455 and -oMas) when testing with -bh.
2457 PH/24 Re-jigged the order of the tests in the default configuration so that the
2458 tests for valid domains and recipients precede the DNS black list and CSA
2459 tests, on the grounds that those ones are more expensive.
2461 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2462 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2463 This bug exists in every version of Exim that I still have, right back to
2466 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2467 However, an attempt to turn on case-sensitivity in a regex key by
2468 including (?-i) didn't work because the subject string was already
2469 lowercased, and the effects were non-intuitive. It turns out that a
2470 one-line patch can be used to allow (?-i) to work as expected.
2476 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2477 other effects) broke the use of negated acl sub-conditions.
2479 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2482 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2483 "Deny" causes Exim to reject the incoming connection with a 554 error.
2484 Unfortunately, if there is a major crisis, such as a disk failure,
2485 tcp-wrappers gives "deny", whereas what one would like would be some
2486 kind of temporary error. A kludge has been added to help with this.
2487 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2488 554 error is used if errno is still zero or contains ENOENT (which occurs
2489 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2492 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2494 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2495 errors. Otherwise a message that provokes a temporary error (when other
2496 messages do not) can cause a whole host to time out.
2498 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2499 addresses were routed directly to files or pipes from a redirect router.
2500 File deliveries just didn't batch; pipe deliveries might have suffered
2503 PH/06 A failure to get a lock for a hints database would erroneously always say
2504 "Failed to get write lock", even when it was really a read lock.
2506 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2507 of 0600. This has been changed to use the value of the lockfile_mode
2508 option (which defaults to 0600).
2510 PH/08 Applied small patch from the Sieve maintainer.
2512 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2513 folder from quota calculations, a direct delivery into this folder messed
2514 up the contents of the maildirsize file. This was because the regex was
2515 used only to exclude .Trash (or whatever) when the size of the mailbox
2516 was calculated. There was no check that a delivery was happening into an
2517 excluded directory. This bug has been fixed by ignoring all quota
2518 processing for deliveries into excluded directories.
2520 PH/10 Added the maildirfolder_create_regex option to appendfile.
2526 PH/01 The code for finding all the local interface addresses on a FreeBSD
2527 system running IPv6 was broken. This may well have applied to all BSD
2528 systems, as well as to others that have similar system calls. The broken
2529 code found IPv4 interfaces correctly, but gave incorrect values for the
2530 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2531 that it would not match correctly against @[] and not recognize the IPv6
2534 PH/02 The ipliteral router was not recognizing addresses of the form user@
2535 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2537 PH/03 Added disable_ipv6.
2539 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2540 decoded form, because it is most often used to construct To: headers
2541 lines in autoreplies, and the decoded form may well be syntactically
2542 invalid. However, $reply_address has leading white space removed, and all
2543 newlines turned into spaces so that the autoreply transport does not
2546 PH/05 If group was specified without a user on a router, and no group or user
2547 was specified on a transport, the group from the router was ignored.
2549 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2550 for visible compile-time settings that can be used to change these
2551 numbers, for those that want even more. Backwards compatibility with old
2552 spool files has been maintained. However, going back to a previous Exim
2553 release will lost any variables that are in spool files.
2555 PH/07 Two small changes when running in the test harness: increase delay when
2556 passing a TCP/IP connection to a new process, in case the original
2557 process has to generate a bounce, and remove special handling of
2558 127.0.0.2 (sic), which is no longer necessary.
2560 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2561 be the same on different OS.
2563 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2566 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2567 whether --show-vars was specified or not
2569 JJ/02 exipick: Added support for new ACL variable spool format introduced
2572 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2573 syntactically invalid From: or Reply-to: line, and a filter used this to
2574 generate an autoreply, and therefore failed to obtain an address for the
2575 autoreply, Exim could try to deliver to a non-existent relative file
2576 name, causing unrelated and misleading errors. What now happens is that
2577 it logs this as a hard delivery error, but does not attempt to create a
2580 PH/11 The exinext utility has a -C option for testing purposes, but although
2581 the given file was scanned by exinext itself; it wasn't being passed on
2582 when Exim was called.
2584 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2585 an end-of-file indication when reading a command response.
2587 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2588 compiled. In many other places in Exim, IPv6 addresses are always
2589 recognized, so I have changed this. It also means that IPv4 domain
2590 literals of the form [IPV4:n.n.n.n] are now always recognized.
2592 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2593 used if the router is not running as root, for example, when verifying at
2594 ACL time, or when using -bh. The debugging output from this situation was
2595 non-existent - all you got was a failure to exec. I have made two
2598 (a) Failures to set uid/gid, the current directory, or a process leader
2599 in a subprocess such as that created by queryprogram now generate
2600 suitable debugging ouput when -d is set.
2602 (b) The queryprogram router detects when it is not running as root,
2603 outputs suitable debugging information if -d is set, and then runs
2604 the subprocess without attempting to change uid/gid.
2606 PH/15 Minor change to Makefile for building test_host (undocumented testing
2609 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2610 additional section of a DNS packet that returns MX or SRV records.
2611 Instead, it always explicitly searches for A/AAAA records. This avoids
2612 major problems that occur when a DNS server includes only records of one
2613 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2614 fixed another bug: if SRV records were looked up and the corresponding
2615 address records were *not* found in the additional section, the port
2616 values from the SRV records were lost.
2618 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2619 using the correct key (the original address) when searching the retry
2620 rules in order to find which one to use for generating the retry hint.
2622 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2623 adding the default one. Similarly, if it contains a Reply-To: header, the
2624 errors_reply_to option, if set, is not used.
2626 PH/19 When calculating a retry time, Exim used to measure the "time since
2627 failure" by looking at the "first failed" field in the retry record. Now
2628 it does not use this if it is later than than the arrival time of the
2629 message. Instead it uses the arrival time. This makes for better
2630 behaviour in cases where some deliveries succeed, thus re-setting the
2631 "first failed" field. An example is a quota failure for a huge message
2632 when small messages continue to be delivered. Without this change, the
2633 "time since failure" will always be short, possible causing more frequent
2634 delivery attempts for the huge message than are intended.
2635 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2637 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2638 $1, $2, $3) because the numerical variables can be reset during some
2639 expansion items (e.g. "match"), thereby losing the authentication data.
2641 PH/21 Make -bV show the size of off_t variables so that the test suite can
2642 decide whether to run tests for quotas > 2G.
2644 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2645 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2646 filecount value is greater than 2G or if a quota value is greater than 2G
2647 on a system where the size of off_t is not greater than 4, a panic error
2650 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2651 never match. The debug and -bh output now contains an explicit error
2652 message indicating a malformed IPv4 address or mask.
2654 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2655 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2656 PH/23 above applies.
2658 PH/25 Do not write to syslog when running in the test harness. The only
2659 occasion when this arises is a failure to open the main or panic logs
2660 (for which there is an explicit test).
2662 PH/26 Added the /no_tell option to "control=freeze".
2664 PH/27 If a host name lookup failed very early in a connection, for example, if
2665 the IP address matched host_lookup and the reverse lookup yielded a name
2666 that did not have a forward lookup, an error message of the form "no IP
2667 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2668 could be logged. Now it outputs the IP address instead of "NULL".
2670 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2671 allows the sender and the authenticated sender to be set when
2672 submitting a message from within Exim. Since child_open_exim() is
2673 documented for local_scan(), the new function should be too.
2675 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2676 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2677 results in an empty string is now treated as unset.
2679 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2681 PH/31 Added sender_verify_fail logging option.
2683 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2684 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2685 tidied the source and removed it altogether.
2687 PH/33 When a queue run was abandoned because the load average was too high, a
2688 log line was always written; now it is written only if the queue_run log
2689 selector is set. In addition, the log line for abandonment now contains
2690 information about the queue run such as the pid. This is always present
2691 in "start" and "stop" lines but was omitted from the "abandon" line.
2693 PH/34 Omit spaces between a header name and the colon in the error message that
2694 is given when verify = headers_syntax fails (if there are lots of them,
2695 the message gets confusing).
2697 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2698 names, as there are now some PTR records that contain slashes. This check
2699 is only to protect against broken name servers that fall over on strange
2700 characters, so the fact that it applies to all lookups doesn't matter.
2702 PH/36 Now that the new test suite is complete, we can remove some of the
2703 special code in Exim that was needed for the old test suite. For example,
2704 sorting DNS records because real resolvers return them in an arbitrary
2705 order. The new test suite's fake resolver always returns records in the
2708 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2709 bounces) except when queue_only is set, to avoid logging races between
2710 the different processes.
2712 PH/38 Panic-die if .include specifies a non-absolute path.
2714 PH/39 A tweak to the "H" retry rule from its user.
2716 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2717 a label. They prevented compilation on older perls.
2719 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2720 a warning to be raised on newish perls.
2722 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2723 on queue. Changes to match documented behaviour of showing count of
2724 messages matching specified criteria.
2726 PH/40 Changed the default ident timeout from 30s to 5s.
2728 PH/41 Added support for the use of login_cap features, on those BSD systems
2729 that have them, for controlling the resources used by pipe deliveries.
2731 PH/42 The content-scanning code uses fopen() to create files in which to put
2732 message data. Previously it was not paying any attention to the mode of
2733 the files. Exim runs with umask(0) because the rest of the code creates
2734 files with open(), and sets the required mode explicitly. Thus, these
2735 files were ending up world-writeable. This was not a big issue, because,
2736 being within the spool directory, they were not world-accessible. I have
2737 created a function called modefopen, which takes an additional mode
2738 argument. It sets umask(777), creates the file, chmods it to the required
2739 mode, then resets the umask. All the relevant calls to fopen() in the
2740 content scanning code have been changed to use this function.
2742 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2743 to 24 hours. This avoids potential overflow problems when processing G
2744 and H retry rules. I suspect nobody ever tinkers with this value.
2746 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2748 PH/45 When the plaintext authenticator is running as a client, the server's
2749 challenges are checked to ensure they are valid base64 strings. By
2750 default, the authentication attempt is cancelled if an invalid string is
2751 received. Setting client_ignore_invalid_base64 true ignores these errors.
2752 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2753 they are received. Thus, the responses can be made to depend on the
2754 challenges. If an invalid string is ignored, an empty string is placed in
2757 PH/46 Messages that are created by the autoreply transport now contains a
2758 References: header, in accordance with RFCs 2822 and 3834.
2760 PH/47 Added authenticated_sender_force to the smtp transport.
2762 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2764 PH/49 Installed latest patch from the Sieve maintainer.
2766 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2767 was also set, the appendfile transport was unnecessarily scanning a
2768 directory of message files (e.g. for maildir delivery) to find the count
2769 of files (along with the size), even though it did not need this
2770 information. It now does the scan only if it needs to find either the
2771 size of the count of files.
2773 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2775 PH/52 Two bugs concerned with error handling when the smtp transport is
2778 (i) Exim was not creating retry information for temporary errors given
2779 for individual recipients after the DATA command when the smtp transport
2780 was used in LMTP mode. This meant that they could be retried too
2781 frequently, and not timed out correctly.
2783 (ii) Exim was setting the flag that allows error details to be returned
2784 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2785 recipients that were returned after the DATA command.
2787 PH/53 This is related to PH/52, but is more general: for any failing address,
2788 when detailed error information was permitted to be returned to the
2789 sender, but the error was temporary, then after the final timeout, only
2790 "retry timeout exceeded" was returned. Now it returns the full error as
2791 well as "retry timeout exceeded".
2793 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2794 do this, and (what is worse) MTAs that accept it.
2796 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2797 will now be deprecated.
2799 PH/56 New os.c-cygwin from the Cygwin maintainer.
2801 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2802 formats (previously only available in exim formats via -bpr, -bpru,
2803 and -bpra. Now also available in native and exiqgrep formats)
2805 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2806 with very large, slow to parse queues
2808 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2810 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2812 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2813 responses to authentication challenges, though it was showing the
2814 challenges; (ii) I've removed the CR characters from the debug output for
2817 PH/58 Allow for the insertion of a newline as well as a space when a string
2818 is turned into more than one encoded-word during RFC 2047 encoding. The
2819 Sieve code now uses this.
2821 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2822 data_4xx, lost_connection, tls_required.
2824 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2825 message was being sent as an SMTP response.
2827 PH/61 Add -l and -k options to exicyclog.
2829 PH/62 When verifying, if an address was redirected to one new address, so that
2830 verification continued, and the new address failed or deferred after
2831 having set something in $address_data, the value of $address_data was not
2832 passed back to the ACL. This was different to the case when no
2833 redirection occurred. The value is now passed back in both cases.
2835 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2836 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2837 use login_cap.h, so on its own it isn't the distinguishing feature. The
2838 new name refers directly to the setclassresources() function.
2840 PH/65 Added configuration files for NetBSD3.
2842 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2844 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2845 is preferred over IPv4.
2847 PH/68 The bounce_return_message and bounce_return_body options were not being
2848 honoured for bounces generated during the reception of non-SMTP messages.
2849 In particular, this applied to messages rejected by the ACL. This bug has
2850 been fixed. However, if bounce_return_message is true and bounce_return_
2851 body is false, the headers that are returned for a non-SMTP message
2852 include only those that have been read before the error was detected.
2853 (In the case of an ACL rejection, they have all been read.)
2855 PH/69 The HTML version of the specification is now built in a directory called
2856 spec_html instead of spec.html, because the latter looks like a path with
2857 a MIME-type, and this confuses some software.
2859 PH/70 Catch two compiler warnings in sieve.c.
2861 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2862 function verify_get_ident() calls ip_connect() to connect a socket, but
2863 if the "connect()" function timed out, ip_connect() used to close the
2864 socket. However, verify_get_ident() also closes the socket later, and in
2865 between Exim writes to the log, which may get opened at this point. When
2866 the socket was closed in ip_connect(), the log could get the same file
2867 descriptor number as the socket. This naturally causes chaos. The fix is
2868 not to close the socket in ip_connect(); the socket should be closed by
2869 the function that creates it. There was only one place in the code where
2870 this was missing, in the iplookup router, which I don't think anybody now
2871 uses, but I've fixed it anyway.
2873 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2874 well as to direct DNS lookups. Otherwise the handling of names in host
2875 lists is inconsistent and therefore confusing.
2881 PH/01 Two changes to the default runtime configuration:
2883 (1) Move the checks for relay_from_hosts and authenticated clients from
2884 after to before the (commented out) DNS black list checks.
2886 (2) Add control=submission to the relay_from_hosts and authenticated
2887 clients checks, on the grounds that messages accepted by these
2888 statements are most likely to be submissions.
2890 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2892 (1) Generate an error if the third argument for the ${prvs expansion is
2895 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2898 (3) Reset the variables that are obtained from the first argument of
2899 ${prvscheck and used in the second argument before leaving the code,
2900 because their memory is reclaimed, so using them afterwards may do
2903 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2904 one (it's much easier than Tom thought :-).
2906 (5) Because of (4), we can now allow for the use of $prvscheck_result
2907 inside the third argument.
2909 PH/03 For some reason, the default setting of PATH when running a command from
2910 a pipe transport was just "/usr/bin". I have changed it to
2913 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2914 anything to be listed in the output from -bV.
2916 PH/05 When a filter generated an autoreply, the entire To: header line was
2917 quoted in the delivery log line, like this:
2919 => >A.N.Other <ano@some.domain> <original@ddress> ...
2921 This has been changed so that it extracts the operative address. There
2922 may be more than one such address. If so, they are comma-separated, like
2925 => >ano@some.domain,ona@other.domain <original@ddress> ...
2927 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2928 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2929 not being looked up in the rDNS to get a host name, Exim was showing the
2930 IP address twice in Received: lines, even though the IP addresses were
2931 identical. For example:
2933 Received: from [1.2.3.4] (helo=[1.2.3.4])
2935 However, if the real host name was known, it was omitting the HELO data
2936 if it matched the actual IP address. This has been tidied up so that it
2937 doesn't show the same IP address twice.
2939 PH/07 When both +timestamp and +memory debugging was on, the value given by
2940 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2941 called by the debug printing, thereby overwriting the timestamp buffer.
2942 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2944 PH/08 When the original message was included in an autoreply transport, it
2945 always said "this is a copy of the message, including all the headers",
2946 even if body_only or headers_only was set. It now gives an appropriate
2949 PH/09 Applied a patch from the Sieve maintainer which:
2951 o fixes some comments
2952 o adds the (disabled) notify extension core
2953 o adds some debug output for the result of if/elsif tests
2954 o points to the current vacation draft in the documentation
2955 and documents the missing references header update
2959 o fixes a bug in processing the envelope test (when testing
2960 multiple envelope elements, the last element determinted the
2963 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2964 Electronic Mail") by including:
2966 Auto-submitted: auto-generated
2968 in the messages that it generates (bounce messages and others, such as
2969 warnings). In the case of bounce messages for non-SMTP mesages, there was
2970 also a typo: it was using "Auto_submitted" (underscore instead of
2971 hyphen). Since every message generated by Exim is necessarily in response
2972 to another message, thes have all been changed to:
2974 Auto-Submitted: auto-replied
2976 in accordance with these statements in the RFC:
2978 The auto-replied keyword:
2980 - SHOULD be used on messages sent in direct response to another
2981 message by an automatic process,
2983 - MUST NOT be used on manually-generated messages,
2985 - MAY be used on Delivery Status Notifications (DSNs) and Message
2986 Disposition Notifications (MDNs),
2988 - MUST NOT be used on messages generated by automatic or periodic
2989 processes, except for messages which are automatic responses to
2992 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2993 to the default Received: header definition.
2995 PH/12 Added log selector acl_warn_skipped (default on).
2997 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2998 variables because (a) they are in the wrong storage pool and (b) even if
2999 they were copied, it wouldn't work properly because of the caching.
3001 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3002 checking when decoding. Apparently there are clients that generate
3003 overlong encoded strings. Why am I not surprised?
3005 PH/15 If the first argument of "${if match_address" was not empty, but did not
3006 contain an "@" character, Exim crashed. Now it writes a panic log message
3007 and treats the condition as false.
3009 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3011 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3012 extension "envelope-auth". The code is finished and in agreement with
3013 other implementations, but there is no documentation so far and in fact,
3014 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3015 not changing the active code.
3017 Print executed "if" and "elsif" statements when debugging is used. This
3018 helps a great deal to understand what a filter does.
3020 Document more things not specified clearly in RFC3028. I had all this
3021 sorted out, when out of a sudden new issues came to my mind. Oops."
3023 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3026 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3027 canonical form (as documented). However, after a host name lookup from
3028 the IP address, check_host() was doing a simple string comparison with
3029 addresses acquired from the DNS when checking that the found name did
3030 have the original IP as one of its addresses. Since any found IPv6
3031 addresses are likely to be in abbreviated form, the comparison could
3032 fail. Luckily, there already exists a function for doing the comparison
3033 by converting both addresses to binary, so now that is used instead of
3034 the text comparison.
3036 PH/20 There was another similar case to PH/19, when a complete host name was
3037 given in a host list; looking up its IP address could give an abbreviated
3038 form, whereas the current host's name might or might not be abbreviated.
3039 The same fix has been applied.
3045 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3046 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3049 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3050 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3052 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3054 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3055 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3056 Also, for "Auto-Submitted", treat anything other than "no" as
3057 non-personal, in accordance with RFC 3834. (Previously it treated
3058 anything starting "auto-" as non-personal.)
3060 TF/01 The control=submission/name=... option had a problem with syntax
3061 errors if the name included a slash character. The /name= option
3062 now slurps the rest of the string, so it can include any characters
3063 but it must come last in the list of options (after /sender_retain
3066 PH/05 Some modifications to the interface to the fake nameserver for the new
3074 TK/01 Added the "success_on_redirect" address verification option. See
3075 NewStuff for rationale and an example.
3077 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3079 PH/02 Patch to exigrep to allow it to work on syslog lines.
3081 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3082 fread() to skip over the body file's header line, because in Cygwin the
3083 header line is locked and is inaccessible.
3085 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3086 co-exist for some time) to make it clear that it is the Exim ID that is
3087 referenced, not the Message-ID: header line.
3089 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3090 string_format() function, because snprintf() does not exist on all
3093 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3094 ${stat: expansion item.
3096 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3097 protocol synchronization error", to keep the pedants happy.
3099 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3100 well as for IRIX systems, when gcc is being used. See the host.c source
3103 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3105 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3108 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3109 transport and to the smtp transport in LMTP mode.
3111 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3113 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3114 run a filter in a subprocess. This could lead to confusion in subsequent
3115 lookups in the parent process. There should also be a search_tidyup() at
3116 the end of the subprocess.
3118 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3119 only if the host matched helo_try_verify_hosts, which caused the
3120 verification to occur when the EHLO/HELO command was issued. The ACL just
3121 tested the remembered result. Now, if a previous verification attempt has
3122 not happened, "verify = helo" does it there and then.
3124 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3126 TK/03 Fix log output including CR from clamd.
3128 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3129 exist provoked a memory error which could cause a segfault.
3131 PH/15 Installed PCRE 6.2
3133 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3135 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3136 of the problem. Specifically, suggested +O2 rather than +O1 for the
3139 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3141 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3142 record, the port was not correctly logged when the outgoing_port log
3143 selector was set (it logged the transort's default port).
3145 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3146 fallback_hosts, and "hosts" in the smtp transport.
3148 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3149 host errors such as "Connection refused".
3151 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3152 authentication with radiusclient 0.4.9:
3154 - Error returned from rc_read_config was caught wrongly
3155 - Username/password not passed on to radius server due to wrong length.
3157 The presumption is that some radiusclient API changes for 4.51/PH/17
3158 were not taken care of correctly. The code is still untested by me (my
3159 Linux distribution still has 0.3.2 of radiusclient), but it was
3160 contributed by a Radius user.
3162 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3163 expanding the "port" option of the smtp transport.
3165 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3166 while reading a MIME header. Thanks to Tom Hughes for a patch.
3168 PH/24 Include config.h inside local_scan.h so that configuration settings are
3171 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3172 that in an ACL for RCPT (for example), you can examine exactly what was
3175 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3176 commands, but it was not correctly comparing the address with the actual
3177 client host address. Thus, it would show the EHLO address in Received:
3178 header lines when this was not necessary.
3180 PH/27 Added the % operator to ${eval:}.
3182 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3183 it should be ignoring failures (because with -C, for example, it has lost
3184 privilege). It wasn't ignoring creation failures other than "already
3187 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3190 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3191 because an input line was too long, either on its own, or by virtue of
3192 too many continuations, the temporary file was not being removed, and the
3193 return code was incorrect.
3195 PH/31 Missing "BOOL" in function definition in filtertest.c.
3197 PH/32 Applied Sieve patches from the maintainer.
3199 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3201 PH/33 Added "verify = not_blind".
3203 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3204 Local/Makefile (with some defaults set). These are used in built scripts
3205 such as exicyclog, but they have never been used in the exim_install
3206 script (though there are many overriding facilities there). I have
3207 arranged that the exim_install script now takes note of these two
3210 PH/35 Installed configuration files for Dragonfly.
3212 PH/36 When a locally submitted message by a trusted user did not contain a
3213 From: header, and the sender address was obtained from -f or from an SMTP
3214 MAIL command, and the trusted user did not use -F to supply a sender
3215 name, $originator_name was incorrectly used when constructing a From:
3216 header. Furthermore, $originator_name was used for submission mode
3217 messages from external hosts without From: headers in a similar way,
3218 which is clearly wrong.
3220 PH/37 Added control=suppress_local_fixups.
3222 PH/38 When log_selector = +received_sender was set, and the addition of the
3223 sender made the log line's construction buffer exactly full, or one byte
3224 less than full, an overflow happened when the terminating "\n" was
3227 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3228 when the result of a list match is failure because a DNS lookup failed.
3230 PH/40 RM_COMMAND is now used in the building process.
3232 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3233 the "build-* directories that it finds.
3235 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3236 address in a domain literal was a prefix of an interface address.
3238 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3239 when verifying a sender address, unless rewrite_headers is false.
3241 PH/44 Wrote a long comment about why errors_to addresses are verified as
3242 recipients, not senders.
3244 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3245 the ratelimit ACL was added.
3247 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3249 PH/46 Added extra information about PostgreSQL errors to the error string.
3251 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3252 avoiding the need to install special zones in a real server. This is
3253 backwards compatible; if it can't find the fake resolver, it drops back.
3254 Thus, both old and new test suites can be run.
3256 TF/02 Added util/ratelimit.pl
3258 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3259 clock is set back in time.
3261 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3262 Candler <B.Candler@pobox.com>.
3264 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3265 OK for addresses that are the result of redirections.
3267 PH/48 A number of further additions for the benefit of the new test suite,
3268 including a fake gethostbyname() that interfaces to the fake DNS resolver
3271 TF/06 The fix for widen_domains has also been applied to qualify_single and
3272 search_parents which are the other dnslookup options that can cause
3275 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3278 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3280 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3281 service" if it encountered an MX record with an empty target hostname.
3282 The message is now "an MX or SRV record indicated no SMTP service".
3284 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3285 if the DNS of the sending site is misconfigured. This is quite a
3286 common situation. This change restores the behaviour of treating a
3287 helo verification defer as a failure.
3289 PH/51 If self=fail was set on a router, the bounce message did not include the
3290 actual error message.
3296 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3298 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3299 command itself ended in error, the underlying message about the transport
3300 filter timeout was being overwritten with the pipe command error. Now the
3301 underlying error message should be appended to the second error message.
3303 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3305 PH/02 Reduce the amount of output that "make" produces by default. Full output
3306 can still be requested.
3308 PH/03 The warning log line about a condition test deferring for a "warn" verb
3309 was being output only once per connection, rather than after each
3310 occurrence (because it was using the same function as for successful
3311 "warn" verbs). This seems wrong, so I have changed it.
3313 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3314 it should not have, which might have caused a crash in the right
3315 circumstances, but probably never did.
3317 PH/04 Installed a modified version of Tony Finch's patch to make submission
3318 mode fix the return path as well as the Sender: header line, and to
3319 add a /name= option so that you can make the user's friendly name appear
3322 TF/03 Added the control = fakedefer ACL modifier.
3324 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3325 Mark Lowes for thorough testing.
3327 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3329 TK/03 Merged latest SRS patch from Miles Wilton.
3331 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3332 with the definition in sysexits.h (which is #included earlier).
3333 Fortunately, Exim does not actually use EX_OK. The code used to try to
3334 preserve the sysexits.h value, by assumimg that macro definitions were
3335 scanned for macro replacements. I have been disabused of this notion,
3336 so now the code just undefines EX_OK before #including unistd.h.
3338 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3339 in the smtp transport. When a block could not be written in a single
3340 write() function, the timeout was being re-applied to each part-write.
3341 This seems wrong - if the receiver was accepting one byte at a time it
3342 would take for ever. The timeout is now adjusted when this happens. It
3343 doesn't have to be particularly precise.
3345 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3346 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3348 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3349 without a domain if the check to <postmaster@domain> fails.
3351 SC/01 Eximstats: added -xls and the ability to specify output files
3352 (patch written by Frank Heydlauf).
3354 SC/02 Eximstats: use FileHandles for outputing results.
3356 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3358 SC/04 Eximstats: fixed display of large numbers with -nvr option
3360 SC/05 Eximstats: fixed merging of reports with empty tables.
3362 SC/06 Eximstats: added the -include_original_destination flag
3364 SC/07 Eximstats: removed tabs and trailing whitespace.
3366 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3368 TK/06 MBOX spool code: Add real "From " MBOX separator line
3369 so the .eml file is really in mbox format (even though
3370 most programs do not really care). Patch from Alex Miller.
3372 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3373 The latter is generated from $received_to and is only set if the
3374 message has one envelope recipient. SA can use these headers,
3375 obviously out-of-the-box. Patch from Alex Miller.
3377 PH/08 The ${def test on a variable was returning false if the variable's
3378 value was "0", contrary to what the specification has always said!
3379 The result should be true unless the variable is empty.
3381 PH/09 The syntax error of a character other than { following "${if
3382 def:variable_name" (after optional whitespace) was not being diagnosed.
3383 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3384 accidental colon was present, for example, could give incorrect results.
3386 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3387 result is used (not including appendfile, where other changes are about
3390 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3391 This involved changing a lot of size variables from int to off_t. It
3392 should work with maildirs and everything.
3394 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3395 spamd dying while we are connected to it.
3397 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3400 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3401 function for BDB 4.3.
3403 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3405 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3406 the sources for PCRE have been re-arranged and more files are now
3409 PH/15 The code I had for printing potentially long long variables in PH/11
3410 above was not the best (it lost precision). The length of off_t variables
3411 is now inspected at build time, and an appropriate printing format (%ld
3412 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3413 to be "long long int" or "long int". This is needed for the internal
3414 formatting function string_vformat().
3416 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3417 the configuration file to be ":syslog", then the script "guesses" where
3418 the logs files are, rather than using the compiled in default. In our
3419 case the guess is not the same as the compiled default, so the script
3420 suddenly stopped working when I started to use syslog. The patch checks
3421 to see if log_file_path is "". If so, it attempts to read it from exim
3422 with no configuration file to get the compiled in version, before it
3423 falls back to the previous guessing code."
3425 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3426 implementing BATV in an Exim configuration. See NewStuff for the gory
3429 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3430 Makefile that are specific to HP-UX.
3432 PH/18 If the "use_postmaster" option was set for a recipient callout together
3433 with the "random" option, the postmaster address was used as the MAIL
3434 FROM address for the random test, but not for the subsequent recipient
3435 test. It is now used for both.
3437 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3438 patch removes a few documentation additions to RFC 3028, because the
3439 latest draft now contains them. It adds the new en;ascii-case comparator
3440 and a new error check for 8bit text in MIME parts. Comparator and
3441 require names are now matched exactly. I enabled the subaddress
3442 extension, but it is not well tested yet (read: it works for me)."
3444 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3445 rework some of the code of TK/09 above to avoid the hardwired use of
3446 "%lld" and "long long". Replaced the call to snprintf() with a call to
3449 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3450 records point to non-existent hosts", "retry timeout exceeded", and
3451 "retry time not reached for any host after a long failure period".
3453 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3454 experimental DomainKeys support:
3456 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3457 (2) On an error such as an illegally used "control", the wrong name for
3458 the control was given.
3460 These problems did NOT occur unless DomainKeys support was compiled.
3462 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3464 PH/24 Added ${if match_ip condition.
3466 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3467 databases so that it will be absolutely obvious if a crash occurs in the
3468 DB library. This is a regular occurrence (often caused by mis-matched
3471 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3472 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3473 picked up on a user's system that detects such things. There doesn't seem
3474 to be a gcc warning option for this - only an attribute that has to be
3475 put on the function's prototype. It seems that in Fedora Core 4 they have
3476 set this on a number of new functions. No doubt there will be more in due
3479 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3480 specify a transport. However, if an address that was verified by such a
3481 router was the subject of a callout, Exim crashed because it tried to
3482 read the rcpt_include_affixes from the non-existent transport. Now it
3483 just assumes that the setting of that option is false. This bug was
3484 introduced by 4.51/PH/31.
3486 PH/28 Changed -d+all to exclude +memory, because that information is very
3487 rarely of interest, but it makes the output a lot bigger. People tend to
3488 do -d+all out of habit.
3490 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3491 code in os-type was giving problems when libc.so lives in lib64, like on
3494 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3495 aren't the modern standard, and it seems that some systems' include files
3496 don't always have them. Exim was already checking for some of the newer
3497 ones like T_AAAA, and defining it itself. I've added checks for all the
3498 record types that Exim uses.
3500 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3501 not automatically generating a new one, as it is supposed to. This
3502 prevented TLS from working. If the file did exist, but contained invalid
3503 data, a new version was generated, as expected. It was only the case of a
3504 non-existent file that was broken.
3506 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3507 with a change in libdomainkeys > 0.64.
3509 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3510 from DNS. If the selector record carries the flag, it now has
3511 precedence over the domain-wide flag.
3513 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3515 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3516 the use of a port name that isn't defined in /etc/services) occurred, the
3517 message was deferred as in a normal delivery, and thus remained on the
3518 spool, instead of being failed because of the mua_wrapper setting. This
3519 is now fixed, and I tidied up some of the mua_wrapper messages at the
3522 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3523 the messages in a hash of arrays rather than using individual hashes.
3524 This is a bit cleaner and results in dramatic memory savings, albeit
3525 at a slight CPU cost.
3527 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3528 as requested by Marc Sherman.
3530 SC/10 Eximstats: added histograms for user specified patterns as requested
3533 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3535 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3536 fopen() in the content-scanning modules that did not already have it.
3542 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3543 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3545 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3547 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3549 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3550 to test Sieve filters that use "vacation".
3552 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3553 that changes the way the GnuTLS parameters are stored in the cache file.
3554 The new format can be generated externally. For backward compatibility,
3555 if the data in the cache doesn't make sense, Exim assumes it has read an
3556 old-format file, and it generates new data and writes a new file. This
3557 means that you can't go back to an older release without removing the
3560 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3561 work if there are any delivery delays because "one_time" forces the
3562 parent to be marked "delivered", so its unseen clone is never tried
3563 again. For this reason, Exim now forbids the simultaneous setting of
3566 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3567 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3568 entry if you want to know the details. The fix, however, neglected to
3569 consider the case when local delivery batching is involved. The test for
3570 "previously delivered" was not happening when checking to see if an
3571 address could be batched with a previous (undelivered) one; under
3572 certain circumstances this could lead to multiple deliveries to the same
3575 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3576 in its include files, and this causes problems building Exim.
3578 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3579 header_syntax) but Exim was just ignoring anything given after a slash.
3580 In particular, this caused confusion with an attempt to use "verify =
3581 reverse_host_lookup/defer_ok". An error is now given when options are
3582 supplied for verify items that do not have them. (Maybe reverse_host_
3583 lookup should have a defer_ok option, but that's a different point.)
3585 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3586 defined by RFC 821) to 2048, because there were problems with some AUTH
3587 commands, and RFC 1869 says the size should be increased for extended
3588 SMTP commands that take arguments.
3590 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3593 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3594 "unknown" error; now it says that the functionality isn't in the binary.
3596 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3597 an address' error message when a string expansion fails (syntax or
3598 whatever). Otherwise the password may appear in the log. Following change
3599 PH/42 below, there is no longer a chance of it appearing in a bounce
3602 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3604 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3605 of its addresses was ever tried. (Bugzilla bug #2.)
3607 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3608 the result incorrectly in the debug output. (It correctly added a newline
3609 to what was transported.)
3611 TF/01 Added $received_time.
3613 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3614 commented out examples of how to interface to a virus scanner and to
3615 SpamAssassin. Also added commented examples of av_scanner and
3616 spamd_address settings.
3618 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3619 and controls are allowed in which ACLs. There were a couple of minor
3620 errors. Some of the entries in the conditions table (which is a table of
3621 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3622 negation of where the condition IS allowed.
3624 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3626 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3627 header file does not have a version number, so I've had to invent a new
3628 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3629 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3630 radiusclient), but it was contributed by a Radius user.
3632 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3633 files or directories, for OpenSSL.
3635 PH/19 When an Exim process that is running as root has to create an Exim log
3636 file, it does so in a subprocess that runs as exim:exim so as to get the
3637 ownership right at creation (otherwise, other Exim processes might see
3638 the file with the wrong ownership). There was no test for failure of this
3639 fork() call, which would lead to the process getting stuck as it waited
3640 for a non-existent subprocess. Forks do occasionally fail when resources
3641 run out. I reviewed all the other calls to fork(); they all seem to check
3644 PH/20 When checking for unexpected SMTP input at connect time (before writing
3645 the banner), Exim was not dealing correctly with a non-positive return
3646 from the read() function. If the client had disconnected by this time,
3647 the result was a log entry for a synchronization error with an empty
3648 string after "input=" when read() returned zero. If read() returned -1
3649 (an event I could not check), uninitialized data bytes were printed.
3650 There were reports of junk text (parts of files, etc) appearing after
3653 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3655 PH/22 Added support for macro redefinition, and (re)definition in between
3656 driver and ACL definitions.
3658 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3659 forgetting to use the resulting value; it was using the unexpanded value.
3661 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3662 hadn't been configured. The fix is from Juergen Kreileder, who
3663 understands it better than I do:
3665 "Here's what I see happening with three configured cyrus_sasl
3666 authenticators configured (plain, login, cram-md5):
3668 On startup auth_cyrus_sasl_init() gets called for each of these.
3669 This means three calls to sasl_listmech() without a specified mech_list.
3670 => SASL tests which mechs of all available mechs actually work
3671 => three warnings about OTP not working
3672 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3674 With the patch, sasl_listmech() also gets called three times. But now
3675 SASL's mech_list option is set to the server_mech specified in the the
3676 authenticator. Or in other words, the answer from sasl_listmech()
3677 gets limited to just the mech you're testing for (which is different
3679 => the return list contains just 'plain' or 'login', 'cram-md5' or
3680 nothing depending on the value of ob->server_mech.
3682 I've just tested the patch: Authentication still works fine,
3683 unavailable mechs specified in the exim configuration are still
3684 caught, and the auth.log warnings about OTP are gone."
3686 PH/25 When debugging is enabled, the contents of the command line are added
3687 to the debugging output, even when log_selector=+arguments is not
3690 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3691 answer is "GNU", and only if the return is "GNU/something" is the answer
3694 PH/27 $acl_verify_message is now set immediately after the failure of a
3695 verification in an ACL, and so is available in subsequent modifiers. In
3696 particular, the message can be preserved by coding like this:
3698 warn !verify = sender
3699 set acl_m0 = $acl_verify_message
3701 Previously, $acl_verify_message was set only while expanding "message"
3702 and "log_message" when a very denied access.
3704 PH/28 Modified OS/os.c-Linux with
3706 -#ifndef OS_LOAD_AVERAGE
3707 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3709 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3710 nomenclature these days.)
3712 PH/29 Installed patch from the Sieve maintainer that adds the options
3713 sieve_useraddress and sieve_subaddress to the redirect router.
3715 PH/30 In these circumstances:
3716 . Two addresses routed to the same list of hosts;
3717 . First host does not offer TLS;
3718 . First host accepts first address;
3719 . First host gives temporary error to second address;
3720 . Second host offers TLS and a TLS session is established;
3721 . Second host accepts second address.
3722 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3723 and peerdn, if requested) that were in fact used only for the second
3726 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3727 attention to any local part prefix or suffix that was matched by the
3728 router that accepted the address. It now behaves in the same way as it
3729 does for delivery: the affixes are removed from the local part unless
3730 rcpt_include_affixes is set on the transport.
3732 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3733 timeout during the DATA phase of an incoming message.
3735 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3736 applied a patch sanctioned by the Sieve maintainer.
3738 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3739 the uid or gid is negative. A case of a negative gid caused this to be
3740 noticed. The fix allows for either to be negative.
3742 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3743 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3746 PH/36 The change PH/12 above was broken. Fixed it.
3748 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3749 the grounds that routing the same address twice would always produce the
3750 same answer. This might have been true once, but it is certainly no
3751 longer true now. Routing a child address may depend on the previous
3752 routing that produced that child. Some complicated redirection strategies
3753 went wrong when messages had multiple recipients, and made Exim's
3754 behaviour dependent on the order in which the addresses were given.
3756 I have moved the duplicate checking until after the routing is complete.
3757 Exim scans the addresses that are assigned to local and remote
3758 transports, and removes any duplicates. This means that more work will be
3759 done, as duplicates will always all be routed, but duplicates are
3760 presumably rare, so I don't expect this is of any significance.
3762 For deliveries to pipes, files, and autoreplies, the duplicate checking
3763 still happens during the routing process, since they are not going to be
3766 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3767 It corrects a timeout issue with spamd. This is Ian's comment: "The
3768 background is that sometimes spamd either never reads data from a
3769 connection it has accepted, or it never writes response data. The exiscan
3770 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3771 blindly assumes that writes won't block so it may never time out."
3773 PH/39 Allow G after quota size as well as K and M.
3775 PH/40 The value set for $authenticated_id in an authenticator may not contain
3776 binary zeroes or newlines because the value is written to log lines and
3777 to spool files. There was no check on this. Now the value is run through
3778 the string_printing() function so that such characters are converted to
3779 printable escape sequences.
3781 PH/41 $message_linecount is a new variable that contains the total number of
3782 lines in the message. Compare $body_linecount, which is the count for the
3785 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3786 bounce and delay warning messages, except in certain special cases, which
3789 (a) An SMTP error message from a remote host;
3790 (b) A message specified in a :fail: redirection;
3791 (c) A message specified in a "fail" command in a system filter;
3792 (d) A message specified in a FAIL return from the queryprogram router;
3793 (e) A message specified by the cannot_route_message router option.
3795 In these cases only, Exim does include the error details in bounce and
3796 warning messages. There are also a few cases where bland messages such
3797 as "unrouteable address" or "local delivery error" are given.
3799 PH/43 $value is now also set for the "else" part of a ${run expansion.
3801 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3802 being worked on, but at least Exim now implements the latest version to
3805 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3806 process to complete was treated as a delivery failure, a timeout while
3807 writing the message to the pipe was logged, but erroneously treated as a
3808 successful delivery. Such timeouts include transport filter timeouts. For
3809 consistency with the overall process timeout, these timeouts are now
3810 treated as errors, giving rise to delivery failures by default. However,
3811 there is now a new Boolean option for the pipe transport called
3812 timeout_defer, which, if set TRUE, converts the failures into defers for
3813 both kinds of timeout. A transport filter timeout is now identified in
3816 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3817 systems where "make" and "gmake" are different, calling "gmake" at top
3818 level broke things. I've arranged for the value of $(MAKE) to be passed
3819 from the Makefile to this script so that it can call the same version of
3823 A note about Exim versions 4.44 and 4.50
3824 ----------------------------------------
3826 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3827 changes of various kinds. As a consequence, a big documentation update was
3828 needed. This delayed the release for rather longer than seemed good, especially
3829 in the light of a couple of (minor) security issues. Therefore, the changes
3830 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3831 release. So 4.44 and 4.50 are in effect two different branches that both start
3834 I have left the 4.50 change log unchanged; it contains all the changes since
3835 4.43. The change log for 4.44 is below; many of its items are identical to
3836 those for 4.50. This seems to be the most sensible way to preserve the
3837 historical information.
3843 1. Minor wording change to the doc/README.SIEVE file.
3845 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3846 computation of the current number of files was incorrect.
3848 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3849 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3852 4. Give more explanation in the error message when the command for a transport
3853 filter fails to execute.
3855 5. There are several places where Exim runs a non-Exim command in a
3856 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3857 was being done only for the command run by the queryprogram router. It is
3858 now done for all such subprocesses. The other cases are: ${run, transport
3859 filters, and the commands run by the lmtp and pipe transports.
3861 6. Added CONFIGURE_GROUP build-time option.
3863 7. Some older OS have a limit of 256 on the maximum number of file
3864 descriptors. Exim was using setrlimit() to set 1000 as a large value
3865 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3866 systems. I've change it so that if it can't get 1000, it tries for 256.
3868 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3869 was an oversight, and furthermore, ever since the addition of extra
3870 controls (e.g. 4.43/32), the checks on when to allow different forms of
3871 "control" were broken. There should now be diagnostics for all cases when a
3872 control that does not make sense is encountered.
3874 9. Added the /retain_sender option to "control=submission".
3876 10. $recipients is now available in the predata ACL (oversight).
3878 11. Tidy the search cache before the fork to do a delivery from a message
3879 received from the command line. Otherwise the child will trigger a lookup
3880 failure and thereby defer the delivery if it tries to use (for example) a
3881 cached ldap connection that the parent has called unbind on.
3883 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3884 of $address_data from the recipient verification was clobbered by the
3885 sender verification.
3887 13. The value of address_data from a sender verification is now available in
3888 $sender_address_data in subsequent conditions in the ACL statement.
3890 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3892 15. Added a new option "connect=<time>" to callout options, to set a different
3895 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3896 was its contents. (It was OK if the option was not defined at all.)
3898 17. A "Completed" log line is now written for messages that are removed from
3899 the spool by the -Mrm option.
3901 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3902 information about exactly what failed.
3904 19. Added -dd to debug only the daemon process.
3906 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3907 handles timeouts, both on the server side and network timeouts. Renamed the
3908 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3910 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3911 It is now set to "smtps".
3913 22. $host_address is now set to the target address during the checking of
3914 ignore_target_hosts.
3916 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3917 being passed; this would have caused $sender_host_name to have been used if
3918 matching the list had actually called for a host name (not very likely,
3919 since this list is usually IP addresses). A host name is now passed as
3922 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3923 code that specifies a non-restarting handler (typically sigaction() in
3924 modern systems) in an attempt to fix a rare and obscure crash bug.
3926 25. Narrowed the window for a race in the daemon that could cause it to ignore
3927 SIGCHLD signals. This is not a major problem, because they are used only to
3928 wake it up if nothing else does.
3930 26. A malformed maildirsize file could cause Exim to calculate negative values
3931 for the mailbox size or file count. Odd effects could occur as a result.
3932 The maildirsize information is now recalculated if the size or filecount
3935 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3936 support for a long time. Removed HAVE_SYS_VFS_H.
3938 28. Installed the latest version of exipick from John Jetmore.
3940 29. In an address list, if the pattern was not a regular expression, an empty
3941 subject address (from a bounce message) matched only if the pattern was an
3942 empty string. Non-empty patterns were not even tested. This was the wrong
3943 because it is perfectly reasonable to use an empty address as part of a
3944 database query. An empty address is now tested by patterns that are
3945 lookups. However, all the other forms of pattern expect the subject to
3946 contain a local part and a domain, and therefore, for them, an empty
3947 address still always fails if the pattern is not itself empty.
3949 30. Exim went into a mad DNS loop when attempting to do a callout where the
3950 host was specified on an smtp transport, and looking it up yielded more
3951 than one IP address.
3953 31. Re-factored the code for checking spool and log partition space into a
3954 function that finds that data and another that does the check. The former
3955 is then used to implement four new variables: $spool_space, $log_space,
3956 $spool_inodes, and $log_inodes.
3958 32. The RFC2047 encoding function was originally intended for short strings
3959 such as real names; it was not keeping to the 75-character limit for
3960 encoded words that the RFC imposes. It now respects the limit, and
3961 generates multiple encoded words if necessary. To be on the safe side, I
3962 have increased the buffer size for the ${rfc2047: expansion operator from
3965 33. It is now permitted to omit both strings after an "if" condition; if the
3966 condition is true, the result is "true". As before, when the second string
3967 is omitted, a false condition yields an empty string. This makes it less
3968 cumbersome to write custom ACL and router conditions.
3970 34. Failure to deliver a bounce message always caused it to be frozen, even if
3971 there was an errors_to setting on the router. The errors_to setting is now
3974 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3975 canonical form (fully expanded) before being placed in
3976 $sender_host_address.
3978 36. The table in the code that translates DNS record types into text (T_A to
3979 "A" for instance) was missing entries for NS and CNAME. It is just possible
3980 that this could have caused confusion if both these types were looked up
3981 for the same domain, because the text type is used as part of Exim's
3982 per-process caching. But the chance of anyone hitting this buglet seems
3985 37. The dnsdb lookup has been extended in a number of ways.
3987 (1) There is a new type, "zns", which walks up the domain tree until it
3988 finds some nameserver records. It should be used with care.
3990 (2) There is a new type, "mxh", which is like "mx" except that it returns
3991 just the host names, not the priorities.
3993 (3) It is now possible to give a list of domains (or IP addresses) to be
3994 looked up. The behaviour when one of the lookups defers can be
3995 controlled by a keyword.
3997 (4) It is now possible to specify the separator character for use when
3998 multiple records are returned.
4000 38. The dnslists ACL condition has been extended: it is now possible to supply
4001 a list of IP addresses and/or domains to be looked up in a particular DNS
4004 39. Added log_selector=+queue_time_overall.
4006 40. When running the queue in the test harness, wait just a tad after forking a
4007 delivery process, to get repeatability of debugging output.
4009 41. Include certificate and key file names in error message when GnuTLS fails
4010 to set them up, because the GnuTLS error message doesn't include the name
4011 of the failing file when there is a problem reading it.
4013 42. Allow both -bf and -bF in the same test run.
4015 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4017 44. The "Exiscan patch" is now merged into the mainline Exim source.
4019 45. Sometimes the final signoff response after QUIT could fail to get
4020 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4021 < 0 before doing a fflush(). This bug looks as though it goes back to the
4022 introduction of TLS in release 3.20, but "sometimes" must have been rare
4023 because the tests only now provoked it.
4025 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4026 (this can affect the format of dates).
4028 47. exim_tidydb, when checking for the continued existence of a message for
4029 which it has found a message-specific retry record, was not finding
4030 messages that were in split spool directories. Consequently, it was
4031 deleting retry records that should have stayed in existence.
4033 48. Steve fixed some bugs in eximstats.
4035 49. The SPA authentication driver was not abandoning authentication and moving
4036 on to the next authenticator when an expansion was forced to fail,
4037 contradicting the general specification for all authenticators. Instead it
4038 was generating a temporary error. It now behaves as specified.
4040 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4041 (the order specifies the preference for clients). The order is now AES256,
4042 AES128, 3DES, ARCFOUR128.
4044 51. Small patch to Sieve code - explicitly set From: when generating an
4047 52. Exim crashed if a remote delivery caused a very long error message to be
4048 recorded - for instance if somebody sent an entire SpamAssassin report back
4049 as a large number of 550 error lines. This bug was coincidentally fixed by
4050 increasing the size of one of Exim's internal buffers (big_buffer) that
4051 happened as part of the Exiscan merge. However, to be on the safe side, I
4052 have made the code more robust (and fixed the comments that describe what
4055 53. Now that there can be additional text after "Completed" in log lines (if
4056 the queue_time_overall log selector is set), a one-byte patch to exigrep
4057 was needed to allow it to recognize "Completed" as not the last thing in
4060 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4061 patch that reportedly fixes this has been added. I am not expert enough to
4062 create a test for it. This is what the patch creator wrote:
4064 "I found a little strange behaviour of ldap code when working with
4065 Windows 2003 AD Domain, where users was placed in more than one
4066 Organization Units. When I tried to give exim partial DN, the exit code
4067 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4068 But simultaneously result of request was absolutely normal ldap result,
4069 so I produce this patch..."
4071 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4072 REFERENCE, so I have modified the code to exclude the patch when that macro
4075 55. Some experimental protocols are using DNS PTR records for new purposes. The
4076 keys for these records are domain names, not reversed IP addresses. The
4077 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4078 leaves it alone. Component reversal etc. now happens only for IP addresses.
4081 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4083 57. Double the size of the debug message buffer (to 2048) so that more of very
4084 long debug lines gets shown.
4086 58. The exicyclog utility now does better if the number of log files to keep
4087 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4089 59. Two changes related to the smtp_active_hostname option:
4091 (1) $smtp_active_hostname is now available as a variable.
4092 (2) The default for smtp_banner uses $smtp_active_hostname instead
4093 of $primary_hostname.
4095 60. The host_aton() function is supposed to be passed a string that is known
4096 to be a valid IP address. However, in the case of IPv6 addresses, it was
4097 not checking this. This is a hostage to fortune. Exim now panics and dies
4098 if the condition is not met. A case was found where this could be provoked
4099 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4100 components; fortuitously, this particular loophole had already been fixed
4101 by change 4.50/55 above.
4103 If there are any other similar loopholes, the new check in host_aton()
4104 itself should stop them being exploited. The report I received stated that
4105 data on the command line could provoke the exploit when Exim was running as
4106 exim, but did not say which command line option was involved. All I could
4107 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4108 running as the user.
4111 61. There was a buffer overflow vulnerability in the SPA authentication code
4112 (which came originally from the Samba project). I have added a test to the
4113 spa_base64_to_bits() function which I hope fixes it.
4116 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4117 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4119 63. The daemon start-up calls getloadavg() while still root for those OS that
4120 need the first call to be done as root, but it missed one case: when
4121 deliver_queue_load_max is set with deliver_drop_privilege. This is
4122 necessary for the benefit of the queue runner, because there is no re-exec
4123 when deliver_drop_privilege is set.
4125 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4126 This has been fixed.
4128 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4129 was in use, was not putting the data itself into the right store pool;
4130 consequently, it could be overwritten for a subsequent message in the same
4131 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4134 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4136 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4137 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4138 the function were treating the return as a boolean value, which happened to
4139 work because 0=false and not-0=true, but is not correct code.
4141 68. The host_aton() function was not handling scoped IPv6 addresses (those
4142 with, for example, "%eth0" on the end) correctly.
4144 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4145 negated items (that is, ~something) in unsigned ints. Some compilers
4146 apparently mutter when there is no cast.
4148 70. If an address verification called from an ACL failed, and did not produce a
4149 user-specific message (i.e. there was only a "system" message), nothing was
4150 put in $acl_verify_message. In this situation, it now puts the system
4153 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4154 session; change 4.31/43 added the unwanted input to the log line - except
4155 that it did not do this in the start of session case. It now does.
4157 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4158 This is wrong and can cause the other end to generate a synchronization
4159 error if it is another Exim or anything else that does the synchronization
4160 check. A QUIT command is no longer sent after a timeout.
4162 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4163 during host lookups.
4165 74. The defer_ok option of callout verification was not working if it was used
4166 when verifying addresses in header lines, that is, for this case:
4168 verify = header_sender/callout=defer_ok
4170 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4171 those file descriptors could be used for SMTP connections. If anything
4172 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4173 could be sent to the SMTP client, causing chaos. The daemon now opens
4174 stdin, stdout, and stderr to /dev/null when it puts itself into the
4177 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4178 log by default. The user can override this with suitable Perl magic.
4180 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4181 the log message when discard triggers, was not working for the DATA ACL or
4182 for the non-SMTP ACL.
4184 78. Error message wording change in sieve.c.
4186 79. If smtp_accept_max_per_host was set, the number of connections could be
4187 restricted to fewer than expected, because the daemon was trying to set up
4188 a new connection before checking whether the processes handling previous
4189 connections had finished. The check for completed processes is now done
4190 earlier. On busy systems, this bug wouldn't be noticed because something
4191 else would have woken the daemon, and it would have reaped the completed
4194 80. If a message was submitted locally by a user whose login name contained one
4195 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4196 It caused a spool format error. I have fixed the spool reading code. A
4197 related problem was that the "from" clause in the Received: line became
4198 illegal because of the space(s). It is now covered by ${quote_local_part.
4200 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4203 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4205 83. Patch from Sieve maintainer to fix unterminated string problem in
4206 "vacation" handling.
4208 84. Some minor changes to the Linux configuration files to help with other
4209 OS variants using glibc.
4211 85. One more patch for Sieve to update vacation handling to latest spec.
4214 ----------------------------------------------------
4215 See the note above about the 4.44 and 4.50 releases.
4216 ----------------------------------------------------
4222 1. Change 4.43/35 introduced a bug that caused file counts to be
4223 incorrectly computed when quota_filecount was set in an appendfile
4226 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4227 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4230 3. Give more explanation in the error message when the command for a transport
4231 filter fails to execute.
4233 4. There are several places where Exim runs a non-Exim command in a
4234 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4235 was being done only for the command run by the queryprogram router. It is
4236 now done for all such subprocesses. The other cases are: ${run, transport
4237 filters, and the commands run by the lmtp and pipe transports.
4239 5. Some older OS have a limit of 256 on the maximum number of file
4240 descriptors. Exim was using setrlimit() to set 1000 as a large value
4241 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4242 systems. I've change it so that if it can't get 1000, it tries for 256.
4244 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4245 was an oversight, and furthermore, ever since the addition of extra
4246 controls (e.g. 4.43/32), the checks on when to allow different forms of
4247 "control" were broken. There should now be diagnostics for all cases when a
4248 control that does not make sense is encountered.
4250 7. $recipients is now available in the predata ACL (oversight).
4252 8. Tidy the search cache before the fork to do a delivery from a message
4253 received from the command line. Otherwise the child will trigger a lookup
4254 failure and thereby defer the delivery if it tries to use (for example) a
4255 cached ldap connection that the parent has called unbind on.
4257 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4258 of $address_data from the recipient verification was clobbered by the
4259 sender verification.
4261 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4262 was its contents. (It was OK if the option was not defined at all.)
4264 11. A "Completed" log line is now written for messages that are removed from
4265 the spool by the -Mrm option.
4267 12. $host_address is now set to the target address during the checking of
4268 ignore_target_hosts.
4270 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4271 being passed; this would have caused $sender_host_name to have been used if
4272 matching the list had actually called for a host name (not very likely,
4273 since this list is usually IP addresses). A host name is now passed as
4276 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4277 code that specifies a non-restarting handler (typically sigaction() in
4278 modern systems) in an attempt to fix a rare and obscure crash bug.
4280 15. Narrowed the window for a race in the daemon that could cause it to ignore
4281 SIGCHLD signals. This is not a major problem, because they are used only to
4282 wake it up if nothing else does.
4284 16. A malformed maildirsize file could cause Exim to calculate negative values
4285 for the mailbox size or file count. Odd effects could occur as a result.
4286 The maildirsize information is now recalculated if the size or filecount
4289 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4290 support for a long time. Removed HAVE_SYS_VFS_H.
4292 18. Updated exipick to current release from John Jetmore.
4294 19. Allow an empty sender to be matched against a lookup in an address list.
4295 Previously the only cases considered were a regular expression, or an
4298 20. Exim went into a mad DNS lookup loop when doing a callout where the
4299 host was specified on the transport, if the DNS lookup yielded more than
4302 21. The RFC2047 encoding function was originally intended for short strings
4303 such as real names; it was not keeping to the 75-character limit for
4304 encoded words that the RFC imposes. It now respects the limit, and
4305 generates multiple encoded words if necessary. To be on the safe side, I
4306 have increased the buffer size for the ${rfc2047: expansion operator from
4309 22. Failure to deliver a bounce message always caused it to be frozen, even if
4310 there was an errors_to setting on the router. The errors_to setting is now
4313 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4314 canonical form (fully expanded) before being placed in
4315 $sender_host_address.
4317 24. Updated eximstats to version 1.33
4319 25. Include certificate and key file names in error message when GnuTLS fails
4320 to set them up, because the GnuTLS error message doesn't include the name
4321 of the failing file when there is a problem reading it.
4323 26. Expand error message when OpenSSL has problems setting up cert/key files.
4326 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4327 (this can affect the format of dates).
4329 28. exim_tidydb, when checking for the continued existence of a message for
4330 which it has found a message-specific retry record, was not finding
4331 messages that were in split spool directories. Consequently, it was
4332 deleting retry records that should have stayed in existence.
4334 29. eximstats updated to version 1.35
4335 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4336 1.35 - bugfix such that pie charts by volume are generated correctly
4338 30. The SPA authentication driver was not abandoning authentication and moving
4339 on to the next authenticator when an expansion was forced to fail,
4340 contradicting the general specification for all authenticators. Instead it
4341 was generating a temporary error. It now behaves as specified.
4343 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4344 (the order specifies the preference for clients). The order is now AES256,
4345 AES128, 3DES, ARCFOUR128.
4347 31. Small patch to Sieve code - explicitly set From: when generating an
4350 32. Exim crashed if a remote delivery caused a very long error message to be
4351 recorded - for instance if somebody sent an entire SpamAssassin report back
4352 as a large number of 550 error lines. This bug was coincidentally fixed by
4353 increasing the size of one of Exim's internal buffers (big_buffer) that
4354 happened as part of the Exiscan merge. However, to be on the safe side, I
4355 have made the code more robust (and fixed the comments that describe what
4358 33. Some experimental protocols are using DNS PTR records for new purposes. The
4359 keys for these records are domain names, not reversed IP addresses. The
4360 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4361 leaves it alone. Component reversal etc. now happens only for IP addresses.
4364 34. The host_aton() function is supposed to be passed a string that is known
4365 to be a valid IP address. However, in the case of IPv6 addresses, it was
4366 not checking this. This is a hostage to fortune. Exim now panics and dies
4367 if the condition is not met. A case was found where this could be provoked
4368 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4369 components; fortuitously, this particular loophole had already been fixed
4370 by change 4.50/55 or 4.44/33 above.
4372 If there are any other similar loopholes, the new check in host_aton()
4373 itself should stop them being exploited. The report I received stated that
4374 data on the command line could provoke the exploit when Exim was running as
4375 exim, but did not say which command line option was involved. All I could
4376 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4377 running as the user.
4380 35. There was a buffer overflow vulnerability in the SPA authentication code
4381 (which came originally from the Samba project). I have added a test to the
4382 spa_base64_to_bits() function which I hope fixes it.
4385 36. The daemon start-up calls getloadavg() while still root for those OS that
4386 need the first call to be done as root, but it missed one case: when
4387 deliver_queue_load_max is set with deliver_drop_privilege. This is
4388 necessary for the benefit of the queue runner, because there is no re-exec
4389 when deliver_drop_privilege is set.
4391 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4392 was in use, was not putting the data itself into the right store pool;
4393 consequently, it could be overwritten for a subsequent message in the same
4394 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4397 38. Sometimes the final signoff response after QUIT could fail to get
4398 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4399 < 0 before doing a fflush(). This bug looks as though it goes back to the
4400 introduction of TLS in release 3.20, but "sometimes" must have been rare
4401 because the tests only now provoked it.
4407 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4408 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4409 Now it returns an integer. A number of calls were still expecting a T/F
4410 return. Fortuitously, in all cases, the tests worked in OK situations,
4411 which is the norm. However, things would have gone wrong on any write
4412 failures on the smtp file descriptor. This function is used when sending
4413 messages over SMTP and also when doing verify callouts.
4415 2. When Exim is called to do synchronous delivery of a locally submitted
4416 message (the -odf or -odi options), it no longer closes stderr before doing
4419 3. Implemented the mua_wrapper option.
4421 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4423 5. Implemented the functions header_remove(), header_testname(),
4424 header_add_at_position(), and receive_remove_recipient(), and exported them
4427 6. If an ACL "warn" statement specified the addition of headers, Exim already
4428 inserted X-ACL-Warn: at the start if there was no header name. However, it
4429 was not making this test for the second and subsequent header lines if
4430 there were newlines in the string. This meant that an invalid header could
4431 be inserted if Exim was badly configured.
4433 7. Allow an ACL "warn" statement to add header lines at the start or after all
4434 the Received: headers, as well as at the end.
4436 8. Added the rcpt_4xx retry error code.
4438 9. Added postmaster_mailfrom=xxx to callout verification option.
4440 10. Added mailfrom=xxxx to the callout verification option, for verify=
4443 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4444 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4446 12. Inserted some casts to stop certain compilers warning when using pointer
4447 differences as field lengths or precisions in printf-type calls (mostly
4448 affecting debugging statements).
4450 13. Added optional readline() support for -be (dynamically loaded).
4452 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4453 same clock tick as a message's arrival, so that its received time was the
4454 same as the "first fail" time on the retry record, and that message
4455 remained on the queue past the ultimate address timeout, every queue runner
4456 would try a delivery (because it was past the ultimate address timeout) but
4457 after another failure, the ultimate address timeout, which should have then
4458 bounced the address, did not kick in. This was a "< instead of <=" error;
4459 in most cases the first failure would have been in the next clock tick
4460 after the received time, and all would be well.
4462 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4463 being recognized when the domain list was tested by the match_domain
4464 condition in an expansion string.
4466 16. Added the ${str2b64: operator.
4468 17. Exim was always calling setrlimit() to set a large limit for the number of
4469 processes, without checking whether the existing limit was already
4470 adequate. (It did check for the limit on file descriptors.) Furthermore,
4471 errors from getrlimit() and setrlimit() were being ignored. Now they are
4472 logged to the main and panic logs, but Exim does carry on, to try to do its
4473 job under whatever limits there are.
4475 18. Imported PCRE 5.0.
4477 19. Trivial typo in log message " temporarily refused connection" (the leading
4480 20. If the log selector return_path_on_delivery was set and an address was
4481 redirected to /dev/null, the delivery process crashed because it assumed
4482 that a return path would always be set for a "successful" delivery. In this
4483 case, the whole delivery is bypassed as an optimization, and therefore no
4486 21. Internal re-arrangement: the function for sending a challenge and reading
4487 a response while authentication was assuming a zero-terminated challenge
4488 string. It's now changed to take a pointer and a length, to allow for
4489 binary data in such strings.
4491 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4493 23. Exim was not respecting finduser_retries when seeking the login of the
4494 uid under which it was called; it was always trying 10 times. (The default
4495 setting of finduser_retries is zero.) Also, it was sleeping after the final
4496 failure, which is pointless.
4498 24. Implemented tls_on_connect_ports.
4500 25. Implemented acl_smtp_predata.
4502 26. If the domain in control=submission is set empty, Exim assumes that the
4503 authenticated id is a complete email address when it generates From: or
4504 Sender: header lines.
4506 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4507 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4508 chown and chgrp in /bin and hostname in /usr/bin.
4510 28. Exim was keeping the "process log" file open after each use, just as it
4511 does for the main log. This opens the possibility of it remaining open for
4512 long periods when the USR1 signal hits a daemon. Occasional processlog
4513 errors were reported, that could have been caused by this. Anyway, it seems
4514 much more sensible not to leave this file open at all, so that is what now
4517 29. The long-running daemon process does not normally write to the log once it
4518 has entered its main loop, and it closes the log before doing so. This is
4519 so that log files can straightforwardly be renamed and moved. However,
4520 there are a couple of unusual error situations where the daemon does write
4521 log entries, and I had neglected to close the log afterwards.
4523 30. The text of an SMTP error response that was received during a remote
4524 delivery was being truncated at 512 bytes. This is too short for some of
4525 the long messages that one sometimes sees. I've increased the limit to
4528 31. It is now possible to make retry rules that apply only when a message has a
4529 specific sender, in particular, an empty sender.
4531 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4532 it possible to be selective about when SMTP synchronization is enforced.
4534 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4536 32. Implemented hosts_connection_nolog.
4538 33. Added an ACL for QUIT.
4540 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4543 35. Added mailbox_size and mailbox_filecount to appendfile.
4545 36. Added control = no_multiline_responses to ACLs.
4547 37. There was a bug in the logic of the code that waits for the clock to tick
4548 in the case where the clock went backwards by a substantial amount such
4549 that the microsecond fraction of "now" was more than the microsecond
4550 fraction of "then" (but the whole seconds number was less).
4552 38. Added support for the libradius Radius client library this is found on
4553 FreeBSD (previously only the radiusclient library was supported).
4559 1. When certain lookups returned multiple values in the form name=value, the
4560 quoting of the values was not always being done properly. Specifically:
4561 (a) If the value started with a double quote, but contained no whitespace,
4563 (b) If the value contained whitespace other than a space character (i.e.
4564 tabs or newlines or carriage returns) it was not quoted.
4565 This fix has been applied to the mysql and pgsql lookups by writing a
4566 separate quoting function and calling it from the lookup code. The fix
4567 should probably also be applied to nisplus, ibase and oracle lookups, but
4568 since I cannot test any of those, I have not disturbed their existing code.
4570 2. A hit in the callout cache for a specific address caused a log line with no
4571 reason for rejecting RCPT. Now it says "Previous (cached) callout
4572 verification failure".
4574 3. There was an off-by-one bug in the queryprogram router. An over-long
4575 return line was truncated at 256 instead of 255 characters, thereby
4576 overflowing its buffer with the terminating zero. As well as fixing this, I
4577 have increased the buffer size to 1024 (and made a note to document this).
4579 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4580 when Exim is waiting for an SMTP response from a remote server, Exim
4581 restarts its select() call on the socket, thereby resetting its timeout.
4582 This is not a problem when such interrupts are rare. Somebody set up a cron
4583 job to run exiwhat every 2 minutes, which is less than the normal select()
4584 timeout (5 or 10 minutes). This meant that the select() timeout never
4585 kicked in because it was always reset. I have fixed this by comparing the
4586 time when an interrupt arrives with the time at the start of the first call
4587 to select(). If more time than the timeout has elapsed, the interrupt is
4588 treated as a timeout.
4590 5. Some internal re-factoring in preparation for the addition of Sieve
4591 extensions (by MH). In particular, the "personal" test is moved to a
4592 separate function, and given an option for scanning Cc: and Bcc: (which is
4593 not set for Exim filters).
4595 6. When Exim created an email address using the login of the caller as the
4596 local part (e.g. when creating a From: or Sender: header line), it was not
4597 quoting the local part when it contained special characters such as @.
4599 7. Installed new OpenBSD configuration files.
4601 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4602 try to make them clearer.
4604 9. Callout options, other than the timeout value, were being ignored when
4605 verifying sender addresses in header lines. For example, when using
4607 verify = header_sender/callout=no_cache
4609 the cache was (incorrectly) being used.
4611 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4612 only the Cygwin environment.
4614 11. When return_path_on_delivery was set as a log selector, if different remote
4615 addresses in the same message used different return paths and parallel
4616 remote delivery occurred, the wrong values would sometimes be logged.
4617 (Whenever a remote delivery process finished, the return path value from
4618 the most recently started remote delivery process was logged.)
4620 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4621 lines when AUTH and/or TLS are in use. This is the "received protocol"
4622 field. Exim used to use "asmtp" for authenticated SMTP, without any
4623 indication (in the protocol name) for TLS use. Now it follows the RFC and
4624 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4625 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4626 names appear in log lines as well as in Received: header lines.
4628 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4629 extensions, and comparison tests, and to fix some bugs.
4631 14. Changes to the "personal" filter test:
4633 (1) The test was buggy in that it was just doing the equivalent of
4634 "contains" tests on header lines. For example, if a user's address was
4635 anne@some.where, the "personal" test would incorrectly be true for
4637 To: susanne@some.where
4639 This test is now done by extracting each address from the header in turn,
4640 and checking the entire address. Other tests that are part of "personal"
4641 are now done using regular expressions (for example, to check local parts
4642 of addresses in From: header lines).
4644 (2) The list of non-personal local parts in From: addresses has been
4645 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4646 taken from the Sieve specification recommendations.
4648 (3) If the message contains any header line starting with "List-" it is
4649 treated as non-personal.
4651 (4) The test for "circular" in the Subject: header line has been removed
4652 because it now seems ill-conceived.
4654 15. Minor typos in src/EDITME comments corrected.
4656 16. Installed latest exipick from John Jetmore.
4658 17. If headers_add on a router specified a text string that was too long for
4659 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4660 of string_sprintf() is now avoided.
4662 18. $message_body_size was not set (it was always zero) when running the DATA
4663 ACL and the local_scan() function.
4665 19. For the "mail" command in an Exim filter, no default was being set for
4666 the once_repeat time, causing a random time value to be used if "once" was
4667 specified. (If the value happened to be <= 0, no repeat happened.) The
4668 default is now 0s, meaning "never repeat". The "vacation" command was OK
4669 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4670 (I found it when inspecting the code).
4672 20. There is now an overall timeout for performing a callout verification. It
4673 defaults to 4 times the callout timeout, which applies to individual SMTP
4674 commands during the callout. The overall timeout applies when there is more
4675 than one host that can be tried. The timeout is checked before trying the
4676 next host. This prevents very long delays if there are a large number of
4677 hosts and all are timing out (e.g. when the network connections are timing
4678 out). The value of the overall timeout can be changed by specifying an
4679 additional sub-option for "callout", called "maxwait". For example:
4681 verify = sender/callout=5s,maxwait=20s
4683 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4684 to the end before writing, but this should make it even safer).
4686 22. Exim was forgetting that it had advertised PIPELINING for the second and
4687 subsequent messages on an SMTP connection. It was also not resetting its
4688 memory on STARTTLS and an internal HELO.
4690 23. When Exim logs an SMTP synchronization error within a session, it now
4691 records whether PIPELINING has been advertised or not.
4693 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4694 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4695 rather than long int.
4697 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4699 26. Added the never_mail option to autoreply.
4705 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4706 crash if the getsockname() call failed; this can happen if a connection is
4707 closed very soon after it is established. The problem was simply in the
4708 order in which certain operations were done, causing Exim to try to write
4709 to the SMTP stream before it had set up the file descriptor. The bug has
4710 been fixed by making things happen in the correct order.
4716 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4717 before the connection was closed, thus losing the rejection response.
4719 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4720 some early Solaris releases, but causes trouble in current releases where
4721 socklen_t is defined.
4723 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4726 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4729 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4730 that releasing the top of it at the end releases what was used for sub-
4731 expansions (unless the block got too big). However, discard this block if
4732 the first thing is a variable or header, so that we can use its block when
4733 it is dynamic (useful for very large $message_headers, for example).
4735 6. Lookups now cache *every* query, not just the most recent. A new, separate
4736 store pool is used for this. It can be recovered when all lookup caches are
4737 flushed. Lookups now release memory at the end of their result strings.
4738 This has involved some general refactoring of the lookup sources.
4740 7. Some code has been added to the store_xxx() functions to reduce the amount
4741 of flapping under certain conditions.
4743 8. log_incoming_interface used to affect only the <= reception log lines. Now
4744 it causes the local interface and port to be added to several more SMTP log
4745 lines, for example "SMTP connection from", and rejection lines.
4747 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4749 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4751 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4752 could be overwritten at the end of the current message (or the start of a
4753 new message if it was set in a HELO ACL). The value is now preserved for
4754 the duration of the SMTP connection.
4756 12. If a transport had a headers_rewrite setting, and a matching header line
4757 contained an unqualified address, that address was qualified, even if it
4758 did not match any rewriting rules. The underlying bug was that the values
4759 of the flags that permit the existence of unqualified sender and recipient
4760 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4761 non-local messages, and by -bnq for local messages) were not being
4762 preserved with the message after it was received.
4764 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4765 "next input=" as part of the text comprising the host identity instead of
4766 the correct text. The code was using the same buffer for two different
4767 strings. However, depending on which order the printing function evaluated
4768 its arguments, the bug did not always show up. Under Linux, for example, my
4769 test suite worked just fine.
4771 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4772 doesn't work with some older versions of Perl. It has been changed to "my",
4773 which in any case is probably the better facility to use.
4775 15. A really picky compiler found some instances of statements for creating
4776 error messages that either had too many or two few arguments for the format
4779 16. The size of the buffer for calls to the DNS resolver has been increased
4780 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4781 for addresses that have a lot of PTR records. This alleviates a problem; it
4782 does not fully solve it.
4784 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4785 buffer now truncates the list and logs the incident, which is the same
4786 action as happens when Exim is looking up a host name and its aliases.
4787 Previously in this situation something unpredictable would happen;
4788 sometimes it was "internal error: store_reset failed".
4790 18. If a server dropped the connection unexpectedly when an Exim client was
4791 using GnuTLS and trying to read a response, the client delivery process
4792 crashed while trying to generate an error log message.
4794 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4795 string, for example:
4797 warn message = H1: something\nH2: something
4799 the text was added as a single header line from Exim's point of view
4800 though it ended up OK in the delivered message. However, searching for the
4801 second and subsequent header lines using $h_h2: did not work. This has been
4802 fixed. Similarly, if a system filter added multiple headers in this way,
4803 the routers could not see them.
4805 20. Expanded the error message when iplsearch is called with an invalid key to
4806 suggest using net-iplsearch in a host list.
4808 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4809 ACLs are no longer actually imposed (and a message to that effect is
4812 22. If a "gecos" field in a passwd entry contained escaped characters, in
4813 particular, if it contained a \" sequence, Exim got it wrong when building
4814 a From: or a Sender: header from that name. A second bug also caused
4815 incorrect handling when an unquoted " was present following a character
4816 that needed quoting.
4818 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4819 was not being matched caselessly.
4821 24. Arranged for all hyphens in the exim.8 source to be escaped with
4824 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4825 properly. Recipient callout cache records were still being keyed to include
4826 the sender, even when use_sender was set false. This led to far more
4827 callouts that were necessary. The sender is no longer included in the key
4828 when use_sender is false.
4830 26. Added "control = submission" modifier to ACLs.
4832 27. Added the ${base62d: operator to decode base 62 numbers.
4834 28. dnsdb lookups can now access SRV records.
4836 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4837 the configuration file.
4839 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4840 (-v) mode. This makes the output for a verbose queue run more intelligible.
4842 31. Added a use_postmaster feature to recipient callouts.
4844 32. Added the $body_zerocount variable, containing the number of binary zero
4845 bytes in the message body.
4847 33. The time of last modification of the "new" subdirectory is now used as the
4848 "mailbox time last read" when there is a quota error for a maildir
4851 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4853 35. Added +ignore_unknown as a special item in host lists.
4855 36. Code for decoding IPv6 addresses in host lists is now included, even if
4856 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4857 address was recognized as an IP address, but was then not correctly decoded
4858 into binary, causing unexpected and incorrect effects when compared with
4865 1. Very minor rewording of debugging text in manualroute to say "list of
4866 hosts" instead of "hostlist".
4868 2. If verify=header_syntax was set, and a header line with an unqualified
4869 address (no domain) and a large number of spaces between the end of the
4870 name and the colon was received, the reception process suffered a buffer
4871 overflow, and (when I tested it) crashed. This was caused by some obsolete
4872 code that should have been removed. The fix is to remove it!
4874 3. When running in the test harness, delay a bit after writing a bounce
4875 message to get a bit more predictability in the log output.
4877 4. Added a call to search_tidyup() just before forking a reception process. In
4878 theory, someone could use a lookup in the expansion of smtp_accept_max_
4879 per_host which, without the tidyup, could leave open a database connection.
4881 5. Added the variables $recipient_data and $sender_data which get set from a
4882 lookup success in an ACL "recipients" or "senders" condition, or a router
4883 "senders" option, similar to $domain_data and $local_part_data.
4885 6. Moved the writing of debug_print from before to after the "senders" test
4888 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4889 problems for message scanning, either using a data ACL, or using
4890 local_scan() because the Received: header was not generated till after they
4891 were called (in order to set the time as the time of reception completion).
4892 I have revised the way this works. The header is now generated after the
4893 body is received, but before the ACL or local_scan() are called. After they
4894 are run, the timestamp in the header is updated.
4900 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4901 before starting a queue runner without re-exec. This happened only when
4902 deliver_drop_privilege was set or when the Exim user was set to root. The
4903 effect of the bug was that timeouts during subsequent deliveries caused
4904 crashes instead of being properly handled. The handler is now left at its
4905 default (and expected) setting.
4907 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4908 message, again when deliver_drop_privilege is set or Exim is run as root.
4909 The bug described in (1) was not present in this case, but the tidying up
4910 of the other signals was missing. I have made the two cases consistent.
4912 3. The ignore_target_hosts setting on a manualroute router was being ignored
4913 for hosts that were looked up using the /MX notation.
4915 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4918 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4919 operated on the sender address. After changing the $sender_address to <>
4920 for the sender address verify, Exim was re-instated it as the original
4921 (before rewriting) address, but remembering that it had rewritten it, so it
4922 wasn't rewriting it again. This bug also had the effect of breaking the
4923 sender address verification caching when the sender address was rewritten.
4925 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4926 This has been changed so that if the ip literal address matches
4927 ignore_target_hosts, the router declines.
4929 7. Added expansion conditions match_domain, match_address, and match_local_
4930 part (NOT match_host).
4932 8. The placeholder for the Received: header didn't have a length field set.
4934 9. Added code to Exim itself and to exim_lock to test for a specific race
4935 condition that could lead to file corruption when using MBX delivery. The
4936 issue is with the lockfile that is created in /tmp. If this file is removed
4937 after a process has opened it but before that process has acquired a lock,
4938 there is the potential for a second process to recreate the file and also
4939 acquire a lock. This could lead to two Exim processes writing to the file
4940 at the same time. The added code performs the same test as UW imapd; it
4941 checks after acquiring the lock that its file descriptor still refers to
4942 the same named file.
4944 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4945 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4948 11. Added the smtp_active_hostname option. If used, this will typically be made
4949 to depend on the incoming interface address. Because $interface_address is
4950 not set up until the daemon has forked a reception process, error responses
4951 that can happen earlier (such as "too many connections") no longer contain
4954 12. If an expansion in a condition on a "warn" statement fails because a lookup
4955 defers, the "warn" statement is abandoned, and the next ACL statement is
4956 processed. Previously this caused the whole ACL to be aborted.
4958 13. Added the iplsearch lookup type.
4960 14. Added ident_timeout as a log selector.
4962 15. Added tls_certificate_verified as a log selector.
4964 16. Added a global option tls_require_ciphers (compare the smtp transport
4965 option of the same name). This controls incoming TLS connections.
4967 17. I finally figured out how to make tls_require_ciphers do a similar thing
4968 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4969 before starting the TLS session.
4971 18. Tabs are now shown as \t in -bP output.
4973 19. If the log selector return_path_on_delivery was set, Exim crashed when
4974 bouncing a message because it had too many Received: header lines.
4976 20. If two routers both had headers_remove settings, and the first one included
4977 a superfluous trailing colon, the final name in the first list and the
4978 first name in the second list were incorrectly joined into one item (with a
4979 colon in the middle).
4985 1. Added -C and -D options to the exinext utility, mainly to make it easier
4986 to include in the automated testing, but these could be helpful when
4987 multiple configurations are in use.
4989 2. The exinext utility was not formatting the output nicely when there was
4990 an alternate port involved in the retry record key, nor when there was a
4991 message id as well (for retries that were specific to a specific message
4992 and a specific host). It was also confused by IPv6 addresses, because of
4993 the additional colons they contain. I have fixed the IPv4 problem, and
4994 patched it up to do a reasonable job for IPv6.
4996 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4997 delivery, the log line now contains "pipelined" if PIPELINING was used.
4999 4. An SMTP transport process used to panic and die if the bind() call to set
5000 an explicit outgoing interface failed. This has been changed; it is now
5001 treated in the same way as a connect() failure.
5003 5. A reference to $sender_host_name in the part of a conditional expansion
5004 that was being skipped was still causing a DNS lookup. This no longer
5007 6. The def: expansion condition was not recognizing references to header lines
5008 that used bh_ and bheader_.
5010 7. Added the _cache feature to named lists.
5012 8. The code for checking quota_filecount in the appendfile transport was
5013 allowing one more file than it should have been.
5015 9. For compatibility with Sendmail, the command line option
5023 and sets the incoming protocol and host name (for trusted callers). The
5024 host name and its colon can be omitted when only the protocol is to be set.
5025 Note the Exim already has two private options, -pd and -ps, that refer to
5026 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5027 "s", but I don't think that's a major issue.
5029 10. A number of refactoring changes to the code, none of which should affect
5032 (a) The number of logging options was getting close to filling up the
5033 32-bit word that was used as a bit map. I have split them into two classes:
5034 those that are passed in the argument to log_write(), and those that are
5035 only ever tested independently outside of that function. These are now in
5036 separate 32-bit words, so there is plenty of room for expansion again.
5037 There is no change in the user interface or the logging behaviour.
5039 (b) When building, for example, log lines, the code previously used a
5040 macro that called string_cat() twice, in order to add two strings. This is
5041 not really sufficiently general. Furthermore, there was one instance where
5042 it was actually wrong because one of the argument was used twice, and in
5043 one call a function was used. (As it happened, calling the function twice
5044 did not affect the overall behaviour.) The macro has been replaced by a
5045 function that can join an arbitrary number of extra strings onto a growing
5048 (c) The code for expansion conditions now uses a table and a binary chop
5049 instead of a serial search (which was left over from when there were very
5050 few conditions). Also, it now recognizes conditions like "pam" even when
5051 the relevant support is not compiled in: a suitably worded error message is
5052 given if an attempt is made to use such a condition.
5054 11. Added ${time_interval:xxxxx}.
5056 12. A bug was causing one of the ddress fields not to be passed back correctly
5057 from remote delivery subprocesses. The field in question was not being
5058 subsequently used, so this caused to problems in practice.
5060 13. Added new log selectors queue_time and deliver_time.
5062 14. Might have fixed a bug in maildirsizefile handling that threw up
5063 "unexpected character" debug warnings, and recalculated the data
5064 unnecessarily. In any case, I expanded the warning message to give more
5067 15. Added the message "Restricted characters in address" to the statements in
5068 the default ACL that block characters like @ and % in local parts.
5070 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5071 Three changes have been made:
5073 (a) There was a serious bug; a negative response to MAIL caused the whole
5074 recipient domain to be cached as invalid, thereby blocking all messages
5075 to all local parts at the same domain, from all senders. This bug has
5076 been fixed. The domain is no longer cached after a negative response to
5077 MAIL if the sender used is not empty.
5079 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5082 (c) A new callout option, "use_sender" has been added for people who want
5083 the modified behaviour.
5089 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5092 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5093 indeed breaks things for older releases.
5095 3. Added additional logging to the case where there is a problem reading data
5096 from a filter that is running in a subprocess using a pipe, in order to
5097 try to track down a specific problem.
5099 4. Testing facility fudge: when running in the test harness and attempting
5100 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5101 getting "No route to host". Convert this to a timeout.
5103 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5106 6. Some OS don't have socklen_t but use size_t instead. This affects the
5107 fifth argument of getsockopt() amongst other things. This is now
5108 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5109 can be set for individual OS. I have set it for SunOS5, OSF1, and
5110 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5111 some earlier ones do not.
5113 7. Change 4.30/15 was not doing the test caselessly.
5115 8. The standard form for an IPv6 address literal was being rejected by address
5116 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5117 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5118 this, as well as the form without the "IPv6" on the front (but only when
5119 address literals are enabled, of course).
5121 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5123 10. Exim crashed if a message with an empty sender address specified by -f
5124 encountered a router with an errors_to setting. This could be provoked only
5125 by a command such as
5129 where an empty string was supplied; "<>" did not hit this bug.
5131 11. Installed PCRE release 4.5.
5133 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5134 remained set. It is now erased.
5136 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5137 times from message ids (which are base 36 rather than the normal 62).
5139 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5140 were being counted as actual protocol errors, and logged if the log
5141 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5142 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5143 and DATA following a set of rejected RCPTs do not count as protocol errors.
5144 In other words, Exim assumes they were pipelined, though this may not
5145 actually be the case. Of course, in all cases the client gets an
5146 appropriate error code.
5148 15. If a lookup fails in an ACL condition, a message about the failure may
5149 be available; it is used if testing the ACL cannot continue, because most
5150 such messages specify what the cause of the deferral is. However, some
5151 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5152 that caused an old message to be retained and used if a later statement
5153 caused a defer, replacing the real cause of the deferral.
5155 16. If an IP address had so many PTR records that the DNS lookup buffer
5156 was not large enough to hold them, Exim could crash while trying to process
5157 the truncated data. It now detects and logs this case.
5159 17. Further to 4.21/58, another change has been made: if (and only if) the
5160 first line of a message (the first header line) ends with CRLF, a bare LF
5161 in a subsequent header line has a space inserted after it, so as not to
5162 terminate the header.
5164 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5165 unnecessarily, used atoi() instead of strtol(), and didn't check the
5166 termination when getting file sizes from file names by regex.
5168 19. Completely re-implemented the support for maildirsize files, in the light
5169 of a number of problems with the previous contributed implementation
5170 (4.30/29). In particular:
5172 . If the quota is zero, the maildirsize file is maintained, but no quota is
5175 . If the maildir directory does not exist, it is created before any attempt
5176 to write a maildirsize file.
5178 . The quota value in the file is just a cache; if the quota is changed in
5179 the transport, the new value overrides.
5181 . A regular expression is available for excluding directories from the
5184 20. The autoreply transport checks the characters in options that define the
5185 message's headers; it allows continued headers, but it was checking with
5186 isspace() after an embedded newline instead of explicitly looking for a
5189 21. If all the "regular" hosts to which an address was routed had passed their
5190 expiry times, and had not reached their retry times, the address was
5191 bounced, even if fallback hosts were defined. Now Exim should go on to try
5194 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5195 equivalent code in the SMTP transport. Some hosts send humungous responses
5196 to HELO/EHLO, more than 1024 it seems.
5198 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5199 gives compiler warnings in some environments. I've now done it "properly",
5202 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5203 (because of problems with the built-in one) was declared to return uschar *
5204 instead of char *, causing compiler failure.
5206 25. Fixed a file descriptor leak when processing alias/forward files.
5208 26. Fixed a minor format string issue in dbfn.c.
5210 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5212 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5213 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5216 29. When a sender address is verified, it is cached, to save repeating the test
5217 when there is more than one recipient in a message. However, when the
5218 verification involves a callout, it is possible for different callout
5219 options to be set for different recipients. It is too complicated to keep
5220 track of this in the cache, so now Exim always runs a verification when a
5221 callout is required, relying on the callout cache for the optimization.
5222 The overhead is duplication of the address routing, but this should not be
5225 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5226 to be verified with callout=postmaster, and the main callout worked but the
5227 postmaster check failed, the verification correctly failed. However, if a
5228 subsequent RCPT command asked for sender verification *without* the
5229 postmaster check, incorrect caching caused this verification also to fail,
5232 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5233 it was not caching the DNS options (qualify_single, search_parents) that
5234 were used when the lookup failed. A subsequent lookup with different
5235 options therefore always gave the same answer, though there were cases
5236 where it should not have. (Example: a "domains = !$mx_any" option on a
5237 dnslookup router: the "domains" option is always processed without any
5238 widening, but the router might have qualify_single set.) Now Exim uses the
5239 cached value only when the same options are set.
5241 32. Added John Jetmore's "exipick" utility to the distribution.
5243 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5244 than a timeout (e.g. a certificate is required, and is not provided), an
5245 Exim server now closes the connection immediately. Previously it waited for
5246 the client to close - but if the client is SSL, it seems that they each
5247 wait for each other, leading to a delay before one of them times out.
5249 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5250 maintained 0.8.x compatibility because I don't think many are using it, and
5251 it is clearly obsolete.
5253 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5256 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5257 client certificate was expired. A simple patch fixes this, though I don't
5258 understand the full logic of why the verify callback is called multiple
5261 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5262 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5263 which causes problems with some clients (such as the Certicom SSL Plus
5264 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5265 disables the coutermeasure allowing Eudora to connect."
5267 38. Exim was not checking that a write() to a log file succeeded. This could
5268 lead to Bad Things if a log got too big, in particular if it hit a file
5269 size limit. Exim now panics and dies if it cannot write to a log file, just
5270 as it does if it cannot open a log file.
5272 39. Modified OS/Makefile-Linux so that it now contains
5274 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5276 The two -D definitions ensure that Exim is compiled with large file
5277 support, which makes it possible to handle log files that are bigger than
5280 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5281 instance) a domain was checked against a named list that involved a lookup,
5282 causing $domain_data to be set, then another domain was checked against the
5283 same list, then the first domain was re-checked, the value of $domain_data
5284 after the final check could be wrong. In particular, if the second check
5285 failed, it could be set empty. This bug probably also applied to
5288 41. The strip_trailing_dot option was not being applied to the address given
5289 with the -f command-line option.
5291 42. The code for reading a message's header from the spool was incrementing
5292 $received_count, but never initializing it. This meant that the value was
5293 incorrect (doubled) while delivering a message in the same process in which
5294 it was received. In the most common configuration of Exim, this never
5295 happens - a fresh exec is done - but it can happen when
5296 deliver_drop_privilege is set.
5298 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5299 it now includes up to 150 characters of the unexpected data in the log
5302 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5303 and building data strings. The size of both of these buffers was 10 000
5304 bytes - far larger than anybody would *ever* want, thought I. Needless to
5305 say, somebody hit the limit. I have increased the maximum line length to
5306 20 000 and the maximum data length of concatenated lines to 100 000. I have
5307 also fixed two bugs, because there was no checking on these buffers. Tsk,
5308 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5309 buffer is too small.
5311 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5312 lsearch lookups. Now it does.
5314 46. When parsing a route_list item in a manualroute router, a fixed-length
5315 buffer was used for the list of hosts. I made this 1024 bytes long,
5316 thinking that nobody would ever have a list of hosts that long. Wrong.
5317 Somebody had a whole pile of complicated expansion conditions, and the
5318 string was silently truncated, leading to an expansion error. It turns out
5319 that it is easier to change to an unlimited length (owing to other changes
5320 that have happened since this code was originally written) than to build
5321 structure for giving a limitation error. The length of the item that
5322 expands into the list of hosts is now unlimited.
5324 47. The lsearch lookup could not handle data where the length of text line was
5325 more than 4095 characters. Such lines were truncated, leading to shortened
5326 data being returned. It should now handle lines of any length.
5328 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5329 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5332 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5333 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5334 no longer happens in comment lines. A list of replacements is now placed
5335 at the head of all of the source files, except those whose only change is
5336 to replace PERL_COMMAND in the very first #! line.
5338 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5339 messages on the queue, with a bottom-up merge sort, using code contributed
5340 by Michael Haardt. This should make operations like -bp somewhat faster on
5341 large queues. It won't affect queue runners, except when queue_run_in_order
5344 51. Installed eximstats 1.31 in the distribution.
5346 52. Added support for SRV lookups to the dnslookup router.
5348 53. If an ACL referred to $message_body or $message_body_end, the value was not
5349 reset for any messages that followed in the same SMTP session.
5351 54. The store-handling optimization for building very long strings was not
5352 differentiating between the different store pools. I don't think this
5353 actually made any difference in practice, but I've tidied it.
5355 55. While running the routers to verify a sender address, $sender_address
5356 was still set to the sender address. This is wrong, because when routing to
5357 send a bounce to the sender, it would be empty. Therefore, I have changed
5358 it so that, while verifying a sender address, $sender_address is set to <>.
5359 (There is no change to what happens when verifying a recipient address.)
5361 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5362 target A or AAAA records (if not already returned) without resetting the
5363 qualify_single or search_parents options of the DNS resolver. These are
5364 inappropriate in this case because the targets of MX and SRV records must
5365 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5366 target that, when qualified, matched something in the local domain. These
5367 two options are now turned off when doing these lookups.
5369 57. It seems that at least some releases of Reiserfs (which does not have the
5370 concept of a fixed number of inodes) returns zero and not -1 for the
5371 number of available inodes. This interacted badly with check_spool_inodes,
5372 which assumed that -1 was the "no such thing" setting. What I have done is
5373 to check that the total number of inodes is greater than zero before doing
5374 the test of how many are available.
5376 58. When a "warn" ACL statement has a log_message modifier, the message is
5377 remembered, and not repeated. This is to avoid a lot of repetition when a
5378 message has many recipients that cause the same warning to be written.
5379 Howewer, Exim was preserving the list of already written lines for an
5380 entire SMTP session, which doesn't seem right. The memory is now reset if a
5381 new message is started.
5383 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5384 debugging output unless log_rewrite was also set.
5386 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5387 of (int)(handle) when we know that handle contains (void *)(-1).
5389 61. The Exim daemon panic-logs an error return when it closes the incoming
5390 connection. However "connection reset by peer" seems to be common, and
5391 isn't really an error worthy of noting specially, so that particular error
5394 62. When Exim is trying to find all the local interfaces, it used to panic and
5395 die if the ioctl to get the interface flags failed. However, it seems that
5396 on at least one OS (Solaris 9) it is possible to have an interface that is
5397 included in the list of interfaces, but for which you get a failure error
5398 for this call. This happens when the interface is not "plumbed" into a
5399 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5400 failure of the "get flags" call assumes that the interface is down.
5402 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5403 makes life easier for people who are doing arithmetic on fields extracted
5404 from dates, where you often get leading zeros that should not be
5405 interpreted as octal.
5407 64. Added qualify_domain to the redirect router, to override the global
5410 65. If a pathologically long header line contained very many addresses (the
5411 report of this problem mentioned 10 000) and each of them was rewritten,
5412 Exim could use up a very large amount of memory. (It kept on making new
5413 copies of the header line as it rewrote, and never released the old ones.)
5414 At the expense of a bit more processing, the header rewriting function has
5415 been changed so that it no longer eats memory in this way.
5417 66. The generation of the Received: header has been moved from the time that a
5418 message starts to be received, to the time that it finishes. The timestamp
5419 in the Received: header should now be very close to that of the <= log
5420 line. There are two side-effects of this change:
5422 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5423 logged header lines no longer include the local Received: line, because
5424 it has not yet been created. The same applies to a copy of the message
5425 that is returned to a non-SMTP sender when a message is rejected.
5427 (b) When a filter file is tested using -bf, no additional Received: header
5428 is added to the test message. After some thought, I decided that this
5431 This change does not affect the value of $received_for. It is still set
5432 after address rewriting, but before local_scan() is called.
5434 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5436 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5437 gave an unhelpful panic error message, and a defer error. I have managed to
5438 change this behaviour so that it now rejects any supplied certificate,
5439 which seems right, as the list of acceptable certificates is empty.
5441 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5442 gave an unhelpful defer error. I have not managed to make this reject any
5443 supplied certificates, but the error message it gives is "no certificate
5444 supplied", which is not helpful.
5446 70. exigrep's output now also includes lines that are not associated with any
5447 message, but which match the given pattern. Implemented by a patch from
5448 Martin Sluka, which also tidied up the Perl a bit.
5450 71. Recipient callout verification, like sender verification, was using <> in
5451 the MAIL FROM command. This isn't really the right thing, since the actual
5452 sender may affect whether the remote host accepts the recipient or not. I
5453 have changed it to use the actual sender in the callout; this means that
5454 the cache record is now keyed on a recipient/sender pair, not just the
5455 recipient address. There doesn't seem to be a real danger of callout loops,
5456 since a callout by the remote host to check the sender would use <>.
5457 [SEE ABOVE: changed after hitting problems.]
5459 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5460 temporary errors. However, in the case of such a code being given after
5461 the end of a data transmission (i.e. after ".") Exim was failing to write
5462 a retry record for the message. (Yes, there was some broken host that was
5463 actually sending 8xx at this point.)
5465 73. An unknown lookup type in a host list could cause Exim to panic-die when
5466 the list was checked. (An example that provoked this was putting <; in the
5467 middle of a list instead of at the start.) If this happened during a DATA
5468 ACL check, a -D file could be left lying around. This kind of configuration
5469 error no longer causes Exim to die; instead it causes a defer errror. The
5470 incident is still logged to the main and panic logs.
5472 74. Buglet left over from Exim 3 conversion. The message "too many messages
5473 in one connection" was written to the rejectlog but not the mainlog, except
5474 when address rewriting (yes!) was being logged.
5476 75. Added write_rejectlog option.
5478 76. When a system filter was run not as root (that is, when system_filter_user
5479 was set), the values of the $n variables were not being returned to the
5480 main process; thus, they were not subsequently available in the $sn
5483 77. Added +return_path_on_delivery log selector.
5485 78. A connection timeout was being treated differently from recipients deferred
5486 when testing hosts_max_try with a message that was older than the host's
5487 retry timeout. (The host should not be counted, thus allowing all hosts to
5488 be tried at least once before bouncing.) This may have been the cause of an
5489 occasionally reported bug whereby a message would remain on the queue
5490 longer than the retry timeout, but would be bounced if a delivery was
5491 forced. I say "may" because I never totally pinned down the problem;
5492 setting up timeout/retry tests is difficult. See also the next item.
5494 79. The ultimate address timeout was not being applied to errors that involved
5495 a combination of host plus message (for example, a timeout on a MAIL
5496 command). When an address resolved to a number of possible hosts, and they
5497 were not all tried for each delivery (e.g. because of hosts_max_try), a
5498 message could remain on the queue longer than the retry timeout.
5500 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5503 81. Fixed an obscure SMTP outgoing bug which required at least the following
5504 conditions: (a) there was another message waiting for the same server;
5505 (b) the server returned 5xx to all RCPT commands in the first message so
5506 that the message was not completed; (c) the server dropped the connection
5507 or gave a negative response to the RSET that Exim sends to abort the
5508 transaction. The observed case was a dropped connection after DATA that had
5509 been sent in pipelining mode. That is, the server had advertised PIPELINING
5510 but was not implementing it correctly. The effect of the bug was incorrect
5511 behaviour, such as trying another host, and this could lead to a crash.
5517 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5518 and daemon.c were passed as pointers to ints; they should have been
5519 pointers to socklen_t variables (which are typically unsigned ints).
5521 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5524 3. Fixed a really odd bug that affected only the testing scheme; patching a
5525 certain fixed string in the binary changed the value of another string that
5526 happened to be identical to the end of the original first string.
5528 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5529 name", it returns that address as the IP address. On some operating
5530 systems (e.g. Solaris), it also passes back the IP address string as the
5531 "host name". However, on others (e.g. Linux), it passes back an empty
5532 string. Exim wasn't checking for this, and was changing the host name to an
5533 empty string, assuming it had been canonicized.
5535 5. Although rare, it is permitted to have more than one PTR record for a given
5536 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5537 all the names associated with an address, because they do in Solaris.
5538 However, it seems that they do not in Linux for data that comes from the
5539 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5540 I found this out when I moved to a new Linux workstation and tried to run
5541 the Exim test suite.
5543 To get round this problem I have changed the code so that it now does its
5544 own call to the DNS to look up PTR records when searching for a host name.
5545 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5546 addresses that are only in /etc/hosts are still found.
5548 This behaviour is, however, controlled by an option called host_lookup_
5549 order, which defaults to "bydns:byaddr". If people want to use the other
5550 order, or indeed, just use one or the other means of lookup, they can
5551 specify it in this variable.
5553 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5554 some operating systems, this comes back from gethostbyaddr() as an empty
5555 string, and this is what Exim used to test for. However, it seems that in
5556 other systems, "." is yielded. Exim now tests for this case too.
5558 7. The values of check_spool_space and check_log_space are now held internally
5559 as a number of kilobytes instead of an absolute number of bytes. If a
5560 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5561 kilobyte. This means that much larger values can be stored.
5563 8. Exim monitor: an attempt to get the action menu when not actually pointing
5564 at a message produces an empty menu entitled "No message selected". This
5565 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5566 no entries in it ("Shell widget menu has zero width and/or height"). So I
5567 have added a single, blank menu entry in this case.
5569 9. Added ${quote_local_part.
5571 10. MIME decoding is now applied to the contents of Subject: header lines when
5574 11. Now that a reference to $sender_host_address automatically causes a reverse
5575 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5576 host lookup before query-style lookups in lists that might use this
5577 variable. This has therefore been abolished, and the "net-" prefix is no
5578 longer necessary for query-style lookups.
5580 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5581 have been a typo for LFLAGS, so it has been changed.
5583 13. The install script calls Exim with "-C /dev/null" in order to find the
5584 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5585 to be output. Howeve, since Exim outputs its version number before the
5586 error, it didn't break the script. It just looked ugly. I fixed this by
5587 always allowing "-C /dev/null" if the caller is root.
5589 14. Ignore overlarge ACL variable number when reading spool file - insurance
5590 against a later release with more variables having written the file.
5592 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5593 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5594 this, as well as the form without the "IPv6" on the front.
5596 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5597 OS/Makefile-Darwin file.
5599 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5600 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5602 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5603 variable. However, some OS use ints for inodes. I've added cast to long int
5604 to get rid of the compiler warning.
5606 19. I had forgotten to lock out "/../" in configuration file names when
5607 ALT_CONFIG_PREFIX was set.
5609 20. Routers used for verification do not need to specify transports. However,
5610 if such a router generated a host list, and callout was configured, Exim
5611 crashed, because it could not find a port number from the (non-existent)
5612 transport. It now assumes port 25 in this circumstance.
5614 21. Added the -t option to exigrep.
5616 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5617 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5618 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5619 them is set, LOOKUP_LSEARCH is forced.
5621 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5623 24. Added sender and host information to the "rejected by local_scan()" log
5624 line; previously there was no indication of these.
5626 25. Added .include_if_exists.
5628 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5629 Linux. It turns out that not all file systems support this. Apparently some
5630 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5631 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5632 not supported on the file descriptor, is now ignored when Exim is trying to
5633 sync a directory. This applies only to Linux.
5635 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5637 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5638 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5641 29. Added support for maildirsize files from supplied patch (modified a bit).
5643 30. The use of :fail: followed by an empty string could lead Exim to respond to
5644 sender verification failures with (e.g.):
5646 550 Verification failed for <xxx>
5647 550 Sender verify failed
5649 where the first response line was missing the '-' that indicates it is not
5650 the final line of the response.
5652 31. The loop for finding the name of the user that called Exim had a hardwired
5653 limit of 10; it now uses the value of finduser_retries, which is used for
5654 all other user lookups.
5656 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5659 33. Exim was neglecting to zero errno before one call of strtol() when
5660 expanding a string and expecting an integer value. On some systems this
5661 resulted in spurious "integer overflow" errors. Also, it was casting the
5662 result into an int without checking.
5664 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5665 did not work. The code looks as if it has *never* worked, though it appears
5666 to have been documented since at least releast 1.62. I have made it work.
5668 35. The "timeout_DNS" error in retry rules, also documented since at least
5669 1.62, also never worked. As it isn't clear exactly what this means, and
5670 clearly it isn't a major issue, I have abolished the feature by treating it
5671 as "timeout", and writing a warning to the main and panic logs.
5673 36. The display of retry rules for -brt wasn't always showing the error code
5676 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5677 timeout_connect_A, timeout_connect_MX.
5679 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5680 to the empty sender.
5682 39. The daemon was not analysing the content of -oX till after it had closed
5683 stderr and disconnected from the controlling terminal. This meant that any
5684 syntax errors were only noted on the panic log, and the return code from
5685 the command was 0. By re-arranging the code a little, I've made the
5686 decoding happen first, so such errors now appear on stderr, and the return
5687 code is 1. However, the actual setting up of the sockets still happens in
5688 the disconnected process, so errors there are still only recorded on the
5691 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5692 connections (as happens on some IP stacks) was logged at start up time as
5693 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5694 it from "IPv6 and IPv4", which means that two separate sockets are being
5697 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5698 says whether AF_INET or AF_INET6 was passed as an argument.
5700 42. Exiwhat output was messed up when time zones were included in log
5703 43. Exiwhat now gives more information about the daemon's listening ports,
5704 and whether -tls-on-connect was used.
5706 44. The "port" option of the smtp transport is now expanded.
5708 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5709 silently ignored. Now an error message is written to the main and panic
5712 46. There's a new ACL modifier called "logwrite" which writes to a log file
5713 as soon as it is encountered.
5715 47. Added $local_user_uid and $local_user_gid at routing time.
5717 48. Exim crashed when trying to verify a sender address that was being
5720 49. Exim was recognizing only a space character after ".include". It now also
5721 recognizes a tab character.
5723 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5724 extracting the relevant information from the specification. The man page no
5725 longer contains scrambled data for the -d option, and I've added a section
5726 at the front about calling Exim under different names.
5728 51. Added "extra_headers" argument to the "mail" command in filter files.
5730 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5733 53. Installed eximstats 1.29.
5735 54. Added transport_filter_timeout as a generic transport option.
5737 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5738 Cc: header lines. This was required by RFC 822, but it not required by RFC
5741 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5742 incoming messages that did not have them. Now it does so only if the
5743 message originates locally, that is, if there is no associated remote host
5744 address. When Resent- header lines are present, this applies to the Resent-
5745 lines rather than the non-Resent- lines.
5747 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5748 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5750 58. Messages for configuration errors now include the name of the main
5751 configuration file - useful now that there may be more than one file in a
5752 list (.included file names were always shown).
5754 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5755 for those rare installations that do not start the daemon as root or run it
5756 setuid root. I've cut out the call to initgroups() if the daemon is not
5759 60. The Exim user and group can now be bound into the binary as text strings
5760 that are looked up at the start of Exim's processing.
5762 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5764 62. Added $mailstore_basename variable.
5766 63. Installed patch to sieve.c from Michael Haardt.
5768 64. When Exim failed to open the panic log after failing to open the main log,
5769 the original message it was trying to log was written to stderr and debug
5770 output, but if they were not available (the usual case in production), it
5771 was lost. Now it is written to syslog before the two lines that record the
5772 failures to open the logs.
5774 65. Users' Exim filters run in subprocesses under the user's uid. It is
5775 possible for a "deliver" command or an alias in a "personal" command to
5776 provoke an address rewrite. If logging of address rewriting is configured,
5777 this fails because the process is not running as root or exim. There may be
5778 a better way of dealing with this, but for the moment (because 4.30 needs
5779 to be released), I have disabled address rewrite logging when running a
5780 filter in a non-root, non-exim process.
5786 1. The buildconfig auxiliary program wasn't quoting the value set for
5787 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5788 not defined. This bug was masked in 4.22 by the effect that was fixed in
5791 2. Some messages that were rejected after a message id was allocated were
5792 shown as "incomplete" by exigrep. It no longer does this for messages that
5793 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5795 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5796 have allow_domain_literals set, the ID did not get logged in the <= line.
5797 Domain literals are now always recognized in Message-ID: header lines.
5799 4. The first argument for a ${extract expansion item is the key name or field
5800 number. Leading and trailing spaces in this item were not being ignored,
5801 causing some misleading effects.
5803 5. When deliver_drop_privilege was set, single queue runner processes started
5804 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5805 same command in the process it spins off) were not dropping privilege.
5807 6. When the daemon running as "exim" started a queue runner, it always
5808 re-executed Exim in the spun-off process. This is a waste of effort when
5809 deliver_drop_privilege is set. The new process now just calls the
5810 queue-runner function directly.
5816 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5819 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5820 ignored. Though the use of -oP was forcing the writing of a pid file, it
5821 was always written to the default place.
5823 3. If the message "no IP address found for host xxxx" is generated during
5824 incoming verification, it is now followed by identification of the incoming
5825 connection (so you can more easily find what provoked it).
5827 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5829 5. Added some features to "harden" Exim a bit more against certain attacks:
5831 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5832 be put in Local/Makefile. This is like the never_users runtime option,
5833 but it cannot be overridden. The default setting is "root".
5835 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5836 prefix string with which any file named in a -C command line option
5839 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5840 is retained for -C and -D only if the caller of Exim is root. Without
5841 it, the exim user may also use -C and -D and retain privilege.
5843 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5844 command line option is disabled.
5846 6. Macro names set by the -D option must start with an upper case letter, just
5847 like macro names defined in the configuration file.
5849 7. Added "dereference=" facility to LDAP.
5851 8. Two instances of the typo "uknown" in the source files are fixed.
5853 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5854 the Configure-Makefile script screwed up while processing it.
5856 10. Incorporated PCRE 4.4.
5858 11. The SMTP synchronization check was not operating right at the start of an
5859 SMTP session. For example, it could not catch a HELO sent before the client
5860 waited for the greeting. There is now a check for outstanding input at the
5861 point when the greeting is written. Because of the duplex, asynchronous
5862 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5863 way, but not yet received, when the check is performed.
5865 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5866 on TCP/IP sockets, because this apparently causes some broken clients to
5869 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5870 unchanged) from the Cygwin maintainer.
5872 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5873 was supported instead of testing for mbx. Effectively a typo.
5875 15. The spa authenticator server code was not checking that the input it
5876 received was valid base64.
5878 16. The debug output line for the "set" modifier in ACLs was not showing the
5879 name of the variable that was being set.
5881 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5883 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5884 lookup on its own. Something else was needed to trigger the lookup. For
5885 example, a match in host_lookup or the need for a host name in a host list.
5886 Now, if $sender_host_name is referenced and the host name has not yet been
5887 looked up, a lookup is performed. If the lookup fails, the variable remains
5888 empty, and $host_lookup_failed is set to "1".
5890 19. Added "eqi" as a case-independent comparison operator.
5892 20. The saslauthd authentication condition could segfault if neither service
5893 nor realm was specified.
5895 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5896 error message that was logged was misleading, and incoming SMTP
5897 connections were dropped. The message is now more accurate, and temporary
5898 errors are given to SMTP connections.
5900 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5901 (except RSET) with a 421 error, until QUIT is received. However, it was
5902 failing to send a response to QUIT.
5904 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5905 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5906 ACL is now run after the helo_try_verify_hosts code.
5908 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5909 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5910 case-independent, so other case variants are also recognized). Apparently
5911 some systems use these upper case variants.
5913 25. If more than two messages were waiting for the same host, and a transport
5914 filter was specified for the transport, Exim sent two messages over the
5915 same TCP/IP connection, and then failed with "socket operation on non-
5916 socket" when it tried to send the third.
5918 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5920 27. The extern definition of crypt16() in expand.c was not being excluded when
5921 the OS had its own crypt16() function.
5923 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5924 as a preferred synonym for return_size_limit, both as an option and as an
5927 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5929 30. Changed the default configuration ACL to relax the local part checking rule
5930 for addresses that are not in any local domains. For these addresses,
5931 slashes and pipe symbols are allowed within local parts, but the sequence
5932 /../ is explicitly forbidden.
5934 31. SPA server authentication was not clearing the challenge buffer before
5937 32. log_message in a "warn" ACL statement was writing to the reject log as
5938 well as to the main log, which contradicts the documentation and doesn't
5939 seem right (because no rejection is happening). So I have stopped it.
5941 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5942 However, I am unable to do any testing of this.
5944 34. Fixed an infelicity in the appendfile transport. When checking directories
5945 for a mailbox, to see if any needed to be created, it was accidentally
5946 using path names with one or more superfluous leading slashes; tracing
5947 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5949 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5950 contents are added to the log line that is written for every discarded
5951 recipient. (Previously a log_message setting was ignored.)
5953 36. The ${quote: operator now quotes the string if it is empty.
5955 37. The install script runs exim in order to find its version number. If for
5956 some reason other than non-existence or emptiness, which it checks, it
5957 could not run './exim', it was installing it with an empty version number,
5958 i.e. as "exim-". This error state is now caught, and the installation is
5961 38. An argument was missing from the function that creates an error message
5962 when Exim fails to connect to the socket for saslauthd authentication.
5963 This could cause Exim to crash, or give a corrupted message.
5965 39. Added isip, isip4, and isip6 to ${if conditions.
5967 40. The ACL variables $acl_xx are now saved with the message, and can be
5968 accessed later in routers, transports, and filters.
5970 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5971 strings in the file are not string-expanded.
5973 42. If a MAIL command specified a SIZE value that was too large to fit into an
5974 int variable, the check against message_size_limit failed. Such values are
5975 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5976 day this will have to be increased, but I don't think I want to be around
5977 when emails are that large.
5984 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5985 iconv() is not standard in FreeBSD.
5987 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5988 IPv6 enabled. The observed symptom was a segmentation fault on return from
5989 the function os_common_find_running_interfaces() in src/os.c.
5991 3. In the check_special_case() function in daemon.c I had used "errno" as an
5992 argument name, which causes warnings on some systems. This was basically a
5993 typo, since it was named "eno" in the comments!
5995 4. The code that waits for the clock to tick (at a resolution of some fraction
5996 of a second) so as to ensure message-id uniqueness was always waiting for
5997 at least one whole tick, when it could have waited for less. [This is
5998 almost certainly not relevant at current processor speeds, where it is
5999 unlikely to ever wait at all. But we try to future-proof.]
6001 5. The function that sleeps for a time interval that includes fractions of a
6002 second contained a race. It did not block SIGALRM between setting the
6003 timer, and suspending (a couple of lines later). If the interval was short
6004 and the sigsuspend() was delayed until after it had expired, the suspension
6005 never ended. On busy systems this could lead to processes getting stuck for
6008 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6009 process, before it forks any delivery processes. The open lookup caching
6010 mechanism meant that the open file or database connection was passed into
6011 the delivery process. The problem was that delivery processes always tidy
6012 up cached lookup data. This could cause a problem for the next delivery
6013 process started by the queue runner, because the external queue runner
6014 process does not know about the closure. So the next delivery process
6015 still has data in the lookup cache. In the case of a file lookup, there was
6016 no problem because closing a file descriptor in a subprocess doesn't affect
6017 the parent. However, if the lookup was caching a connection to a database,
6018 the connection was closed, and the second delivery process was likely to
6019 see errors such as "PGSQL: query failed: server closed the connection
6020 unexpectedly". The problem has been fixed by closing all cached lookups
6021 in a queue runner before running a delivery process.
6023 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6024 seem to have the "const" qualifier which it has on other OS. I've
6027 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6028 *of the same type* (client or server) with the same public name that an
6029 error should be diagnosed.
6031 9. When Exim looked up a host name for an IP address, but failed to find the
6032 original IP address when looking up the host name (a safety check), it
6033 output the message "<ip address> does not match any IP for NULL", which was
6034 confusing, to say the least. The bug was that the host name should have
6035 appeared instead of "NULL".
6037 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6038 user that is built into the binary, and the -C or -D options is used, root
6039 privilege is dropped before the configuration file is read. In addition,
6040 logging is switched to stderr instead of the normal log files. If the
6041 configuration then re-defines the Exim user, the unprivileged environment
6042 is probably not what is expected, so Exim logs a panic warning message (but
6045 However, if deliver_drop_privilege is set, the unprivileged state may well
6046 be exactly what is intended, so the warning has been cut out in that case,
6047 and Exim is allowed to try to write to its normal log files.
6053 1. smtp_return_error_details was not giving details for temporary sender
6054 or receiver verification errors.
6056 2. Diagnose a configuration error if two authenticators have the same public
6059 3. Exim used not to create the message log file for a message until the first
6060 delivery attempt. This could be confusing when incoming messages were held
6061 for policy or load reasons. The message log file is now created at the time
6062 the message is received, and an initial "Received" line is written to it.
6064 4. The automatically generated man page for command line options had a minor
6065 bug that caused no ill effects; however, a more serious problem was that
6066 the procedure for building the man page automatically didn't always
6067 operate. Consequently, release 4.20 contains an out-of-date version. This
6068 shouldn't happen again.
6070 5. When building Exim with embedded Perl support, the script that builds the
6071 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6072 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6074 6. The freeze_tell option was not being used for messages that were frozen on
6075 arrival, either by an ACL or by local_scan().
6077 7. Added the smtp_incomplete_transaction log selector.
6079 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6080 was accepting AUTH without a new EHLO.
6082 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6083 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6086 10. Logging of TCP/IP connections (when configured) now happens in the main
6087 daemon process instead of the child process, so that the TCP/IP connection
6088 count is more accurate (but it can never be perfect).
6090 11. The use of "drop" in a nested ACL was not being handled correctly in the
6091 outer ACL. Now, if condition failure induced by the nested "drop" causes
6092 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6093 or "require"), the connection is dropped.
6095 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6096 that yield "discard" can now be used with an "accept" or a "discard" verb,
6097 but an error is generated for any others (because I can't see a useful way
6098 to define what should happen).
6100 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6101 are now processed in the same way as lines in the Exim configuration file.
6102 In particular, continuation lines are supported.
6104 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6106 15. Added -ti meaning -t -i.
6108 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6109 domains, and warn by logging if others are found.
6111 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6112 returned from GIFCONF when Exim is trying to find the list of interfaces on
6113 a host. The code in os.c has been modified to copy each ifreq to an aligned
6114 structure in all cases.
6116 Also, in some cases, the returned ifreq's were being copied to a 'struct
6117 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6118 means the last couple of bytes of an IPv6 address could be chopped if the
6119 ifreq contained only a normal sockaddr (14 bytes storage).
6121 18. Named domain lists were not supported in the hosts_treat_as_local option.
6122 An entry such as +xxxx was not recognized, and was treated as a literal
6125 19. Ensure that header lines added by a DATA ACL are included in the reject log
6126 if the ACL subsequently rejects the message.
6128 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6129 MD5 (which is deprecated).
6131 21. When testing a filter file using -bf, Exim was writing a message when it
6132 took the sender from a "From " line in the message, but it was not doing so
6133 when it took $return_path from a Return-Path: header line. It now does.
6135 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6136 with a valid header line field name (a series of printing characters
6137 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6139 23. Changed "disc" in the source to "disk" to conform to the documentation and
6140 the book and for uniformity.
6142 24. Ignore Sendmail's -Ooption=value command line item.
6144 25. When execve() failed while trying to run a command in a pipe transport,
6145 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6146 could be confused with a return value of 69 from the command itself. This
6147 has been changed to 127, the value the shell returns if it is asked to run
6148 a non-existent command. The wording for the related log line suggests a
6149 non-existent command as the problem.
6151 26. If received_header_text expands to an empty string, do not add a Received:
6152 header line to the message. (Well, it adds a token one on the spool, but
6153 marks it "old" so that it doesn't get used or transmitted.)
6155 27. Installed eximstats 1.28 (addition of -nt option).
6157 28. There was no check for failure on the call to getsockname() in the daemon
6158 code. This can fail if there is a shortage of resources on the system, with
6159 ENOMEM, for example. A temporary error is now given on failure.
6161 29. Contrary to the C standard, it seems that in some environments, the
6162 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6163 program. Exim now does this explicitly; it affects the formatting of
6164 timestamps using strftime().
6166 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6167 complaints. I've now initialized all the variables, to avoid this.
6169 32. Header lines added by a system filter were not being "seen" during
6170 transport-time rewrites.
6172 33. The info_callback() function passed to OpenSSL is set up with type void
6173 (*)(SSL *, int, int), as described somewhere. However, when calling the
6174 function (actually a macro) that sets it up, the type void(*)() is
6175 expected. I've put in a cast to prevent warnings from picky compilers.
6177 34. If a DNS black list lookup found a CNAME record, but there were no A
6178 records associated with the domain it pointed at, Exim crashed.
6180 35. If a DNS black list lookup returned more than one A record, Exim ignored
6181 all but the first. It now scans all returned addresses if a particular IP
6182 value is being sought. In this situation, the contents of the
6183 $dnslist_value variable are a list of all the addresses, separated by a
6186 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6187 to accept a host name and all its aliases if the forward lookup for any of
6188 them yielded the IP address of the incoming connection. Now it accepts only
6189 those names whose forward lookup yields the correct IP address. Any other
6190 names are discarded. This closes a loophole whereby a rogue DNS
6191 administrator could create reverse DNS records to break through a
6192 wildcarded host restriction in an ACL.
6194 37. If a user filter or a system filter that ran in a subprocess used any of
6195 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6196 the wrong values were passed to the pipe command ($thisaddress had the
6197 value of $0, $0 had the value of $1, etc). This bug was introduced by
6198 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6200 38. Improved the line breaking for long SMTP error messages from ACLs.
6201 Previously, if there was no break point between 40 and 75 characters, Exim
6202 left the rest of the message alone. Two changes have been made: (a) I've
6203 reduced the minimum length to 35 characters; (b) if it can't find a break
6204 point between 35 and 75 characters, it looks ahead and uses the first one
6205 that it finds. This may give the occasional overlong line, but at least the
6206 remaining text gets split now.
6208 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6209 file descriptors might be low, and that setting 1000 would always raise it.
6210 It turns out that in some environments, the limit is already over 1000 and
6211 that lowering it causes trouble. So now Exim takes care not to decrease it.
6213 40. When delivering a message, the value of $return_path is set to $sender_
6214 address at the start of routing (routers may change the value). By an
6215 oversight, this default was not being set up when an address was tested by
6216 -bt or -bv, which affected the outcome if any router or filter referred to
6219 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6220 log only when "message" or "log_message" are set. However, if one of the
6221 conditions was an address verification, or a call to a nested ACL, the
6222 messages generated by the underlying test were being passed through. This
6223 no longer happens. The underlying message is available in $acl_verify_
6224 message for both "message" and "log_message" expansions, so it can be
6225 passed through if needed.
6227 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6228 new expansion $bh_ to give the encoded byte string without charset
6229 translation. Translation happens only if iconv() is available; HAVE_ICONV
6230 indicates this at build time. HEADERS_CHARSET gives the charset to
6231 translate to; headers_charset can change it in the configuration, and
6232 "headers charset" can change it in an individual filter file.
6234 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6235 that creates RFC 2047 encoded "words" labels them as that charset instead
6236 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6237 expansion operator; (ii) when Exim creates a From: line for a local
6238 message; (iii) when a header line is rewritten to include a "phrase" part.
6240 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6241 buggy, causing it to skip the first lines of messages whose message ID
6242 ended in 'D'. This would not have bitten before Exim release 4.14, because
6243 message IDs were unlikely to end in 'D' before then. The effect was to have
6244 incorrect size information for certain domains.
6246 45. #include "config.h" was missing at the start of the crypt16.c module. This
6247 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6250 46. If there was a timeout during a "random" callout check, Exim treated it as
6251 a failure of the random address, and carried on sending RSET and the real
6252 address. If the delay was just some slowness somewhere, the response to the
6253 original RCPT would be taken as a response to RSET and so on, causing
6254 mayhem of various kinds.
6256 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6257 when I implemented it. It didn't allow for the fact that some option values
6258 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6259 the right test for positive values.
6261 48. Domain names in DNS records are case-independent. Exim always looks them up
6262 in lower case. Some resolvers return domain names in exactly the case they
6263 appear in the zone file, that is, they may contain uppercase letters. Not
6264 all resolvers do this - some return always lower case. Exim was treating a
6265 change of case by a resolver as a change of domain, similar to a widening
6266 of a domain abbreviation. This triggered its re-routing code and so it was
6267 trying to route what was effectively the same domain again. This normally
6268 caused routing to fail (because the router wouldn't handle the domain
6269 twice). Now Exim checks for this case specially, and just changes the
6270 casing of the domain that it ultimately uses when it transmits the message
6273 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6276 50. If a filter generated a file delivery with a non-absolute name (possible if
6277 no home directory exists for the router), the forbid_file option was not
6280 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6281 the existing equality matching.
6283 52. Exim was using ints instead of ino_t variables in some places where it was
6284 dealing with inode numbers.
6286 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6287 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6288 called TMPDIR, and if it finds it is different, it changes its value.
6290 54. The smtp_printf() function is now made available to local_scan() so
6291 additional output lines can be written before returning. There is also an
6292 smtp_fflush() function to enable the detection of a dropped connection.
6293 The variables smtp_input and smtp_batched_input are exported to
6296 55. Changed the default runtime configuration: the message "Unknown user"
6297 has been removed from the ACL, and instead placed on the localuser router,
6298 using the cannot_route_message feature. This means that any verification
6299 failures that generate their own messages won't get overridden. Similarly,
6300 the "Unrouteable address" message that was in the ACL for unverifiable
6301 relay addresses has also been removed.
6303 56. Added hosts_avoid_esmtp to the smtp transport.
6305 57. The exicyclog script was not checking for the esoteric option
6306 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6307 will work only if exicyclog is run under the appropriate euid.
6309 58. Following a discussion on the list, the rules by which Exim recognises line
6310 endings on incoming messages have been changed. The -dropcr and drop_cr
6311 options are now no-ops, retained only for backwards compatibility. The
6312 following line terminators are recognized: LF CRLF CR. However, special
6313 processing applies to CR:
6315 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6316 nor a local message in the state where . is a terminator.
6318 (ii) If a bare CR is encountered in a header line, an extra space is added
6319 after the line terminator so as not to end the header. The reasoning
6320 behind this is that bare CRs in header lines are most likely either
6321 to be mistakes, or people trying to play silly games.
6323 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6324 was being incorrectly given as 18 bytes larger than it should have been.
6325 This is a VOB (very old bug).
6327 60. This may never have affected anything current, but just in case it has:
6328 When the local host is found other than at the start of a list of hosts,
6329 the local host, those with the same MX, and any that follow, are discarded.
6330 When the list in question was part of a longer list of hosts, the following
6331 hosts (not currently being processed) were also being discarded. This no
6332 longer happens. I'm not sure if this situation could ever has previously
6335 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6338 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6339 header. This is something that is recommended in a new Internet Draft, and
6340 is something that is documented as being done by Sendmail. There are two
6341 possible values. For messages generated by the autoreply transport, Exim
6344 Auto-Submitted: auto-replied
6346 whereas for all other generated messages (e.g. bounces) it adds
6348 Auto-Submitted: auto-generated
6350 63. The "personal" condition in filters now includes a test for the
6351 Auto-Submitted: header. If it contains the string "auto-" the message it
6352 not considered personal.
6354 64. Added rcpt_include_affixes as a generic transport option.
6356 65. Added queue_only_override (default true).
6358 66. Added the syslog_duplication option.
6360 67. If what should have been the first header line of a message consisted of
6361 a space followed by a colon, Exim was mis-interpreting it as a header line.
6362 It isn't of course - it is syntactically invalid and should therefore be
6363 treated as the start of the message body. The misbehaviour could have
6364 caused a number of strange effects, including loss of data in subsequent
6365 header lines, and spool format errors.
6367 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6368 client host had authenticated. This control can now be exercised by an ACL
6369 for more flexibility.
6371 69. By default, callouts do not happen when testing with -bh. There is now a
6372 variant, -bhc, which does actually run the callout code, including
6373 consulting and updating the callout cache.
6375 70. Added support for saslauthd authentication, courtesy of Alexander
6378 71. If statvfs() failed on the spool or log directories while checking their
6379 size for availability, Exim confusingly gave the error "space shortage".
6380 Furthermore, in debugging mode it crashed with a floating point exception.
6381 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6382 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6383 serious problem, Exim now writes to the main and panic logs when this
6384 happens, with details of the failure. It then refuses to accept the
6385 incoming message, giving the message "spool directory problem" or "log
6386 directory problem" with a 421 code for SMTP messages.
6388 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6389 0, 1, and 2 exist, because some OS complain for execs without them (see
6390 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6391 descriptors. However, the code omitted to check that the open succeeded,
6392 causing mysterious errors if for some reason the permissions on /dev/null
6393 got screwed. Now Exim writes a message to the main and panic logs, and
6394 bombs out if it can't open /dev/null.
6396 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6397 interact so that it is all more flexible. It is supposed to remain
6398 backwards compatible. Also added extra_local_interfaces.
6400 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6401 to bomb out with an assertion failure - to the client this appears as a
6402 connection drop. This problem occurs in the part of the code that was taken
6403 from the Samba project. Fortunately, the assertion is in a very simple
6404 function, so I have fixed this by reproducing the function inline in the
6405 one place where it is called, and arranging for authentication to fail
6406 instead of killing the process with assert().
6408 75. The SPA client code was not working when the server requested OEM rather
6409 than Unicode encoding.
6411 76. Added code to make require_files with a specific uid setting more usable in
6412 the case where statting the file as root fails - usually a non-root-mounted
6413 NFS file system. When this happens and the failure is EACCES, Exim now
6414 forks a subprocess and does the per-uid checking as the relevant uid.
6416 77. Added process_log_path.
6418 78. If log_file_path was not explicitly set, a setting of check_log_space or
6419 check_log_inodes was ignored.
6421 79. If a space check for the spool or log partitions fails, the incident is now
6422 logged. Of course, in the latter case the data may get lost...
6424 80. Added the %p formatting code to string_format() so that it can be used to
6425 print addresses in debug_print(). Adjusted all the address printing in the
6426 debugging in store.c to use %p rather than %d.
6428 81. There was a concern that a line of code in smtp_in.c could overflow a
6429 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6430 initially expressed, the concern was not well-founded, because trailing
6431 spaces are removed early. However, if the trailing spaces were followed by
6432 a NULL, they did not get removed, so the overflow was possible. Two fixes
6435 (a) I re-wrote the offending code in a cleaner fashion.
6436 (b) If an incoming SMTP command contains a NULL character, it is rejected
6439 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6440 runs initgroups(), so that if the Exim user is in any additional groups,
6441 they will be used during message reception.
6447 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob