1 Change log file for Exim from version 4.21
2 -------------------------------------------
7 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
8 and 3.4.4 - once the server is enabled to respond to an OCSP request
9 it does even when not requested, resulting in a stapling non-aware
10 client dropping the TLS connection.
12 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
13 support variable-length bit vectors. No functional change.
15 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
16 The I= interface field on outgoing lines is now after the H= remote
17 host field, same as incoming lines. There is a separate
18 outgoing_interface log selector which allows you to disable the
21 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
22 If not running log_selector +smtp_connection the mainlog would be held
23 open indefinitely after a "too many connections" event, including to a
24 deleted file after a log rotate. Leave the per net connection logging
25 leaving it open for efficiency as that will be quickly detected by the
26 check on the next write.
28 HS/01 Bug 1671: Fix post transport crash.
29 Processing the wait-<transport> messages could crash the delivery
30 process if the message IDs didn't exist for some reason. When
31 using 'split_spool_directory=yes' the construction of the spool
32 file name failed already, exposing the same netto behaviour.
34 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
35 mime_regex ACL conditions.
37 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
38 to DSN fail messages (bounces): remote IP, remote greeting, remote response
39 to HELO, local diagnostic string.
41 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
42 log line to debug. Even when configured with a tls authenticator many
43 client connections are expected to not authenticate in this way, so
44 an authenticate fail is not an error.
46 HS/02 Add the Exim version string to the process info. This way exiwhat
47 gives some more detail about the running daemon.
49 JH/06 Bug 1395: time-limit cacheing of DNS lookups, to the TTL value. This may
50 matter for fast-change records such as DNSBLs.
52 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
53 retry record, even if constant. There may be multiple transports with
54 different interface settings and the retry behaviour needs to be kept
57 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
59 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
61 JH/10 Bug 840: fix log_defer_output option of pipe transport
63 JH/11 Bug 830: use same host for all RCPTS of a message, even under
64 hosts_randomize. This matters a lot when combined with mua_wrapper.
66 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
67 ${quote_pgsql:<string>} operator.
69 JH/13 Bug 1708: avoid misaligned access in cached lookup.
71 JH/14 Change header file name for freeradius-client. Relevant if compiling
72 with Radius support; from the Gentoo tree and checked under Fedora.
74 JH/15 Bug 1712: Introduce $prdr_requested flag variable
76 JH/16 Bug 1714: Permit an empty string as expansion result for transport
77 option transport_filter, meaning no filtering.
79 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
81 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
82 defaults to "*" (all hosts). The variable is now available when not built
83 with TLS, default unset, mainly to enable keeping the testuite sane.
84 If a server certificate is not supplied (via tls_certificate) an error is
85 logged, and clients will find TLS connections fail on startup. Presumably
86 they will retry in-clear.
87 Packagers of Exim are strongly encouraged to create a server certificate
90 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
91 with the $config_file variable.
93 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
94 in transport context, after the attempt, and per-recipient. The latter type
95 is per host attempted. The event data is the error message, and the errno
96 information encodes the lookup type (A vs. MX) used for the (first) host,
97 and the trailing two digits of the smtp 4xx reponse.
99 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
100 to write to mainlog (or rejectlog, paniclog) in the window between file
101 creation and permissions/ownership being changed. Particularly affects
102 installations where exicyclog is run as root, rather than exim user;
103 result is that the running daemon panics and dies.
105 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
107 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
108 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
109 "pri" and "weight". Note that the previous implicit priority given by the
110 list order is no longer honoured.
112 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalisation
115 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
116 by defining SUPPORT_SOCKS.
118 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
119 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
120 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
121 variables are renamed to proxy_{local,external}_{address,port}.
123 JH/27 Move Internationalisation support from Experimental to mainline, enabled
124 for a build by defining SUPPORT_I18N
126 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
127 of the query string, and make ${quote_redis:} do that quoting.
129 JH/29 Move Events support from Experimental to mainline, enabled by default
130 and removable for a build by defining DISABLE_EVENT.
132 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
134 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
135 cached by the daemon.
137 JH/32 Move Redis support from Experimental to mainline, enabled for a build
138 by defining LOOKUP_REDIS. The libhiredis library is required.
140 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
141 keys are given for lookup.
143 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
144 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
145 only supported when built with TLS support. The PolarSSL SHA routines
146 are still used when the TLS library is too old for convenient support.
148 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
149 openssl_options), for security. OpenSSL forces this from version 1.1.0
150 server-side so match that on older versions.
152 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
153 allocation for $value could be released as the expansion processing
154 concluded, but leaving the global pointer active for it.
156 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
157 and to use the domains and local_parts ACL conditions.
159 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
160 incorrectly not doubled on cutthrough transmission, hence seen as a
161 body-termination at the receiving system - resulting in truncated mails.
162 Commonly the sender saw a TCP-level error, and retransmitted the message
163 via the normal store-and-forward channel. This could result in duplicates
164 received - but deduplicating mailstores were liable to retain only the
165 initial truncated version.
167 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
169 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
171 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
172 we're in there, support oversigning also; bug 1309.
174 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
176 HS/04 Add support for keep_environment and add_environment options.
178 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
179 either intentional arithmetic overflow during PRNG, or testing config-
182 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
183 delivery resulted in actual delivery. Cancel cutthrough before DATA
186 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
187 rejects a recipient: pass the reject to the originator.
193 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
196 JH/02 The smtp transport option "multi_domain" is now expanded.
198 JH/03 The smtp transport now requests PRDR by default, if the server offers
201 JH/04 Certificate name checking on server certificates, when exim is a client,
202 is now done by default. The transport option tls_verify_cert_hostnames
203 can be used to disable this per-host. The build option
204 EXPERIMENTAL_CERTNAMES is withdrawn.
206 JH/05 The value of the tls_verify_certificates smtp transport and main options
207 default to the word "system" to access the system default CA bundle.
208 For GnuTLS, only version 3.0.20 or later.
210 JH/06 Verification of the server certificate for a TLS connection is now tried
211 (but not required) by default. The verification status is now logged by
212 default, for both outbound TLS and client-certificate supplying inbound
215 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
218 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
219 Status Notification (bounce) messages are now MIME format per RFC 3464.
220 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
221 under the control of the dsn_advertise_hosts option, and routers may
222 have a dsn_lasthop option.
224 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
225 default, modifiable by a malware= option. The list separator for
226 the options can now be changed in the usual way. Bug 68.
228 JH/10 The smtp_receive_timeout main option is now expanded before use.
230 JH/11 The incoming_interface log option now also enables logging of the
231 local interface on delivery outgoing connections.
233 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
234 if the interface and destination host and port all match.
236 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
239 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
240 Patch from Andrew Lewis.
242 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
243 now supports optional time-restrictions, weighting, and priority
244 modifiers per server. Patch originally by <rommer@active.by>.
246 JH/16 The spamd_address main option now supports a mixed list of local
247 and remote servers. Remote servers can be IPv6 addresses, and
248 specify a port-range.
250 JH/17 Bug 68: The spamd_address main option now supports an optional
251 timeout value per server.
253 JH/18 Bug 1581: Router and transport options headers_add/remove can
254 now have the list separator specified.
256 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
259 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
262 JH/21 Support for the A6 type of dns record is withdrawn.
264 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
265 rather than the verbs used.
267 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
268 from 255 to 1024 chars.
270 JH/24 Verification callouts now attempt to use TLS by default.
272 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
273 are generic router options now. The defaults didn't change.
275 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
276 Original patch from Alexander Shikoff, worked over by JH.
278 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
279 files if ZCAT_COMMAND is not executable.
281 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
283 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
285 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
286 Normally benign, it bites when the pair was led to by a CNAME;
287 modern usage is to not canoicalize the domain to a CNAME target
288 (and we were inconsistent anyway for A-only vs AAAA+A).
290 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
292 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
293 when evaluating $sender_host_dnssec.
295 JH/31 Check the HELO verification lookup for DNSSEC, adding new
296 $sender_helo_dnssec variable.
298 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
300 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
302 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
304 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
305 documented as working, but never had. Support all but $spam_report.
307 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
308 added for tls authenticator.
313 TL/01 When running the test suite, the README says that variables such as
314 no_msglog_check are global and can be placed anywhere in a specific
315 test's script, however it was observed that placement needed to be near
316 the beginning for it to behave that way. Changed the runtest perl
317 script to read through the entire script once to detect and set these
318 variables, reset to the beginning of the script, and then run through
319 the script parsing/test process like normal.
321 TL/02 The BSD's have an arc4random API. One of the functions to induce
322 adding randomness was arc4random_stir(), but it has been removed in
323 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
324 function when detected.
326 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
327 cause callback expansion.
329 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
330 syntax errors in an expansion can be treated as a string instead of
331 logging or causing an error, due to the internal use of bool_lax
332 instead of bool when processing it.
334 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
335 server certificates when making smtp deliveries.
337 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
339 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
341 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
343 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
344 Merged patch from Sebastian Wiedenroth.
346 JH/05 Fix results-pipe from transport process. Several recipients, combined
347 with certificate use, exposed issues where response data items split
348 over buffer boundaries were not parsed properly. This eventually
349 resulted in duplicates being sent. This issue only became common enough
350 to notice due to the introduction of conection certificate information,
351 the item size being so much larger. Found and fixed by Wolfgang Breyha.
353 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
354 size buffer was used, resulting in syntax errors when an expansion
357 JH/07 Add support for directories of certificates when compiled with a GnuTLS
358 version 3.3.6 or later.
360 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
361 is EXPERIMENTAL_EVENT, the main-configuration and transport options
362 both become "event_action", the variables become $event_name, $event_data
363 and $event_defer_errno. There is a new variable $verify_mode, usable in
364 routers, transports and related events. The tls:cert event is now also
365 raised for inbound connections, if the main configuration event_action
368 TL/06 In test suite, disable OCSP for old versions of openssl which contained
369 early OCSP support, but no stapling (appears to be less than 1.0.0).
371 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
372 server certificate names available under the smtp transport option
373 "tls_verify_cert_hostname" now do not permit multi-component wildcard
376 JH/10 Time-related extraction expansions from certificates now use the main
377 option "timezone" setting for output formatting, and are consistent
378 between OpenSSL and GnuTLS compilations. Bug 1541.
380 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
381 encoded parameter in the incoming message. Bug 1558.
383 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
384 include certificate info, eximon was claiming there were spoolfile
387 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
389 JH/14 Log delivery-related information more consistently, using the sequence
390 "H=<name> [<ip>]" wherever possible.
392 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
393 are problematic for Debian distribution, omit them from the release
396 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
398 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
401 JH/17 Fix a null-indirection in certextract expansions when a nondefault
402 output list separator was used.
407 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
408 checkers that were complaining about end of non-void function with no
411 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
412 This was a regression intruduced in 4.83 by another bugfix.
414 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
416 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
417 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
423 TF/01 Correctly close the server side of TLS when forking for delivery.
425 When a message was received over SMTP with TLS, Exim failed to clear up
426 the incoming connection properly after forking off the child process to
427 deliver the message. In some situations the subsequent outgoing
428 delivery connection happened to have the same fd number as the incoming
429 connection previously had. Exim would try to use TLS and fail, logging
430 a "Bad file descriptor" error.
432 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
433 utilities have not been installed.
435 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
436 temporary space as the ACL may create new global variables.
438 TL/01 LDAP support uses per connection or global context settings, depending
439 upon the detected version of the libraries at build time.
441 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
442 to extract and use the src ip:port in logging and expansions as if it
443 were a direct connection from the outside internet. PPv2 support was
444 updated based on HAProxy spec change in May 2014.
446 JH/02 Add ${listextract {number}{list}{success}{fail}}.
448 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
449 Properly escape header and check for NULL return.
451 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
454 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
456 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
457 characters in header names, implemented as a verify condition.
458 Contributed by Michael Fischer v. Mollard.
460 TL/05 Rename SPF condition results err_perm and err_temp to standardized
461 results permerror and temperror. Previous values are deprecated but
462 still accepted. In a future release, err_perm and err_temp will be
463 completely removed, which will be a backward incompatibility if the
464 ACL tests for either of these two old results. Patch contributed by
465 user bes-internal on the mailing list.
467 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
469 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
470 selectors, in both main and reject logs.
472 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
475 JH/07 Add malware type "sock" for talking to simple daemon.
477 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
479 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
480 routers/transports under cutthrough routing.
482 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
483 numbers. Touch up "bool" conditional to keep the same definition.
485 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
487 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
489 JH/12 Expand items in router/transport headers_add or headers_remove lists
490 individually rather than the list as a whole. Bug 1452.
492 Required for reasonable handling of multiple headers_ options when
493 they may be empty; requires that headers_remove items with embedded
494 colons must have them doubled (or the list-separator changed).
496 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
497 view the policy declared in the DMARC record. Currently, $dmarc_status
498 is a combined value of both the record presence and the result of the
501 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
503 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
504 dnslookup router and the smtp transport (applying to the forward
507 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
508 of ldap servers used for a specific lookup. Patch provided by Heiko
511 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
512 New variable $lookup_dnssec_authenticated for observability.
514 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
515 Patch submitted by Lars Timman.
517 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
519 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
520 Requires trusted mode and valid format message id, aborts otherwise.
521 Patch contributed by Heiko Schlichting.
523 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
524 certextract with support for various fields. Bug 1358.
526 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
527 is requested by default, modifiable by smtp transport option
530 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
531 operate on certificate variables to give certificate fingerprints
532 Also new ${sha256:cert_variable}.
534 JH/23 The PRDR feature is moved from being Experimental into the mainline.
536 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
537 Christian Aistleitner.
539 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
541 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
542 file. Patch from Wolfgang Breyha.
544 JH/25 Expand the coverage of the delivery $host and $host_address to
545 client authenticators run in verify callout. Bug 1476.
547 JH/26 Port service names are now accepted for tls_on_connect_ports, to
548 align with daemon_smtp_ports. Bug 72.
550 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
551 support and error reporting did not work properly.
553 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
554 and is readable. Patch from Andrew Colin Kissa.
556 TL/14 Enhance documentation of ${run expansion and how it parses the
557 commandline after expansion, particularly in the case when an
558 unquoted variable expansion results in an empty value.
560 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
562 PP/02 Fix internal collision of T_APL on systems which support RFC3123
563 by renaming away from it. Addresses GH issue 15, reported by
566 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
568 TL/15 SECURITY: prevent double expansion in math comparison functions
569 (can expand unsanitized data). Not remotely exploitable.
576 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
578 PP/02 Make -n do something, by making it not do something.
579 When combined with -bP, the name of an option is not output.
581 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
584 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
585 $sender_host_name and config options to manage this, and basic check
588 PP/05 DSCP support for outbound connections and control modifier for inbound.
590 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
591 (Only plugin which currently uses this is kerberos4, which nobody should
592 be using, but we should make it available and other future plugins might
593 conceivably use it, even though it would break NAT; stuff *should* be
594 using channel bindings instead).
596 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
597 name; added for Sendmail compatibility; requires admin caller.
598 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
599 just ignore it); requires trusted caller.
600 Also parse but ignore: -Ac -Am -X<logfile>
603 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
605 TL/02 Add +smtp_confirmation as a default logging option.
607 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
608 Patch by Magnus Holmgren from 2007-02-20.
610 TL/04 Bugzilla 1281 - Spec typo.
611 Bugzilla 1283 - Spec typo.
612 Bugzilla 1290 - Spec grammar fixes.
614 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
616 TL/06 Add Experimental DMARC support using libopendmarc libraries.
618 TL/07 Fix an out of order global option causing a segfault. Reported to dev
619 mailing list by by Dmitry Isaikin.
621 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
623 JH/02 Support "G" suffix to numbers in ${if comparisons.
625 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
627 NM/01 Bugzilla 1197 - Spec typo
628 Bugzilla 1196 - Spec examples corrections
630 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
632 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
633 gnutls_enable_pkcs11, but renamed to more accurately indicate its
636 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
637 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
639 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
640 "acl {{name}{arg}...}", and optional args on acl condition
643 JH/05 Permit multiple router/transport headers_add/remove lines.
645 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
647 JH/07 Avoid using a waiting database for a single-message-only transport.
648 Performance patch from Paul Fisher. Bugzilla 1262.
650 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
653 JH/09 Add $headers_added variable, with content from use of ACL modifier
654 add_header (but not yet added to the message). Bugzilla 199.
656 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
657 Pulled from Bugzilla 817 by Wolfgang Breyha.
659 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
661 (nb: this is the same fix as in Exim 4.80.1)
663 JH/11 Add A= logging on delivery lines, and a client_set_id option on
666 JH/12 Add optional authenticated_sender logging to A= and a log_selector
669 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
671 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
672 advertise SMTP AUTH mechanism to us, instead of a generic
673 protocol violation error. Also, make Exim more robust to bad
674 data from the Dovecot auth socket.
676 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
678 When a queue runner is handling a message, Exim first routes the
679 recipient addresses, during which it prunes them based on the retry
680 hints database. After that it attempts to deliver the message to
681 any remaining recipients. It then updates the hints database using
684 So if a recipient address works intermittently, it can get repeatedly
685 deferred at routing time. The retry hints record remains fresh so the
686 address never reaches the final cutoff time.
688 This is a fairly common occurrence when a user is bumping up against
689 their storage quota. Exim had some logic in its local delivery code
690 to deal with this. However it did not apply to per-recipient defers
691 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
693 This change adds a proper retry rule check during routing so that the
694 final cutoff time is checked against the message's age. We only do
695 this check if there is an address retry record and there is not a
696 domain retry record; this implies that previous attempts to handle
697 the address had the retry_use_local_parts option turned on. We use
698 this as an approximation for the destination being like a local
699 delivery, as in LMTP.
701 I suspect this new check makes the old local delivery cutoff check
702 redundant, but I have not verified this so I left the code in place.
704 TF/02 Correct gecos expansion when From: is a prefix of the username.
706 Test 0254 submits a message to Exim with the header
710 When I ran the test suite under the user fanf2, Exim expanded
711 the header to contain my full name, whereas it should have added
712 a Resent-Sender: header. It erroneously treats any prefix of the
713 username as equal to the username.
715 This change corrects that bug.
717 GF/01 DCC debug and logging tidyup
718 Error conditions log to paniclog rather than rejectlog.
719 Debug lines prefixed by "DCC: " to remove any ambiguity.
721 TF/03 Avoid unnecessary rebuilds of lookup-related code.
723 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
724 Bug spotted by Jeremy Harris; was flawed since initial commit.
725 Would have resulted in OCSP responses post-SNI triggering an Exim
726 NULL dereference and crash.
728 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
730 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
731 Bug detection, analysis and fix by Samuel Thibault.
732 Bugzilla 1331, Debian bug #698092.
734 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
736 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
737 Server implementation by Todd Lyons, client by JH.
738 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
739 config variable "prdr_enable" controls whether the server
740 advertises the facility. If the client requests PRDR a new
741 acl_data_smtp_prdr ACL is called once for each recipient, after
742 the body content is received and before the acl_smtp_data ACL.
743 The client is controlled by bolth of: a hosts_try_prdr option
744 on the smtp transport, and the server advertisement.
745 Default client logging of deliveries and rejections involving
746 PRDR are flagged with the string "PRDR".
748 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
749 fclose(). Diagnosis by Todd Lyons.
751 PP/17 Update configure.default to handle IPv6 localhost better.
752 Patch by Alain Williams (plus minor tweaks).
755 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
756 This is now consistent with GnuTLS, and is now documented: the
757 previous undocumented portable approach to treating the option as
758 unset was to force an expansion failure. That still works, and
759 an empty string is now equivalent.
761 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
762 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
763 not performing validation itself.
765 PP/20 Added force_command boolean option to pipe transport.
766 Patch from Nick Koston, of cPanel Inc.
768 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
771 TF/04 Added udpsend ACL modifer and hexquote expansion operator
773 PP/21 Fix eximon continuous updating with timestamped log-files.
774 Broken in a format-string cleanup in 4.80, missed when I repaired the
775 other false fix of the same issue.
776 Report and fix from Heiko Schlichting.
779 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
780 Report from Prashanth Katuri.
782 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
783 It's SecureTransport, so affects any MacOS clients which use the
784 system-integrated TLS libraries, including email clients.
786 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
787 using a MIME ACL for non-SMTP local injection.
788 Report and assistance in diagnosis by Warren Baker.
790 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
792 JH/16 Fix comparisons for 64b. Bugzilla 1385.
794 TL/09 Add expansion variable $authenticated_fail_id to keep track of
795 last id that failed so it may be referenced in subsequent ACL's.
797 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
800 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
801 ldap library initialization, allowing self-signed CA's to be
802 used. Also properly sets require_cert option later in code by
803 using NULL (global ldap config) instead of ldap handle (per
804 session). Bug diagnosis and testing by alxgomz.
806 TL/12 Enhanced documentation in the ratelimit.pl script provided in
807 the src/util/ subdirectory.
809 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
810 renamed to Transport Post Delivery Action by Jeremy Harris, as
813 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
814 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
815 redis_servers = needs to be configured which will be used by the redis
816 lookup. Patch from Warren Baker, of The Packet Hub.
818 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
820 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
821 hostname or reverse DNS when processing a host list. Used suggestions
822 from multiple comments on this bug.
824 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
826 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
827 Missed a few lines, added it to make the runtest require no keyboard
830 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
831 contains upper case chars. Make router use caseful_local_part.
833 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
834 support when GnuTLS has been built with p11-kit.
840 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
842 This, or similar/improved, will also be change PP/11 of 4.82.
848 PP/01 Handle short writes when writing local log-files.
849 In practice, only affects FreeBSD (8 onwards).
850 Bugzilla 1053, with thanks to Dmitry Isaikin.
852 NM/01 Bugzilla 949 - Documentation tweak
854 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
857 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
859 PP/02 Implemented gsasl authenticator.
861 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
863 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
864 `pkg-config foo` for cflags/libs.
866 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
867 with rest of GSASL and with heimdal_gssapi.
869 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
870 `pkg-config foo` for cflags/libs for the TLS implementation.
872 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
873 properties get this fed in as external SSF. A number of robustness
874 and debugging improvements to the cyrus_sasl authenticator.
876 PP/08 cyrus_sasl server now expands the server_realm option.
878 PP/09 Bugzilla 1214 - Log authentication information in reject log.
879 Patch by Jeremy Harris.
881 PP/10 Added dbmjz lookup type.
883 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
885 PP/12 MAIL args handles TAB as well as SP, for better interop with
886 non-compliant senders.
887 Analysis and variant patch by Todd Lyons.
889 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
890 Bug report from Lars Müller <lars@samba.org> (via SUSE),
891 Patch from Dirk Mueller <dmueller@suse.com>
893 PP/13 tls_peerdn now print-escaped for spool files.
894 Observed some $tls_peerdn in wild which contained \n, which resulted
895 in spool file corruption.
897 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
898 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
899 or write after TLS renegotiation, which otherwise led to messages
902 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
903 as a tracking header (ie: a signed header comes before the signature).
904 Patch from Wolfgang Breyha.
906 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
907 comma-sep list; embedded commas doubled.
909 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
911 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
913 Report and patch from Dmitry Banschikov.
915 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
916 Removed SSL_clear() after SSL_new() which led to protocol negotiation
917 failures. We appear to now support TLS1.1+ with Exim.
919 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
920 lets Exim select keys and certificates based upon TLS SNI from client.
921 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
922 before an outbound SMTP session. New log_selector, +tls_sni.
924 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
925 NULL dereference. Report and patch from Alun Jones.
927 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
928 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
929 Not seeing resolver debug output on NetBSD, but suspect this is a
930 resolver implementation change.
932 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
933 Left warnings. Added "eximon gdb" invocation mode.
935 PP/21 Defaulting "accept_8bitmime" to true, not false.
937 PP/22 Added -bw for inetd wait mode support.
939 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
940 locate the relevant includes and libraries. Made this the default.
942 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
943 Bugzilla 1246, report and most of solution from Tomasz Kusy.
945 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
946 This may cause build issues on older platforms.
948 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
949 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
950 gnutls_require_mac & gnutls_require_protocols (no longer supported).
951 Added SNI support via GnuTLS too.
952 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
954 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
956 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
957 Applied second patch from Janne, implementing suggestion to default
958 multiple-strings-in-record handling to match SPF spec.
960 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
962 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
963 read-only, out of scope).
964 Patch from Wolfgang Breyha, report from Stuart Northfield.
966 PP/29 Fix three issues highlighted by clang analyser static analysis.
967 Only crash-plausible issue would require the Cambridge-specific
968 iplookup router and a misconfiguration.
969 Report from Marcin Mirosław.
971 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
973 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
974 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
975 As part of this, removing so much warning spew let me fix some minor
976 real issues in debug logging.
978 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
979 assignment on my part. Fixed.
981 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
982 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
983 Janne Snabb (who went above and beyond: thank you).
985 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
986 string otherwise requires a connection and a bunch more work and it's
987 relatively easy to get wrong. Should also expose TLS library linkage
990 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
991 64-bit ${eval} (JH/03).
993 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
994 GNU libc to support some of the 64-bit stuff, should not lead to
995 conflicts. Defined before os.h is pulled in, so if a given platform
996 needs to override this, it can.
998 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
999 protection layer was required, which is not implemented.
1000 Bugzilla 1254, patch from Wolfgang Breyha.
1002 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1003 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1004 tls_dhparam take prime identifiers. Also unbreak combination of
1005 OpenSSL+DH_params+TLSSNI.
1007 PP/39 Disable SSLv2 by default in OpenSSL support.
1013 PP/01 Solaris build fix for Oracle's LDAP libraries.
1014 Bugzilla 1109, patch from Stephen Usher.
1016 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1018 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1021 TF/02 Fix a couple more cases where we did not log the error message
1022 when unlink() failed. See also change 4.74-TF/03.
1024 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1025 lock up or crash if it happened to be inside a call to libc when it
1026 got a SIGUSR1 from exiwhat.
1028 The SIGUSR1 handler appends the current process status to the process
1029 log which is later printed by exiwhat. It used to use the general
1030 purpose logging code to do this, but several functions it calls are
1031 not safe for signals.
1033 The new output code in the SIGUSR1 handler is specific to the process
1034 log, and simple enough that it's easy to inspect for signal safety.
1035 Removing some special cases also simplifies the general logging code.
1036 Removing the spurious timestamps from the process log simplifies
1039 TF/04 Improved ratelimit ACL condition.
1041 The /noupdate option has been deprecated in favour of /readonly which
1042 has clearer semantics. The /leaky, /strict, and /readonly update modes
1043 are mutually exclusive. The update mode is no longer included in the
1044 database key; it just determines when the database is updated. (This
1045 means that when you upgrde Exim will forget old rate measurements.)
1047 Exim now checks that the per_* options are used with an update mode that
1048 makes sense for the current ACL. For example, when Exim is processing a
1049 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1050 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1051 must specify per_mail/readonly. If you omit the update mode it defaults to
1052 /leaky where that makes sense (as before) or /readonly where required.
1054 The /noupdate option is now undocumented but still supported for
1055 backwards compatibility. It is equivalent to /readonly except that in
1056 ACLs where /readonly is required you may specify /leaky/noupdate or
1057 /strict/noupdate which are treated the same as /readonly.
1059 A useful new feature is the /count= option. This is a generalization
1060 of the per_byte option, so that you can measure the throughput of other
1061 aggregate values. For example, the per_byte option is now equivalent
1062 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1064 The per_rcpt option has been generalized using the /count= mechanism
1065 (though it's more complicated than the per_byte equivalence). When it is
1066 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1067 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1068 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1069 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1070 Note that using per_rcpt with a non-readonly update mode in more than
1071 one ACL will cause the recipients to be double-counted. (The per_mail
1072 and per_byte options don't have this problem.)
1074 The handling of very low rates has changed slightly. If the computed rate
1075 is less than the event's count (usually one) then this event is the first
1076 after a long gap. In this case the rate is set to the same as this event's
1077 count, so that the first message of a spam run is counted properly.
1079 The major new feature is a mechanism for counting the rate of unique
1080 events. The new per_addr option counts the number of different
1081 recipients that someone has sent messages to in the last time period. It
1082 behaves like per_rcpt if all the recipient addresses are different, but
1083 duplicate recipient addresses do not increase the measured rate. Like
1084 the /count= option this is a general mechanism, so the per_addr option
1085 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1086 example, measure the rate that a client uses different sender addresses
1087 with the options per_mail/unique=$sender_address. There are further
1088 details in the main documentation.
1090 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1092 TF/06 Removed a few PCRE remnants.
1094 TF/07 Automatically extract Exim's version number from tags in the git
1095 repository when doing development or release builds.
1097 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1098 Bugzilla 879. Patch from Paul Fisher.
1100 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1101 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1104 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1106 PP/05 Variable $av_failed, true if the AV scanner deferred.
1107 Bugzilla 1078. Patch from John Horne.
1109 PP/06 Stop make process more reliably on build failure.
1110 Bugzilla 1087. Patch from Heiko Schlittermann.
1112 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1113 Bugzilla 1089. Patch from Heiko Schlittermann.
1115 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1116 Bugzilla 1131. Patch from Holger Weiß.
1118 PP/09 Handle IPv6 addresses with SPF.
1119 Bugzilla 860. Patch from Wolfgang Breyha.
1121 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1123 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1126 PP/11 match_* no longer expand right-hand-side by default.
1127 New compile-time build option, EXPAND_LISTMATCH_RHS.
1128 New expansion conditions, "inlist", "inlisti".
1130 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1132 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1134 PP/14 fix log_write() format string regression from TF/03.
1135 Bugzilla 1152. Patch from Dmitry Isaikin.
1141 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1143 PP/02 Harmonised TLS library version reporting; only show if debugging.
1144 Layout now matches that introduced for other libraries in 4.74 PP/03.
1146 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1148 PP/04 New "dns_use_edns0" global option.
1150 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1153 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1154 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1156 TK/01 Updated PolarSSL code to 0.14.2.
1157 Bugzilla 1097. Patch from Andreas Metzler.
1159 PP/07 Catch divide-by-zero in ${eval:...}.
1160 Fixes bugzilla 1102.
1162 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1165 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1166 format-string attack -- SECURITY: remote arbitrary code execution.
1168 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1169 time unintentionally subject to list matching rules, letting the header
1170 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1171 arbitrary string expansion). This allowed for information disclosure.
1173 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1174 INT_MIN/-1 -- value coerced to INT_MAX.
1180 NM/01 Workround for PCRE version dependency in version reporting
1183 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1184 This fixes portability to compilers other than gcc, notably
1185 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1187 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1188 makefiles for portability to HP-UX and POSIX correctness.
1190 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1191 Also via indented variable definition in the Makefile.
1192 (Debugging by Oliver Heesakkers).
1194 PP/02 Restore caching of spamd results with expanded spamd_address.
1195 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1197 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1198 Improves build reliability. Fix from: Frank Elsner
1200 NM/02 Fix wide character breakage in the rfc2047 coding
1201 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1203 NM/03 Allow underscore in dnslist lookups
1204 Fixes bug 1026. Patch from Graeme Fowler
1206 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1207 Code patches from Adam Ciarcinski of NetBSD.
1209 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1212 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1213 is logged, to avoid truncation. Patch from John Horne.
1215 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1216 Patch from Jakob Hirsch.
1218 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1219 SQL string expansion failure details.
1220 Patch from Andrey Oktyabrski.
1222 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1223 Patch from Simon Arlott.
1225 PP/09 New lookups functionality failed to compile on old gcc which rejects
1226 extern declarations in function scope.
1227 Patch from Oliver Fleischmann
1229 PP/10 Use sig_atomic_t for flags set from signal handlers.
1230 Check getgroups() return and improve debugging.
1231 Fixed developed for diagnosis in bug 927 (which turned out to be
1234 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1235 Patch from Mark Zealey.
1237 PP/12 Bugzilla 1056: Improved spamd server selection.
1238 Patch from Mark Zealey.
1240 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1241 Based on patch from Heiko Schlittermann.
1243 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1244 Patch from Uwe Doering, sign-off by Michael Haardt.
1246 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1247 variable declaration deep within a block. Bug and patch from
1250 PP/15 lookups-Makefile IRIX compatibilty coercion.
1252 PP/16 Make DISABLE_DKIM build knob functional.
1254 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1255 Patch by Simon Arlott
1257 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1258 variable argument macros. Our copy now differs from upstream.
1264 TF/01 Failure to get a lock on a hints database can have serious
1265 consequences so log it to the panic log.
1267 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1268 controlled using the smtp_confirmation log selector.
1270 TF/03 Include the error message when we fail to unlink a spool file.
1272 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1273 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1274 for maintaining out-of-tree patches for some time.
1276 PP/01 Bugzilla 139: Documentation and portability issues.
1277 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1278 Handle per-OS dynamic-module compilation flags.
1280 PP/02 Let /dev/null have normal permissions.
1281 The 4.73 fixes were a little too stringent and complained about the
1282 permissions on /dev/null. Exempt it from some checks.
1283 Reported by Andreas M. Kirchwitz.
1285 PP/03 Report version information for many libraries, including
1286 Exim version information for dynamically loaded libraries. Created
1287 version.h, now support a version extension string for distributors
1288 who patch heavily. Dynamic module ABI change.
1290 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1291 privilege escalation vulnerability whereby the Exim run-time user
1292 can cause root to append content of the attacker's choosing to
1295 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1298 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1299 If dropping privileges for untrusted macros, we disabled normal logging
1300 on the basis that it would fail; for the Exim run-time user, this is not
1301 the case, and it resulted in successful deliveries going unlogged.
1302 Fixed. Reported by Andreas Metzler.
1308 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1309 only prepend for the Resent-* case. Fixes regression introduced in
1310 Exim 4.70 by NM/22 for Bugzilla 607.
1312 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1313 increasing numbers of administrators be bitten by this.
1315 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1317 PP/03 Bugzilla 994: added openssl_options main configuration option.
1319 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1321 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1323 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1325 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1326 without a peer certificate, leading to a segfault because of an
1327 assumption that peers always have certificates. Be a little more
1328 paranoid. Problem reported by Martin Tscholak.
1330 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1331 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1332 NB: ClamAV planning to remove STREAM in "middle of 2010".
1333 CL also introduces -bmalware, various -d+acl logging additions and
1334 more caution in buffer sizes.
1336 PP/09 Implemented reverse_ip expansion operator.
1338 PP/10 Bugzilla 937: provide a "debug" ACL control.
1340 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1342 PP/12 Bugzilla 973: Implement --version.
1344 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1346 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1348 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1350 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1351 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1352 ignore trailing whitespace.
1354 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1356 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1359 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1360 Notification from Dr Andrew Aitchison.
1362 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1363 ExtendedDetectionInfo response format.
1364 Notification from John Horne.
1366 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1369 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1370 XSL and documented dependency on system catalogs, with examples of how
1373 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1376 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1377 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1378 configuration file which is writeable by the Exim user or group.
1380 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1381 of configuration files to cover files specified with the -C option if
1382 they are going to be used with root privileges, not just the default
1385 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1386 option (effectively making it always true).
1388 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1389 files to be used while preserving root privileges.
1391 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1392 that rogue child processes cannot use them.
1394 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1395 run-time user, instead of root.
1397 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1398 Exim run-time user without dropping privileges.
1400 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1401 result string, instead of calling string_vformat() twice with the same
1404 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1405 for other users. Others should always drop root privileges if they use
1406 -C on the command line, even for a whitelisted configure file.
1408 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1410 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1416 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1417 $data_path, and $header_path variables; fixed documentation bugs and
1420 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1421 exipick to access non-standard spools, including the "frozen" queue
1424 NM/01 Bugzilla 965: Support mysql stored procedures.
1425 Patch from Alain Williams
1427 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1429 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1430 Patch from Andreas Metzler
1432 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1433 Patch from Kirill Miazine
1435 NM/05 Bugzilla 671: Added umask to procmail example.
1437 JJ/03 installed exipick 20100323.0, fixing doc bug
1439 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1440 directory. Notification and patch from Dan Rosenberg.
1442 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1444 TK/02 Improve log output when DKIM signing operation fails.
1446 MH/01 Treat the transport option dkim_domain as a colon separated
1447 list, not as a single string, and sign the message with each element,
1448 omitting multiple occurences of the same signer.
1450 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1451 Bugzilla 985, 986. Patch by Simon Arlott
1453 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1454 Patch by Simon Arlott
1456 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1457 MBX locking. Notification from Dan Rosenberg.
1463 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1465 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1467 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1469 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1471 NM/04 Bugzilla 915: Flag broken perl installation during build.
1477 TK/01 Added patch by Johannes Berg that expands the main option
1478 "spamd_address" if it starts with a dollar sign.
1480 TK/02 Write list of recipients to X-Envelope-Sender header when building
1481 the mbox-format spool file for content scanning (suggested by Jakob
1484 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1485 (http://www.dcc-servers.net/) support via dccifd. Activated by
1486 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1488 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1489 by Mark Daniel Reidel <mr@df.eu>.
1491 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1492 When building exim an external PCRE library is now needed -
1493 PCRE is a system library on the majority of modern systems.
1494 See entry on PCRE_LIBS in EDITME file.
1496 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1497 conversation. Added nologin parameter to request.
1498 Patch contributed by Kirill Miazine.
1500 TF/01 Do not log submission mode rewrites if they do not change the address.
1502 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1504 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1505 log files in place. Contributed by Roberto Lima.
1507 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1509 TF/03 Bugzilla 615: When checking the local_parts router precondition
1510 after a local_part_suffix or local_part_prefix option, Exim now
1511 does not use the address's named list lookup cache, since this
1512 contains cached lookups for the whole local part.
1514 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1515 Robert Millan. Documentation is in experimental-spec.txt.
1517 TF/04 Bugzilla 668: Fix parallel build (make -j).
1519 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1521 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1522 Patch provided by Jan Srzednicki.
1524 TF/05 Leading white space used to be stripped from $spam_report which
1525 wrecked the formatting. Now it is preserved.
1527 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1528 that they are available at delivery time.
1530 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1532 TF/08 TLS error reporting now respects the incoming_interface and
1533 incoming_port log selectors.
1535 TF/09 Produce a more useful error message if an SMTP transport's hosts
1536 setting expands to an empty string.
1538 NM/06 Bugzilla 744: EXPN did not work under TLS.
1539 Patch provided by Phil Pennock.
1541 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1542 Patch provided by Richard Godbee.
1544 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1545 acl_smtp_notquit, added index entry.
1547 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1548 Patch provided by Eugene Bujak.
1550 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1551 accept(). Patch provided by Maxim Dounin.
1553 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1554 Patch provided by Phil Pennock.
1556 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1558 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1559 Patch provided by Brad "anomie" Jorsch.
1561 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1562 Patch provided by Dean Brooks.
1564 TK/05 Add native DKIM support (does not depend on external libraries).
1566 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1567 Patch provided by Graeme Fowler.
1569 NM/16 Bugzilla 851: Documentation example syntax fix.
1571 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1573 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1576 NM/19 Bugzilla 745: TLS version reporting.
1577 Patch provided by Phil Pennock.
1579 NM/20 Bugzilla 167: bool: condition support.
1580 Patch provided by Phil Pennock.
1582 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1583 clients. Patch provided by Phil Pennock.
1585 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1586 Patch provided by Brad "anomie" Jorsch.
1588 NM/23 Bugzilla 687: Fix misparses in eximstats.
1589 Patch provided by Heiko Schlittermann.
1591 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1592 Patch provided by Heiko Schlittermann.
1594 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1595 plus update to original patch.
1597 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1599 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1600 Patch provided by David Brownlee.
1602 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1604 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1606 NM/30 Bugzilla 888: TLS documentation bugfixes.
1608 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1610 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1611 Unlike the original bugzilla I have changed all shell scripts in src tree.
1613 NM/33 Bugzilla 898: Transport filter timeout fix.
1614 Patch by Todd Rinaldo.
1616 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1617 Patch by Serge Demonchaux.
1619 NM/35 Bugzilla 39: Base64 decode bug fixes.
1620 Patch by Jakob Hirsch.
1622 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1624 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1626 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1628 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1634 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1635 ALT-N's libdkim that I have put here:
1636 http://duncanthrax.net/exim-experimental/
1638 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1639 were called 'true' and it seems that C99 defines that as a reserved
1640 keyword to be used with 'bool' variable types. That means you could
1641 not include C99-style headers which use bools without triggering
1642 build errors in sieve.c.
1644 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1645 as mailq or other aliases. Changed the --help handling significantly
1646 to do whats expected. exim_usage() emits usage/help information.
1648 SC/01 Added the -bylocaldomain option to eximstats.
1650 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1652 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1654 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1660 PH/01 Another patch from the Sieve maintainer.
1662 PH/02 When an IPv6 address is converted to a string for single-key lookup
1663 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1664 dots are used instead of colons so that keys in lsearch files need not
1665 contain colons. This was done some time before quoting was made available
1666 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1667 (notated using the quote facility) so as to distinguish them from IPv4
1668 keys. This meant that lookups for IP addresses in host lists did not work
1669 for iplsearch lookups.
1671 This has been fixed by arranging for IPv6 addresses to be expressed with
1672 colons if the lookup type is iplsearch. This is not incompatible, because
1673 previously such lookups could never work.
1675 The situation is now rather anomolous, since one *can* have colons in
1676 ordinary lsearch keys. However, making the change in all cases is
1677 incompatible and would probably break a number of configurations.
1679 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1682 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1683 conversion specification without a maximum field width, thereby enabling
1684 a rogue spamd server to cause a buffer overflow. While nobody in their
1685 right mind would setup Exim to query an untrusted spamd server, an
1686 attacker that gains access to a server running spamd could potentially
1687 exploit this vulnerability to run arbitrary code as the Exim user.
1689 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1690 $primary_hostname instead of what libspf2 thinks the hosts name is.
1692 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1693 a directory entry by the name of the lookup key. Previously, if a
1694 symlink pointed to a non-existing file or a file in a directory that
1695 Exim lacked permissions to read, a lookup for a key matching that
1696 symlink would fail. Now it is enough that a matching directory entry
1697 exists, symlink or not. (Bugzilla 503.)
1699 PH/03 The body_linecount and body_zerocount variables are now exported in the
1702 PH/04 Added the $dnslist_matched variable.
1704 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1705 This means they are set thereafter only if the connection becomes
1708 PH/06 Added the client_condition to authenticators so that some can be skipped
1709 by clients under certain conditions.
1711 PH/07 The error message for a badly-placed control=no_multiline_responses left
1712 "_responses" off the end of the name.
1714 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1716 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1717 (without spaces) instead of just copying the configuration text.
1719 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1721 PH/11 Added $max_received_linelength.
1723 PH/12 Added +ignore_defer and +include_defer to host lists.
1725 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1726 way in which PCRE > 7.0 is built.
1728 PH/14 Implemented queue_only_load_latch.
1730 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1731 MAIL command. The effect was to mangle the value on 64-bit systems.
1733 PH/16 Another patch from the Sieve maintainer.
1735 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1737 PH/18 If a system quota error occurred while trying to create the file for
1738 a maildir delivery, the message "Mailbox is full" was not appended to the
1739 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1740 only to a quota excession during the actual writing of the file.
1742 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1743 characters?) which causes problems in log lines. The DN values are now
1744 passed through string_printing() before being added to log lines.
1746 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1747 and InterBase are left for another time.)
1749 PH/21 Added message_body_newlines option.
1751 PH/22 Guard against possible overflow in moan_check_errorcopy().
1753 PH/23 POSIX allows open() to be a macro; guard against that.
1755 PH/24 If the recipient of an error message contained an @ in the local part
1756 (suitably quoted, of course), incorrect values were put in $domain and
1757 $local_part during the evaluation of errors_copy.
1763 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1764 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1767 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1768 issue a MAIL command.
1770 PH/02 In an ACL statement such as
1772 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1774 if a client was not listed at all, or was listed with a value other than
1775 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1776 the condition was not true (as it should be), so access was not denied.
1777 The bug was that the ! inversion was incorrectly passed on to the second
1778 item. This has been fixed.
1780 PH/03 Added additional dnslists conditions == and =& which are different from
1781 = and & when the dns lookup returns more than one IP address.
1783 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1784 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1786 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1787 FSYNC, which compiles an option called disable_fsync that allows for
1788 bypassing fsync(). The documentation is heavily laced with warnings.
1790 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1792 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1793 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1794 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1795 including adding "make clean"; (3) Added -fPIC when compiling the test
1796 dynamically loaded module, to get rid of a warning.
1798 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1799 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1800 = 0s. The bug is otherwise harmless.
1802 PH/07 There was a bug in the dovecot authenticator such that the value of
1803 $auth1 could be overwritten, and so not correctly preserved, after a
1804 successful authentication. This usually meant that the value preserved by
1805 the server_setid option was incorrect.
1807 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1809 PH/09 Installed PCRE release 7.0.
1811 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1812 run for batched SMTP input. It is now run at the start of every message
1813 in the batch. While fixing this I discovered that the process information
1814 (output by running exiwhat) was not always getting set for -bs and -bS
1815 input. This is fixed, and it now also says "batched" for BSMTP.
1817 PH/11 Added control=no_pipelining.
1819 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1820 patch, slightly modified), and move the expansion of helo_data till after
1821 the connection is made in the smtp transport (so it can use these
1824 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1826 PH/14 Added log_selector = +pid.
1828 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1830 PH/16 Add ${if forany and ${if forall.
1832 PH/17 Added dsn_from option to vary the From: line in DSNs.
1834 PH/18 Flush SMTP output before performing a callout, unless control =
1835 no_callout_flush is set.
1837 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1838 was true (the default) a successful delivery failed to delete the retry
1839 item, thus causing premature timeout of the address. The bug is now
1842 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1844 PH/21 Long custom messages for fakedefer and fakereject are now split up
1845 into multiline reponses in the same way that messages for "deny" and
1846 other ACL rejections are.
1848 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1849 with slight modification.
1851 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1852 draft, changing the syntax and factoring some duplicate code".
1854 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1855 for deliveries of the second and subsequent messages over the same SMTP
1858 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1859 ${reduce, with only minor "tidies".
1861 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1863 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1864 expansion side effects.
1866 PH/27 When a message times out after an over-quota error from an Exim-imposed
1867 quota, the bounce message says "mailbox is full". This message was not
1868 being given when it was a system quota that was exceeded. It now should
1871 MH/03 Made $recipients available in local_scan(). local_scan() already has
1872 better access to the recipient list through recipients_list[], but
1873 $recipients can be useful in postmaster-provided expansion strings.
1875 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1876 in the case of a MAIL command with additional options following the
1877 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1878 were accidentally chopped off.
1880 PH/29 SMTP synchronization checks are implemented when a command is read -
1881 there is a check that no more input is waiting when there shouldn't be
1882 any. However, for some commands, a delay in an ACL can mean that it is
1883 some time before the response is written. In this time, more input might
1884 arrive, invalidly. So now there are extra checks after an ACL has run for
1885 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1886 pipelining has not been advertised.
1888 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1890 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1891 returned in the SMTP response when smtp_return_error_details was false.
1892 This has been fixed.
1894 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1895 instead of the C I/O that was originally supplied, because problems were
1896 reported on Solaris.
1898 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1899 Exim which did not show up earlier: it was assuming that a call to
1900 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1901 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1902 was a macro that became an assignment, so it seemed to work. This has
1903 changed to a proper function call with a void return, hence the compile
1904 error. Exim's code has been fixed.
1906 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1909 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1911 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1913 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1914 because it turns out to be too big or there is a timeout) there is no
1915 "Completed" line in the log. When some messages of this type were
1916 selected by exigrep, they were listed as "not completed". Others were
1917 picked up by some special patterns. I have improved the selection
1918 criteria to be more general.
1920 PH/38 The host_find_failed option in the manualroute router can now be set
1921 to "ignore", to completely ignore a host whose IP address cannot be
1922 found. If all hosts are ignored, the behaviour is controlled by the new
1923 host_all_ignored option.
1925 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1926 homing or because of multiple MX records with /mx) generated more than
1927 one IP address, and the following item turned out to be the local host,
1928 all the secondary addresses of the first item were incorrectly removed
1929 from the list, along with the local host and any following hosts (which
1930 is what is supposed to happen).
1932 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1933 whoever called Exim into the -H file. In the case of the daemon it was
1934 behaving confusingly. When first started, it used values for whoever
1935 started the daemon, but after a SIGHUP it used the Exim user (because it
1936 calls itself on a restart). I have changed the code so that it now always
1939 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1940 message are rejected with the same error (e.g. no authentication or bad
1941 sender address), and a DATA command is nevertheless sent (as can happen
1942 with PIPELINING or a stupid MUA), the error message that was given to the
1943 RCPT commands is included in the rejection of the DATA command. This is
1944 intended to be helpful for MUAs that show only the final error to their
1947 PH/42 Another patch from the Sieve maintainer.
1949 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1950 Eximstats - Fixed some broken HTML links and added missing column headers
1952 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1953 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1955 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1957 PH/43 Yet another patch from the Sieve maintainer.
1959 PH/44 I found a way to check for a TCP/IP connection going away before sending
1960 the response to the final '.' that terminates a message, but only in the
1961 case where the client has not sent further data following the '.'
1962 (unfortunately, this is allowed). However, in many cases there won't be
1963 any further data because there won't be any more messages to send. A call
1964 to select() can be used: if it shows that the input is "ready", there is
1965 either input waiting, or the socket has been closed. An attempt to read
1966 the next input character can distinguish the two cases. Previously, Exim
1967 would have sent an OK response which the client would never have see.
1968 This could lead to message repetition. This fix should cure that, at
1969 least in a lot of common cases.
1971 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1972 advertised in response to EHLO.
1978 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1979 fixed by 4.65/MH/01 (is this a record?) are fixed:
1981 (i) An empty string was always treated as zero by the numeric comparison
1982 operators. This behaviour has been restored.
1984 (ii) It is documented that the numeric comparison operators always treat
1985 their arguments as decimal numbers. This was broken in that numbers
1986 starting with 0 were being interpreted as octal.
1988 While fixing these problems I realized that there was another issue that
1989 hadn't been noticed. Values of message_size_limit (both the global option
1990 and the transport option) were treated as octal if they started with 0.
1991 The documentation was vague. These values are now always treated as
1992 decimal, and I will make that clear in the documentation.
1998 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1999 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2002 MH/01 Don't check that the operands of numeric comparison operators are
2003 integers when their expansion is in "skipping" mode (fixes bug
2004 introduced by 4.64-PH/07).
2006 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2007 child addresses, Exim now panics and dies. Previously, because the count
2008 is held in a short int, deliveries were likely to be lost. As such a
2009 large number of recipients for a single message is ridiculous
2010 (performance will be very, very poor), I have chosen to impose a limit
2011 rather than extend the field.
2017 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2018 leftover -K file (the existence of which was triggered by #402).
2019 While we were at it, introduced process PID as part of the -K
2020 filename. This should rule out race conditions when creating
2023 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2024 processing considerably. Previous code took too long for large mails,
2025 triggering a timeout which in turn triggers #401.
2027 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2028 in the DK code in transports.c. sendfile() is not really portable,
2029 hence the _LINUX specificness.
2031 TF/01 In the add_headers option to the mail command in an Exim filter,
2032 there was a bug that Exim would claim a syntax error in any
2033 header after the first one which had an odd number of characters
2036 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2037 callout verification, Exim cached a "reject" for the entire domain. This
2038 is correct for most verifications, but it is not correct for a recipient
2039 verification with use_sender or use_postmaster set, because in that case
2040 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2041 case of MAIL FROM:<> rejection from other early rejections (e.g.
2042 rejection of HELO). When verifying a recipient using a non-null MAIL
2043 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2044 Whatever the result of the callout, the value of the domain cache is
2045 left unchanged (for any other kind of callout, getting as far as trying
2046 RCPT means that the domain itself is ok).
2048 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2051 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2052 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2053 session, the daemon ignores ECONNECT errors and logs others; it now
2054 ignores EPIPE as well.
2056 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2057 (quoted-printable decoding).
2059 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2060 later the small subsequent patch to fix an introduced bug.
2062 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2064 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2066 PH/08 An error is now given if message_size_limit is specified negative.
2068 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2069 to be given (somewhat) arbitrary names.
2071 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2074 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2075 miscellaneous code fixes
2077 PH/10 Added the log_reject_target ACL modifier to specify where to log
2080 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2081 hostname. This is wrong, because it relates to the incoming message (and
2082 probably the interface on which it is arriving) and not to the outgoing
2083 callout (which could be using a different interface). This has been
2084 changed to use the value of the helo_data option from the smtp transport
2085 instead - this is what is used when a message is actually being sent. If
2086 there is no remote transport (possible with a router that sets up host
2087 addresses), $smtp_active_hostname is used.
2089 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2090 tweaks were necessary in order to get it to work (see also 21 below):
2091 (a) The code assumed that strncpy() returns a negative number on buffer
2092 overflow, which isn't the case. Replaced with Exim's string_format()
2094 (b) There were several signed/unsigned issues. I just did the minimum
2095 hacking in of casts. There is scope for a larger refactoring.
2096 (c) The code used strcasecmp() which is not a standard C function.
2097 Replaced with Exim's strcmpic() function.
2098 (d) The code set only $1; it now sets $auth1 as well.
2099 (e) A simple test gave the error "authentication client didn't specify
2100 service in request". It would seem that Dovecot has changed its
2101 interface. Fortunately there's a specification; I followed it and
2102 changed what the client sends and it appears to be working now.
2104 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2107 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2108 address A is aliased to B and C, where B exists and C does not. Without
2109 -v the output is "A verified" because verification stops after a
2110 successful redirection if more than one address is generated. However,
2111 with -v the child addresses are also verified. Exim was outputting "A
2112 failed to verify" and then showing the successful verification for C,
2113 with its parentage. It now outputs "B failed to verify", showing B's
2114 parentage before showing the successful verification of C.
2116 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2117 look up a TXT record in a specific list after matching in a combined
2120 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2121 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2122 they consult the DNS. I had assumed they would set it the way they
2123 wanted; and indeed my experiments on Linux seem to show that in some
2124 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2125 To be on the safe side, however, I have now made the interface to
2126 host_find_byname() similar to host_find_bydns(), with an argument
2127 containing the DNS resolver options. The host_find_byname() function now
2128 sets these options at its start, just as host_find_bydns() does. The smtp
2129 transport options dns_qualify_single and dns_search_parents are passed to
2130 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2131 of host_find_byname() use the default settings of RES_DEFNAMES
2132 (qualify_single) but not RES_DNSRCH (search_parents).
2134 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2135 spool_read_header() do less string testing, by means of a preliminary
2136 switch on the second character of optional "-foo" lines. (This is
2137 overdue, caused by the large number of possibilities that now exist.
2138 Originally there were few.) While I was there, I also converted the
2139 str(n)cmp tests so they don't re-test the leading "-" and the first
2140 character, in the hope this might squeeze out yet more improvement.
2142 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2143 flag allowing group syntax was set by the header_syntax check but not
2144 turned off, possible causing trouble later; (2) The flag was not being
2145 set at all for the header_verify test, causing "group"-style headers to
2146 be rejected. I have now set it in this case, and also caused header_
2147 verify to ignore an empty address taken from a group. While doing this, I
2148 came across some other cases where the code for allowing group syntax
2149 while scanning a header line wasn't quite right (mostly, not resetting
2150 the flag correctly in the right place). These bugs could have caused
2151 trouble for malformed header lines. I hope it is now all correct.
2153 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2154 with the "reply" argument non-NULL. The code, however (which originally
2155 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2156 but it didn't always do it. This confused somebody who was copying the
2157 code for some other use. I have removed all the tests.
2159 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2160 feature that was used to support insecure browsers during the U.S. crypto
2161 embargo. It requires special client support, and Exim is probably the
2162 only MTA that supported it -- and would never use it because real RSA is
2163 always available. This code has been removed, because it had the bad
2164 effect of slowing Exim down by computing (never used) parameters for the
2165 RSA_EXPORT functionality.
2167 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2168 authenticator to fail if there's a tab character in the incoming data
2169 (there should never be unless someone is messing about, as it's supposed
2170 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2171 if the connection is using TLS or if the remote IP is the same as the
2172 local IP, and the "valid-client-cert option" if a client certificate has
2175 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2176 authenticators. This can be used for authorization after authentication
2177 succeeds. (In the case of plaintext, it servers for both authentication
2180 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2181 if any retry times were supplied.
2183 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2184 connection, where there is no client IP address to check. In this
2185 situation, the verify now always succeeds.
2187 PH/25 Applied John Jetmore's -Mset patch.
2189 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2191 PH/27 In a string expansion for a processed (not raw) header when multiple
2192 headers of the same name were present, leading whitespace was being
2193 removed from all of them, but trailing whitespace was being removed only
2194 from the last one. Now trailing whitespace is removed from each header
2195 before concatenation. Completely empty headers in a concatenation (as
2196 before) are ignored.
2198 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2199 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2201 PH/29 [Removed. This was a change that I later backed out, and forgot to
2202 correct the ChangeLog entry (that I had efficiently created) before
2203 committing the later change.]
2205 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2206 address errors (4xx response to RCPT) over the same connection as other
2207 messages routed to the same hosts. Such deliveries are always "forced",
2208 so retry times are not inspected. This resulted in far too many retries
2209 for the affected addresses. The effect occurred only when there were more
2210 hosts than the hosts_max_try setting in the smtp transport when it had
2211 the 4xx errors. Those hosts that it had tried were not added to the list
2212 of hosts for which the message was waiting, so if all were tried, there
2213 was no problem. Two fixes have been applied:
2215 (i) If there are any address or message errors in an SMTP delivery, none
2216 of the hosts (tried or untried) are now added to the list of hosts
2217 for which the message is waiting, so the message should not be a
2218 candidate for sending over the same connection that was used for a
2219 successful delivery of some other message. This seems entirely
2220 reasonable: after all the message is NOT "waiting for some host".
2221 This is so "obvious" that I'm not sure why it wasn't done
2222 previously. Hope I haven't missed anything, but it can't do any
2223 harm, as the worst effect is to miss an optimization.
2225 (ii) If, despite (i), such a delivery is accidentally attempted, the
2226 routing retry time is respected, so at least it doesn't keep
2227 hammering the server.
2229 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2230 in ${readsocket because some servers need this prod.
2232 PH/32 Added some extra debug output when updating a wait-xxx database.
2234 PH/33 The hint "could be header name not terminated by colon", which has been
2235 given for certain expansion errors for a long time, was not being given
2236 for the ${if def:h_colon_omitted{... case.
2238 PH/34 The spec says: "With one important exception, whenever a domain list is
2239 being scanned, $domain contains the subject domain." There was at least
2240 one case where this was not true.
2242 PH/35 The error "getsockname() failed: connection reset by peer" was being
2243 written to the panic log as well as the main log, but it isn't really
2244 panic-worthy as it just means the connection died rather early on. I have
2245 removed the panic log writing for the ECONNRESET error when getsockname()
2248 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2249 runs only) independently of the message's sender address. This meant
2250 that, if the 4xx error was in fact related to the sender, a different
2251 message to the same recipient with a different sender could confuse
2252 things. In particualar, this can happen when sending to a greylisting
2253 server, but other circumstances could also provoke similar problems.
2254 I have changed the default so that the retry time for these errors is now
2255 based a combination of the sender and recipient addresses. This change
2256 can be overridden by setting address_retry_include_sender=false in the
2259 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2260 remote server are returned as part of bounce messages. This was not
2261 happening for LMTP over a pipe (the lmtp transport), but now it is the
2262 same for both kinds of LMTP.
2264 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2265 in header lines that were in fact CNAMEs. This is no longer the case.
2267 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2268 and queue runs started by the daemon processed all messages. This has
2269 been fixed so that -R and -S can now usefully be given with -q<time>.
2271 PH/40 Import PCRE release 6.7 (fixes some bugs).
2273 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2275 PH/42 Give an error if -q is specified more than once.
2277 PH/43 Renamed the variables $interface_address and $interface_port as
2278 $received_ip_address and $received_port, to make it clear that these
2279 values apply to message reception, and not to the outgoing interface when
2280 a message is delivered. (The old names remain recognized, of course.)
2282 PH/44 There was no timeout on the connect() call when using a Unix domain
2283 socket in the ${readsocket expansion. There now is.
2285 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2286 be meaningful with "accept".
2288 SC/01 Eximstats V1.43
2289 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2291 SC/02 Eximstats V1.44
2292 Use a glob alias rather than an array ref in the generated
2293 parser. This improves both readability and performance.
2295 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2296 Collect SpamAssassin and rejection statistics.
2297 Don't display local sender or destination tables unless
2298 there is data to show.
2299 Added average volumes into the top table text output.
2301 SC/04 Eximstats V1.46
2302 Collect data on the number of addresses (recipients)
2303 as well as the number of messages.
2305 SC/05 Eximstats V1.47
2306 Added 'Message too big' to the list of mail rejection
2307 reasons (thanks to Marco Gaiarin).
2309 SC/06 Eximstats V1.48
2310 Mainlog lines which have GMT offsets and are too short to
2311 have a flag are now skipped.
2313 SC/07 Eximstats V1.49 (Alain Williams)
2314 Added the -emptyok flag.
2316 SC/08 Eximstats V1.50
2317 Fixes for obtaining the IP address from reject messages.
2319 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2320 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2321 whitesspace changes from 4.64-PH/27
2323 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2326 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2327 are found, allow negative numbers in numeric criteria)
2329 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2331 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2332 to match changes made in 4.64-PH/43
2334 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2336 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2337 "domain" is actually an IP address, and force a failure. This locks out
2338 those revolvers/nameservers that support "A-for-A" lookups, in
2339 contravention of the specifications.
2341 PH/48 When a host name was looked up from an IP address, and the subsequent
2342 forward lookup of the name timed out, the host name was left in
2343 $sender_host_name, contrary to the specification.
2345 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2346 restricted to single-key lookups, Exim was not diagnosing an error if
2347 * or *@ was used with a query-style lookup.
2349 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2351 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2352 long ago, but noone interested enough thought of it. Let's just say that
2353 the "1.1" means that there are some new functions that weren't there at
2354 some point in the past.
2356 PH/51 Error processing for expansion failure of helo_data from an smtp
2357 transport during callout processing was broken.
2359 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2360 tested/used via the -bh/-bhc/-bs options.
2362 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2363 bug, fixed in subsequent PCRE releases).
2365 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2366 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2368 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2374 SC/01 Use a glob alias rather than an array ref in eximstats generated
2375 parser. This improves both readability and performance.
2377 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2378 Don't display local sender or destination tables in eximstats unless
2379 there is data to show.
2380 Added average volumes into the eximstats top table text output.
2382 SC/03 Collect data on the number of addresses (recipients) as well
2383 as the number of messages in eximstats.
2385 TF/01 Correct an error in the documentation for the redirect router. Exim
2386 does (usually) call initgroups() when daemonizing.
2388 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2389 with consistent privilege compared to when running as a daemon.
2391 TF/03 Note in the spec that $authenticated_id is not set for local
2392 submissions from trusted users.
2394 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2395 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2397 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2398 by adding some example configuration directives to the default
2399 configuration file. A little bit of work is required to uncomment the
2400 directives and define how usernames and passwords are checked, but
2401 there is now a framework to start from.
2403 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2404 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2405 without this. I don't know how relevant this is to other LDAP libraries.
2407 PH/02 Add the verb name to the "unknown ACL verb" error.
2409 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2411 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2413 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2414 directory not expanded when it should be if an expanded home directory
2415 was set for the address (which is overridden by the transport).
2417 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2420 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2421 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2422 because it is too late at that time, and has no effect.
2424 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2425 security issue with \' (bugzilla #107). I could not use the
2426 PQescapeStringConn() function, because it needs a PGconn value as one of
2429 PH/08 When testing addresses using -bt, indicate those final addresses that
2430 are duplicates that would not cause an additional delivery. At least one
2431 person was confused, thinking that -bt output corresponded to deliveries.
2432 (Suppressing duplicates isn't a good idea as you lose the information
2433 about possibly different redirections that led to the duplicates.)
2435 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2436 systems where poll() doesn't work, in particular OS X.
2438 PH/10 Added more information to debugging output for retry time not reached.
2440 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2441 operations in malware.c.
2443 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2446 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2447 syslog_duplication set false, log lines that would normally be written
2448 both the the main log and to the reject log were not written to syslog at
2451 PH/14 In the default configuration, change the use of "message" in ACL warn
2452 statements to "add_header".
2454 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2455 not followed by a command (e.g. "seen endif").
2457 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2458 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2461 PH/17 Added extra conditions to the default value of delay_warning_condition
2465 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2466 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2467 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2470 The Auto-Submitted: and various List- headers are standardised, whereas I
2471 don't think Precedence: ever was.
2473 PH/18 Refactored debugging code in route_finduser() to show more information,
2474 in particular, the error code if getpwnam() issues one.
2476 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2477 This is apparently needed in addition to the PH/07 change above to avoid
2478 any possible encoding problems.
2480 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2481 but not after initializing Perl.
2483 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2484 output them only if debugging. By default they are written stderr,
2485 apparently, which is not desirable.
2487 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2490 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2493 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2495 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2496 authenticated or an ident call has been made. Suppress the default
2497 values for $authenticated_id and $authenticated_sender (but permit -oMai
2498 and -oMas) when testing with -bh.
2500 PH/24 Re-jigged the order of the tests in the default configuration so that the
2501 tests for valid domains and recipients precede the DNS black list and CSA
2502 tests, on the grounds that those ones are more expensive.
2504 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2505 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2506 This bug exists in every version of Exim that I still have, right back to
2509 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2510 However, an attempt to turn on case-sensitivity in a regex key by
2511 including (?-i) didn't work because the subject string was already
2512 lowercased, and the effects were non-intuitive. It turns out that a
2513 one-line patch can be used to allow (?-i) to work as expected.
2519 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2520 other effects) broke the use of negated acl sub-conditions.
2522 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2525 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2526 "Deny" causes Exim to reject the incoming connection with a 554 error.
2527 Unfortunately, if there is a major crisis, such as a disk failure,
2528 tcp-wrappers gives "deny", whereas what one would like would be some
2529 kind of temporary error. A kludge has been added to help with this.
2530 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2531 554 error is used if errno is still zero or contains ENOENT (which occurs
2532 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2535 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2537 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2538 errors. Otherwise a message that provokes a temporary error (when other
2539 messages do not) can cause a whole host to time out.
2541 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2542 addresses were routed directly to files or pipes from a redirect router.
2543 File deliveries just didn't batch; pipe deliveries might have suffered
2546 PH/06 A failure to get a lock for a hints database would erroneously always say
2547 "Failed to get write lock", even when it was really a read lock.
2549 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2550 of 0600. This has been changed to use the value of the lockfile_mode
2551 option (which defaults to 0600).
2553 PH/08 Applied small patch from the Sieve maintainer.
2555 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2556 folder from quota calculations, a direct delivery into this folder messed
2557 up the contents of the maildirsize file. This was because the regex was
2558 used only to exclude .Trash (or whatever) when the size of the mailbox
2559 was calculated. There was no check that a delivery was happening into an
2560 excluded directory. This bug has been fixed by ignoring all quota
2561 processing for deliveries into excluded directories.
2563 PH/10 Added the maildirfolder_create_regex option to appendfile.
2569 PH/01 The code for finding all the local interface addresses on a FreeBSD
2570 system running IPv6 was broken. This may well have applied to all BSD
2571 systems, as well as to others that have similar system calls. The broken
2572 code found IPv4 interfaces correctly, but gave incorrect values for the
2573 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2574 that it would not match correctly against @[] and not recognize the IPv6
2577 PH/02 The ipliteral router was not recognizing addresses of the form user@
2578 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2580 PH/03 Added disable_ipv6.
2582 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2583 decoded form, because it is most often used to construct To: headers
2584 lines in autoreplies, and the decoded form may well be syntactically
2585 invalid. However, $reply_address has leading white space removed, and all
2586 newlines turned into spaces so that the autoreply transport does not
2589 PH/05 If group was specified without a user on a router, and no group or user
2590 was specified on a transport, the group from the router was ignored.
2592 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2593 for visible compile-time settings that can be used to change these
2594 numbers, for those that want even more. Backwards compatibility with old
2595 spool files has been maintained. However, going back to a previous Exim
2596 release will lost any variables that are in spool files.
2598 PH/07 Two small changes when running in the test harness: increase delay when
2599 passing a TCP/IP connection to a new process, in case the original
2600 process has to generate a bounce, and remove special handling of
2601 127.0.0.2 (sic), which is no longer necessary.
2603 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2604 be the same on different OS.
2606 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2609 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2610 whether --show-vars was specified or not
2612 JJ/02 exipick: Added support for new ACL variable spool format introduced
2615 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2616 syntactically invalid From: or Reply-to: line, and a filter used this to
2617 generate an autoreply, and therefore failed to obtain an address for the
2618 autoreply, Exim could try to deliver to a non-existent relative file
2619 name, causing unrelated and misleading errors. What now happens is that
2620 it logs this as a hard delivery error, but does not attempt to create a
2623 PH/11 The exinext utility has a -C option for testing purposes, but although
2624 the given file was scanned by exinext itself; it wasn't being passed on
2625 when Exim was called.
2627 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2628 an end-of-file indication when reading a command response.
2630 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2631 compiled. In many other places in Exim, IPv6 addresses are always
2632 recognized, so I have changed this. It also means that IPv4 domain
2633 literals of the form [IPV4:n.n.n.n] are now always recognized.
2635 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2636 used if the router is not running as root, for example, when verifying at
2637 ACL time, or when using -bh. The debugging output from this situation was
2638 non-existent - all you got was a failure to exec. I have made two
2641 (a) Failures to set uid/gid, the current directory, or a process leader
2642 in a subprocess such as that created by queryprogram now generate
2643 suitable debugging ouput when -d is set.
2645 (b) The queryprogram router detects when it is not running as root,
2646 outputs suitable debugging information if -d is set, and then runs
2647 the subprocess without attempting to change uid/gid.
2649 PH/15 Minor change to Makefile for building test_host (undocumented testing
2652 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2653 additional section of a DNS packet that returns MX or SRV records.
2654 Instead, it always explicitly searches for A/AAAA records. This avoids
2655 major problems that occur when a DNS server includes only records of one
2656 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2657 fixed another bug: if SRV records were looked up and the corresponding
2658 address records were *not* found in the additional section, the port
2659 values from the SRV records were lost.
2661 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2662 using the correct key (the original address) when searching the retry
2663 rules in order to find which one to use for generating the retry hint.
2665 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2666 adding the default one. Similarly, if it contains a Reply-To: header, the
2667 errors_reply_to option, if set, is not used.
2669 PH/19 When calculating a retry time, Exim used to measure the "time since
2670 failure" by looking at the "first failed" field in the retry record. Now
2671 it does not use this if it is later than than the arrival time of the
2672 message. Instead it uses the arrival time. This makes for better
2673 behaviour in cases where some deliveries succeed, thus re-setting the
2674 "first failed" field. An example is a quota failure for a huge message
2675 when small messages continue to be delivered. Without this change, the
2676 "time since failure" will always be short, possible causing more frequent
2677 delivery attempts for the huge message than are intended.
2678 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2680 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2681 $1, $2, $3) because the numerical variables can be reset during some
2682 expansion items (e.g. "match"), thereby losing the authentication data.
2684 PH/21 Make -bV show the size of off_t variables so that the test suite can
2685 decide whether to run tests for quotas > 2G.
2687 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2688 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2689 filecount value is greater than 2G or if a quota value is greater than 2G
2690 on a system where the size of off_t is not greater than 4, a panic error
2693 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2694 never match. The debug and -bh output now contains an explicit error
2695 message indicating a malformed IPv4 address or mask.
2697 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2698 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2699 PH/23 above applies.
2701 PH/25 Do not write to syslog when running in the test harness. The only
2702 occasion when this arises is a failure to open the main or panic logs
2703 (for which there is an explicit test).
2705 PH/26 Added the /no_tell option to "control=freeze".
2707 PH/27 If a host name lookup failed very early in a connection, for example, if
2708 the IP address matched host_lookup and the reverse lookup yielded a name
2709 that did not have a forward lookup, an error message of the form "no IP
2710 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2711 could be logged. Now it outputs the IP address instead of "NULL".
2713 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2714 allows the sender and the authenticated sender to be set when
2715 submitting a message from within Exim. Since child_open_exim() is
2716 documented for local_scan(), the new function should be too.
2718 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2719 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2720 results in an empty string is now treated as unset.
2722 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2724 PH/31 Added sender_verify_fail logging option.
2726 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2727 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2728 tidied the source and removed it altogether.
2730 PH/33 When a queue run was abandoned because the load average was too high, a
2731 log line was always written; now it is written only if the queue_run log
2732 selector is set. In addition, the log line for abandonment now contains
2733 information about the queue run such as the pid. This is always present
2734 in "start" and "stop" lines but was omitted from the "abandon" line.
2736 PH/34 Omit spaces between a header name and the colon in the error message that
2737 is given when verify = headers_syntax fails (if there are lots of them,
2738 the message gets confusing).
2740 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2741 names, as there are now some PTR records that contain slashes. This check
2742 is only to protect against broken name servers that fall over on strange
2743 characters, so the fact that it applies to all lookups doesn't matter.
2745 PH/36 Now that the new test suite is complete, we can remove some of the
2746 special code in Exim that was needed for the old test suite. For example,
2747 sorting DNS records because real resolvers return them in an arbitrary
2748 order. The new test suite's fake resolver always returns records in the
2751 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2752 bounces) except when queue_only is set, to avoid logging races between
2753 the different processes.
2755 PH/38 Panic-die if .include specifies a non-absolute path.
2757 PH/39 A tweak to the "H" retry rule from its user.
2759 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2760 a label. They prevented compilation on older perls.
2762 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2763 a warning to be raised on newish perls.
2765 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2766 on queue. Changes to match documented behaviour of showing count of
2767 messages matching specified criteria.
2769 PH/40 Changed the default ident timeout from 30s to 5s.
2771 PH/41 Added support for the use of login_cap features, on those BSD systems
2772 that have them, for controlling the resources used by pipe deliveries.
2774 PH/42 The content-scanning code uses fopen() to create files in which to put
2775 message data. Previously it was not paying any attention to the mode of
2776 the files. Exim runs with umask(0) because the rest of the code creates
2777 files with open(), and sets the required mode explicitly. Thus, these
2778 files were ending up world-writeable. This was not a big issue, because,
2779 being within the spool directory, they were not world-accessible. I have
2780 created a function called modefopen, which takes an additional mode
2781 argument. It sets umask(777), creates the file, chmods it to the required
2782 mode, then resets the umask. All the relevant calls to fopen() in the
2783 content scanning code have been changed to use this function.
2785 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2786 to 24 hours. This avoids potential overflow problems when processing G
2787 and H retry rules. I suspect nobody ever tinkers with this value.
2789 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2791 PH/45 When the plaintext authenticator is running as a client, the server's
2792 challenges are checked to ensure they are valid base64 strings. By
2793 default, the authentication attempt is cancelled if an invalid string is
2794 received. Setting client_ignore_invalid_base64 true ignores these errors.
2795 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2796 they are received. Thus, the responses can be made to depend on the
2797 challenges. If an invalid string is ignored, an empty string is placed in
2800 PH/46 Messages that are created by the autoreply transport now contains a
2801 References: header, in accordance with RFCs 2822 and 3834.
2803 PH/47 Added authenticated_sender_force to the smtp transport.
2805 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2807 PH/49 Installed latest patch from the Sieve maintainer.
2809 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2810 was also set, the appendfile transport was unnecessarily scanning a
2811 directory of message files (e.g. for maildir delivery) to find the count
2812 of files (along with the size), even though it did not need this
2813 information. It now does the scan only if it needs to find either the
2814 size of the count of files.
2816 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2818 PH/52 Two bugs concerned with error handling when the smtp transport is
2821 (i) Exim was not creating retry information for temporary errors given
2822 for individual recipients after the DATA command when the smtp transport
2823 was used in LMTP mode. This meant that they could be retried too
2824 frequently, and not timed out correctly.
2826 (ii) Exim was setting the flag that allows error details to be returned
2827 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2828 recipients that were returned after the DATA command.
2830 PH/53 This is related to PH/52, but is more general: for any failing address,
2831 when detailed error information was permitted to be returned to the
2832 sender, but the error was temporary, then after the final timeout, only
2833 "retry timeout exceeded" was returned. Now it returns the full error as
2834 well as "retry timeout exceeded".
2836 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2837 do this, and (what is worse) MTAs that accept it.
2839 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2840 will now be deprecated.
2842 PH/56 New os.c-cygwin from the Cygwin maintainer.
2844 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2845 formats (previously only available in exim formats via -bpr, -bpru,
2846 and -bpra. Now also available in native and exiqgrep formats)
2848 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2849 with very large, slow to parse queues
2851 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2853 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2855 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2856 responses to authentication challenges, though it was showing the
2857 challenges; (ii) I've removed the CR characters from the debug output for
2860 PH/58 Allow for the insertion of a newline as well as a space when a string
2861 is turned into more than one encoded-word during RFC 2047 encoding. The
2862 Sieve code now uses this.
2864 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2865 data_4xx, lost_connection, tls_required.
2867 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2868 message was being sent as an SMTP response.
2870 PH/61 Add -l and -k options to exicyclog.
2872 PH/62 When verifying, if an address was redirected to one new address, so that
2873 verification continued, and the new address failed or deferred after
2874 having set something in $address_data, the value of $address_data was not
2875 passed back to the ACL. This was different to the case when no
2876 redirection occurred. The value is now passed back in both cases.
2878 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2879 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2880 use login_cap.h, so on its own it isn't the distinguishing feature. The
2881 new name refers directly to the setclassresources() function.
2883 PH/65 Added configuration files for NetBSD3.
2885 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2887 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2888 is preferred over IPv4.
2890 PH/68 The bounce_return_message and bounce_return_body options were not being
2891 honoured for bounces generated during the reception of non-SMTP messages.
2892 In particular, this applied to messages rejected by the ACL. This bug has
2893 been fixed. However, if bounce_return_message is true and bounce_return_
2894 body is false, the headers that are returned for a non-SMTP message
2895 include only those that have been read before the error was detected.
2896 (In the case of an ACL rejection, they have all been read.)
2898 PH/69 The HTML version of the specification is now built in a directory called
2899 spec_html instead of spec.html, because the latter looks like a path with
2900 a MIME-type, and this confuses some software.
2902 PH/70 Catch two compiler warnings in sieve.c.
2904 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2905 function verify_get_ident() calls ip_connect() to connect a socket, but
2906 if the "connect()" function timed out, ip_connect() used to close the
2907 socket. However, verify_get_ident() also closes the socket later, and in
2908 between Exim writes to the log, which may get opened at this point. When
2909 the socket was closed in ip_connect(), the log could get the same file
2910 descriptor number as the socket. This naturally causes chaos. The fix is
2911 not to close the socket in ip_connect(); the socket should be closed by
2912 the function that creates it. There was only one place in the code where
2913 this was missing, in the iplookup router, which I don't think anybody now
2914 uses, but I've fixed it anyway.
2916 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2917 well as to direct DNS lookups. Otherwise the handling of names in host
2918 lists is inconsistent and therefore confusing.
2924 PH/01 Two changes to the default runtime configuration:
2926 (1) Move the checks for relay_from_hosts and authenticated clients from
2927 after to before the (commented out) DNS black list checks.
2929 (2) Add control=submission to the relay_from_hosts and authenticated
2930 clients checks, on the grounds that messages accepted by these
2931 statements are most likely to be submissions.
2933 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2935 (1) Generate an error if the third argument for the ${prvs expansion is
2938 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2941 (3) Reset the variables that are obtained from the first argument of
2942 ${prvscheck and used in the second argument before leaving the code,
2943 because their memory is reclaimed, so using them afterwards may do
2946 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2947 one (it's much easier than Tom thought :-).
2949 (5) Because of (4), we can now allow for the use of $prvscheck_result
2950 inside the third argument.
2952 PH/03 For some reason, the default setting of PATH when running a command from
2953 a pipe transport was just "/usr/bin". I have changed it to
2956 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2957 anything to be listed in the output from -bV.
2959 PH/05 When a filter generated an autoreply, the entire To: header line was
2960 quoted in the delivery log line, like this:
2962 => >A.N.Other <ano@some.domain> <original@ddress> ...
2964 This has been changed so that it extracts the operative address. There
2965 may be more than one such address. If so, they are comma-separated, like
2968 => >ano@some.domain,ona@other.domain <original@ddress> ...
2970 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2971 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2972 not being looked up in the rDNS to get a host name, Exim was showing the
2973 IP address twice in Received: lines, even though the IP addresses were
2974 identical. For example:
2976 Received: from [1.2.3.4] (helo=[1.2.3.4])
2978 However, if the real host name was known, it was omitting the HELO data
2979 if it matched the actual IP address. This has been tidied up so that it
2980 doesn't show the same IP address twice.
2982 PH/07 When both +timestamp and +memory debugging was on, the value given by
2983 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2984 called by the debug printing, thereby overwriting the timestamp buffer.
2985 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2987 PH/08 When the original message was included in an autoreply transport, it
2988 always said "this is a copy of the message, including all the headers",
2989 even if body_only or headers_only was set. It now gives an appropriate
2992 PH/09 Applied a patch from the Sieve maintainer which:
2994 o fixes some comments
2995 o adds the (disabled) notify extension core
2996 o adds some debug output for the result of if/elsif tests
2997 o points to the current vacation draft in the documentation
2998 and documents the missing references header update
3002 o fixes a bug in processing the envelope test (when testing
3003 multiple envelope elements, the last element determinted the
3006 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3007 Electronic Mail") by including:
3009 Auto-submitted: auto-generated
3011 in the messages that it generates (bounce messages and others, such as
3012 warnings). In the case of bounce messages for non-SMTP mesages, there was
3013 also a typo: it was using "Auto_submitted" (underscore instead of
3014 hyphen). Since every message generated by Exim is necessarily in response
3015 to another message, thes have all been changed to:
3017 Auto-Submitted: auto-replied
3019 in accordance with these statements in the RFC:
3021 The auto-replied keyword:
3023 - SHOULD be used on messages sent in direct response to another
3024 message by an automatic process,
3026 - MUST NOT be used on manually-generated messages,
3028 - MAY be used on Delivery Status Notifications (DSNs) and Message
3029 Disposition Notifications (MDNs),
3031 - MUST NOT be used on messages generated by automatic or periodic
3032 processes, except for messages which are automatic responses to
3035 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3036 to the default Received: header definition.
3038 PH/12 Added log selector acl_warn_skipped (default on).
3040 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3041 variables because (a) they are in the wrong storage pool and (b) even if
3042 they were copied, it wouldn't work properly because of the caching.
3044 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3045 checking when decoding. Apparently there are clients that generate
3046 overlong encoded strings. Why am I not surprised?
3048 PH/15 If the first argument of "${if match_address" was not empty, but did not
3049 contain an "@" character, Exim crashed. Now it writes a panic log message
3050 and treats the condition as false.
3052 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3054 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3055 extension "envelope-auth". The code is finished and in agreement with
3056 other implementations, but there is no documentation so far and in fact,
3057 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3058 not changing the active code.
3060 Print executed "if" and "elsif" statements when debugging is used. This
3061 helps a great deal to understand what a filter does.
3063 Document more things not specified clearly in RFC3028. I had all this
3064 sorted out, when out of a sudden new issues came to my mind. Oops."
3066 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3069 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3070 canonical form (as documented). However, after a host name lookup from
3071 the IP address, check_host() was doing a simple string comparison with
3072 addresses acquired from the DNS when checking that the found name did
3073 have the original IP as one of its addresses. Since any found IPv6
3074 addresses are likely to be in abbreviated form, the comparison could
3075 fail. Luckily, there already exists a function for doing the comparison
3076 by converting both addresses to binary, so now that is used instead of
3077 the text comparison.
3079 PH/20 There was another similar case to PH/19, when a complete host name was
3080 given in a host list; looking up its IP address could give an abbreviated
3081 form, whereas the current host's name might or might not be abbreviated.
3082 The same fix has been applied.
3088 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3089 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3092 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3093 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3095 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3097 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3098 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3099 Also, for "Auto-Submitted", treat anything other than "no" as
3100 non-personal, in accordance with RFC 3834. (Previously it treated
3101 anything starting "auto-" as non-personal.)
3103 TF/01 The control=submission/name=... option had a problem with syntax
3104 errors if the name included a slash character. The /name= option
3105 now slurps the rest of the string, so it can include any characters
3106 but it must come last in the list of options (after /sender_retain
3109 PH/05 Some modifications to the interface to the fake nameserver for the new
3117 TK/01 Added the "success_on_redirect" address verification option. See
3118 NewStuff for rationale and an example.
3120 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3122 PH/02 Patch to exigrep to allow it to work on syslog lines.
3124 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3125 fread() to skip over the body file's header line, because in Cygwin the
3126 header line is locked and is inaccessible.
3128 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3129 co-exist for some time) to make it clear that it is the Exim ID that is
3130 referenced, not the Message-ID: header line.
3132 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3133 string_format() function, because snprintf() does not exist on all
3136 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3137 ${stat: expansion item.
3139 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3140 protocol synchronization error", to keep the pedants happy.
3142 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3143 well as for IRIX systems, when gcc is being used. See the host.c source
3146 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3148 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3151 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3152 transport and to the smtp transport in LMTP mode.
3154 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3156 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3157 run a filter in a subprocess. This could lead to confusion in subsequent
3158 lookups in the parent process. There should also be a search_tidyup() at
3159 the end of the subprocess.
3161 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3162 only if the host matched helo_try_verify_hosts, which caused the
3163 verification to occur when the EHLO/HELO command was issued. The ACL just
3164 tested the remembered result. Now, if a previous verification attempt has
3165 not happened, "verify = helo" does it there and then.
3167 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3169 TK/03 Fix log output including CR from clamd.
3171 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3172 exist provoked a memory error which could cause a segfault.
3174 PH/15 Installed PCRE 6.2
3176 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3178 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3179 of the problem. Specifically, suggested +O2 rather than +O1 for the
3182 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3184 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3185 record, the port was not correctly logged when the outgoing_port log
3186 selector was set (it logged the transort's default port).
3188 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3189 fallback_hosts, and "hosts" in the smtp transport.
3191 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3192 host errors such as "Connection refused".
3194 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3195 authentication with radiusclient 0.4.9:
3197 - Error returned from rc_read_config was caught wrongly
3198 - Username/password not passed on to radius server due to wrong length.
3200 The presumption is that some radiusclient API changes for 4.51/PH/17
3201 were not taken care of correctly. The code is still untested by me (my
3202 Linux distribution still has 0.3.2 of radiusclient), but it was
3203 contributed by a Radius user.
3205 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3206 expanding the "port" option of the smtp transport.
3208 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3209 while reading a MIME header. Thanks to Tom Hughes for a patch.
3211 PH/24 Include config.h inside local_scan.h so that configuration settings are
3214 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3215 that in an ACL for RCPT (for example), you can examine exactly what was
3218 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3219 commands, but it was not correctly comparing the address with the actual
3220 client host address. Thus, it would show the EHLO address in Received:
3221 header lines when this was not necessary.
3223 PH/27 Added the % operator to ${eval:}.
3225 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3226 it should be ignoring failures (because with -C, for example, it has lost
3227 privilege). It wasn't ignoring creation failures other than "already
3230 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3233 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3234 because an input line was too long, either on its own, or by virtue of
3235 too many continuations, the temporary file was not being removed, and the
3236 return code was incorrect.
3238 PH/31 Missing "BOOL" in function definition in filtertest.c.
3240 PH/32 Applied Sieve patches from the maintainer.
3242 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3244 PH/33 Added "verify = not_blind".
3246 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3247 Local/Makefile (with some defaults set). These are used in built scripts
3248 such as exicyclog, but they have never been used in the exim_install
3249 script (though there are many overriding facilities there). I have
3250 arranged that the exim_install script now takes note of these two
3253 PH/35 Installed configuration files for Dragonfly.
3255 PH/36 When a locally submitted message by a trusted user did not contain a
3256 From: header, and the sender address was obtained from -f or from an SMTP
3257 MAIL command, and the trusted user did not use -F to supply a sender
3258 name, $originator_name was incorrectly used when constructing a From:
3259 header. Furthermore, $originator_name was used for submission mode
3260 messages from external hosts without From: headers in a similar way,
3261 which is clearly wrong.
3263 PH/37 Added control=suppress_local_fixups.
3265 PH/38 When log_selector = +received_sender was set, and the addition of the
3266 sender made the log line's construction buffer exactly full, or one byte
3267 less than full, an overflow happened when the terminating "\n" was
3270 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3271 when the result of a list match is failure because a DNS lookup failed.
3273 PH/40 RM_COMMAND is now used in the building process.
3275 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3276 the "build-* directories that it finds.
3278 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3279 address in a domain literal was a prefix of an interface address.
3281 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3282 when verifying a sender address, unless rewrite_headers is false.
3284 PH/44 Wrote a long comment about why errors_to addresses are verified as
3285 recipients, not senders.
3287 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3288 the ratelimit ACL was added.
3290 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3292 PH/46 Added extra information about PostgreSQL errors to the error string.
3294 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3295 avoiding the need to install special zones in a real server. This is
3296 backwards compatible; if it can't find the fake resolver, it drops back.
3297 Thus, both old and new test suites can be run.
3299 TF/02 Added util/ratelimit.pl
3301 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3302 clock is set back in time.
3304 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3305 Candler <B.Candler@pobox.com>.
3307 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3308 OK for addresses that are the result of redirections.
3310 PH/48 A number of further additions for the benefit of the new test suite,
3311 including a fake gethostbyname() that interfaces to the fake DNS resolver
3314 TF/06 The fix for widen_domains has also been applied to qualify_single and
3315 search_parents which are the other dnslookup options that can cause
3318 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3321 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3323 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3324 service" if it encountered an MX record with an empty target hostname.
3325 The message is now "an MX or SRV record indicated no SMTP service".
3327 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3328 if the DNS of the sending site is misconfigured. This is quite a
3329 common situation. This change restores the behaviour of treating a
3330 helo verification defer as a failure.
3332 PH/51 If self=fail was set on a router, the bounce message did not include the
3333 actual error message.
3339 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3341 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3342 command itself ended in error, the underlying message about the transport
3343 filter timeout was being overwritten with the pipe command error. Now the
3344 underlying error message should be appended to the second error message.
3346 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3348 PH/02 Reduce the amount of output that "make" produces by default. Full output
3349 can still be requested.
3351 PH/03 The warning log line about a condition test deferring for a "warn" verb
3352 was being output only once per connection, rather than after each
3353 occurrence (because it was using the same function as for successful
3354 "warn" verbs). This seems wrong, so I have changed it.
3356 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3357 it should not have, which might have caused a crash in the right
3358 circumstances, but probably never did.
3360 PH/04 Installed a modified version of Tony Finch's patch to make submission
3361 mode fix the return path as well as the Sender: header line, and to
3362 add a /name= option so that you can make the user's friendly name appear
3365 TF/03 Added the control = fakedefer ACL modifier.
3367 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3368 Mark Lowes for thorough testing.
3370 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3372 TK/03 Merged latest SRS patch from Miles Wilton.
3374 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3375 with the definition in sysexits.h (which is #included earlier).
3376 Fortunately, Exim does not actually use EX_OK. The code used to try to
3377 preserve the sysexits.h value, by assumimg that macro definitions were
3378 scanned for macro replacements. I have been disabused of this notion,
3379 so now the code just undefines EX_OK before #including unistd.h.
3381 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3382 in the smtp transport. When a block could not be written in a single
3383 write() function, the timeout was being re-applied to each part-write.
3384 This seems wrong - if the receiver was accepting one byte at a time it
3385 would take for ever. The timeout is now adjusted when this happens. It
3386 doesn't have to be particularly precise.
3388 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3389 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3391 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3392 without a domain if the check to <postmaster@domain> fails.
3394 SC/01 Eximstats: added -xls and the ability to specify output files
3395 (patch written by Frank Heydlauf).
3397 SC/02 Eximstats: use FileHandles for outputing results.
3399 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3401 SC/04 Eximstats: fixed display of large numbers with -nvr option
3403 SC/05 Eximstats: fixed merging of reports with empty tables.
3405 SC/06 Eximstats: added the -include_original_destination flag
3407 SC/07 Eximstats: removed tabs and trailing whitespace.
3409 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3411 TK/06 MBOX spool code: Add real "From " MBOX separator line
3412 so the .eml file is really in mbox format (even though
3413 most programs do not really care). Patch from Alex Miller.
3415 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3416 The latter is generated from $received_to and is only set if the
3417 message has one envelope recipient. SA can use these headers,
3418 obviously out-of-the-box. Patch from Alex Miller.
3420 PH/08 The ${def test on a variable was returning false if the variable's
3421 value was "0", contrary to what the specification has always said!
3422 The result should be true unless the variable is empty.
3424 PH/09 The syntax error of a character other than { following "${if
3425 def:variable_name" (after optional whitespace) was not being diagnosed.
3426 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3427 accidental colon was present, for example, could give incorrect results.
3429 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3430 result is used (not including appendfile, where other changes are about
3433 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3434 This involved changing a lot of size variables from int to off_t. It
3435 should work with maildirs and everything.
3437 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3438 spamd dying while we are connected to it.
3440 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3443 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3444 function for BDB 4.3.
3446 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3448 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3449 the sources for PCRE have been re-arranged and more files are now
3452 PH/15 The code I had for printing potentially long long variables in PH/11
3453 above was not the best (it lost precision). The length of off_t variables
3454 is now inspected at build time, and an appropriate printing format (%ld
3455 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3456 to be "long long int" or "long int". This is needed for the internal
3457 formatting function string_vformat().
3459 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3460 the configuration file to be ":syslog", then the script "guesses" where
3461 the logs files are, rather than using the compiled in default. In our
3462 case the guess is not the same as the compiled default, so the script
3463 suddenly stopped working when I started to use syslog. The patch checks
3464 to see if log_file_path is "". If so, it attempts to read it from exim
3465 with no configuration file to get the compiled in version, before it
3466 falls back to the previous guessing code."
3468 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3469 implementing BATV in an Exim configuration. See NewStuff for the gory
3472 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3473 Makefile that are specific to HP-UX.
3475 PH/18 If the "use_postmaster" option was set for a recipient callout together
3476 with the "random" option, the postmaster address was used as the MAIL
3477 FROM address for the random test, but not for the subsequent recipient
3478 test. It is now used for both.
3480 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3481 patch removes a few documentation additions to RFC 3028, because the
3482 latest draft now contains them. It adds the new en;ascii-case comparator
3483 and a new error check for 8bit text in MIME parts. Comparator and
3484 require names are now matched exactly. I enabled the subaddress
3485 extension, but it is not well tested yet (read: it works for me)."
3487 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3488 rework some of the code of TK/09 above to avoid the hardwired use of
3489 "%lld" and "long long". Replaced the call to snprintf() with a call to
3492 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3493 records point to non-existent hosts", "retry timeout exceeded", and
3494 "retry time not reached for any host after a long failure period".
3496 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3497 experimental DomainKeys support:
3499 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3500 (2) On an error such as an illegally used "control", the wrong name for
3501 the control was given.
3503 These problems did NOT occur unless DomainKeys support was compiled.
3505 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3507 PH/24 Added ${if match_ip condition.
3509 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3510 databases so that it will be absolutely obvious if a crash occurs in the
3511 DB library. This is a regular occurrence (often caused by mis-matched
3514 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3515 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3516 picked up on a user's system that detects such things. There doesn't seem
3517 to be a gcc warning option for this - only an attribute that has to be
3518 put on the function's prototype. It seems that in Fedora Core 4 they have
3519 set this on a number of new functions. No doubt there will be more in due
3522 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3523 specify a transport. However, if an address that was verified by such a
3524 router was the subject of a callout, Exim crashed because it tried to
3525 read the rcpt_include_affixes from the non-existent transport. Now it
3526 just assumes that the setting of that option is false. This bug was
3527 introduced by 4.51/PH/31.
3529 PH/28 Changed -d+all to exclude +memory, because that information is very
3530 rarely of interest, but it makes the output a lot bigger. People tend to
3531 do -d+all out of habit.
3533 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3534 code in os-type was giving problems when libc.so lives in lib64, like on
3537 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3538 aren't the modern standard, and it seems that some systems' include files
3539 don't always have them. Exim was already checking for some of the newer
3540 ones like T_AAAA, and defining it itself. I've added checks for all the
3541 record types that Exim uses.
3543 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3544 not automatically generating a new one, as it is supposed to. This
3545 prevented TLS from working. If the file did exist, but contained invalid
3546 data, a new version was generated, as expected. It was only the case of a
3547 non-existent file that was broken.
3549 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3550 with a change in libdomainkeys > 0.64.
3552 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3553 from DNS. If the selector record carries the flag, it now has
3554 precedence over the domain-wide flag.
3556 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3558 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3559 the use of a port name that isn't defined in /etc/services) occurred, the
3560 message was deferred as in a normal delivery, and thus remained on the
3561 spool, instead of being failed because of the mua_wrapper setting. This
3562 is now fixed, and I tidied up some of the mua_wrapper messages at the
3565 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3566 the messages in a hash of arrays rather than using individual hashes.
3567 This is a bit cleaner and results in dramatic memory savings, albeit
3568 at a slight CPU cost.
3570 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3571 as requested by Marc Sherman.
3573 SC/10 Eximstats: added histograms for user specified patterns as requested
3576 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3578 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3579 fopen() in the content-scanning modules that did not already have it.
3585 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3586 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3588 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3590 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3592 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3593 to test Sieve filters that use "vacation".
3595 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3596 that changes the way the GnuTLS parameters are stored in the cache file.
3597 The new format can be generated externally. For backward compatibility,
3598 if the data in the cache doesn't make sense, Exim assumes it has read an
3599 old-format file, and it generates new data and writes a new file. This
3600 means that you can't go back to an older release without removing the
3603 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3604 work if there are any delivery delays because "one_time" forces the
3605 parent to be marked "delivered", so its unseen clone is never tried
3606 again. For this reason, Exim now forbids the simultaneous setting of
3609 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3610 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3611 entry if you want to know the details. The fix, however, neglected to
3612 consider the case when local delivery batching is involved. The test for
3613 "previously delivered" was not happening when checking to see if an
3614 address could be batched with a previous (undelivered) one; under
3615 certain circumstances this could lead to multiple deliveries to the same
3618 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3619 in its include files, and this causes problems building Exim.
3621 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3622 header_syntax) but Exim was just ignoring anything given after a slash.
3623 In particular, this caused confusion with an attempt to use "verify =
3624 reverse_host_lookup/defer_ok". An error is now given when options are
3625 supplied for verify items that do not have them. (Maybe reverse_host_
3626 lookup should have a defer_ok option, but that's a different point.)
3628 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3629 defined by RFC 821) to 2048, because there were problems with some AUTH
3630 commands, and RFC 1869 says the size should be increased for extended
3631 SMTP commands that take arguments.
3633 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3636 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3637 "unknown" error; now it says that the functionality isn't in the binary.
3639 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3640 an address' error message when a string expansion fails (syntax or
3641 whatever). Otherwise the password may appear in the log. Following change
3642 PH/42 below, there is no longer a chance of it appearing in a bounce
3645 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3647 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3648 of its addresses was ever tried. (Bugzilla bug #2.)
3650 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3651 the result incorrectly in the debug output. (It correctly added a newline
3652 to what was transported.)
3654 TF/01 Added $received_time.
3656 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3657 commented out examples of how to interface to a virus scanner and to
3658 SpamAssassin. Also added commented examples of av_scanner and
3659 spamd_address settings.
3661 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3662 and controls are allowed in which ACLs. There were a couple of minor
3663 errors. Some of the entries in the conditions table (which is a table of
3664 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3665 negation of where the condition IS allowed.
3667 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3669 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3670 header file does not have a version number, so I've had to invent a new
3671 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3672 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3673 radiusclient), but it was contributed by a Radius user.
3675 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3676 files or directories, for OpenSSL.
3678 PH/19 When an Exim process that is running as root has to create an Exim log
3679 file, it does so in a subprocess that runs as exim:exim so as to get the
3680 ownership right at creation (otherwise, other Exim processes might see
3681 the file with the wrong ownership). There was no test for failure of this
3682 fork() call, which would lead to the process getting stuck as it waited
3683 for a non-existent subprocess. Forks do occasionally fail when resources
3684 run out. I reviewed all the other calls to fork(); they all seem to check
3687 PH/20 When checking for unexpected SMTP input at connect time (before writing
3688 the banner), Exim was not dealing correctly with a non-positive return
3689 from the read() function. If the client had disconnected by this time,
3690 the result was a log entry for a synchronization error with an empty
3691 string after "input=" when read() returned zero. If read() returned -1
3692 (an event I could not check), uninitialized data bytes were printed.
3693 There were reports of junk text (parts of files, etc) appearing after
3696 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3698 PH/22 Added support for macro redefinition, and (re)definition in between
3699 driver and ACL definitions.
3701 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3702 forgetting to use the resulting value; it was using the unexpanded value.
3704 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3705 hadn't been configured. The fix is from Juergen Kreileder, who
3706 understands it better than I do:
3708 "Here's what I see happening with three configured cyrus_sasl
3709 authenticators configured (plain, login, cram-md5):
3711 On startup auth_cyrus_sasl_init() gets called for each of these.
3712 This means three calls to sasl_listmech() without a specified mech_list.
3713 => SASL tests which mechs of all available mechs actually work
3714 => three warnings about OTP not working
3715 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3717 With the patch, sasl_listmech() also gets called three times. But now
3718 SASL's mech_list option is set to the server_mech specified in the the
3719 authenticator. Or in other words, the answer from sasl_listmech()
3720 gets limited to just the mech you're testing for (which is different
3722 => the return list contains just 'plain' or 'login', 'cram-md5' or
3723 nothing depending on the value of ob->server_mech.
3725 I've just tested the patch: Authentication still works fine,
3726 unavailable mechs specified in the exim configuration are still
3727 caught, and the auth.log warnings about OTP are gone."
3729 PH/25 When debugging is enabled, the contents of the command line are added
3730 to the debugging output, even when log_selector=+arguments is not
3733 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3734 answer is "GNU", and only if the return is "GNU/something" is the answer
3737 PH/27 $acl_verify_message is now set immediately after the failure of a
3738 verification in an ACL, and so is available in subsequent modifiers. In
3739 particular, the message can be preserved by coding like this:
3741 warn !verify = sender
3742 set acl_m0 = $acl_verify_message
3744 Previously, $acl_verify_message was set only while expanding "message"
3745 and "log_message" when a very denied access.
3747 PH/28 Modified OS/os.c-Linux with
3749 -#ifndef OS_LOAD_AVERAGE
3750 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3752 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3753 nomenclature these days.)
3755 PH/29 Installed patch from the Sieve maintainer that adds the options
3756 sieve_useraddress and sieve_subaddress to the redirect router.
3758 PH/30 In these circumstances:
3759 . Two addresses routed to the same list of hosts;
3760 . First host does not offer TLS;
3761 . First host accepts first address;
3762 . First host gives temporary error to second address;
3763 . Second host offers TLS and a TLS session is established;
3764 . Second host accepts second address.
3765 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3766 and peerdn, if requested) that were in fact used only for the second
3769 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3770 attention to any local part prefix or suffix that was matched by the
3771 router that accepted the address. It now behaves in the same way as it
3772 does for delivery: the affixes are removed from the local part unless
3773 rcpt_include_affixes is set on the transport.
3775 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3776 timeout during the DATA phase of an incoming message.
3778 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3779 applied a patch sanctioned by the Sieve maintainer.
3781 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3782 the uid or gid is negative. A case of a negative gid caused this to be
3783 noticed. The fix allows for either to be negative.
3785 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3786 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3789 PH/36 The change PH/12 above was broken. Fixed it.
3791 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3792 the grounds that routing the same address twice would always produce the
3793 same answer. This might have been true once, but it is certainly no
3794 longer true now. Routing a child address may depend on the previous
3795 routing that produced that child. Some complicated redirection strategies
3796 went wrong when messages had multiple recipients, and made Exim's
3797 behaviour dependent on the order in which the addresses were given.
3799 I have moved the duplicate checking until after the routing is complete.
3800 Exim scans the addresses that are assigned to local and remote
3801 transports, and removes any duplicates. This means that more work will be
3802 done, as duplicates will always all be routed, but duplicates are
3803 presumably rare, so I don't expect this is of any significance.
3805 For deliveries to pipes, files, and autoreplies, the duplicate checking
3806 still happens during the routing process, since they are not going to be
3809 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3810 It corrects a timeout issue with spamd. This is Ian's comment: "The
3811 background is that sometimes spamd either never reads data from a
3812 connection it has accepted, or it never writes response data. The exiscan
3813 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3814 blindly assumes that writes won't block so it may never time out."
3816 PH/39 Allow G after quota size as well as K and M.
3818 PH/40 The value set for $authenticated_id in an authenticator may not contain
3819 binary zeroes or newlines because the value is written to log lines and
3820 to spool files. There was no check on this. Now the value is run through
3821 the string_printing() function so that such characters are converted to
3822 printable escape sequences.
3824 PH/41 $message_linecount is a new variable that contains the total number of
3825 lines in the message. Compare $body_linecount, which is the count for the
3828 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3829 bounce and delay warning messages, except in certain special cases, which
3832 (a) An SMTP error message from a remote host;
3833 (b) A message specified in a :fail: redirection;
3834 (c) A message specified in a "fail" command in a system filter;
3835 (d) A message specified in a FAIL return from the queryprogram router;
3836 (e) A message specified by the cannot_route_message router option.
3838 In these cases only, Exim does include the error details in bounce and
3839 warning messages. There are also a few cases where bland messages such
3840 as "unrouteable address" or "local delivery error" are given.
3842 PH/43 $value is now also set for the "else" part of a ${run expansion.
3844 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3845 being worked on, but at least Exim now implements the latest version to
3848 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3849 process to complete was treated as a delivery failure, a timeout while
3850 writing the message to the pipe was logged, but erroneously treated as a
3851 successful delivery. Such timeouts include transport filter timeouts. For
3852 consistency with the overall process timeout, these timeouts are now
3853 treated as errors, giving rise to delivery failures by default. However,
3854 there is now a new Boolean option for the pipe transport called
3855 timeout_defer, which, if set TRUE, converts the failures into defers for
3856 both kinds of timeout. A transport filter timeout is now identified in
3859 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3860 systems where "make" and "gmake" are different, calling "gmake" at top
3861 level broke things. I've arranged for the value of $(MAKE) to be passed
3862 from the Makefile to this script so that it can call the same version of
3866 A note about Exim versions 4.44 and 4.50
3867 ----------------------------------------
3869 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3870 changes of various kinds. As a consequence, a big documentation update was
3871 needed. This delayed the release for rather longer than seemed good, especially
3872 in the light of a couple of (minor) security issues. Therefore, the changes
3873 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3874 release. So 4.44 and 4.50 are in effect two different branches that both start
3877 I have left the 4.50 change log unchanged; it contains all the changes since
3878 4.43. The change log for 4.44 is below; many of its items are identical to
3879 those for 4.50. This seems to be the most sensible way to preserve the
3880 historical information.
3886 1. Minor wording change to the doc/README.SIEVE file.
3888 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3889 computation of the current number of files was incorrect.
3891 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3892 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3895 4. Give more explanation in the error message when the command for a transport
3896 filter fails to execute.
3898 5. There are several places where Exim runs a non-Exim command in a
3899 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3900 was being done only for the command run by the queryprogram router. It is
3901 now done for all such subprocesses. The other cases are: ${run, transport
3902 filters, and the commands run by the lmtp and pipe transports.
3904 6. Added CONFIGURE_GROUP build-time option.
3906 7. Some older OS have a limit of 256 on the maximum number of file
3907 descriptors. Exim was using setrlimit() to set 1000 as a large value
3908 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3909 systems. I've change it so that if it can't get 1000, it tries for 256.
3911 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3912 was an oversight, and furthermore, ever since the addition of extra
3913 controls (e.g. 4.43/32), the checks on when to allow different forms of
3914 "control" were broken. There should now be diagnostics for all cases when a
3915 control that does not make sense is encountered.
3917 9. Added the /retain_sender option to "control=submission".
3919 10. $recipients is now available in the predata ACL (oversight).
3921 11. Tidy the search cache before the fork to do a delivery from a message
3922 received from the command line. Otherwise the child will trigger a lookup
3923 failure and thereby defer the delivery if it tries to use (for example) a
3924 cached ldap connection that the parent has called unbind on.
3926 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3927 of $address_data from the recipient verification was clobbered by the
3928 sender verification.
3930 13. The value of address_data from a sender verification is now available in
3931 $sender_address_data in subsequent conditions in the ACL statement.
3933 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3935 15. Added a new option "connect=<time>" to callout options, to set a different
3938 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3939 was its contents. (It was OK if the option was not defined at all.)
3941 17. A "Completed" log line is now written for messages that are removed from
3942 the spool by the -Mrm option.
3944 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3945 information about exactly what failed.
3947 19. Added -dd to debug only the daemon process.
3949 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3950 handles timeouts, both on the server side and network timeouts. Renamed the
3951 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3953 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3954 It is now set to "smtps".
3956 22. $host_address is now set to the target address during the checking of
3957 ignore_target_hosts.
3959 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3960 being passed; this would have caused $sender_host_name to have been used if
3961 matching the list had actually called for a host name (not very likely,
3962 since this list is usually IP addresses). A host name is now passed as
3965 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3966 code that specifies a non-restarting handler (typically sigaction() in
3967 modern systems) in an attempt to fix a rare and obscure crash bug.
3969 25. Narrowed the window for a race in the daemon that could cause it to ignore
3970 SIGCHLD signals. This is not a major problem, because they are used only to
3971 wake it up if nothing else does.
3973 26. A malformed maildirsize file could cause Exim to calculate negative values
3974 for the mailbox size or file count. Odd effects could occur as a result.
3975 The maildirsize information is now recalculated if the size or filecount
3978 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3979 support for a long time. Removed HAVE_SYS_VFS_H.
3981 28. Installed the latest version of exipick from John Jetmore.
3983 29. In an address list, if the pattern was not a regular expression, an empty
3984 subject address (from a bounce message) matched only if the pattern was an
3985 empty string. Non-empty patterns were not even tested. This was the wrong
3986 because it is perfectly reasonable to use an empty address as part of a
3987 database query. An empty address is now tested by patterns that are
3988 lookups. However, all the other forms of pattern expect the subject to
3989 contain a local part and a domain, and therefore, for them, an empty
3990 address still always fails if the pattern is not itself empty.
3992 30. Exim went into a mad DNS loop when attempting to do a callout where the
3993 host was specified on an smtp transport, and looking it up yielded more
3994 than one IP address.
3996 31. Re-factored the code for checking spool and log partition space into a
3997 function that finds that data and another that does the check. The former
3998 is then used to implement four new variables: $spool_space, $log_space,
3999 $spool_inodes, and $log_inodes.
4001 32. The RFC2047 encoding function was originally intended for short strings
4002 such as real names; it was not keeping to the 75-character limit for
4003 encoded words that the RFC imposes. It now respects the limit, and
4004 generates multiple encoded words if necessary. To be on the safe side, I
4005 have increased the buffer size for the ${rfc2047: expansion operator from
4008 33. It is now permitted to omit both strings after an "if" condition; if the
4009 condition is true, the result is "true". As before, when the second string
4010 is omitted, a false condition yields an empty string. This makes it less
4011 cumbersome to write custom ACL and router conditions.
4013 34. Failure to deliver a bounce message always caused it to be frozen, even if
4014 there was an errors_to setting on the router. The errors_to setting is now
4017 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4018 canonical form (fully expanded) before being placed in
4019 $sender_host_address.
4021 36. The table in the code that translates DNS record types into text (T_A to
4022 "A" for instance) was missing entries for NS and CNAME. It is just possible
4023 that this could have caused confusion if both these types were looked up
4024 for the same domain, because the text type is used as part of Exim's
4025 per-process caching. But the chance of anyone hitting this buglet seems
4028 37. The dnsdb lookup has been extended in a number of ways.
4030 (1) There is a new type, "zns", which walks up the domain tree until it
4031 finds some nameserver records. It should be used with care.
4033 (2) There is a new type, "mxh", which is like "mx" except that it returns
4034 just the host names, not the priorities.
4036 (3) It is now possible to give a list of domains (or IP addresses) to be
4037 looked up. The behaviour when one of the lookups defers can be
4038 controlled by a keyword.
4040 (4) It is now possible to specify the separator character for use when
4041 multiple records are returned.
4043 38. The dnslists ACL condition has been extended: it is now possible to supply
4044 a list of IP addresses and/or domains to be looked up in a particular DNS
4047 39. Added log_selector=+queue_time_overall.
4049 40. When running the queue in the test harness, wait just a tad after forking a
4050 delivery process, to get repeatability of debugging output.
4052 41. Include certificate and key file names in error message when GnuTLS fails
4053 to set them up, because the GnuTLS error message doesn't include the name
4054 of the failing file when there is a problem reading it.
4056 42. Allow both -bf and -bF in the same test run.
4058 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4060 44. The "Exiscan patch" is now merged into the mainline Exim source.
4062 45. Sometimes the final signoff response after QUIT could fail to get
4063 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4064 < 0 before doing a fflush(). This bug looks as though it goes back to the
4065 introduction of TLS in release 3.20, but "sometimes" must have been rare
4066 because the tests only now provoked it.
4068 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4069 (this can affect the format of dates).
4071 47. exim_tidydb, when checking for the continued existence of a message for
4072 which it has found a message-specific retry record, was not finding
4073 messages that were in split spool directories. Consequently, it was
4074 deleting retry records that should have stayed in existence.
4076 48. Steve fixed some bugs in eximstats.
4078 49. The SPA authentication driver was not abandoning authentication and moving
4079 on to the next authenticator when an expansion was forced to fail,
4080 contradicting the general specification for all authenticators. Instead it
4081 was generating a temporary error. It now behaves as specified.
4083 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4084 (the order specifies the preference for clients). The order is now AES256,
4085 AES128, 3DES, ARCFOUR128.
4087 51. Small patch to Sieve code - explicitly set From: when generating an
4090 52. Exim crashed if a remote delivery caused a very long error message to be
4091 recorded - for instance if somebody sent an entire SpamAssassin report back
4092 as a large number of 550 error lines. This bug was coincidentally fixed by
4093 increasing the size of one of Exim's internal buffers (big_buffer) that
4094 happened as part of the Exiscan merge. However, to be on the safe side, I
4095 have made the code more robust (and fixed the comments that describe what
4098 53. Now that there can be additional text after "Completed" in log lines (if
4099 the queue_time_overall log selector is set), a one-byte patch to exigrep
4100 was needed to allow it to recognize "Completed" as not the last thing in
4103 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4104 patch that reportedly fixes this has been added. I am not expert enough to
4105 create a test for it. This is what the patch creator wrote:
4107 "I found a little strange behaviour of ldap code when working with
4108 Windows 2003 AD Domain, where users was placed in more than one
4109 Organization Units. When I tried to give exim partial DN, the exit code
4110 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4111 But simultaneously result of request was absolutely normal ldap result,
4112 so I produce this patch..."
4114 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4115 REFERENCE, so I have modified the code to exclude the patch when that macro
4118 55. Some experimental protocols are using DNS PTR records for new purposes. The
4119 keys for these records are domain names, not reversed IP addresses. The
4120 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4121 leaves it alone. Component reversal etc. now happens only for IP addresses.
4124 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4126 57. Double the size of the debug message buffer (to 2048) so that more of very
4127 long debug lines gets shown.
4129 58. The exicyclog utility now does better if the number of log files to keep
4130 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4132 59. Two changes related to the smtp_active_hostname option:
4134 (1) $smtp_active_hostname is now available as a variable.
4135 (2) The default for smtp_banner uses $smtp_active_hostname instead
4136 of $primary_hostname.
4138 60. The host_aton() function is supposed to be passed a string that is known
4139 to be a valid IP address. However, in the case of IPv6 addresses, it was
4140 not checking this. This is a hostage to fortune. Exim now panics and dies
4141 if the condition is not met. A case was found where this could be provoked
4142 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4143 components; fortuitously, this particular loophole had already been fixed
4144 by change 4.50/55 above.
4146 If there are any other similar loopholes, the new check in host_aton()
4147 itself should stop them being exploited. The report I received stated that
4148 data on the command line could provoke the exploit when Exim was running as
4149 exim, but did not say which command line option was involved. All I could
4150 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4151 running as the user.
4154 61. There was a buffer overflow vulnerability in the SPA authentication code
4155 (which came originally from the Samba project). I have added a test to the
4156 spa_base64_to_bits() function which I hope fixes it.
4159 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4160 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4162 63. The daemon start-up calls getloadavg() while still root for those OS that
4163 need the first call to be done as root, but it missed one case: when
4164 deliver_queue_load_max is set with deliver_drop_privilege. This is
4165 necessary for the benefit of the queue runner, because there is no re-exec
4166 when deliver_drop_privilege is set.
4168 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4169 This has been fixed.
4171 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4172 was in use, was not putting the data itself into the right store pool;
4173 consequently, it could be overwritten for a subsequent message in the same
4174 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4177 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4179 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4180 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4181 the function were treating the return as a boolean value, which happened to
4182 work because 0=false and not-0=true, but is not correct code.
4184 68. The host_aton() function was not handling scoped IPv6 addresses (those
4185 with, for example, "%eth0" on the end) correctly.
4187 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4188 negated items (that is, ~something) in unsigned ints. Some compilers
4189 apparently mutter when there is no cast.
4191 70. If an address verification called from an ACL failed, and did not produce a
4192 user-specific message (i.e. there was only a "system" message), nothing was
4193 put in $acl_verify_message. In this situation, it now puts the system
4196 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4197 session; change 4.31/43 added the unwanted input to the log line - except
4198 that it did not do this in the start of session case. It now does.
4200 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4201 This is wrong and can cause the other end to generate a synchronization
4202 error if it is another Exim or anything else that does the synchronization
4203 check. A QUIT command is no longer sent after a timeout.
4205 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4206 during host lookups.
4208 74. The defer_ok option of callout verification was not working if it was used
4209 when verifying addresses in header lines, that is, for this case:
4211 verify = header_sender/callout=defer_ok
4213 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4214 those file descriptors could be used for SMTP connections. If anything
4215 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4216 could be sent to the SMTP client, causing chaos. The daemon now opens
4217 stdin, stdout, and stderr to /dev/null when it puts itself into the
4220 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4221 log by default. The user can override this with suitable Perl magic.
4223 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4224 the log message when discard triggers, was not working for the DATA ACL or
4225 for the non-SMTP ACL.
4227 78. Error message wording change in sieve.c.
4229 79. If smtp_accept_max_per_host was set, the number of connections could be
4230 restricted to fewer than expected, because the daemon was trying to set up
4231 a new connection before checking whether the processes handling previous
4232 connections had finished. The check for completed processes is now done
4233 earlier. On busy systems, this bug wouldn't be noticed because something
4234 else would have woken the daemon, and it would have reaped the completed
4237 80. If a message was submitted locally by a user whose login name contained one
4238 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4239 It caused a spool format error. I have fixed the spool reading code. A
4240 related problem was that the "from" clause in the Received: line became
4241 illegal because of the space(s). It is now covered by ${quote_local_part.
4243 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4246 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4248 83. Patch from Sieve maintainer to fix unterminated string problem in
4249 "vacation" handling.
4251 84. Some minor changes to the Linux configuration files to help with other
4252 OS variants using glibc.
4254 85. One more patch for Sieve to update vacation handling to latest spec.
4257 ----------------------------------------------------
4258 See the note above about the 4.44 and 4.50 releases.
4259 ----------------------------------------------------
4265 1. Change 4.43/35 introduced a bug that caused file counts to be
4266 incorrectly computed when quota_filecount was set in an appendfile
4269 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4270 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4273 3. Give more explanation in the error message when the command for a transport
4274 filter fails to execute.
4276 4. There are several places where Exim runs a non-Exim command in a
4277 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4278 was being done only for the command run by the queryprogram router. It is
4279 now done for all such subprocesses. The other cases are: ${run, transport
4280 filters, and the commands run by the lmtp and pipe transports.
4282 5. Some older OS have a limit of 256 on the maximum number of file
4283 descriptors. Exim was using setrlimit() to set 1000 as a large value
4284 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4285 systems. I've change it so that if it can't get 1000, it tries for 256.
4287 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4288 was an oversight, and furthermore, ever since the addition of extra
4289 controls (e.g. 4.43/32), the checks on when to allow different forms of
4290 "control" were broken. There should now be diagnostics for all cases when a
4291 control that does not make sense is encountered.
4293 7. $recipients is now available in the predata ACL (oversight).
4295 8. Tidy the search cache before the fork to do a delivery from a message
4296 received from the command line. Otherwise the child will trigger a lookup
4297 failure and thereby defer the delivery if it tries to use (for example) a
4298 cached ldap connection that the parent has called unbind on.
4300 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4301 of $address_data from the recipient verification was clobbered by the
4302 sender verification.
4304 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4305 was its contents. (It was OK if the option was not defined at all.)
4307 11. A "Completed" log line is now written for messages that are removed from
4308 the spool by the -Mrm option.
4310 12. $host_address is now set to the target address during the checking of
4311 ignore_target_hosts.
4313 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4314 being passed; this would have caused $sender_host_name to have been used if
4315 matching the list had actually called for a host name (not very likely,
4316 since this list is usually IP addresses). A host name is now passed as
4319 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4320 code that specifies a non-restarting handler (typically sigaction() in
4321 modern systems) in an attempt to fix a rare and obscure crash bug.
4323 15. Narrowed the window for a race in the daemon that could cause it to ignore
4324 SIGCHLD signals. This is not a major problem, because they are used only to
4325 wake it up if nothing else does.
4327 16. A malformed maildirsize file could cause Exim to calculate negative values
4328 for the mailbox size or file count. Odd effects could occur as a result.
4329 The maildirsize information is now recalculated if the size or filecount
4332 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4333 support for a long time. Removed HAVE_SYS_VFS_H.
4335 18. Updated exipick to current release from John Jetmore.
4337 19. Allow an empty sender to be matched against a lookup in an address list.
4338 Previously the only cases considered were a regular expression, or an
4341 20. Exim went into a mad DNS lookup loop when doing a callout where the
4342 host was specified on the transport, if the DNS lookup yielded more than
4345 21. The RFC2047 encoding function was originally intended for short strings
4346 such as real names; it was not keeping to the 75-character limit for
4347 encoded words that the RFC imposes. It now respects the limit, and
4348 generates multiple encoded words if necessary. To be on the safe side, I
4349 have increased the buffer size for the ${rfc2047: expansion operator from
4352 22. Failure to deliver a bounce message always caused it to be frozen, even if
4353 there was an errors_to setting on the router. The errors_to setting is now
4356 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4357 canonical form (fully expanded) before being placed in
4358 $sender_host_address.
4360 24. Updated eximstats to version 1.33
4362 25. Include certificate and key file names in error message when GnuTLS fails
4363 to set them up, because the GnuTLS error message doesn't include the name
4364 of the failing file when there is a problem reading it.
4366 26. Expand error message when OpenSSL has problems setting up cert/key files.
4369 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4370 (this can affect the format of dates).
4372 28. exim_tidydb, when checking for the continued existence of a message for
4373 which it has found a message-specific retry record, was not finding
4374 messages that were in split spool directories. Consequently, it was
4375 deleting retry records that should have stayed in existence.
4377 29. eximstats updated to version 1.35
4378 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4379 1.35 - bugfix such that pie charts by volume are generated correctly
4381 30. The SPA authentication driver was not abandoning authentication and moving
4382 on to the next authenticator when an expansion was forced to fail,
4383 contradicting the general specification for all authenticators. Instead it
4384 was generating a temporary error. It now behaves as specified.
4386 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4387 (the order specifies the preference for clients). The order is now AES256,
4388 AES128, 3DES, ARCFOUR128.
4390 31. Small patch to Sieve code - explicitly set From: when generating an
4393 32. Exim crashed if a remote delivery caused a very long error message to be
4394 recorded - for instance if somebody sent an entire SpamAssassin report back
4395 as a large number of 550 error lines. This bug was coincidentally fixed by
4396 increasing the size of one of Exim's internal buffers (big_buffer) that
4397 happened as part of the Exiscan merge. However, to be on the safe side, I
4398 have made the code more robust (and fixed the comments that describe what
4401 33. Some experimental protocols are using DNS PTR records for new purposes. The
4402 keys for these records are domain names, not reversed IP addresses. The
4403 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4404 leaves it alone. Component reversal etc. now happens only for IP addresses.
4407 34. The host_aton() function is supposed to be passed a string that is known
4408 to be a valid IP address. However, in the case of IPv6 addresses, it was
4409 not checking this. This is a hostage to fortune. Exim now panics and dies
4410 if the condition is not met. A case was found where this could be provoked
4411 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4412 components; fortuitously, this particular loophole had already been fixed
4413 by change 4.50/55 or 4.44/33 above.
4415 If there are any other similar loopholes, the new check in host_aton()
4416 itself should stop them being exploited. The report I received stated that
4417 data on the command line could provoke the exploit when Exim was running as
4418 exim, but did not say which command line option was involved. All I could
4419 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4420 running as the user.
4423 35. There was a buffer overflow vulnerability in the SPA authentication code
4424 (which came originally from the Samba project). I have added a test to the
4425 spa_base64_to_bits() function which I hope fixes it.
4428 36. The daemon start-up calls getloadavg() while still root for those OS that
4429 need the first call to be done as root, but it missed one case: when
4430 deliver_queue_load_max is set with deliver_drop_privilege. This is
4431 necessary for the benefit of the queue runner, because there is no re-exec
4432 when deliver_drop_privilege is set.
4434 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4435 was in use, was not putting the data itself into the right store pool;
4436 consequently, it could be overwritten for a subsequent message in the same
4437 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4440 38. Sometimes the final signoff response after QUIT could fail to get
4441 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4442 < 0 before doing a fflush(). This bug looks as though it goes back to the
4443 introduction of TLS in release 3.20, but "sometimes" must have been rare
4444 because the tests only now provoked it.
4450 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4451 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4452 Now it returns an integer. A number of calls were still expecting a T/F
4453 return. Fortuitously, in all cases, the tests worked in OK situations,
4454 which is the norm. However, things would have gone wrong on any write
4455 failures on the smtp file descriptor. This function is used when sending
4456 messages over SMTP and also when doing verify callouts.
4458 2. When Exim is called to do synchronous delivery of a locally submitted
4459 message (the -odf or -odi options), it no longer closes stderr before doing
4462 3. Implemented the mua_wrapper option.
4464 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4466 5. Implemented the functions header_remove(), header_testname(),
4467 header_add_at_position(), and receive_remove_recipient(), and exported them
4470 6. If an ACL "warn" statement specified the addition of headers, Exim already
4471 inserted X-ACL-Warn: at the start if there was no header name. However, it
4472 was not making this test for the second and subsequent header lines if
4473 there were newlines in the string. This meant that an invalid header could
4474 be inserted if Exim was badly configured.
4476 7. Allow an ACL "warn" statement to add header lines at the start or after all
4477 the Received: headers, as well as at the end.
4479 8. Added the rcpt_4xx retry error code.
4481 9. Added postmaster_mailfrom=xxx to callout verification option.
4483 10. Added mailfrom=xxxx to the callout verification option, for verify=
4486 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4487 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4489 12. Inserted some casts to stop certain compilers warning when using pointer
4490 differences as field lengths or precisions in printf-type calls (mostly
4491 affecting debugging statements).
4493 13. Added optional readline() support for -be (dynamically loaded).
4495 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4496 same clock tick as a message's arrival, so that its received time was the
4497 same as the "first fail" time on the retry record, and that message
4498 remained on the queue past the ultimate address timeout, every queue runner
4499 would try a delivery (because it was past the ultimate address timeout) but
4500 after another failure, the ultimate address timeout, which should have then
4501 bounced the address, did not kick in. This was a "< instead of <=" error;
4502 in most cases the first failure would have been in the next clock tick
4503 after the received time, and all would be well.
4505 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4506 being recognized when the domain list was tested by the match_domain
4507 condition in an expansion string.
4509 16. Added the ${str2b64: operator.
4511 17. Exim was always calling setrlimit() to set a large limit for the number of
4512 processes, without checking whether the existing limit was already
4513 adequate. (It did check for the limit on file descriptors.) Furthermore,
4514 errors from getrlimit() and setrlimit() were being ignored. Now they are
4515 logged to the main and panic logs, but Exim does carry on, to try to do its
4516 job under whatever limits there are.
4518 18. Imported PCRE 5.0.
4520 19. Trivial typo in log message " temporarily refused connection" (the leading
4523 20. If the log selector return_path_on_delivery was set and an address was
4524 redirected to /dev/null, the delivery process crashed because it assumed
4525 that a return path would always be set for a "successful" delivery. In this
4526 case, the whole delivery is bypassed as an optimization, and therefore no
4529 21. Internal re-arrangement: the function for sending a challenge and reading
4530 a response while authentication was assuming a zero-terminated challenge
4531 string. It's now changed to take a pointer and a length, to allow for
4532 binary data in such strings.
4534 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4536 23. Exim was not respecting finduser_retries when seeking the login of the
4537 uid under which it was called; it was always trying 10 times. (The default
4538 setting of finduser_retries is zero.) Also, it was sleeping after the final
4539 failure, which is pointless.
4541 24. Implemented tls_on_connect_ports.
4543 25. Implemented acl_smtp_predata.
4545 26. If the domain in control=submission is set empty, Exim assumes that the
4546 authenticated id is a complete email address when it generates From: or
4547 Sender: header lines.
4549 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4550 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4551 chown and chgrp in /bin and hostname in /usr/bin.
4553 28. Exim was keeping the "process log" file open after each use, just as it
4554 does for the main log. This opens the possibility of it remaining open for
4555 long periods when the USR1 signal hits a daemon. Occasional processlog
4556 errors were reported, that could have been caused by this. Anyway, it seems
4557 much more sensible not to leave this file open at all, so that is what now
4560 29. The long-running daemon process does not normally write to the log once it
4561 has entered its main loop, and it closes the log before doing so. This is
4562 so that log files can straightforwardly be renamed and moved. However,
4563 there are a couple of unusual error situations where the daemon does write
4564 log entries, and I had neglected to close the log afterwards.
4566 30. The text of an SMTP error response that was received during a remote
4567 delivery was being truncated at 512 bytes. This is too short for some of
4568 the long messages that one sometimes sees. I've increased the limit to
4571 31. It is now possible to make retry rules that apply only when a message has a
4572 specific sender, in particular, an empty sender.
4574 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4575 it possible to be selective about when SMTP synchronization is enforced.
4577 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4579 32. Implemented hosts_connection_nolog.
4581 33. Added an ACL for QUIT.
4583 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4586 35. Added mailbox_size and mailbox_filecount to appendfile.
4588 36. Added control = no_multiline_responses to ACLs.
4590 37. There was a bug in the logic of the code that waits for the clock to tick
4591 in the case where the clock went backwards by a substantial amount such
4592 that the microsecond fraction of "now" was more than the microsecond
4593 fraction of "then" (but the whole seconds number was less).
4595 38. Added support for the libradius Radius client library this is found on
4596 FreeBSD (previously only the radiusclient library was supported).
4602 1. When certain lookups returned multiple values in the form name=value, the
4603 quoting of the values was not always being done properly. Specifically:
4604 (a) If the value started with a double quote, but contained no whitespace,
4606 (b) If the value contained whitespace other than a space character (i.e.
4607 tabs or newlines or carriage returns) it was not quoted.
4608 This fix has been applied to the mysql and pgsql lookups by writing a
4609 separate quoting function and calling it from the lookup code. The fix
4610 should probably also be applied to nisplus, ibase and oracle lookups, but
4611 since I cannot test any of those, I have not disturbed their existing code.
4613 2. A hit in the callout cache for a specific address caused a log line with no
4614 reason for rejecting RCPT. Now it says "Previous (cached) callout
4615 verification failure".
4617 3. There was an off-by-one bug in the queryprogram router. An over-long
4618 return line was truncated at 256 instead of 255 characters, thereby
4619 overflowing its buffer with the terminating zero. As well as fixing this, I
4620 have increased the buffer size to 1024 (and made a note to document this).
4622 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4623 when Exim is waiting for an SMTP response from a remote server, Exim
4624 restarts its select() call on the socket, thereby resetting its timeout.
4625 This is not a problem when such interrupts are rare. Somebody set up a cron
4626 job to run exiwhat every 2 minutes, which is less than the normal select()
4627 timeout (5 or 10 minutes). This meant that the select() timeout never
4628 kicked in because it was always reset. I have fixed this by comparing the
4629 time when an interrupt arrives with the time at the start of the first call
4630 to select(). If more time than the timeout has elapsed, the interrupt is
4631 treated as a timeout.
4633 5. Some internal re-factoring in preparation for the addition of Sieve
4634 extensions (by MH). In particular, the "personal" test is moved to a
4635 separate function, and given an option for scanning Cc: and Bcc: (which is
4636 not set for Exim filters).
4638 6. When Exim created an email address using the login of the caller as the
4639 local part (e.g. when creating a From: or Sender: header line), it was not
4640 quoting the local part when it contained special characters such as @.
4642 7. Installed new OpenBSD configuration files.
4644 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4645 try to make them clearer.
4647 9. Callout options, other than the timeout value, were being ignored when
4648 verifying sender addresses in header lines. For example, when using
4650 verify = header_sender/callout=no_cache
4652 the cache was (incorrectly) being used.
4654 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4655 only the Cygwin environment.
4657 11. When return_path_on_delivery was set as a log selector, if different remote
4658 addresses in the same message used different return paths and parallel
4659 remote delivery occurred, the wrong values would sometimes be logged.
4660 (Whenever a remote delivery process finished, the return path value from
4661 the most recently started remote delivery process was logged.)
4663 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4664 lines when AUTH and/or TLS are in use. This is the "received protocol"
4665 field. Exim used to use "asmtp" for authenticated SMTP, without any
4666 indication (in the protocol name) for TLS use. Now it follows the RFC and
4667 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4668 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4669 names appear in log lines as well as in Received: header lines.
4671 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4672 extensions, and comparison tests, and to fix some bugs.
4674 14. Changes to the "personal" filter test:
4676 (1) The test was buggy in that it was just doing the equivalent of
4677 "contains" tests on header lines. For example, if a user's address was
4678 anne@some.where, the "personal" test would incorrectly be true for
4680 To: susanne@some.where
4682 This test is now done by extracting each address from the header in turn,
4683 and checking the entire address. Other tests that are part of "personal"
4684 are now done using regular expressions (for example, to check local parts
4685 of addresses in From: header lines).
4687 (2) The list of non-personal local parts in From: addresses has been
4688 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4689 taken from the Sieve specification recommendations.
4691 (3) If the message contains any header line starting with "List-" it is
4692 treated as non-personal.
4694 (4) The test for "circular" in the Subject: header line has been removed
4695 because it now seems ill-conceived.
4697 15. Minor typos in src/EDITME comments corrected.
4699 16. Installed latest exipick from John Jetmore.
4701 17. If headers_add on a router specified a text string that was too long for
4702 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4703 of string_sprintf() is now avoided.
4705 18. $message_body_size was not set (it was always zero) when running the DATA
4706 ACL and the local_scan() function.
4708 19. For the "mail" command in an Exim filter, no default was being set for
4709 the once_repeat time, causing a random time value to be used if "once" was
4710 specified. (If the value happened to be <= 0, no repeat happened.) The
4711 default is now 0s, meaning "never repeat". The "vacation" command was OK
4712 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4713 (I found it when inspecting the code).
4715 20. There is now an overall timeout for performing a callout verification. It
4716 defaults to 4 times the callout timeout, which applies to individual SMTP
4717 commands during the callout. The overall timeout applies when there is more
4718 than one host that can be tried. The timeout is checked before trying the
4719 next host. This prevents very long delays if there are a large number of
4720 hosts and all are timing out (e.g. when the network connections are timing
4721 out). The value of the overall timeout can be changed by specifying an
4722 additional sub-option for "callout", called "maxwait". For example:
4724 verify = sender/callout=5s,maxwait=20s
4726 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4727 to the end before writing, but this should make it even safer).
4729 22. Exim was forgetting that it had advertised PIPELINING for the second and
4730 subsequent messages on an SMTP connection. It was also not resetting its
4731 memory on STARTTLS and an internal HELO.
4733 23. When Exim logs an SMTP synchronization error within a session, it now
4734 records whether PIPELINING has been advertised or not.
4736 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4737 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4738 rather than long int.
4740 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4742 26. Added the never_mail option to autoreply.
4748 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4749 crash if the getsockname() call failed; this can happen if a connection is
4750 closed very soon after it is established. The problem was simply in the
4751 order in which certain operations were done, causing Exim to try to write
4752 to the SMTP stream before it had set up the file descriptor. The bug has
4753 been fixed by making things happen in the correct order.
4759 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4760 before the connection was closed, thus losing the rejection response.
4762 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4763 some early Solaris releases, but causes trouble in current releases where
4764 socklen_t is defined.
4766 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4769 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4772 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4773 that releasing the top of it at the end releases what was used for sub-
4774 expansions (unless the block got too big). However, discard this block if
4775 the first thing is a variable or header, so that we can use its block when
4776 it is dynamic (useful for very large $message_headers, for example).
4778 6. Lookups now cache *every* query, not just the most recent. A new, separate
4779 store pool is used for this. It can be recovered when all lookup caches are
4780 flushed. Lookups now release memory at the end of their result strings.
4781 This has involved some general refactoring of the lookup sources.
4783 7. Some code has been added to the store_xxx() functions to reduce the amount
4784 of flapping under certain conditions.
4786 8. log_incoming_interface used to affect only the <= reception log lines. Now
4787 it causes the local interface and port to be added to several more SMTP log
4788 lines, for example "SMTP connection from", and rejection lines.
4790 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4792 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4794 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4795 could be overwritten at the end of the current message (or the start of a
4796 new message if it was set in a HELO ACL). The value is now preserved for
4797 the duration of the SMTP connection.
4799 12. If a transport had a headers_rewrite setting, and a matching header line
4800 contained an unqualified address, that address was qualified, even if it
4801 did not match any rewriting rules. The underlying bug was that the values
4802 of the flags that permit the existence of unqualified sender and recipient
4803 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4804 non-local messages, and by -bnq for local messages) were not being
4805 preserved with the message after it was received.
4807 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4808 "next input=" as part of the text comprising the host identity instead of
4809 the correct text. The code was using the same buffer for two different
4810 strings. However, depending on which order the printing function evaluated
4811 its arguments, the bug did not always show up. Under Linux, for example, my
4812 test suite worked just fine.
4814 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4815 doesn't work with some older versions of Perl. It has been changed to "my",
4816 which in any case is probably the better facility to use.
4818 15. A really picky compiler found some instances of statements for creating
4819 error messages that either had too many or two few arguments for the format
4822 16. The size of the buffer for calls to the DNS resolver has been increased
4823 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4824 for addresses that have a lot of PTR records. This alleviates a problem; it
4825 does not fully solve it.
4827 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4828 buffer now truncates the list and logs the incident, which is the same
4829 action as happens when Exim is looking up a host name and its aliases.
4830 Previously in this situation something unpredictable would happen;
4831 sometimes it was "internal error: store_reset failed".
4833 18. If a server dropped the connection unexpectedly when an Exim client was
4834 using GnuTLS and trying to read a response, the client delivery process
4835 crashed while trying to generate an error log message.
4837 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4838 string, for example:
4840 warn message = H1: something\nH2: something
4842 the text was added as a single header line from Exim's point of view
4843 though it ended up OK in the delivered message. However, searching for the
4844 second and subsequent header lines using $h_h2: did not work. This has been
4845 fixed. Similarly, if a system filter added multiple headers in this way,
4846 the routers could not see them.
4848 20. Expanded the error message when iplsearch is called with an invalid key to
4849 suggest using net-iplsearch in a host list.
4851 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4852 ACLs are no longer actually imposed (and a message to that effect is
4855 22. If a "gecos" field in a passwd entry contained escaped characters, in
4856 particular, if it contained a \" sequence, Exim got it wrong when building
4857 a From: or a Sender: header from that name. A second bug also caused
4858 incorrect handling when an unquoted " was present following a character
4859 that needed quoting.
4861 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4862 was not being matched caselessly.
4864 24. Arranged for all hyphens in the exim.8 source to be escaped with
4867 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4868 properly. Recipient callout cache records were still being keyed to include
4869 the sender, even when use_sender was set false. This led to far more
4870 callouts that were necessary. The sender is no longer included in the key
4871 when use_sender is false.
4873 26. Added "control = submission" modifier to ACLs.
4875 27. Added the ${base62d: operator to decode base 62 numbers.
4877 28. dnsdb lookups can now access SRV records.
4879 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4880 the configuration file.
4882 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4883 (-v) mode. This makes the output for a verbose queue run more intelligible.
4885 31. Added a use_postmaster feature to recipient callouts.
4887 32. Added the $body_zerocount variable, containing the number of binary zero
4888 bytes in the message body.
4890 33. The time of last modification of the "new" subdirectory is now used as the
4891 "mailbox time last read" when there is a quota error for a maildir
4894 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4896 35. Added +ignore_unknown as a special item in host lists.
4898 36. Code for decoding IPv6 addresses in host lists is now included, even if
4899 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4900 address was recognized as an IP address, but was then not correctly decoded
4901 into binary, causing unexpected and incorrect effects when compared with
4908 1. Very minor rewording of debugging text in manualroute to say "list of
4909 hosts" instead of "hostlist".
4911 2. If verify=header_syntax was set, and a header line with an unqualified
4912 address (no domain) and a large number of spaces between the end of the
4913 name and the colon was received, the reception process suffered a buffer
4914 overflow, and (when I tested it) crashed. This was caused by some obsolete
4915 code that should have been removed. The fix is to remove it!
4917 3. When running in the test harness, delay a bit after writing a bounce
4918 message to get a bit more predictability in the log output.
4920 4. Added a call to search_tidyup() just before forking a reception process. In
4921 theory, someone could use a lookup in the expansion of smtp_accept_max_
4922 per_host which, without the tidyup, could leave open a database connection.
4924 5. Added the variables $recipient_data and $sender_data which get set from a
4925 lookup success in an ACL "recipients" or "senders" condition, or a router
4926 "senders" option, similar to $domain_data and $local_part_data.
4928 6. Moved the writing of debug_print from before to after the "senders" test
4931 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4932 problems for message scanning, either using a data ACL, or using
4933 local_scan() because the Received: header was not generated till after they
4934 were called (in order to set the time as the time of reception completion).
4935 I have revised the way this works. The header is now generated after the
4936 body is received, but before the ACL or local_scan() are called. After they
4937 are run, the timestamp in the header is updated.
4943 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4944 before starting a queue runner without re-exec. This happened only when
4945 deliver_drop_privilege was set or when the Exim user was set to root. The
4946 effect of the bug was that timeouts during subsequent deliveries caused
4947 crashes instead of being properly handled. The handler is now left at its
4948 default (and expected) setting.
4950 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4951 message, again when deliver_drop_privilege is set or Exim is run as root.
4952 The bug described in (1) was not present in this case, but the tidying up
4953 of the other signals was missing. I have made the two cases consistent.
4955 3. The ignore_target_hosts setting on a manualroute router was being ignored
4956 for hosts that were looked up using the /MX notation.
4958 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4961 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4962 operated on the sender address. After changing the $sender_address to <>
4963 for the sender address verify, Exim was re-instated it as the original
4964 (before rewriting) address, but remembering that it had rewritten it, so it
4965 wasn't rewriting it again. This bug also had the effect of breaking the
4966 sender address verification caching when the sender address was rewritten.
4968 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4969 This has been changed so that if the ip literal address matches
4970 ignore_target_hosts, the router declines.
4972 7. Added expansion conditions match_domain, match_address, and match_local_
4973 part (NOT match_host).
4975 8. The placeholder for the Received: header didn't have a length field set.
4977 9. Added code to Exim itself and to exim_lock to test for a specific race
4978 condition that could lead to file corruption when using MBX delivery. The
4979 issue is with the lockfile that is created in /tmp. If this file is removed
4980 after a process has opened it but before that process has acquired a lock,
4981 there is the potential for a second process to recreate the file and also
4982 acquire a lock. This could lead to two Exim processes writing to the file
4983 at the same time. The added code performs the same test as UW imapd; it
4984 checks after acquiring the lock that its file descriptor still refers to
4985 the same named file.
4987 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4988 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4991 11. Added the smtp_active_hostname option. If used, this will typically be made
4992 to depend on the incoming interface address. Because $interface_address is
4993 not set up until the daemon has forked a reception process, error responses
4994 that can happen earlier (such as "too many connections") no longer contain
4997 12. If an expansion in a condition on a "warn" statement fails because a lookup
4998 defers, the "warn" statement is abandoned, and the next ACL statement is
4999 processed. Previously this caused the whole ACL to be aborted.
5001 13. Added the iplsearch lookup type.
5003 14. Added ident_timeout as a log selector.
5005 15. Added tls_certificate_verified as a log selector.
5007 16. Added a global option tls_require_ciphers (compare the smtp transport
5008 option of the same name). This controls incoming TLS connections.
5010 17. I finally figured out how to make tls_require_ciphers do a similar thing
5011 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5012 before starting the TLS session.
5014 18. Tabs are now shown as \t in -bP output.
5016 19. If the log selector return_path_on_delivery was set, Exim crashed when
5017 bouncing a message because it had too many Received: header lines.
5019 20. If two routers both had headers_remove settings, and the first one included
5020 a superfluous trailing colon, the final name in the first list and the
5021 first name in the second list were incorrectly joined into one item (with a
5022 colon in the middle).
5028 1. Added -C and -D options to the exinext utility, mainly to make it easier
5029 to include in the automated testing, but these could be helpful when
5030 multiple configurations are in use.
5032 2. The exinext utility was not formatting the output nicely when there was
5033 an alternate port involved in the retry record key, nor when there was a
5034 message id as well (for retries that were specific to a specific message
5035 and a specific host). It was also confused by IPv6 addresses, because of
5036 the additional colons they contain. I have fixed the IPv4 problem, and
5037 patched it up to do a reasonable job for IPv6.
5039 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5040 delivery, the log line now contains "pipelined" if PIPELINING was used.
5042 4. An SMTP transport process used to panic and die if the bind() call to set
5043 an explicit outgoing interface failed. This has been changed; it is now
5044 treated in the same way as a connect() failure.
5046 5. A reference to $sender_host_name in the part of a conditional expansion
5047 that was being skipped was still causing a DNS lookup. This no longer
5050 6. The def: expansion condition was not recognizing references to header lines
5051 that used bh_ and bheader_.
5053 7. Added the _cache feature to named lists.
5055 8. The code for checking quota_filecount in the appendfile transport was
5056 allowing one more file than it should have been.
5058 9. For compatibility with Sendmail, the command line option
5066 and sets the incoming protocol and host name (for trusted callers). The
5067 host name and its colon can be omitted when only the protocol is to be set.
5068 Note the Exim already has two private options, -pd and -ps, that refer to
5069 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5070 "s", but I don't think that's a major issue.
5072 10. A number of refactoring changes to the code, none of which should affect
5075 (a) The number of logging options was getting close to filling up the
5076 32-bit word that was used as a bit map. I have split them into two classes:
5077 those that are passed in the argument to log_write(), and those that are
5078 only ever tested independently outside of that function. These are now in
5079 separate 32-bit words, so there is plenty of room for expansion again.
5080 There is no change in the user interface or the logging behaviour.
5082 (b) When building, for example, log lines, the code previously used a
5083 macro that called string_cat() twice, in order to add two strings. This is
5084 not really sufficiently general. Furthermore, there was one instance where
5085 it was actually wrong because one of the argument was used twice, and in
5086 one call a function was used. (As it happened, calling the function twice
5087 did not affect the overall behaviour.) The macro has been replaced by a
5088 function that can join an arbitrary number of extra strings onto a growing
5091 (c) The code for expansion conditions now uses a table and a binary chop
5092 instead of a serial search (which was left over from when there were very
5093 few conditions). Also, it now recognizes conditions like "pam" even when
5094 the relevant support is not compiled in: a suitably worded error message is
5095 given if an attempt is made to use such a condition.
5097 11. Added ${time_interval:xxxxx}.
5099 12. A bug was causing one of the ddress fields not to be passed back correctly
5100 from remote delivery subprocesses. The field in question was not being
5101 subsequently used, so this caused to problems in practice.
5103 13. Added new log selectors queue_time and deliver_time.
5105 14. Might have fixed a bug in maildirsizefile handling that threw up
5106 "unexpected character" debug warnings, and recalculated the data
5107 unnecessarily. In any case, I expanded the warning message to give more
5110 15. Added the message "Restricted characters in address" to the statements in
5111 the default ACL that block characters like @ and % in local parts.
5113 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5114 Three changes have been made:
5116 (a) There was a serious bug; a negative response to MAIL caused the whole
5117 recipient domain to be cached as invalid, thereby blocking all messages
5118 to all local parts at the same domain, from all senders. This bug has
5119 been fixed. The domain is no longer cached after a negative response to
5120 MAIL if the sender used is not empty.
5122 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5125 (c) A new callout option, "use_sender" has been added for people who want
5126 the modified behaviour.
5132 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5135 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5136 indeed breaks things for older releases.
5138 3. Added additional logging to the case where there is a problem reading data
5139 from a filter that is running in a subprocess using a pipe, in order to
5140 try to track down a specific problem.
5142 4. Testing facility fudge: when running in the test harness and attempting
5143 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5144 getting "No route to host". Convert this to a timeout.
5146 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5149 6. Some OS don't have socklen_t but use size_t instead. This affects the
5150 fifth argument of getsockopt() amongst other things. This is now
5151 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5152 can be set for individual OS. I have set it for SunOS5, OSF1, and
5153 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5154 some earlier ones do not.
5156 7. Change 4.30/15 was not doing the test caselessly.
5158 8. The standard form for an IPv6 address literal was being rejected by address
5159 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5160 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5161 this, as well as the form without the "IPv6" on the front (but only when
5162 address literals are enabled, of course).
5164 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5166 10. Exim crashed if a message with an empty sender address specified by -f
5167 encountered a router with an errors_to setting. This could be provoked only
5168 by a command such as
5172 where an empty string was supplied; "<>" did not hit this bug.
5174 11. Installed PCRE release 4.5.
5176 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5177 remained set. It is now erased.
5179 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5180 times from message ids (which are base 36 rather than the normal 62).
5182 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5183 were being counted as actual protocol errors, and logged if the log
5184 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5185 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5186 and DATA following a set of rejected RCPTs do not count as protocol errors.
5187 In other words, Exim assumes they were pipelined, though this may not
5188 actually be the case. Of course, in all cases the client gets an
5189 appropriate error code.
5191 15. If a lookup fails in an ACL condition, a message about the failure may
5192 be available; it is used if testing the ACL cannot continue, because most
5193 such messages specify what the cause of the deferral is. However, some
5194 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5195 that caused an old message to be retained and used if a later statement
5196 caused a defer, replacing the real cause of the deferral.
5198 16. If an IP address had so many PTR records that the DNS lookup buffer
5199 was not large enough to hold them, Exim could crash while trying to process
5200 the truncated data. It now detects and logs this case.
5202 17. Further to 4.21/58, another change has been made: if (and only if) the
5203 first line of a message (the first header line) ends with CRLF, a bare LF
5204 in a subsequent header line has a space inserted after it, so as not to
5205 terminate the header.
5207 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5208 unnecessarily, used atoi() instead of strtol(), and didn't check the
5209 termination when getting file sizes from file names by regex.
5211 19. Completely re-implemented the support for maildirsize files, in the light
5212 of a number of problems with the previous contributed implementation
5213 (4.30/29). In particular:
5215 . If the quota is zero, the maildirsize file is maintained, but no quota is
5218 . If the maildir directory does not exist, it is created before any attempt
5219 to write a maildirsize file.
5221 . The quota value in the file is just a cache; if the quota is changed in
5222 the transport, the new value overrides.
5224 . A regular expression is available for excluding directories from the
5227 20. The autoreply transport checks the characters in options that define the
5228 message's headers; it allows continued headers, but it was checking with
5229 isspace() after an embedded newline instead of explicitly looking for a
5232 21. If all the "regular" hosts to which an address was routed had passed their
5233 expiry times, and had not reached their retry times, the address was
5234 bounced, even if fallback hosts were defined. Now Exim should go on to try
5237 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5238 equivalent code in the SMTP transport. Some hosts send humungous responses
5239 to HELO/EHLO, more than 1024 it seems.
5241 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5242 gives compiler warnings in some environments. I've now done it "properly",
5245 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5246 (because of problems with the built-in one) was declared to return uschar *
5247 instead of char *, causing compiler failure.
5249 25. Fixed a file descriptor leak when processing alias/forward files.
5251 26. Fixed a minor format string issue in dbfn.c.
5253 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5255 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5256 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5259 29. When a sender address is verified, it is cached, to save repeating the test
5260 when there is more than one recipient in a message. However, when the
5261 verification involves a callout, it is possible for different callout
5262 options to be set for different recipients. It is too complicated to keep
5263 track of this in the cache, so now Exim always runs a verification when a
5264 callout is required, relying on the callout cache for the optimization.
5265 The overhead is duplication of the address routing, but this should not be
5268 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5269 to be verified with callout=postmaster, and the main callout worked but the
5270 postmaster check failed, the verification correctly failed. However, if a
5271 subsequent RCPT command asked for sender verification *without* the
5272 postmaster check, incorrect caching caused this verification also to fail,
5275 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5276 it was not caching the DNS options (qualify_single, search_parents) that
5277 were used when the lookup failed. A subsequent lookup with different
5278 options therefore always gave the same answer, though there were cases
5279 where it should not have. (Example: a "domains = !$mx_any" option on a
5280 dnslookup router: the "domains" option is always processed without any
5281 widening, but the router might have qualify_single set.) Now Exim uses the
5282 cached value only when the same options are set.
5284 32. Added John Jetmore's "exipick" utility to the distribution.
5286 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5287 than a timeout (e.g. a certificate is required, and is not provided), an
5288 Exim server now closes the connection immediately. Previously it waited for
5289 the client to close - but if the client is SSL, it seems that they each
5290 wait for each other, leading to a delay before one of them times out.
5292 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5293 maintained 0.8.x compatibility because I don't think many are using it, and
5294 it is clearly obsolete.
5296 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5299 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5300 client certificate was expired. A simple patch fixes this, though I don't
5301 understand the full logic of why the verify callback is called multiple
5304 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5305 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5306 which causes problems with some clients (such as the Certicom SSL Plus
5307 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5308 disables the coutermeasure allowing Eudora to connect."
5310 38. Exim was not checking that a write() to a log file succeeded. This could
5311 lead to Bad Things if a log got too big, in particular if it hit a file
5312 size limit. Exim now panics and dies if it cannot write to a log file, just
5313 as it does if it cannot open a log file.
5315 39. Modified OS/Makefile-Linux so that it now contains
5317 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5319 The two -D definitions ensure that Exim is compiled with large file
5320 support, which makes it possible to handle log files that are bigger than
5323 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5324 instance) a domain was checked against a named list that involved a lookup,
5325 causing $domain_data to be set, then another domain was checked against the
5326 same list, then the first domain was re-checked, the value of $domain_data
5327 after the final check could be wrong. In particular, if the second check
5328 failed, it could be set empty. This bug probably also applied to
5331 41. The strip_trailing_dot option was not being applied to the address given
5332 with the -f command-line option.
5334 42. The code for reading a message's header from the spool was incrementing
5335 $received_count, but never initializing it. This meant that the value was
5336 incorrect (doubled) while delivering a message in the same process in which
5337 it was received. In the most common configuration of Exim, this never
5338 happens - a fresh exec is done - but it can happen when
5339 deliver_drop_privilege is set.
5341 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5342 it now includes up to 150 characters of the unexpected data in the log
5345 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5346 and building data strings. The size of both of these buffers was 10 000
5347 bytes - far larger than anybody would *ever* want, thought I. Needless to
5348 say, somebody hit the limit. I have increased the maximum line length to
5349 20 000 and the maximum data length of concatenated lines to 100 000. I have
5350 also fixed two bugs, because there was no checking on these buffers. Tsk,
5351 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5352 buffer is too small.
5354 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5355 lsearch lookups. Now it does.
5357 46. When parsing a route_list item in a manualroute router, a fixed-length
5358 buffer was used for the list of hosts. I made this 1024 bytes long,
5359 thinking that nobody would ever have a list of hosts that long. Wrong.
5360 Somebody had a whole pile of complicated expansion conditions, and the
5361 string was silently truncated, leading to an expansion error. It turns out
5362 that it is easier to change to an unlimited length (owing to other changes
5363 that have happened since this code was originally written) than to build
5364 structure for giving a limitation error. The length of the item that
5365 expands into the list of hosts is now unlimited.
5367 47. The lsearch lookup could not handle data where the length of text line was
5368 more than 4095 characters. Such lines were truncated, leading to shortened
5369 data being returned. It should now handle lines of any length.
5371 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5372 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5375 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5376 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5377 no longer happens in comment lines. A list of replacements is now placed
5378 at the head of all of the source files, except those whose only change is
5379 to replace PERL_COMMAND in the very first #! line.
5381 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5382 messages on the queue, with a bottom-up merge sort, using code contributed
5383 by Michael Haardt. This should make operations like -bp somewhat faster on
5384 large queues. It won't affect queue runners, except when queue_run_in_order
5387 51. Installed eximstats 1.31 in the distribution.
5389 52. Added support for SRV lookups to the dnslookup router.
5391 53. If an ACL referred to $message_body or $message_body_end, the value was not
5392 reset for any messages that followed in the same SMTP session.
5394 54. The store-handling optimization for building very long strings was not
5395 differentiating between the different store pools. I don't think this
5396 actually made any difference in practice, but I've tidied it.
5398 55. While running the routers to verify a sender address, $sender_address
5399 was still set to the sender address. This is wrong, because when routing to
5400 send a bounce to the sender, it would be empty. Therefore, I have changed
5401 it so that, while verifying a sender address, $sender_address is set to <>.
5402 (There is no change to what happens when verifying a recipient address.)
5404 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5405 target A or AAAA records (if not already returned) without resetting the
5406 qualify_single or search_parents options of the DNS resolver. These are
5407 inappropriate in this case because the targets of MX and SRV records must
5408 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5409 target that, when qualified, matched something in the local domain. These
5410 two options are now turned off when doing these lookups.
5412 57. It seems that at least some releases of Reiserfs (which does not have the
5413 concept of a fixed number of inodes) returns zero and not -1 for the
5414 number of available inodes. This interacted badly with check_spool_inodes,
5415 which assumed that -1 was the "no such thing" setting. What I have done is
5416 to check that the total number of inodes is greater than zero before doing
5417 the test of how many are available.
5419 58. When a "warn" ACL statement has a log_message modifier, the message is
5420 remembered, and not repeated. This is to avoid a lot of repetition when a
5421 message has many recipients that cause the same warning to be written.
5422 Howewer, Exim was preserving the list of already written lines for an
5423 entire SMTP session, which doesn't seem right. The memory is now reset if a
5424 new message is started.
5426 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5427 debugging output unless log_rewrite was also set.
5429 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5430 of (int)(handle) when we know that handle contains (void *)(-1).
5432 61. The Exim daemon panic-logs an error return when it closes the incoming
5433 connection. However "connection reset by peer" seems to be common, and
5434 isn't really an error worthy of noting specially, so that particular error
5437 62. When Exim is trying to find all the local interfaces, it used to panic and
5438 die if the ioctl to get the interface flags failed. However, it seems that
5439 on at least one OS (Solaris 9) it is possible to have an interface that is
5440 included in the list of interfaces, but for which you get a failure error
5441 for this call. This happens when the interface is not "plumbed" into a
5442 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5443 failure of the "get flags" call assumes that the interface is down.
5445 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5446 makes life easier for people who are doing arithmetic on fields extracted
5447 from dates, where you often get leading zeros that should not be
5448 interpreted as octal.
5450 64. Added qualify_domain to the redirect router, to override the global
5453 65. If a pathologically long header line contained very many addresses (the
5454 report of this problem mentioned 10 000) and each of them was rewritten,
5455 Exim could use up a very large amount of memory. (It kept on making new
5456 copies of the header line as it rewrote, and never released the old ones.)
5457 At the expense of a bit more processing, the header rewriting function has
5458 been changed so that it no longer eats memory in this way.
5460 66. The generation of the Received: header has been moved from the time that a
5461 message starts to be received, to the time that it finishes. The timestamp
5462 in the Received: header should now be very close to that of the <= log
5463 line. There are two side-effects of this change:
5465 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5466 logged header lines no longer include the local Received: line, because
5467 it has not yet been created. The same applies to a copy of the message
5468 that is returned to a non-SMTP sender when a message is rejected.
5470 (b) When a filter file is tested using -bf, no additional Received: header
5471 is added to the test message. After some thought, I decided that this
5474 This change does not affect the value of $received_for. It is still set
5475 after address rewriting, but before local_scan() is called.
5477 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5479 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5480 gave an unhelpful panic error message, and a defer error. I have managed to
5481 change this behaviour so that it now rejects any supplied certificate,
5482 which seems right, as the list of acceptable certificates is empty.
5484 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5485 gave an unhelpful defer error. I have not managed to make this reject any
5486 supplied certificates, but the error message it gives is "no certificate
5487 supplied", which is not helpful.
5489 70. exigrep's output now also includes lines that are not associated with any
5490 message, but which match the given pattern. Implemented by a patch from
5491 Martin Sluka, which also tidied up the Perl a bit.
5493 71. Recipient callout verification, like sender verification, was using <> in
5494 the MAIL FROM command. This isn't really the right thing, since the actual
5495 sender may affect whether the remote host accepts the recipient or not. I
5496 have changed it to use the actual sender in the callout; this means that
5497 the cache record is now keyed on a recipient/sender pair, not just the
5498 recipient address. There doesn't seem to be a real danger of callout loops,
5499 since a callout by the remote host to check the sender would use <>.
5500 [SEE ABOVE: changed after hitting problems.]
5502 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5503 temporary errors. However, in the case of such a code being given after
5504 the end of a data transmission (i.e. after ".") Exim was failing to write
5505 a retry record for the message. (Yes, there was some broken host that was
5506 actually sending 8xx at this point.)
5508 73. An unknown lookup type in a host list could cause Exim to panic-die when
5509 the list was checked. (An example that provoked this was putting <; in the
5510 middle of a list instead of at the start.) If this happened during a DATA
5511 ACL check, a -D file could be left lying around. This kind of configuration
5512 error no longer causes Exim to die; instead it causes a defer errror. The
5513 incident is still logged to the main and panic logs.
5515 74. Buglet left over from Exim 3 conversion. The message "too many messages
5516 in one connection" was written to the rejectlog but not the mainlog, except
5517 when address rewriting (yes!) was being logged.
5519 75. Added write_rejectlog option.
5521 76. When a system filter was run not as root (that is, when system_filter_user
5522 was set), the values of the $n variables were not being returned to the
5523 main process; thus, they were not subsequently available in the $sn
5526 77. Added +return_path_on_delivery log selector.
5528 78. A connection timeout was being treated differently from recipients deferred
5529 when testing hosts_max_try with a message that was older than the host's
5530 retry timeout. (The host should not be counted, thus allowing all hosts to
5531 be tried at least once before bouncing.) This may have been the cause of an
5532 occasionally reported bug whereby a message would remain on the queue
5533 longer than the retry timeout, but would be bounced if a delivery was
5534 forced. I say "may" because I never totally pinned down the problem;
5535 setting up timeout/retry tests is difficult. See also the next item.
5537 79. The ultimate address timeout was not being applied to errors that involved
5538 a combination of host plus message (for example, a timeout on a MAIL
5539 command). When an address resolved to a number of possible hosts, and they
5540 were not all tried for each delivery (e.g. because of hosts_max_try), a
5541 message could remain on the queue longer than the retry timeout.
5543 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5546 81. Fixed an obscure SMTP outgoing bug which required at least the following
5547 conditions: (a) there was another message waiting for the same server;
5548 (b) the server returned 5xx to all RCPT commands in the first message so
5549 that the message was not completed; (c) the server dropped the connection
5550 or gave a negative response to the RSET that Exim sends to abort the
5551 transaction. The observed case was a dropped connection after DATA that had
5552 been sent in pipelining mode. That is, the server had advertised PIPELINING
5553 but was not implementing it correctly. The effect of the bug was incorrect
5554 behaviour, such as trying another host, and this could lead to a crash.
5560 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5561 and daemon.c were passed as pointers to ints; they should have been
5562 pointers to socklen_t variables (which are typically unsigned ints).
5564 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5567 3. Fixed a really odd bug that affected only the testing scheme; patching a
5568 certain fixed string in the binary changed the value of another string that
5569 happened to be identical to the end of the original first string.
5571 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5572 name", it returns that address as the IP address. On some operating
5573 systems (e.g. Solaris), it also passes back the IP address string as the
5574 "host name". However, on others (e.g. Linux), it passes back an empty
5575 string. Exim wasn't checking for this, and was changing the host name to an
5576 empty string, assuming it had been canonicized.
5578 5. Although rare, it is permitted to have more than one PTR record for a given
5579 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5580 all the names associated with an address, because they do in Solaris.
5581 However, it seems that they do not in Linux for data that comes from the
5582 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5583 I found this out when I moved to a new Linux workstation and tried to run
5584 the Exim test suite.
5586 To get round this problem I have changed the code so that it now does its
5587 own call to the DNS to look up PTR records when searching for a host name.
5588 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5589 addresses that are only in /etc/hosts are still found.
5591 This behaviour is, however, controlled by an option called host_lookup_
5592 order, which defaults to "bydns:byaddr". If people want to use the other
5593 order, or indeed, just use one or the other means of lookup, they can
5594 specify it in this variable.
5596 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5597 some operating systems, this comes back from gethostbyaddr() as an empty
5598 string, and this is what Exim used to test for. However, it seems that in
5599 other systems, "." is yielded. Exim now tests for this case too.
5601 7. The values of check_spool_space and check_log_space are now held internally
5602 as a number of kilobytes instead of an absolute number of bytes. If a
5603 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5604 kilobyte. This means that much larger values can be stored.
5606 8. Exim monitor: an attempt to get the action menu when not actually pointing
5607 at a message produces an empty menu entitled "No message selected". This
5608 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5609 no entries in it ("Shell widget menu has zero width and/or height"). So I
5610 have added a single, blank menu entry in this case.
5612 9. Added ${quote_local_part.
5614 10. MIME decoding is now applied to the contents of Subject: header lines when
5617 11. Now that a reference to $sender_host_address automatically causes a reverse
5618 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5619 host lookup before query-style lookups in lists that might use this
5620 variable. This has therefore been abolished, and the "net-" prefix is no
5621 longer necessary for query-style lookups.
5623 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5624 have been a typo for LFLAGS, so it has been changed.
5626 13. The install script calls Exim with "-C /dev/null" in order to find the
5627 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5628 to be output. Howeve, since Exim outputs its version number before the
5629 error, it didn't break the script. It just looked ugly. I fixed this by
5630 always allowing "-C /dev/null" if the caller is root.
5632 14. Ignore overlarge ACL variable number when reading spool file - insurance
5633 against a later release with more variables having written the file.
5635 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5636 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5637 this, as well as the form without the "IPv6" on the front.
5639 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5640 OS/Makefile-Darwin file.
5642 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5643 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5645 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5646 variable. However, some OS use ints for inodes. I've added cast to long int
5647 to get rid of the compiler warning.
5649 19. I had forgotten to lock out "/../" in configuration file names when
5650 ALT_CONFIG_PREFIX was set.
5652 20. Routers used for verification do not need to specify transports. However,
5653 if such a router generated a host list, and callout was configured, Exim
5654 crashed, because it could not find a port number from the (non-existent)
5655 transport. It now assumes port 25 in this circumstance.
5657 21. Added the -t option to exigrep.
5659 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5660 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5661 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5662 them is set, LOOKUP_LSEARCH is forced.
5664 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5666 24. Added sender and host information to the "rejected by local_scan()" log
5667 line; previously there was no indication of these.
5669 25. Added .include_if_exists.
5671 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5672 Linux. It turns out that not all file systems support this. Apparently some
5673 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5674 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5675 not supported on the file descriptor, is now ignored when Exim is trying to
5676 sync a directory. This applies only to Linux.
5678 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5680 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5681 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5684 29. Added support for maildirsize files from supplied patch (modified a bit).
5686 30. The use of :fail: followed by an empty string could lead Exim to respond to
5687 sender verification failures with (e.g.):
5689 550 Verification failed for <xxx>
5690 550 Sender verify failed
5692 where the first response line was missing the '-' that indicates it is not
5693 the final line of the response.
5695 31. The loop for finding the name of the user that called Exim had a hardwired
5696 limit of 10; it now uses the value of finduser_retries, which is used for
5697 all other user lookups.
5699 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5702 33. Exim was neglecting to zero errno before one call of strtol() when
5703 expanding a string and expecting an integer value. On some systems this
5704 resulted in spurious "integer overflow" errors. Also, it was casting the
5705 result into an int without checking.
5707 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5708 did not work. The code looks as if it has *never* worked, though it appears
5709 to have been documented since at least releast 1.62. I have made it work.
5711 35. The "timeout_DNS" error in retry rules, also documented since at least
5712 1.62, also never worked. As it isn't clear exactly what this means, and
5713 clearly it isn't a major issue, I have abolished the feature by treating it
5714 as "timeout", and writing a warning to the main and panic logs.
5716 36. The display of retry rules for -brt wasn't always showing the error code
5719 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5720 timeout_connect_A, timeout_connect_MX.
5722 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5723 to the empty sender.
5725 39. The daemon was not analysing the content of -oX till after it had closed
5726 stderr and disconnected from the controlling terminal. This meant that any
5727 syntax errors were only noted on the panic log, and the return code from
5728 the command was 0. By re-arranging the code a little, I've made the
5729 decoding happen first, so such errors now appear on stderr, and the return
5730 code is 1. However, the actual setting up of the sockets still happens in
5731 the disconnected process, so errors there are still only recorded on the
5734 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5735 connections (as happens on some IP stacks) was logged at start up time as
5736 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5737 it from "IPv6 and IPv4", which means that two separate sockets are being
5740 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5741 says whether AF_INET or AF_INET6 was passed as an argument.
5743 42. Exiwhat output was messed up when time zones were included in log
5746 43. Exiwhat now gives more information about the daemon's listening ports,
5747 and whether -tls-on-connect was used.
5749 44. The "port" option of the smtp transport is now expanded.
5751 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5752 silently ignored. Now an error message is written to the main and panic
5755 46. There's a new ACL modifier called "logwrite" which writes to a log file
5756 as soon as it is encountered.
5758 47. Added $local_user_uid and $local_user_gid at routing time.
5760 48. Exim crashed when trying to verify a sender address that was being
5763 49. Exim was recognizing only a space character after ".include". It now also
5764 recognizes a tab character.
5766 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5767 extracting the relevant information from the specification. The man page no
5768 longer contains scrambled data for the -d option, and I've added a section
5769 at the front about calling Exim under different names.
5771 51. Added "extra_headers" argument to the "mail" command in filter files.
5773 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5776 53. Installed eximstats 1.29.
5778 54. Added transport_filter_timeout as a generic transport option.
5780 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5781 Cc: header lines. This was required by RFC 822, but it not required by RFC
5784 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5785 incoming messages that did not have them. Now it does so only if the
5786 message originates locally, that is, if there is no associated remote host
5787 address. When Resent- header lines are present, this applies to the Resent-
5788 lines rather than the non-Resent- lines.
5790 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5791 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5793 58. Messages for configuration errors now include the name of the main
5794 configuration file - useful now that there may be more than one file in a
5795 list (.included file names were always shown).
5797 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5798 for those rare installations that do not start the daemon as root or run it
5799 setuid root. I've cut out the call to initgroups() if the daemon is not
5802 60. The Exim user and group can now be bound into the binary as text strings
5803 that are looked up at the start of Exim's processing.
5805 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5807 62. Added $mailstore_basename variable.
5809 63. Installed patch to sieve.c from Michael Haardt.
5811 64. When Exim failed to open the panic log after failing to open the main log,
5812 the original message it was trying to log was written to stderr and debug
5813 output, but if they were not available (the usual case in production), it
5814 was lost. Now it is written to syslog before the two lines that record the
5815 failures to open the logs.
5817 65. Users' Exim filters run in subprocesses under the user's uid. It is
5818 possible for a "deliver" command or an alias in a "personal" command to
5819 provoke an address rewrite. If logging of address rewriting is configured,
5820 this fails because the process is not running as root or exim. There may be
5821 a better way of dealing with this, but for the moment (because 4.30 needs
5822 to be released), I have disabled address rewrite logging when running a
5823 filter in a non-root, non-exim process.
5829 1. The buildconfig auxiliary program wasn't quoting the value set for
5830 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5831 not defined. This bug was masked in 4.22 by the effect that was fixed in
5834 2. Some messages that were rejected after a message id was allocated were
5835 shown as "incomplete" by exigrep. It no longer does this for messages that
5836 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5838 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5839 have allow_domain_literals set, the ID did not get logged in the <= line.
5840 Domain literals are now always recognized in Message-ID: header lines.
5842 4. The first argument for a ${extract expansion item is the key name or field
5843 number. Leading and trailing spaces in this item were not being ignored,
5844 causing some misleading effects.
5846 5. When deliver_drop_privilege was set, single queue runner processes started
5847 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5848 same command in the process it spins off) were not dropping privilege.
5850 6. When the daemon running as "exim" started a queue runner, it always
5851 re-executed Exim in the spun-off process. This is a waste of effort when
5852 deliver_drop_privilege is set. The new process now just calls the
5853 queue-runner function directly.
5859 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5862 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5863 ignored. Though the use of -oP was forcing the writing of a pid file, it
5864 was always written to the default place.
5866 3. If the message "no IP address found for host xxxx" is generated during
5867 incoming verification, it is now followed by identification of the incoming
5868 connection (so you can more easily find what provoked it).
5870 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5872 5. Added some features to "harden" Exim a bit more against certain attacks:
5874 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5875 be put in Local/Makefile. This is like the never_users runtime option,
5876 but it cannot be overridden. The default setting is "root".
5878 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5879 prefix string with which any file named in a -C command line option
5882 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5883 is retained for -C and -D only if the caller of Exim is root. Without
5884 it, the exim user may also use -C and -D and retain privilege.
5886 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5887 command line option is disabled.
5889 6. Macro names set by the -D option must start with an upper case letter, just
5890 like macro names defined in the configuration file.
5892 7. Added "dereference=" facility to LDAP.
5894 8. Two instances of the typo "uknown" in the source files are fixed.
5896 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5897 the Configure-Makefile script screwed up while processing it.
5899 10. Incorporated PCRE 4.4.
5901 11. The SMTP synchronization check was not operating right at the start of an
5902 SMTP session. For example, it could not catch a HELO sent before the client
5903 waited for the greeting. There is now a check for outstanding input at the
5904 point when the greeting is written. Because of the duplex, asynchronous
5905 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5906 way, but not yet received, when the check is performed.
5908 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5909 on TCP/IP sockets, because this apparently causes some broken clients to
5912 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5913 unchanged) from the Cygwin maintainer.
5915 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5916 was supported instead of testing for mbx. Effectively a typo.
5918 15. The spa authenticator server code was not checking that the input it
5919 received was valid base64.
5921 16. The debug output line for the "set" modifier in ACLs was not showing the
5922 name of the variable that was being set.
5924 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5926 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5927 lookup on its own. Something else was needed to trigger the lookup. For
5928 example, a match in host_lookup or the need for a host name in a host list.
5929 Now, if $sender_host_name is referenced and the host name has not yet been
5930 looked up, a lookup is performed. If the lookup fails, the variable remains
5931 empty, and $host_lookup_failed is set to "1".
5933 19. Added "eqi" as a case-independent comparison operator.
5935 20. The saslauthd authentication condition could segfault if neither service
5936 nor realm was specified.
5938 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5939 error message that was logged was misleading, and incoming SMTP
5940 connections were dropped. The message is now more accurate, and temporary
5941 errors are given to SMTP connections.
5943 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5944 (except RSET) with a 421 error, until QUIT is received. However, it was
5945 failing to send a response to QUIT.
5947 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5948 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5949 ACL is now run after the helo_try_verify_hosts code.
5951 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5952 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5953 case-independent, so other case variants are also recognized). Apparently
5954 some systems use these upper case variants.
5956 25. If more than two messages were waiting for the same host, and a transport
5957 filter was specified for the transport, Exim sent two messages over the
5958 same TCP/IP connection, and then failed with "socket operation on non-
5959 socket" when it tried to send the third.
5961 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5963 27. The extern definition of crypt16() in expand.c was not being excluded when
5964 the OS had its own crypt16() function.
5966 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5967 as a preferred synonym for return_size_limit, both as an option and as an
5970 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5972 30. Changed the default configuration ACL to relax the local part checking rule
5973 for addresses that are not in any local domains. For these addresses,
5974 slashes and pipe symbols are allowed within local parts, but the sequence
5975 /../ is explicitly forbidden.
5977 31. SPA server authentication was not clearing the challenge buffer before
5980 32. log_message in a "warn" ACL statement was writing to the reject log as
5981 well as to the main log, which contradicts the documentation and doesn't
5982 seem right (because no rejection is happening). So I have stopped it.
5984 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5985 However, I am unable to do any testing of this.
5987 34. Fixed an infelicity in the appendfile transport. When checking directories
5988 for a mailbox, to see if any needed to be created, it was accidentally
5989 using path names with one or more superfluous leading slashes; tracing
5990 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5992 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5993 contents are added to the log line that is written for every discarded
5994 recipient. (Previously a log_message setting was ignored.)
5996 36. The ${quote: operator now quotes the string if it is empty.
5998 37. The install script runs exim in order to find its version number. If for
5999 some reason other than non-existence or emptiness, which it checks, it
6000 could not run './exim', it was installing it with an empty version number,
6001 i.e. as "exim-". This error state is now caught, and the installation is
6004 38. An argument was missing from the function that creates an error message
6005 when Exim fails to connect to the socket for saslauthd authentication.
6006 This could cause Exim to crash, or give a corrupted message.
6008 39. Added isip, isip4, and isip6 to ${if conditions.
6010 40. The ACL variables $acl_xx are now saved with the message, and can be
6011 accessed later in routers, transports, and filters.
6013 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6014 strings in the file are not string-expanded.
6016 42. If a MAIL command specified a SIZE value that was too large to fit into an
6017 int variable, the check against message_size_limit failed. Such values are
6018 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6019 day this will have to be increased, but I don't think I want to be around
6020 when emails are that large.
6027 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6028 iconv() is not standard in FreeBSD.
6030 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6031 IPv6 enabled. The observed symptom was a segmentation fault on return from
6032 the function os_common_find_running_interfaces() in src/os.c.
6034 3. In the check_special_case() function in daemon.c I had used "errno" as an
6035 argument name, which causes warnings on some systems. This was basically a
6036 typo, since it was named "eno" in the comments!
6038 4. The code that waits for the clock to tick (at a resolution of some fraction
6039 of a second) so as to ensure message-id uniqueness was always waiting for
6040 at least one whole tick, when it could have waited for less. [This is
6041 almost certainly not relevant at current processor speeds, where it is
6042 unlikely to ever wait at all. But we try to future-proof.]
6044 5. The function that sleeps for a time interval that includes fractions of a
6045 second contained a race. It did not block SIGALRM between setting the
6046 timer, and suspending (a couple of lines later). If the interval was short
6047 and the sigsuspend() was delayed until after it had expired, the suspension
6048 never ended. On busy systems this could lead to processes getting stuck for
6051 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6052 process, before it forks any delivery processes. The open lookup caching
6053 mechanism meant that the open file or database connection was passed into
6054 the delivery process. The problem was that delivery processes always tidy
6055 up cached lookup data. This could cause a problem for the next delivery
6056 process started by the queue runner, because the external queue runner
6057 process does not know about the closure. So the next delivery process
6058 still has data in the lookup cache. In the case of a file lookup, there was
6059 no problem because closing a file descriptor in a subprocess doesn't affect
6060 the parent. However, if the lookup was caching a connection to a database,
6061 the connection was closed, and the second delivery process was likely to
6062 see errors such as "PGSQL: query failed: server closed the connection
6063 unexpectedly". The problem has been fixed by closing all cached lookups
6064 in a queue runner before running a delivery process.
6066 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6067 seem to have the "const" qualifier which it has on other OS. I've
6070 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6071 *of the same type* (client or server) with the same public name that an
6072 error should be diagnosed.
6074 9. When Exim looked up a host name for an IP address, but failed to find the
6075 original IP address when looking up the host name (a safety check), it
6076 output the message "<ip address> does not match any IP for NULL", which was
6077 confusing, to say the least. The bug was that the host name should have
6078 appeared instead of "NULL".
6080 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6081 user that is built into the binary, and the -C or -D options is used, root
6082 privilege is dropped before the configuration file is read. In addition,
6083 logging is switched to stderr instead of the normal log files. If the
6084 configuration then re-defines the Exim user, the unprivileged environment
6085 is probably not what is expected, so Exim logs a panic warning message (but
6088 However, if deliver_drop_privilege is set, the unprivileged state may well
6089 be exactly what is intended, so the warning has been cut out in that case,
6090 and Exim is allowed to try to write to its normal log files.
6096 1. smtp_return_error_details was not giving details for temporary sender
6097 or receiver verification errors.
6099 2. Diagnose a configuration error if two authenticators have the same public
6102 3. Exim used not to create the message log file for a message until the first
6103 delivery attempt. This could be confusing when incoming messages were held
6104 for policy or load reasons. The message log file is now created at the time
6105 the message is received, and an initial "Received" line is written to it.
6107 4. The automatically generated man page for command line options had a minor
6108 bug that caused no ill effects; however, a more serious problem was that
6109 the procedure for building the man page automatically didn't always
6110 operate. Consequently, release 4.20 contains an out-of-date version. This
6111 shouldn't happen again.
6113 5. When building Exim with embedded Perl support, the script that builds the
6114 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6115 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6117 6. The freeze_tell option was not being used for messages that were frozen on
6118 arrival, either by an ACL or by local_scan().
6120 7. Added the smtp_incomplete_transaction log selector.
6122 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6123 was accepting AUTH without a new EHLO.
6125 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6126 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6129 10. Logging of TCP/IP connections (when configured) now happens in the main
6130 daemon process instead of the child process, so that the TCP/IP connection
6131 count is more accurate (but it can never be perfect).
6133 11. The use of "drop" in a nested ACL was not being handled correctly in the
6134 outer ACL. Now, if condition failure induced by the nested "drop" causes
6135 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6136 or "require"), the connection is dropped.
6138 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6139 that yield "discard" can now be used with an "accept" or a "discard" verb,
6140 but an error is generated for any others (because I can't see a useful way
6141 to define what should happen).
6143 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6144 are now processed in the same way as lines in the Exim configuration file.
6145 In particular, continuation lines are supported.
6147 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6149 15. Added -ti meaning -t -i.
6151 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6152 domains, and warn by logging if others are found.
6154 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6155 returned from GIFCONF when Exim is trying to find the list of interfaces on
6156 a host. The code in os.c has been modified to copy each ifreq to an aligned
6157 structure in all cases.
6159 Also, in some cases, the returned ifreq's were being copied to a 'struct
6160 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6161 means the last couple of bytes of an IPv6 address could be chopped if the
6162 ifreq contained only a normal sockaddr (14 bytes storage).
6164 18. Named domain lists were not supported in the hosts_treat_as_local option.
6165 An entry such as +xxxx was not recognized, and was treated as a literal
6168 19. Ensure that header lines added by a DATA ACL are included in the reject log
6169 if the ACL subsequently rejects the message.
6171 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6172 MD5 (which is deprecated).
6174 21. When testing a filter file using -bf, Exim was writing a message when it
6175 took the sender from a "From " line in the message, but it was not doing so
6176 when it took $return_path from a Return-Path: header line. It now does.
6178 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6179 with a valid header line field name (a series of printing characters
6180 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6182 23. Changed "disc" in the source to "disk" to conform to the documentation and
6183 the book and for uniformity.
6185 24. Ignore Sendmail's -Ooption=value command line item.
6187 25. When execve() failed while trying to run a command in a pipe transport,
6188 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6189 could be confused with a return value of 69 from the command itself. This
6190 has been changed to 127, the value the shell returns if it is asked to run
6191 a non-existent command. The wording for the related log line suggests a
6192 non-existent command as the problem.
6194 26. If received_header_text expands to an empty string, do not add a Received:
6195 header line to the message. (Well, it adds a token one on the spool, but
6196 marks it "old" so that it doesn't get used or transmitted.)
6198 27. Installed eximstats 1.28 (addition of -nt option).
6200 28. There was no check for failure on the call to getsockname() in the daemon
6201 code. This can fail if there is a shortage of resources on the system, with
6202 ENOMEM, for example. A temporary error is now given on failure.
6204 29. Contrary to the C standard, it seems that in some environments, the
6205 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6206 program. Exim now does this explicitly; it affects the formatting of
6207 timestamps using strftime().
6209 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6210 complaints. I've now initialized all the variables, to avoid this.
6212 32. Header lines added by a system filter were not being "seen" during
6213 transport-time rewrites.
6215 33. The info_callback() function passed to OpenSSL is set up with type void
6216 (*)(SSL *, int, int), as described somewhere. However, when calling the
6217 function (actually a macro) that sets it up, the type void(*)() is
6218 expected. I've put in a cast to prevent warnings from picky compilers.
6220 34. If a DNS black list lookup found a CNAME record, but there were no A
6221 records associated with the domain it pointed at, Exim crashed.
6223 35. If a DNS black list lookup returned more than one A record, Exim ignored
6224 all but the first. It now scans all returned addresses if a particular IP
6225 value is being sought. In this situation, the contents of the
6226 $dnslist_value variable are a list of all the addresses, separated by a
6229 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6230 to accept a host name and all its aliases if the forward lookup for any of
6231 them yielded the IP address of the incoming connection. Now it accepts only
6232 those names whose forward lookup yields the correct IP address. Any other
6233 names are discarded. This closes a loophole whereby a rogue DNS
6234 administrator could create reverse DNS records to break through a
6235 wildcarded host restriction in an ACL.
6237 37. If a user filter or a system filter that ran in a subprocess used any of
6238 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6239 the wrong values were passed to the pipe command ($thisaddress had the
6240 value of $0, $0 had the value of $1, etc). This bug was introduced by
6241 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6243 38. Improved the line breaking for long SMTP error messages from ACLs.
6244 Previously, if there was no break point between 40 and 75 characters, Exim
6245 left the rest of the message alone. Two changes have been made: (a) I've
6246 reduced the minimum length to 35 characters; (b) if it can't find a break
6247 point between 35 and 75 characters, it looks ahead and uses the first one
6248 that it finds. This may give the occasional overlong line, but at least the
6249 remaining text gets split now.
6251 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6252 file descriptors might be low, and that setting 1000 would always raise it.
6253 It turns out that in some environments, the limit is already over 1000 and
6254 that lowering it causes trouble. So now Exim takes care not to decrease it.
6256 40. When delivering a message, the value of $return_path is set to $sender_
6257 address at the start of routing (routers may change the value). By an
6258 oversight, this default was not being set up when an address was tested by
6259 -bt or -bv, which affected the outcome if any router or filter referred to
6262 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6263 log only when "message" or "log_message" are set. However, if one of the
6264 conditions was an address verification, or a call to a nested ACL, the
6265 messages generated by the underlying test were being passed through. This
6266 no longer happens. The underlying message is available in $acl_verify_
6267 message for both "message" and "log_message" expansions, so it can be
6268 passed through if needed.
6270 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6271 new expansion $bh_ to give the encoded byte string without charset
6272 translation. Translation happens only if iconv() is available; HAVE_ICONV
6273 indicates this at build time. HEADERS_CHARSET gives the charset to
6274 translate to; headers_charset can change it in the configuration, and
6275 "headers charset" can change it in an individual filter file.
6277 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6278 that creates RFC 2047 encoded "words" labels them as that charset instead
6279 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6280 expansion operator; (ii) when Exim creates a From: line for a local
6281 message; (iii) when a header line is rewritten to include a "phrase" part.
6283 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6284 buggy, causing it to skip the first lines of messages whose message ID
6285 ended in 'D'. This would not have bitten before Exim release 4.14, because
6286 message IDs were unlikely to end in 'D' before then. The effect was to have
6287 incorrect size information for certain domains.
6289 45. #include "config.h" was missing at the start of the crypt16.c module. This
6290 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6293 46. If there was a timeout during a "random" callout check, Exim treated it as
6294 a failure of the random address, and carried on sending RSET and the real
6295 address. If the delay was just some slowness somewhere, the response to the
6296 original RCPT would be taken as a response to RSET and so on, causing
6297 mayhem of various kinds.
6299 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6300 when I implemented it. It didn't allow for the fact that some option values
6301 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6302 the right test for positive values.
6304 48. Domain names in DNS records are case-independent. Exim always looks them up
6305 in lower case. Some resolvers return domain names in exactly the case they
6306 appear in the zone file, that is, they may contain uppercase letters. Not
6307 all resolvers do this - some return always lower case. Exim was treating a
6308 change of case by a resolver as a change of domain, similar to a widening
6309 of a domain abbreviation. This triggered its re-routing code and so it was
6310 trying to route what was effectively the same domain again. This normally
6311 caused routing to fail (because the router wouldn't handle the domain
6312 twice). Now Exim checks for this case specially, and just changes the
6313 casing of the domain that it ultimately uses when it transmits the message
6316 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6319 50. If a filter generated a file delivery with a non-absolute name (possible if
6320 no home directory exists for the router), the forbid_file option was not
6323 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6324 the existing equality matching.
6326 52. Exim was using ints instead of ino_t variables in some places where it was
6327 dealing with inode numbers.
6329 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6330 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6331 called TMPDIR, and if it finds it is different, it changes its value.
6333 54. The smtp_printf() function is now made available to local_scan() so
6334 additional output lines can be written before returning. There is also an
6335 smtp_fflush() function to enable the detection of a dropped connection.
6336 The variables smtp_input and smtp_batched_input are exported to
6339 55. Changed the default runtime configuration: the message "Unknown user"
6340 has been removed from the ACL, and instead placed on the localuser router,
6341 using the cannot_route_message feature. This means that any verification
6342 failures that generate their own messages won't get overridden. Similarly,
6343 the "Unrouteable address" message that was in the ACL for unverifiable
6344 relay addresses has also been removed.
6346 56. Added hosts_avoid_esmtp to the smtp transport.
6348 57. The exicyclog script was not checking for the esoteric option
6349 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6350 will work only if exicyclog is run under the appropriate euid.
6352 58. Following a discussion on the list, the rules by which Exim recognises line
6353 endings on incoming messages have been changed. The -dropcr and drop_cr
6354 options are now no-ops, retained only for backwards compatibility. The
6355 following line terminators are recognized: LF CRLF CR. However, special
6356 processing applies to CR:
6358 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6359 nor a local message in the state where . is a terminator.
6361 (ii) If a bare CR is encountered in a header line, an extra space is added
6362 after the line terminator so as not to end the header. The reasoning
6363 behind this is that bare CRs in header lines are most likely either
6364 to be mistakes, or people trying to play silly games.
6366 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6367 was being incorrectly given as 18 bytes larger than it should have been.
6368 This is a VOB (very old bug).
6370 60. This may never have affected anything current, but just in case it has:
6371 When the local host is found other than at the start of a list of hosts,
6372 the local host, those with the same MX, and any that follow, are discarded.
6373 When the list in question was part of a longer list of hosts, the following
6374 hosts (not currently being processed) were also being discarded. This no
6375 longer happens. I'm not sure if this situation could ever has previously
6378 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6381 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6382 header. This is something that is recommended in a new Internet Draft, and
6383 is something that is documented as being done by Sendmail. There are two
6384 possible values. For messages generated by the autoreply transport, Exim
6387 Auto-Submitted: auto-replied
6389 whereas for all other generated messages (e.g. bounces) it adds
6391 Auto-Submitted: auto-generated
6393 63. The "personal" condition in filters now includes a test for the
6394 Auto-Submitted: header. If it contains the string "auto-" the message it
6395 not considered personal.
6397 64. Added rcpt_include_affixes as a generic transport option.
6399 65. Added queue_only_override (default true).
6401 66. Added the syslog_duplication option.
6403 67. If what should have been the first header line of a message consisted of
6404 a space followed by a colon, Exim was mis-interpreting it as a header line.
6405 It isn't of course - it is syntactically invalid and should therefore be
6406 treated as the start of the message body. The misbehaviour could have
6407 caused a number of strange effects, including loss of data in subsequent
6408 header lines, and spool format errors.
6410 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6411 client host had authenticated. This control can now be exercised by an ACL
6412 for more flexibility.
6414 69. By default, callouts do not happen when testing with -bh. There is now a
6415 variant, -bhc, which does actually run the callout code, including
6416 consulting and updating the callout cache.
6418 70. Added support for saslauthd authentication, courtesy of Alexander
6421 71. If statvfs() failed on the spool or log directories while checking their
6422 size for availability, Exim confusingly gave the error "space shortage".
6423 Furthermore, in debugging mode it crashed with a floating point exception.
6424 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6425 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6426 serious problem, Exim now writes to the main and panic logs when this
6427 happens, with details of the failure. It then refuses to accept the
6428 incoming message, giving the message "spool directory problem" or "log
6429 directory problem" with a 421 code for SMTP messages.
6431 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6432 0, 1, and 2 exist, because some OS complain for execs without them (see
6433 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6434 descriptors. However, the code omitted to check that the open succeeded,
6435 causing mysterious errors if for some reason the permissions on /dev/null
6436 got screwed. Now Exim writes a message to the main and panic logs, and
6437 bombs out if it can't open /dev/null.
6439 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6440 interact so that it is all more flexible. It is supposed to remain
6441 backwards compatible. Also added extra_local_interfaces.
6443 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6444 to bomb out with an assertion failure - to the client this appears as a
6445 connection drop. This problem occurs in the part of the code that was taken
6446 from the Samba project. Fortunately, the assertion is in a very simple
6447 function, so I have fixed this by reproducing the function inline in the
6448 one place where it is called, and arranging for authentication to fail
6449 instead of killing the process with assert().
6451 75. The SPA client code was not working when the server requested OEM rather
6452 than Unicode encoding.
6454 76. Added code to make require_files with a specific uid setting more usable in
6455 the case where statting the file as root fails - usually a non-root-mounted
6456 NFS file system. When this happens and the failure is EACCES, Exim now
6457 forks a subprocess and does the per-uid checking as the relevant uid.
6459 77. Added process_log_path.
6461 78. If log_file_path was not explicitly set, a setting of check_log_space or
6462 check_log_inodes was ignored.
6464 79. If a space check for the spool or log partitions fails, the incident is now
6465 logged. Of course, in the latter case the data may get lost...
6467 80. Added the %p formatting code to string_format() so that it can be used to
6468 print addresses in debug_print(). Adjusted all the address printing in the
6469 debugging in store.c to use %p rather than %d.
6471 81. There was a concern that a line of code in smtp_in.c could overflow a
6472 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6473 initially expressed, the concern was not well-founded, because trailing
6474 spaces are removed early. However, if the trailing spaces were followed by
6475 a NULL, they did not get removed, so the overflow was possible. Two fixes
6478 (a) I re-wrote the offending code in a cleaner fashion.
6479 (b) If an incoming SMTP command contains a NULL character, it is rejected
6482 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6483 runs initgroups(), so that if the Exim user is in any additional groups,
6484 they will be used during message reception.
6490 The change log for 4.20 and earlier releases has been archived.