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.
102 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
105 JH/02 The smtp transport option "multi_domain" is now expanded.
107 JH/03 The smtp transport now requests PRDR by default, if the server offers
110 JH/04 Certificate name checking on server certificates, when exim is a client,
111 is now done by default. The transport option tls_verify_cert_hostnames
112 can be used to disable this per-host. The build option
113 EXPERIMENTAL_CERTNAMES is withdrawn.
115 JH/05 The value of the tls_verify_certificates smtp transport and main options
116 default to the word "system" to access the system default CA bundle.
117 For GnuTLS, only version 3.0.20 or later.
119 JH/06 Verification of the server certificate for a TLS connection is now tried
120 (but not required) by default. The verification status is now logged by
121 default, for both outbound TLS and client-certificate supplying inbound
124 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
127 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
128 Status Notification (bounce) messages are now MIME format per RFC 3464.
129 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
130 under the control of the dsn_advertise_hosts option, and routers may
131 have a dsn_lasthop option.
133 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
134 default, modifiable by a malware= option. The list separator for
135 the options can now be changed in the usual way. Bug 68.
137 JH/10 The smtp_receive_timeout main option is now expanded before use.
139 JH/11 The incoming_interface log option now also enables logging of the
140 local interface on delivery outgoing connections.
142 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
143 if the interface and destination host and port all match.
145 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
148 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
149 Patch from Andrew Lewis.
151 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
152 now supports optional time-restrictions, weighting, and priority
153 modifiers per server. Patch originally by <rommer@active.by>.
155 JH/16 The spamd_address main option now supports a mixed list of local
156 and remote servers. Remote servers can be IPv6 addresses, and
157 specify a port-range.
159 JH/17 Bug 68: The spamd_address main option now supports an optional
160 timeout value per server.
162 JH/18 Bug 1581: Router and transport options headers_add/remove can
163 now have the list separator specified.
165 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
168 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
171 JH/21 Support for the A6 type of dns record is withdrawn.
173 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
174 rather than the verbs used.
176 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
177 from 255 to 1024 chars.
179 JH/24 Verification callouts now attempt to use TLS by default.
181 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
182 are generic router options now. The defaults didn't change.
184 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
185 Original patch from Alexander Shikoff, worked over by JH.
187 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
188 files if ZCAT_COMMAND is not executable.
190 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
192 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
194 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
195 Normally benign, it bites when the pair was led to by a CNAME;
196 modern usage is to not canoicalize the domain to a CNAME target
197 (and we were inconsistent anyway for A-only vs AAAA+A).
199 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
201 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
202 when evaluating $sender_host_dnssec.
204 JH/31 Check the HELO verification lookup for DNSSEC, adding new
205 $sender_helo_dnssec variable.
207 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
209 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
211 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
213 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
214 documented as working, but never had. Support all but $spam_report.
216 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
217 added for tls authenticator.
222 TL/01 When running the test suite, the README says that variables such as
223 no_msglog_check are global and can be placed anywhere in a specific
224 test's script, however it was observed that placement needed to be near
225 the beginning for it to behave that way. Changed the runtest perl
226 script to read through the entire script once to detect and set these
227 variables, reset to the beginning of the script, and then run through
228 the script parsing/test process like normal.
230 TL/02 The BSD's have an arc4random API. One of the functions to induce
231 adding randomness was arc4random_stir(), but it has been removed in
232 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
233 function when detected.
235 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
236 cause callback expansion.
238 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
239 syntax errors in an expansion can be treated as a string instead of
240 logging or causing an error, due to the internal use of bool_lax
241 instead of bool when processing it.
243 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
244 server certificates when making smtp deliveries.
246 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
248 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
250 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
252 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
253 Merged patch from Sebastian Wiedenroth.
255 JH/05 Fix results-pipe from transport process. Several recipients, combined
256 with certificate use, exposed issues where response data items split
257 over buffer boundaries were not parsed properly. This eventually
258 resulted in duplicates being sent. This issue only became common enough
259 to notice due to the introduction of conection certificate information,
260 the item size being so much larger. Found and fixed by Wolfgang Breyha.
262 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
263 size buffer was used, resulting in syntax errors when an expansion
266 JH/07 Add support for directories of certificates when compiled with a GnuTLS
267 version 3.3.6 or later.
269 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
270 is EXPERIMENTAL_EVENT, the main-configuration and transport options
271 both become "event_action", the variables become $event_name, $event_data
272 and $event_defer_errno. There is a new variable $verify_mode, usable in
273 routers, transports and related events. The tls:cert event is now also
274 raised for inbound connections, if the main configuration event_action
277 TL/06 In test suite, disable OCSP for old versions of openssl which contained
278 early OCSP support, but no stapling (appears to be less than 1.0.0).
280 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
281 server certificate names available under the smtp transport option
282 "tls_verify_cert_hostname" now do not permit multi-component wildcard
285 JH/10 Time-related extraction expansions from certificates now use the main
286 option "timezone" setting for output formatting, and are consistent
287 between OpenSSL and GnuTLS compilations. Bug 1541.
289 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
290 encoded parameter in the incoming message. Bug 1558.
292 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
293 include certificate info, eximon was claiming there were spoolfile
296 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
298 JH/14 Log delivery-related information more consistently, using the sequence
299 "H=<name> [<ip>]" wherever possible.
301 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
302 are problematic for Debian distribution, omit them from the release
305 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
307 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
310 JH/17 Fix a null-indirection in certextract expansions when a nondefault
311 output list separator was used.
316 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
317 checkers that were complaining about end of non-void function with no
320 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
321 This was a regression intruduced in 4.83 by another bugfix.
323 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
325 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
326 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
332 TF/01 Correctly close the server side of TLS when forking for delivery.
334 When a message was received over SMTP with TLS, Exim failed to clear up
335 the incoming connection properly after forking off the child process to
336 deliver the message. In some situations the subsequent outgoing
337 delivery connection happened to have the same fd number as the incoming
338 connection previously had. Exim would try to use TLS and fail, logging
339 a "Bad file descriptor" error.
341 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
342 utilities have not been installed.
344 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
345 temporary space as the ACL may create new global variables.
347 TL/01 LDAP support uses per connection or global context settings, depending
348 upon the detected version of the libraries at build time.
350 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
351 to extract and use the src ip:port in logging and expansions as if it
352 were a direct connection from the outside internet. PPv2 support was
353 updated based on HAProxy spec change in May 2014.
355 JH/02 Add ${listextract {number}{list}{success}{fail}}.
357 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
358 Properly escape header and check for NULL return.
360 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
363 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
365 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
366 characters in header names, implemented as a verify condition.
367 Contributed by Michael Fischer v. Mollard.
369 TL/05 Rename SPF condition results err_perm and err_temp to standardized
370 results permerror and temperror. Previous values are deprecated but
371 still accepted. In a future release, err_perm and err_temp will be
372 completely removed, which will be a backward incompatibility if the
373 ACL tests for either of these two old results. Patch contributed by
374 user bes-internal on the mailing list.
376 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
378 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
379 selectors, in both main and reject logs.
381 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
384 JH/07 Add malware type "sock" for talking to simple daemon.
386 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
388 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
389 routers/transports under cutthrough routing.
391 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
392 numbers. Touch up "bool" conditional to keep the same definition.
394 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
396 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
398 JH/12 Expand items in router/transport headers_add or headers_remove lists
399 individually rather than the list as a whole. Bug 1452.
401 Required for reasonable handling of multiple headers_ options when
402 they may be empty; requires that headers_remove items with embedded
403 colons must have them doubled (or the list-separator changed).
405 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
406 view the policy declared in the DMARC record. Currently, $dmarc_status
407 is a combined value of both the record presence and the result of the
410 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
412 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
413 dnslookup router and the smtp transport (applying to the forward
416 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
417 of ldap servers used for a specific lookup. Patch provided by Heiko
420 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
421 New variable $lookup_dnssec_authenticated for observability.
423 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
424 Patch submitted by Lars Timman.
426 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
428 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
429 Requires trusted mode and valid format message id, aborts otherwise.
430 Patch contributed by Heiko Schlichting.
432 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
433 certextract with support for various fields. Bug 1358.
435 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
436 is requested by default, modifiable by smtp transport option
439 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
440 operate on certificate variables to give certificate fingerprints
441 Also new ${sha256:cert_variable}.
443 JH/23 The PRDR feature is moved from being Experimental into the mainline.
445 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
446 Christian Aistleitner.
448 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
450 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
451 file. Patch from Wolfgang Breyha.
453 JH/25 Expand the coverage of the delivery $host and $host_address to
454 client authenticators run in verify callout. Bug 1476.
456 JH/26 Port service names are now accepted for tls_on_connect_ports, to
457 align with daemon_smtp_ports. Bug 72.
459 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
460 support and error reporting did not work properly.
462 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
463 and is readable. Patch from Andrew Colin Kissa.
465 TL/14 Enhance documentation of ${run expansion and how it parses the
466 commandline after expansion, particularly in the case when an
467 unquoted variable expansion results in an empty value.
469 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
471 PP/02 Fix internal collision of T_APL on systems which support RFC3123
472 by renaming away from it. Addresses GH issue 15, reported by
475 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
477 TL/15 SECURITY: prevent double expansion in math comparison functions
478 (can expand unsanitized data). Not remotely exploitable.
485 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
487 PP/02 Make -n do something, by making it not do something.
488 When combined with -bP, the name of an option is not output.
490 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
493 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
494 $sender_host_name and config options to manage this, and basic check
497 PP/05 DSCP support for outbound connections and control modifier for inbound.
499 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
500 (Only plugin which currently uses this is kerberos4, which nobody should
501 be using, but we should make it available and other future plugins might
502 conceivably use it, even though it would break NAT; stuff *should* be
503 using channel bindings instead).
505 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
506 name; added for Sendmail compatibility; requires admin caller.
507 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
508 just ignore it); requires trusted caller.
509 Also parse but ignore: -Ac -Am -X<logfile>
512 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
514 TL/02 Add +smtp_confirmation as a default logging option.
516 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
517 Patch by Magnus Holmgren from 2007-02-20.
519 TL/04 Bugzilla 1281 - Spec typo.
520 Bugzilla 1283 - Spec typo.
521 Bugzilla 1290 - Spec grammar fixes.
523 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
525 TL/06 Add Experimental DMARC support using libopendmarc libraries.
527 TL/07 Fix an out of order global option causing a segfault. Reported to dev
528 mailing list by by Dmitry Isaikin.
530 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
532 JH/02 Support "G" suffix to numbers in ${if comparisons.
534 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
536 NM/01 Bugzilla 1197 - Spec typo
537 Bugzilla 1196 - Spec examples corrections
539 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
541 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
542 gnutls_enable_pkcs11, but renamed to more accurately indicate its
545 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
546 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
548 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
549 "acl {{name}{arg}...}", and optional args on acl condition
552 JH/05 Permit multiple router/transport headers_add/remove lines.
554 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
556 JH/07 Avoid using a waiting database for a single-message-only transport.
557 Performance patch from Paul Fisher. Bugzilla 1262.
559 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
562 JH/09 Add $headers_added variable, with content from use of ACL modifier
563 add_header (but not yet added to the message). Bugzilla 199.
565 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
566 Pulled from Bugzilla 817 by Wolfgang Breyha.
568 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
570 (nb: this is the same fix as in Exim 4.80.1)
572 JH/11 Add A= logging on delivery lines, and a client_set_id option on
575 JH/12 Add optional authenticated_sender logging to A= and a log_selector
578 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
580 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
581 advertise SMTP AUTH mechanism to us, instead of a generic
582 protocol violation error. Also, make Exim more robust to bad
583 data from the Dovecot auth socket.
585 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
587 When a queue runner is handling a message, Exim first routes the
588 recipient addresses, during which it prunes them based on the retry
589 hints database. After that it attempts to deliver the message to
590 any remaining recipients. It then updates the hints database using
593 So if a recipient address works intermittently, it can get repeatedly
594 deferred at routing time. The retry hints record remains fresh so the
595 address never reaches the final cutoff time.
597 This is a fairly common occurrence when a user is bumping up against
598 their storage quota. Exim had some logic in its local delivery code
599 to deal with this. However it did not apply to per-recipient defers
600 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
602 This change adds a proper retry rule check during routing so that the
603 final cutoff time is checked against the message's age. We only do
604 this check if there is an address retry record and there is not a
605 domain retry record; this implies that previous attempts to handle
606 the address had the retry_use_local_parts option turned on. We use
607 this as an approximation for the destination being like a local
608 delivery, as in LMTP.
610 I suspect this new check makes the old local delivery cutoff check
611 redundant, but I have not verified this so I left the code in place.
613 TF/02 Correct gecos expansion when From: is a prefix of the username.
615 Test 0254 submits a message to Exim with the header
619 When I ran the test suite under the user fanf2, Exim expanded
620 the header to contain my full name, whereas it should have added
621 a Resent-Sender: header. It erroneously treats any prefix of the
622 username as equal to the username.
624 This change corrects that bug.
626 GF/01 DCC debug and logging tidyup
627 Error conditions log to paniclog rather than rejectlog.
628 Debug lines prefixed by "DCC: " to remove any ambiguity.
630 TF/03 Avoid unnecessary rebuilds of lookup-related code.
632 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
633 Bug spotted by Jeremy Harris; was flawed since initial commit.
634 Would have resulted in OCSP responses post-SNI triggering an Exim
635 NULL dereference and crash.
637 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
639 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
640 Bug detection, analysis and fix by Samuel Thibault.
641 Bugzilla 1331, Debian bug #698092.
643 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
645 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
646 Server implementation by Todd Lyons, client by JH.
647 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
648 config variable "prdr_enable" controls whether the server
649 advertises the facility. If the client requests PRDR a new
650 acl_data_smtp_prdr ACL is called once for each recipient, after
651 the body content is received and before the acl_smtp_data ACL.
652 The client is controlled by bolth of: a hosts_try_prdr option
653 on the smtp transport, and the server advertisement.
654 Default client logging of deliveries and rejections involving
655 PRDR are flagged with the string "PRDR".
657 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
658 fclose(). Diagnosis by Todd Lyons.
660 PP/17 Update configure.default to handle IPv6 localhost better.
661 Patch by Alain Williams (plus minor tweaks).
664 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
665 This is now consistent with GnuTLS, and is now documented: the
666 previous undocumented portable approach to treating the option as
667 unset was to force an expansion failure. That still works, and
668 an empty string is now equivalent.
670 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
671 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
672 not performing validation itself.
674 PP/20 Added force_command boolean option to pipe transport.
675 Patch from Nick Koston, of cPanel Inc.
677 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
680 TF/04 Added udpsend ACL modifer and hexquote expansion operator
682 PP/21 Fix eximon continuous updating with timestamped log-files.
683 Broken in a format-string cleanup in 4.80, missed when I repaired the
684 other false fix of the same issue.
685 Report and fix from Heiko Schlichting.
688 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
689 Report from Prashanth Katuri.
691 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
692 It's SecureTransport, so affects any MacOS clients which use the
693 system-integrated TLS libraries, including email clients.
695 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
696 using a MIME ACL for non-SMTP local injection.
697 Report and assistance in diagnosis by Warren Baker.
699 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
701 JH/16 Fix comparisons for 64b. Bugzilla 1385.
703 TL/09 Add expansion variable $authenticated_fail_id to keep track of
704 last id that failed so it may be referenced in subsequent ACL's.
706 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
709 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
710 ldap library initialization, allowing self-signed CA's to be
711 used. Also properly sets require_cert option later in code by
712 using NULL (global ldap config) instead of ldap handle (per
713 session). Bug diagnosis and testing by alxgomz.
715 TL/12 Enhanced documentation in the ratelimit.pl script provided in
716 the src/util/ subdirectory.
718 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
719 renamed to Transport Post Delivery Action by Jeremy Harris, as
722 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
723 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
724 redis_servers = needs to be configured which will be used by the redis
725 lookup. Patch from Warren Baker, of The Packet Hub.
727 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
729 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
730 hostname or reverse DNS when processing a host list. Used suggestions
731 from multiple comments on this bug.
733 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
735 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
736 Missed a few lines, added it to make the runtest require no keyboard
739 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
740 contains upper case chars. Make router use caseful_local_part.
742 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
743 support when GnuTLS has been built with p11-kit.
749 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
751 This, or similar/improved, will also be change PP/11 of 4.82.
757 PP/01 Handle short writes when writing local log-files.
758 In practice, only affects FreeBSD (8 onwards).
759 Bugzilla 1053, with thanks to Dmitry Isaikin.
761 NM/01 Bugzilla 949 - Documentation tweak
763 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
766 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
768 PP/02 Implemented gsasl authenticator.
770 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
772 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
773 `pkg-config foo` for cflags/libs.
775 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
776 with rest of GSASL and with heimdal_gssapi.
778 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
779 `pkg-config foo` for cflags/libs for the TLS implementation.
781 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
782 properties get this fed in as external SSF. A number of robustness
783 and debugging improvements to the cyrus_sasl authenticator.
785 PP/08 cyrus_sasl server now expands the server_realm option.
787 PP/09 Bugzilla 1214 - Log authentication information in reject log.
788 Patch by Jeremy Harris.
790 PP/10 Added dbmjz lookup type.
792 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
794 PP/12 MAIL args handles TAB as well as SP, for better interop with
795 non-compliant senders.
796 Analysis and variant patch by Todd Lyons.
798 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
799 Bug report from Lars Müller <lars@samba.org> (via SUSE),
800 Patch from Dirk Mueller <dmueller@suse.com>
802 PP/13 tls_peerdn now print-escaped for spool files.
803 Observed some $tls_peerdn in wild which contained \n, which resulted
804 in spool file corruption.
806 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
807 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
808 or write after TLS renegotiation, which otherwise led to messages
811 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
812 as a tracking header (ie: a signed header comes before the signature).
813 Patch from Wolfgang Breyha.
815 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
816 comma-sep list; embedded commas doubled.
818 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
820 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
822 Report and patch from Dmitry Banschikov.
824 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
825 Removed SSL_clear() after SSL_new() which led to protocol negotiation
826 failures. We appear to now support TLS1.1+ with Exim.
828 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
829 lets Exim select keys and certificates based upon TLS SNI from client.
830 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
831 before an outbound SMTP session. New log_selector, +tls_sni.
833 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
834 NULL dereference. Report and patch from Alun Jones.
836 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
837 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
838 Not seeing resolver debug output on NetBSD, but suspect this is a
839 resolver implementation change.
841 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
842 Left warnings. Added "eximon gdb" invocation mode.
844 PP/21 Defaulting "accept_8bitmime" to true, not false.
846 PP/22 Added -bw for inetd wait mode support.
848 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
849 locate the relevant includes and libraries. Made this the default.
851 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
852 Bugzilla 1246, report and most of solution from Tomasz Kusy.
854 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
855 This may cause build issues on older platforms.
857 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
858 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
859 gnutls_require_mac & gnutls_require_protocols (no longer supported).
860 Added SNI support via GnuTLS too.
861 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
863 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
865 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
866 Applied second patch from Janne, implementing suggestion to default
867 multiple-strings-in-record handling to match SPF spec.
869 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
871 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
872 read-only, out of scope).
873 Patch from Wolfgang Breyha, report from Stuart Northfield.
875 PP/29 Fix three issues highlighted by clang analyser static analysis.
876 Only crash-plausible issue would require the Cambridge-specific
877 iplookup router and a misconfiguration.
878 Report from Marcin Mirosław.
880 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
882 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
883 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
884 As part of this, removing so much warning spew let me fix some minor
885 real issues in debug logging.
887 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
888 assignment on my part. Fixed.
890 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
891 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
892 Janne Snabb (who went above and beyond: thank you).
894 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
895 string otherwise requires a connection and a bunch more work and it's
896 relatively easy to get wrong. Should also expose TLS library linkage
899 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
900 64-bit ${eval} (JH/03).
902 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
903 GNU libc to support some of the 64-bit stuff, should not lead to
904 conflicts. Defined before os.h is pulled in, so if a given platform
905 needs to override this, it can.
907 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
908 protection layer was required, which is not implemented.
909 Bugzilla 1254, patch from Wolfgang Breyha.
911 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
912 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
913 tls_dhparam take prime identifiers. Also unbreak combination of
914 OpenSSL+DH_params+TLSSNI.
916 PP/39 Disable SSLv2 by default in OpenSSL support.
922 PP/01 Solaris build fix for Oracle's LDAP libraries.
923 Bugzilla 1109, patch from Stephen Usher.
925 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
927 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
930 TF/02 Fix a couple more cases where we did not log the error message
931 when unlink() failed. See also change 4.74-TF/03.
933 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
934 lock up or crash if it happened to be inside a call to libc when it
935 got a SIGUSR1 from exiwhat.
937 The SIGUSR1 handler appends the current process status to the process
938 log which is later printed by exiwhat. It used to use the general
939 purpose logging code to do this, but several functions it calls are
940 not safe for signals.
942 The new output code in the SIGUSR1 handler is specific to the process
943 log, and simple enough that it's easy to inspect for signal safety.
944 Removing some special cases also simplifies the general logging code.
945 Removing the spurious timestamps from the process log simplifies
948 TF/04 Improved ratelimit ACL condition.
950 The /noupdate option has been deprecated in favour of /readonly which
951 has clearer semantics. The /leaky, /strict, and /readonly update modes
952 are mutually exclusive. The update mode is no longer included in the
953 database key; it just determines when the database is updated. (This
954 means that when you upgrde Exim will forget old rate measurements.)
956 Exim now checks that the per_* options are used with an update mode that
957 makes sense for the current ACL. For example, when Exim is processing a
958 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
959 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
960 must specify per_mail/readonly. If you omit the update mode it defaults to
961 /leaky where that makes sense (as before) or /readonly where required.
963 The /noupdate option is now undocumented but still supported for
964 backwards compatibility. It is equivalent to /readonly except that in
965 ACLs where /readonly is required you may specify /leaky/noupdate or
966 /strict/noupdate which are treated the same as /readonly.
968 A useful new feature is the /count= option. This is a generalization
969 of the per_byte option, so that you can measure the throughput of other
970 aggregate values. For example, the per_byte option is now equivalent
971 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
973 The per_rcpt option has been generalized using the /count= mechanism
974 (though it's more complicated than the per_byte equivalence). When it is
975 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
976 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
977 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
978 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
979 Note that using per_rcpt with a non-readonly update mode in more than
980 one ACL will cause the recipients to be double-counted. (The per_mail
981 and per_byte options don't have this problem.)
983 The handling of very low rates has changed slightly. If the computed rate
984 is less than the event's count (usually one) then this event is the first
985 after a long gap. In this case the rate is set to the same as this event's
986 count, so that the first message of a spam run is counted properly.
988 The major new feature is a mechanism for counting the rate of unique
989 events. The new per_addr option counts the number of different
990 recipients that someone has sent messages to in the last time period. It
991 behaves like per_rcpt if all the recipient addresses are different, but
992 duplicate recipient addresses do not increase the measured rate. Like
993 the /count= option this is a general mechanism, so the per_addr option
994 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
995 example, measure the rate that a client uses different sender addresses
996 with the options per_mail/unique=$sender_address. There are further
997 details in the main documentation.
999 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1001 TF/06 Removed a few PCRE remnants.
1003 TF/07 Automatically extract Exim's version number from tags in the git
1004 repository when doing development or release builds.
1006 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1007 Bugzilla 879. Patch from Paul Fisher.
1009 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1010 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1013 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1015 PP/05 Variable $av_failed, true if the AV scanner deferred.
1016 Bugzilla 1078. Patch from John Horne.
1018 PP/06 Stop make process more reliably on build failure.
1019 Bugzilla 1087. Patch from Heiko Schlittermann.
1021 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1022 Bugzilla 1089. Patch from Heiko Schlittermann.
1024 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1025 Bugzilla 1131. Patch from Holger Weiß.
1027 PP/09 Handle IPv6 addresses with SPF.
1028 Bugzilla 860. Patch from Wolfgang Breyha.
1030 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1032 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1035 PP/11 match_* no longer expand right-hand-side by default.
1036 New compile-time build option, EXPAND_LISTMATCH_RHS.
1037 New expansion conditions, "inlist", "inlisti".
1039 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1041 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1043 PP/14 fix log_write() format string regression from TF/03.
1044 Bugzilla 1152. Patch from Dmitry Isaikin.
1050 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1052 PP/02 Harmonised TLS library version reporting; only show if debugging.
1053 Layout now matches that introduced for other libraries in 4.74 PP/03.
1055 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1057 PP/04 New "dns_use_edns0" global option.
1059 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1062 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1063 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1065 TK/01 Updated PolarSSL code to 0.14.2.
1066 Bugzilla 1097. Patch from Andreas Metzler.
1068 PP/07 Catch divide-by-zero in ${eval:...}.
1069 Fixes bugzilla 1102.
1071 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1074 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1075 format-string attack -- SECURITY: remote arbitrary code execution.
1077 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1078 time unintentionally subject to list matching rules, letting the header
1079 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1080 arbitrary string expansion). This allowed for information disclosure.
1082 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1083 INT_MIN/-1 -- value coerced to INT_MAX.
1089 NM/01 Workround for PCRE version dependency in version reporting
1092 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1093 This fixes portability to compilers other than gcc, notably
1094 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1096 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1097 makefiles for portability to HP-UX and POSIX correctness.
1099 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1100 Also via indented variable definition in the Makefile.
1101 (Debugging by Oliver Heesakkers).
1103 PP/02 Restore caching of spamd results with expanded spamd_address.
1104 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1106 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1107 Improves build reliability. Fix from: Frank Elsner
1109 NM/02 Fix wide character breakage in the rfc2047 coding
1110 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1112 NM/03 Allow underscore in dnslist lookups
1113 Fixes bug 1026. Patch from Graeme Fowler
1115 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1116 Code patches from Adam Ciarcinski of NetBSD.
1118 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1121 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1122 is logged, to avoid truncation. Patch from John Horne.
1124 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1125 Patch from Jakob Hirsch.
1127 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1128 SQL string expansion failure details.
1129 Patch from Andrey Oktyabrski.
1131 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1132 Patch from Simon Arlott.
1134 PP/09 New lookups functionality failed to compile on old gcc which rejects
1135 extern declarations in function scope.
1136 Patch from Oliver Fleischmann
1138 PP/10 Use sig_atomic_t for flags set from signal handlers.
1139 Check getgroups() return and improve debugging.
1140 Fixed developed for diagnosis in bug 927 (which turned out to be
1143 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1144 Patch from Mark Zealey.
1146 PP/12 Bugzilla 1056: Improved spamd server selection.
1147 Patch from Mark Zealey.
1149 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1150 Based on patch from Heiko Schlittermann.
1152 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1153 Patch from Uwe Doering, sign-off by Michael Haardt.
1155 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1156 variable declaration deep within a block. Bug and patch from
1159 PP/15 lookups-Makefile IRIX compatibilty coercion.
1161 PP/16 Make DISABLE_DKIM build knob functional.
1163 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1164 Patch by Simon Arlott
1166 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1167 variable argument macros. Our copy now differs from upstream.
1173 TF/01 Failure to get a lock on a hints database can have serious
1174 consequences so log it to the panic log.
1176 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1177 controlled using the smtp_confirmation log selector.
1179 TF/03 Include the error message when we fail to unlink a spool file.
1181 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1182 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1183 for maintaining out-of-tree patches for some time.
1185 PP/01 Bugzilla 139: Documentation and portability issues.
1186 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1187 Handle per-OS dynamic-module compilation flags.
1189 PP/02 Let /dev/null have normal permissions.
1190 The 4.73 fixes were a little too stringent and complained about the
1191 permissions on /dev/null. Exempt it from some checks.
1192 Reported by Andreas M. Kirchwitz.
1194 PP/03 Report version information for many libraries, including
1195 Exim version information for dynamically loaded libraries. Created
1196 version.h, now support a version extension string for distributors
1197 who patch heavily. Dynamic module ABI change.
1199 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1200 privilege escalation vulnerability whereby the Exim run-time user
1201 can cause root to append content of the attacker's choosing to
1204 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1207 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1208 If dropping privileges for untrusted macros, we disabled normal logging
1209 on the basis that it would fail; for the Exim run-time user, this is not
1210 the case, and it resulted in successful deliveries going unlogged.
1211 Fixed. Reported by Andreas Metzler.
1217 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1218 only prepend for the Resent-* case. Fixes regression introduced in
1219 Exim 4.70 by NM/22 for Bugzilla 607.
1221 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1222 increasing numbers of administrators be bitten by this.
1224 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1226 PP/03 Bugzilla 994: added openssl_options main configuration option.
1228 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1230 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1232 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1234 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1235 without a peer certificate, leading to a segfault because of an
1236 assumption that peers always have certificates. Be a little more
1237 paranoid. Problem reported by Martin Tscholak.
1239 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1240 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1241 NB: ClamAV planning to remove STREAM in "middle of 2010".
1242 CL also introduces -bmalware, various -d+acl logging additions and
1243 more caution in buffer sizes.
1245 PP/09 Implemented reverse_ip expansion operator.
1247 PP/10 Bugzilla 937: provide a "debug" ACL control.
1249 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1251 PP/12 Bugzilla 973: Implement --version.
1253 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1255 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1257 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1259 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1260 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1261 ignore trailing whitespace.
1263 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1265 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1268 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1269 Notification from Dr Andrew Aitchison.
1271 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1272 ExtendedDetectionInfo response format.
1273 Notification from John Horne.
1275 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1278 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1279 XSL and documented dependency on system catalogs, with examples of how
1282 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1285 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1286 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1287 configuration file which is writeable by the Exim user or group.
1289 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1290 of configuration files to cover files specified with the -C option if
1291 they are going to be used with root privileges, not just the default
1294 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1295 option (effectively making it always true).
1297 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1298 files to be used while preserving root privileges.
1300 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1301 that rogue child processes cannot use them.
1303 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1304 run-time user, instead of root.
1306 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1307 Exim run-time user without dropping privileges.
1309 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1310 result string, instead of calling string_vformat() twice with the same
1313 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1314 for other users. Others should always drop root privileges if they use
1315 -C on the command line, even for a whitelisted configure file.
1317 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1319 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1325 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1326 $data_path, and $header_path variables; fixed documentation bugs and
1329 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1330 exipick to access non-standard spools, including the "frozen" queue
1333 NM/01 Bugzilla 965: Support mysql stored procedures.
1334 Patch from Alain Williams
1336 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1338 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1339 Patch from Andreas Metzler
1341 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1342 Patch from Kirill Miazine
1344 NM/05 Bugzilla 671: Added umask to procmail example.
1346 JJ/03 installed exipick 20100323.0, fixing doc bug
1348 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1349 directory. Notification and patch from Dan Rosenberg.
1351 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1353 TK/02 Improve log output when DKIM signing operation fails.
1355 MH/01 Treat the transport option dkim_domain as a colon separated
1356 list, not as a single string, and sign the message with each element,
1357 omitting multiple occurences of the same signer.
1359 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1360 Bugzilla 985, 986. Patch by Simon Arlott
1362 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1363 Patch by Simon Arlott
1365 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1366 MBX locking. Notification from Dan Rosenberg.
1372 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1374 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1376 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1378 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1380 NM/04 Bugzilla 915: Flag broken perl installation during build.
1386 TK/01 Added patch by Johannes Berg that expands the main option
1387 "spamd_address" if it starts with a dollar sign.
1389 TK/02 Write list of recipients to X-Envelope-Sender header when building
1390 the mbox-format spool file for content scanning (suggested by Jakob
1393 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1394 (http://www.dcc-servers.net/) support via dccifd. Activated by
1395 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1397 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1398 by Mark Daniel Reidel <mr@df.eu>.
1400 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1401 When building exim an external PCRE library is now needed -
1402 PCRE is a system library on the majority of modern systems.
1403 See entry on PCRE_LIBS in EDITME file.
1405 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1406 conversation. Added nologin parameter to request.
1407 Patch contributed by Kirill Miazine.
1409 TF/01 Do not log submission mode rewrites if they do not change the address.
1411 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1413 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1414 log files in place. Contributed by Roberto Lima.
1416 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1418 TF/03 Bugzilla 615: When checking the local_parts router precondition
1419 after a local_part_suffix or local_part_prefix option, Exim now
1420 does not use the address's named list lookup cache, since this
1421 contains cached lookups for the whole local part.
1423 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1424 Robert Millan. Documentation is in experimental-spec.txt.
1426 TF/04 Bugzilla 668: Fix parallel build (make -j).
1428 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1430 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1431 Patch provided by Jan Srzednicki.
1433 TF/05 Leading white space used to be stripped from $spam_report which
1434 wrecked the formatting. Now it is preserved.
1436 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1437 that they are available at delivery time.
1439 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1441 TF/08 TLS error reporting now respects the incoming_interface and
1442 incoming_port log selectors.
1444 TF/09 Produce a more useful error message if an SMTP transport's hosts
1445 setting expands to an empty string.
1447 NM/06 Bugzilla 744: EXPN did not work under TLS.
1448 Patch provided by Phil Pennock.
1450 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1451 Patch provided by Richard Godbee.
1453 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1454 acl_smtp_notquit, added index entry.
1456 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1457 Patch provided by Eugene Bujak.
1459 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1460 accept(). Patch provided by Maxim Dounin.
1462 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1463 Patch provided by Phil Pennock.
1465 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1467 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1468 Patch provided by Brad "anomie" Jorsch.
1470 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1471 Patch provided by Dean Brooks.
1473 TK/05 Add native DKIM support (does not depend on external libraries).
1475 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1476 Patch provided by Graeme Fowler.
1478 NM/16 Bugzilla 851: Documentation example syntax fix.
1480 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1482 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1485 NM/19 Bugzilla 745: TLS version reporting.
1486 Patch provided by Phil Pennock.
1488 NM/20 Bugzilla 167: bool: condition support.
1489 Patch provided by Phil Pennock.
1491 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1492 clients. Patch provided by Phil Pennock.
1494 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1495 Patch provided by Brad "anomie" Jorsch.
1497 NM/23 Bugzilla 687: Fix misparses in eximstats.
1498 Patch provided by Heiko Schlittermann.
1500 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1501 Patch provided by Heiko Schlittermann.
1503 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1504 plus update to original patch.
1506 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1508 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1509 Patch provided by David Brownlee.
1511 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1513 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1515 NM/30 Bugzilla 888: TLS documentation bugfixes.
1517 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1519 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1520 Unlike the original bugzilla I have changed all shell scripts in src tree.
1522 NM/33 Bugzilla 898: Transport filter timeout fix.
1523 Patch by Todd Rinaldo.
1525 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1526 Patch by Serge Demonchaux.
1528 NM/35 Bugzilla 39: Base64 decode bug fixes.
1529 Patch by Jakob Hirsch.
1531 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1533 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1535 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1537 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1543 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1544 ALT-N's libdkim that I have put here:
1545 http://duncanthrax.net/exim-experimental/
1547 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1548 were called 'true' and it seems that C99 defines that as a reserved
1549 keyword to be used with 'bool' variable types. That means you could
1550 not include C99-style headers which use bools without triggering
1551 build errors in sieve.c.
1553 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1554 as mailq or other aliases. Changed the --help handling significantly
1555 to do whats expected. exim_usage() emits usage/help information.
1557 SC/01 Added the -bylocaldomain option to eximstats.
1559 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1561 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1563 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1569 PH/01 Another patch from the Sieve maintainer.
1571 PH/02 When an IPv6 address is converted to a string for single-key lookup
1572 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1573 dots are used instead of colons so that keys in lsearch files need not
1574 contain colons. This was done some time before quoting was made available
1575 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1576 (notated using the quote facility) so as to distinguish them from IPv4
1577 keys. This meant that lookups for IP addresses in host lists did not work
1578 for iplsearch lookups.
1580 This has been fixed by arranging for IPv6 addresses to be expressed with
1581 colons if the lookup type is iplsearch. This is not incompatible, because
1582 previously such lookups could never work.
1584 The situation is now rather anomolous, since one *can* have colons in
1585 ordinary lsearch keys. However, making the change in all cases is
1586 incompatible and would probably break a number of configurations.
1588 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1591 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1592 conversion specification without a maximum field width, thereby enabling
1593 a rogue spamd server to cause a buffer overflow. While nobody in their
1594 right mind would setup Exim to query an untrusted spamd server, an
1595 attacker that gains access to a server running spamd could potentially
1596 exploit this vulnerability to run arbitrary code as the Exim user.
1598 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1599 $primary_hostname instead of what libspf2 thinks the hosts name is.
1601 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1602 a directory entry by the name of the lookup key. Previously, if a
1603 symlink pointed to a non-existing file or a file in a directory that
1604 Exim lacked permissions to read, a lookup for a key matching that
1605 symlink would fail. Now it is enough that a matching directory entry
1606 exists, symlink or not. (Bugzilla 503.)
1608 PH/03 The body_linecount and body_zerocount variables are now exported in the
1611 PH/04 Added the $dnslist_matched variable.
1613 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1614 This means they are set thereafter only if the connection becomes
1617 PH/06 Added the client_condition to authenticators so that some can be skipped
1618 by clients under certain conditions.
1620 PH/07 The error message for a badly-placed control=no_multiline_responses left
1621 "_responses" off the end of the name.
1623 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1625 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1626 (without spaces) instead of just copying the configuration text.
1628 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1630 PH/11 Added $max_received_linelength.
1632 PH/12 Added +ignore_defer and +include_defer to host lists.
1634 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1635 way in which PCRE > 7.0 is built.
1637 PH/14 Implemented queue_only_load_latch.
1639 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1640 MAIL command. The effect was to mangle the value on 64-bit systems.
1642 PH/16 Another patch from the Sieve maintainer.
1644 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1646 PH/18 If a system quota error occurred while trying to create the file for
1647 a maildir delivery, the message "Mailbox is full" was not appended to the
1648 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1649 only to a quota excession during the actual writing of the file.
1651 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1652 characters?) which causes problems in log lines. The DN values are now
1653 passed through string_printing() before being added to log lines.
1655 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1656 and InterBase are left for another time.)
1658 PH/21 Added message_body_newlines option.
1660 PH/22 Guard against possible overflow in moan_check_errorcopy().
1662 PH/23 POSIX allows open() to be a macro; guard against that.
1664 PH/24 If the recipient of an error message contained an @ in the local part
1665 (suitably quoted, of course), incorrect values were put in $domain and
1666 $local_part during the evaluation of errors_copy.
1672 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1673 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1676 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1677 issue a MAIL command.
1679 PH/02 In an ACL statement such as
1681 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1683 if a client was not listed at all, or was listed with a value other than
1684 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1685 the condition was not true (as it should be), so access was not denied.
1686 The bug was that the ! inversion was incorrectly passed on to the second
1687 item. This has been fixed.
1689 PH/03 Added additional dnslists conditions == and =& which are different from
1690 = and & when the dns lookup returns more than one IP address.
1692 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1693 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1695 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1696 FSYNC, which compiles an option called disable_fsync that allows for
1697 bypassing fsync(). The documentation is heavily laced with warnings.
1699 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1701 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1702 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1703 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1704 including adding "make clean"; (3) Added -fPIC when compiling the test
1705 dynamically loaded module, to get rid of a warning.
1707 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1708 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1709 = 0s. The bug is otherwise harmless.
1711 PH/07 There was a bug in the dovecot authenticator such that the value of
1712 $auth1 could be overwritten, and so not correctly preserved, after a
1713 successful authentication. This usually meant that the value preserved by
1714 the server_setid option was incorrect.
1716 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1718 PH/09 Installed PCRE release 7.0.
1720 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1721 run for batched SMTP input. It is now run at the start of every message
1722 in the batch. While fixing this I discovered that the process information
1723 (output by running exiwhat) was not always getting set for -bs and -bS
1724 input. This is fixed, and it now also says "batched" for BSMTP.
1726 PH/11 Added control=no_pipelining.
1728 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1729 patch, slightly modified), and move the expansion of helo_data till after
1730 the connection is made in the smtp transport (so it can use these
1733 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1735 PH/14 Added log_selector = +pid.
1737 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1739 PH/16 Add ${if forany and ${if forall.
1741 PH/17 Added dsn_from option to vary the From: line in DSNs.
1743 PH/18 Flush SMTP output before performing a callout, unless control =
1744 no_callout_flush is set.
1746 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1747 was true (the default) a successful delivery failed to delete the retry
1748 item, thus causing premature timeout of the address. The bug is now
1751 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1753 PH/21 Long custom messages for fakedefer and fakereject are now split up
1754 into multiline reponses in the same way that messages for "deny" and
1755 other ACL rejections are.
1757 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1758 with slight modification.
1760 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1761 draft, changing the syntax and factoring some duplicate code".
1763 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1764 for deliveries of the second and subsequent messages over the same SMTP
1767 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1768 ${reduce, with only minor "tidies".
1770 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1772 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1773 expansion side effects.
1775 PH/27 When a message times out after an over-quota error from an Exim-imposed
1776 quota, the bounce message says "mailbox is full". This message was not
1777 being given when it was a system quota that was exceeded. It now should
1780 MH/03 Made $recipients available in local_scan(). local_scan() already has
1781 better access to the recipient list through recipients_list[], but
1782 $recipients can be useful in postmaster-provided expansion strings.
1784 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1785 in the case of a MAIL command with additional options following the
1786 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1787 were accidentally chopped off.
1789 PH/29 SMTP synchronization checks are implemented when a command is read -
1790 there is a check that no more input is waiting when there shouldn't be
1791 any. However, for some commands, a delay in an ACL can mean that it is
1792 some time before the response is written. In this time, more input might
1793 arrive, invalidly. So now there are extra checks after an ACL has run for
1794 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1795 pipelining has not been advertised.
1797 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1799 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1800 returned in the SMTP response when smtp_return_error_details was false.
1801 This has been fixed.
1803 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1804 instead of the C I/O that was originally supplied, because problems were
1805 reported on Solaris.
1807 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1808 Exim which did not show up earlier: it was assuming that a call to
1809 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1810 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1811 was a macro that became an assignment, so it seemed to work. This has
1812 changed to a proper function call with a void return, hence the compile
1813 error. Exim's code has been fixed.
1815 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1818 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1820 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1822 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1823 because it turns out to be too big or there is a timeout) there is no
1824 "Completed" line in the log. When some messages of this type were
1825 selected by exigrep, they were listed as "not completed". Others were
1826 picked up by some special patterns. I have improved the selection
1827 criteria to be more general.
1829 PH/38 The host_find_failed option in the manualroute router can now be set
1830 to "ignore", to completely ignore a host whose IP address cannot be
1831 found. If all hosts are ignored, the behaviour is controlled by the new
1832 host_all_ignored option.
1834 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1835 homing or because of multiple MX records with /mx) generated more than
1836 one IP address, and the following item turned out to be the local host,
1837 all the secondary addresses of the first item were incorrectly removed
1838 from the list, along with the local host and any following hosts (which
1839 is what is supposed to happen).
1841 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1842 whoever called Exim into the -H file. In the case of the daemon it was
1843 behaving confusingly. When first started, it used values for whoever
1844 started the daemon, but after a SIGHUP it used the Exim user (because it
1845 calls itself on a restart). I have changed the code so that it now always
1848 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1849 message are rejected with the same error (e.g. no authentication or bad
1850 sender address), and a DATA command is nevertheless sent (as can happen
1851 with PIPELINING or a stupid MUA), the error message that was given to the
1852 RCPT commands is included in the rejection of the DATA command. This is
1853 intended to be helpful for MUAs that show only the final error to their
1856 PH/42 Another patch from the Sieve maintainer.
1858 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1859 Eximstats - Fixed some broken HTML links and added missing column headers
1861 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1862 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1864 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1866 PH/43 Yet another patch from the Sieve maintainer.
1868 PH/44 I found a way to check for a TCP/IP connection going away before sending
1869 the response to the final '.' that terminates a message, but only in the
1870 case where the client has not sent further data following the '.'
1871 (unfortunately, this is allowed). However, in many cases there won't be
1872 any further data because there won't be any more messages to send. A call
1873 to select() can be used: if it shows that the input is "ready", there is
1874 either input waiting, or the socket has been closed. An attempt to read
1875 the next input character can distinguish the two cases. Previously, Exim
1876 would have sent an OK response which the client would never have see.
1877 This could lead to message repetition. This fix should cure that, at
1878 least in a lot of common cases.
1880 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1881 advertised in response to EHLO.
1887 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1888 fixed by 4.65/MH/01 (is this a record?) are fixed:
1890 (i) An empty string was always treated as zero by the numeric comparison
1891 operators. This behaviour has been restored.
1893 (ii) It is documented that the numeric comparison operators always treat
1894 their arguments as decimal numbers. This was broken in that numbers
1895 starting with 0 were being interpreted as octal.
1897 While fixing these problems I realized that there was another issue that
1898 hadn't been noticed. Values of message_size_limit (both the global option
1899 and the transport option) were treated as octal if they started with 0.
1900 The documentation was vague. These values are now always treated as
1901 decimal, and I will make that clear in the documentation.
1907 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1908 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1911 MH/01 Don't check that the operands of numeric comparison operators are
1912 integers when their expansion is in "skipping" mode (fixes bug
1913 introduced by 4.64-PH/07).
1915 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1916 child addresses, Exim now panics and dies. Previously, because the count
1917 is held in a short int, deliveries were likely to be lost. As such a
1918 large number of recipients for a single message is ridiculous
1919 (performance will be very, very poor), I have chosen to impose a limit
1920 rather than extend the field.
1926 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1927 leftover -K file (the existence of which was triggered by #402).
1928 While we were at it, introduced process PID as part of the -K
1929 filename. This should rule out race conditions when creating
1932 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1933 processing considerably. Previous code took too long for large mails,
1934 triggering a timeout which in turn triggers #401.
1936 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1937 in the DK code in transports.c. sendfile() is not really portable,
1938 hence the _LINUX specificness.
1940 TF/01 In the add_headers option to the mail command in an Exim filter,
1941 there was a bug that Exim would claim a syntax error in any
1942 header after the first one which had an odd number of characters
1945 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1946 callout verification, Exim cached a "reject" for the entire domain. This
1947 is correct for most verifications, but it is not correct for a recipient
1948 verification with use_sender or use_postmaster set, because in that case
1949 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1950 case of MAIL FROM:<> rejection from other early rejections (e.g.
1951 rejection of HELO). When verifying a recipient using a non-null MAIL
1952 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1953 Whatever the result of the callout, the value of the domain cache is
1954 left unchanged (for any other kind of callout, getting as far as trying
1955 RCPT means that the domain itself is ok).
1957 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1960 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1961 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1962 session, the daemon ignores ECONNECT errors and logs others; it now
1963 ignores EPIPE as well.
1965 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1966 (quoted-printable decoding).
1968 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1969 later the small subsequent patch to fix an introduced bug.
1971 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1973 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1975 PH/08 An error is now given if message_size_limit is specified negative.
1977 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1978 to be given (somewhat) arbitrary names.
1980 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1983 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1984 miscellaneous code fixes
1986 PH/10 Added the log_reject_target ACL modifier to specify where to log
1989 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1990 hostname. This is wrong, because it relates to the incoming message (and
1991 probably the interface on which it is arriving) and not to the outgoing
1992 callout (which could be using a different interface). This has been
1993 changed to use the value of the helo_data option from the smtp transport
1994 instead - this is what is used when a message is actually being sent. If
1995 there is no remote transport (possible with a router that sets up host
1996 addresses), $smtp_active_hostname is used.
1998 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1999 tweaks were necessary in order to get it to work (see also 21 below):
2000 (a) The code assumed that strncpy() returns a negative number on buffer
2001 overflow, which isn't the case. Replaced with Exim's string_format()
2003 (b) There were several signed/unsigned issues. I just did the minimum
2004 hacking in of casts. There is scope for a larger refactoring.
2005 (c) The code used strcasecmp() which is not a standard C function.
2006 Replaced with Exim's strcmpic() function.
2007 (d) The code set only $1; it now sets $auth1 as well.
2008 (e) A simple test gave the error "authentication client didn't specify
2009 service in request". It would seem that Dovecot has changed its
2010 interface. Fortunately there's a specification; I followed it and
2011 changed what the client sends and it appears to be working now.
2013 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2016 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2017 address A is aliased to B and C, where B exists and C does not. Without
2018 -v the output is "A verified" because verification stops after a
2019 successful redirection if more than one address is generated. However,
2020 with -v the child addresses are also verified. Exim was outputting "A
2021 failed to verify" and then showing the successful verification for C,
2022 with its parentage. It now outputs "B failed to verify", showing B's
2023 parentage before showing the successful verification of C.
2025 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2026 look up a TXT record in a specific list after matching in a combined
2029 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2030 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2031 they consult the DNS. I had assumed they would set it the way they
2032 wanted; and indeed my experiments on Linux seem to show that in some
2033 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2034 To be on the safe side, however, I have now made the interface to
2035 host_find_byname() similar to host_find_bydns(), with an argument
2036 containing the DNS resolver options. The host_find_byname() function now
2037 sets these options at its start, just as host_find_bydns() does. The smtp
2038 transport options dns_qualify_single and dns_search_parents are passed to
2039 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2040 of host_find_byname() use the default settings of RES_DEFNAMES
2041 (qualify_single) but not RES_DNSRCH (search_parents).
2043 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2044 spool_read_header() do less string testing, by means of a preliminary
2045 switch on the second character of optional "-foo" lines. (This is
2046 overdue, caused by the large number of possibilities that now exist.
2047 Originally there were few.) While I was there, I also converted the
2048 str(n)cmp tests so they don't re-test the leading "-" and the first
2049 character, in the hope this might squeeze out yet more improvement.
2051 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2052 flag allowing group syntax was set by the header_syntax check but not
2053 turned off, possible causing trouble later; (2) The flag was not being
2054 set at all for the header_verify test, causing "group"-style headers to
2055 be rejected. I have now set it in this case, and also caused header_
2056 verify to ignore an empty address taken from a group. While doing this, I
2057 came across some other cases where the code for allowing group syntax
2058 while scanning a header line wasn't quite right (mostly, not resetting
2059 the flag correctly in the right place). These bugs could have caused
2060 trouble for malformed header lines. I hope it is now all correct.
2062 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2063 with the "reply" argument non-NULL. The code, however (which originally
2064 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2065 but it didn't always do it. This confused somebody who was copying the
2066 code for some other use. I have removed all the tests.
2068 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2069 feature that was used to support insecure browsers during the U.S. crypto
2070 embargo. It requires special client support, and Exim is probably the
2071 only MTA that supported it -- and would never use it because real RSA is
2072 always available. This code has been removed, because it had the bad
2073 effect of slowing Exim down by computing (never used) parameters for the
2074 RSA_EXPORT functionality.
2076 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2077 authenticator to fail if there's a tab character in the incoming data
2078 (there should never be unless someone is messing about, as it's supposed
2079 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2080 if the connection is using TLS or if the remote IP is the same as the
2081 local IP, and the "valid-client-cert option" if a client certificate has
2084 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2085 authenticators. This can be used for authorization after authentication
2086 succeeds. (In the case of plaintext, it servers for both authentication
2089 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2090 if any retry times were supplied.
2092 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2093 connection, where there is no client IP address to check. In this
2094 situation, the verify now always succeeds.
2096 PH/25 Applied John Jetmore's -Mset patch.
2098 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2100 PH/27 In a string expansion for a processed (not raw) header when multiple
2101 headers of the same name were present, leading whitespace was being
2102 removed from all of them, but trailing whitespace was being removed only
2103 from the last one. Now trailing whitespace is removed from each header
2104 before concatenation. Completely empty headers in a concatenation (as
2105 before) are ignored.
2107 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2108 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2110 PH/29 [Removed. This was a change that I later backed out, and forgot to
2111 correct the ChangeLog entry (that I had efficiently created) before
2112 committing the later change.]
2114 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2115 address errors (4xx response to RCPT) over the same connection as other
2116 messages routed to the same hosts. Such deliveries are always "forced",
2117 so retry times are not inspected. This resulted in far too many retries
2118 for the affected addresses. The effect occurred only when there were more
2119 hosts than the hosts_max_try setting in the smtp transport when it had
2120 the 4xx errors. Those hosts that it had tried were not added to the list
2121 of hosts for which the message was waiting, so if all were tried, there
2122 was no problem. Two fixes have been applied:
2124 (i) If there are any address or message errors in an SMTP delivery, none
2125 of the hosts (tried or untried) are now added to the list of hosts
2126 for which the message is waiting, so the message should not be a
2127 candidate for sending over the same connection that was used for a
2128 successful delivery of some other message. This seems entirely
2129 reasonable: after all the message is NOT "waiting for some host".
2130 This is so "obvious" that I'm not sure why it wasn't done
2131 previously. Hope I haven't missed anything, but it can't do any
2132 harm, as the worst effect is to miss an optimization.
2134 (ii) If, despite (i), such a delivery is accidentally attempted, the
2135 routing retry time is respected, so at least it doesn't keep
2136 hammering the server.
2138 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2139 in ${readsocket because some servers need this prod.
2141 PH/32 Added some extra debug output when updating a wait-xxx database.
2143 PH/33 The hint "could be header name not terminated by colon", which has been
2144 given for certain expansion errors for a long time, was not being given
2145 for the ${if def:h_colon_omitted{... case.
2147 PH/34 The spec says: "With one important exception, whenever a domain list is
2148 being scanned, $domain contains the subject domain." There was at least
2149 one case where this was not true.
2151 PH/35 The error "getsockname() failed: connection reset by peer" was being
2152 written to the panic log as well as the main log, but it isn't really
2153 panic-worthy as it just means the connection died rather early on. I have
2154 removed the panic log writing for the ECONNRESET error when getsockname()
2157 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2158 runs only) independently of the message's sender address. This meant
2159 that, if the 4xx error was in fact related to the sender, a different
2160 message to the same recipient with a different sender could confuse
2161 things. In particualar, this can happen when sending to a greylisting
2162 server, but other circumstances could also provoke similar problems.
2163 I have changed the default so that the retry time for these errors is now
2164 based a combination of the sender and recipient addresses. This change
2165 can be overridden by setting address_retry_include_sender=false in the
2168 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2169 remote server are returned as part of bounce messages. This was not
2170 happening for LMTP over a pipe (the lmtp transport), but now it is the
2171 same for both kinds of LMTP.
2173 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2174 in header lines that were in fact CNAMEs. This is no longer the case.
2176 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2177 and queue runs started by the daemon processed all messages. This has
2178 been fixed so that -R and -S can now usefully be given with -q<time>.
2180 PH/40 Import PCRE release 6.7 (fixes some bugs).
2182 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2184 PH/42 Give an error if -q is specified more than once.
2186 PH/43 Renamed the variables $interface_address and $interface_port as
2187 $received_ip_address and $received_port, to make it clear that these
2188 values apply to message reception, and not to the outgoing interface when
2189 a message is delivered. (The old names remain recognized, of course.)
2191 PH/44 There was no timeout on the connect() call when using a Unix domain
2192 socket in the ${readsocket expansion. There now is.
2194 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2195 be meaningful with "accept".
2197 SC/01 Eximstats V1.43
2198 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2200 SC/02 Eximstats V1.44
2201 Use a glob alias rather than an array ref in the generated
2202 parser. This improves both readability and performance.
2204 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2205 Collect SpamAssassin and rejection statistics.
2206 Don't display local sender or destination tables unless
2207 there is data to show.
2208 Added average volumes into the top table text output.
2210 SC/04 Eximstats V1.46
2211 Collect data on the number of addresses (recipients)
2212 as well as the number of messages.
2214 SC/05 Eximstats V1.47
2215 Added 'Message too big' to the list of mail rejection
2216 reasons (thanks to Marco Gaiarin).
2218 SC/06 Eximstats V1.48
2219 Mainlog lines which have GMT offsets and are too short to
2220 have a flag are now skipped.
2222 SC/07 Eximstats V1.49 (Alain Williams)
2223 Added the -emptyok flag.
2225 SC/08 Eximstats V1.50
2226 Fixes for obtaining the IP address from reject messages.
2228 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2229 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2230 whitesspace changes from 4.64-PH/27
2232 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2235 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2236 are found, allow negative numbers in numeric criteria)
2238 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2240 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2241 to match changes made in 4.64-PH/43
2243 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2245 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2246 "domain" is actually an IP address, and force a failure. This locks out
2247 those revolvers/nameservers that support "A-for-A" lookups, in
2248 contravention of the specifications.
2250 PH/48 When a host name was looked up from an IP address, and the subsequent
2251 forward lookup of the name timed out, the host name was left in
2252 $sender_host_name, contrary to the specification.
2254 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2255 restricted to single-key lookups, Exim was not diagnosing an error if
2256 * or *@ was used with a query-style lookup.
2258 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2260 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2261 long ago, but noone interested enough thought of it. Let's just say that
2262 the "1.1" means that there are some new functions that weren't there at
2263 some point in the past.
2265 PH/51 Error processing for expansion failure of helo_data from an smtp
2266 transport during callout processing was broken.
2268 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2269 tested/used via the -bh/-bhc/-bs options.
2271 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2272 bug, fixed in subsequent PCRE releases).
2274 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2275 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2277 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2283 SC/01 Use a glob alias rather than an array ref in eximstats generated
2284 parser. This improves both readability and performance.
2286 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2287 Don't display local sender or destination tables in eximstats unless
2288 there is data to show.
2289 Added average volumes into the eximstats top table text output.
2291 SC/03 Collect data on the number of addresses (recipients) as well
2292 as the number of messages in eximstats.
2294 TF/01 Correct an error in the documentation for the redirect router. Exim
2295 does (usually) call initgroups() when daemonizing.
2297 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2298 with consistent privilege compared to when running as a daemon.
2300 TF/03 Note in the spec that $authenticated_id is not set for local
2301 submissions from trusted users.
2303 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2304 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2306 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2307 by adding some example configuration directives to the default
2308 configuration file. A little bit of work is required to uncomment the
2309 directives and define how usernames and passwords are checked, but
2310 there is now a framework to start from.
2312 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2313 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2314 without this. I don't know how relevant this is to other LDAP libraries.
2316 PH/02 Add the verb name to the "unknown ACL verb" error.
2318 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2320 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2322 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2323 directory not expanded when it should be if an expanded home directory
2324 was set for the address (which is overridden by the transport).
2326 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2329 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2330 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2331 because it is too late at that time, and has no effect.
2333 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2334 security issue with \' (bugzilla #107). I could not use the
2335 PQescapeStringConn() function, because it needs a PGconn value as one of
2338 PH/08 When testing addresses using -bt, indicate those final addresses that
2339 are duplicates that would not cause an additional delivery. At least one
2340 person was confused, thinking that -bt output corresponded to deliveries.
2341 (Suppressing duplicates isn't a good idea as you lose the information
2342 about possibly different redirections that led to the duplicates.)
2344 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2345 systems where poll() doesn't work, in particular OS X.
2347 PH/10 Added more information to debugging output for retry time not reached.
2349 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2350 operations in malware.c.
2352 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2355 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2356 syslog_duplication set false, log lines that would normally be written
2357 both the the main log and to the reject log were not written to syslog at
2360 PH/14 In the default configuration, change the use of "message" in ACL warn
2361 statements to "add_header".
2363 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2364 not followed by a command (e.g. "seen endif").
2366 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2367 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2370 PH/17 Added extra conditions to the default value of delay_warning_condition
2374 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2375 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2376 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2379 The Auto-Submitted: and various List- headers are standardised, whereas I
2380 don't think Precedence: ever was.
2382 PH/18 Refactored debugging code in route_finduser() to show more information,
2383 in particular, the error code if getpwnam() issues one.
2385 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2386 This is apparently needed in addition to the PH/07 change above to avoid
2387 any possible encoding problems.
2389 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2390 but not after initializing Perl.
2392 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2393 output them only if debugging. By default they are written stderr,
2394 apparently, which is not desirable.
2396 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2399 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2402 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2404 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2405 authenticated or an ident call has been made. Suppress the default
2406 values for $authenticated_id and $authenticated_sender (but permit -oMai
2407 and -oMas) when testing with -bh.
2409 PH/24 Re-jigged the order of the tests in the default configuration so that the
2410 tests for valid domains and recipients precede the DNS black list and CSA
2411 tests, on the grounds that those ones are more expensive.
2413 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2414 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2415 This bug exists in every version of Exim that I still have, right back to
2418 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2419 However, an attempt to turn on case-sensitivity in a regex key by
2420 including (?-i) didn't work because the subject string was already
2421 lowercased, and the effects were non-intuitive. It turns out that a
2422 one-line patch can be used to allow (?-i) to work as expected.
2428 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2429 other effects) broke the use of negated acl sub-conditions.
2431 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2434 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2435 "Deny" causes Exim to reject the incoming connection with a 554 error.
2436 Unfortunately, if there is a major crisis, such as a disk failure,
2437 tcp-wrappers gives "deny", whereas what one would like would be some
2438 kind of temporary error. A kludge has been added to help with this.
2439 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2440 554 error is used if errno is still zero or contains ENOENT (which occurs
2441 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2444 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2446 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2447 errors. Otherwise a message that provokes a temporary error (when other
2448 messages do not) can cause a whole host to time out.
2450 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2451 addresses were routed directly to files or pipes from a redirect router.
2452 File deliveries just didn't batch; pipe deliveries might have suffered
2455 PH/06 A failure to get a lock for a hints database would erroneously always say
2456 "Failed to get write lock", even when it was really a read lock.
2458 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2459 of 0600. This has been changed to use the value of the lockfile_mode
2460 option (which defaults to 0600).
2462 PH/08 Applied small patch from the Sieve maintainer.
2464 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2465 folder from quota calculations, a direct delivery into this folder messed
2466 up the contents of the maildirsize file. This was because the regex was
2467 used only to exclude .Trash (or whatever) when the size of the mailbox
2468 was calculated. There was no check that a delivery was happening into an
2469 excluded directory. This bug has been fixed by ignoring all quota
2470 processing for deliveries into excluded directories.
2472 PH/10 Added the maildirfolder_create_regex option to appendfile.
2478 PH/01 The code for finding all the local interface addresses on a FreeBSD
2479 system running IPv6 was broken. This may well have applied to all BSD
2480 systems, as well as to others that have similar system calls. The broken
2481 code found IPv4 interfaces correctly, but gave incorrect values for the
2482 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2483 that it would not match correctly against @[] and not recognize the IPv6
2486 PH/02 The ipliteral router was not recognizing addresses of the form user@
2487 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2489 PH/03 Added disable_ipv6.
2491 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2492 decoded form, because it is most often used to construct To: headers
2493 lines in autoreplies, and the decoded form may well be syntactically
2494 invalid. However, $reply_address has leading white space removed, and all
2495 newlines turned into spaces so that the autoreply transport does not
2498 PH/05 If group was specified without a user on a router, and no group or user
2499 was specified on a transport, the group from the router was ignored.
2501 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2502 for visible compile-time settings that can be used to change these
2503 numbers, for those that want even more. Backwards compatibility with old
2504 spool files has been maintained. However, going back to a previous Exim
2505 release will lost any variables that are in spool files.
2507 PH/07 Two small changes when running in the test harness: increase delay when
2508 passing a TCP/IP connection to a new process, in case the original
2509 process has to generate a bounce, and remove special handling of
2510 127.0.0.2 (sic), which is no longer necessary.
2512 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2513 be the same on different OS.
2515 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2518 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2519 whether --show-vars was specified or not
2521 JJ/02 exipick: Added support for new ACL variable spool format introduced
2524 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2525 syntactically invalid From: or Reply-to: line, and a filter used this to
2526 generate an autoreply, and therefore failed to obtain an address for the
2527 autoreply, Exim could try to deliver to a non-existent relative file
2528 name, causing unrelated and misleading errors. What now happens is that
2529 it logs this as a hard delivery error, but does not attempt to create a
2532 PH/11 The exinext utility has a -C option for testing purposes, but although
2533 the given file was scanned by exinext itself; it wasn't being passed on
2534 when Exim was called.
2536 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2537 an end-of-file indication when reading a command response.
2539 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2540 compiled. In many other places in Exim, IPv6 addresses are always
2541 recognized, so I have changed this. It also means that IPv4 domain
2542 literals of the form [IPV4:n.n.n.n] are now always recognized.
2544 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2545 used if the router is not running as root, for example, when verifying at
2546 ACL time, or when using -bh. The debugging output from this situation was
2547 non-existent - all you got was a failure to exec. I have made two
2550 (a) Failures to set uid/gid, the current directory, or a process leader
2551 in a subprocess such as that created by queryprogram now generate
2552 suitable debugging ouput when -d is set.
2554 (b) The queryprogram router detects when it is not running as root,
2555 outputs suitable debugging information if -d is set, and then runs
2556 the subprocess without attempting to change uid/gid.
2558 PH/15 Minor change to Makefile for building test_host (undocumented testing
2561 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2562 additional section of a DNS packet that returns MX or SRV records.
2563 Instead, it always explicitly searches for A/AAAA records. This avoids
2564 major problems that occur when a DNS server includes only records of one
2565 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2566 fixed another bug: if SRV records were looked up and the corresponding
2567 address records were *not* found in the additional section, the port
2568 values from the SRV records were lost.
2570 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2571 using the correct key (the original address) when searching the retry
2572 rules in order to find which one to use for generating the retry hint.
2574 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2575 adding the default one. Similarly, if it contains a Reply-To: header, the
2576 errors_reply_to option, if set, is not used.
2578 PH/19 When calculating a retry time, Exim used to measure the "time since
2579 failure" by looking at the "first failed" field in the retry record. Now
2580 it does not use this if it is later than than the arrival time of the
2581 message. Instead it uses the arrival time. This makes for better
2582 behaviour in cases where some deliveries succeed, thus re-setting the
2583 "first failed" field. An example is a quota failure for a huge message
2584 when small messages continue to be delivered. Without this change, the
2585 "time since failure" will always be short, possible causing more frequent
2586 delivery attempts for the huge message than are intended.
2587 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2589 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2590 $1, $2, $3) because the numerical variables can be reset during some
2591 expansion items (e.g. "match"), thereby losing the authentication data.
2593 PH/21 Make -bV show the size of off_t variables so that the test suite can
2594 decide whether to run tests for quotas > 2G.
2596 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2597 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2598 filecount value is greater than 2G or if a quota value is greater than 2G
2599 on a system where the size of off_t is not greater than 4, a panic error
2602 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2603 never match. The debug and -bh output now contains an explicit error
2604 message indicating a malformed IPv4 address or mask.
2606 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2607 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2608 PH/23 above applies.
2610 PH/25 Do not write to syslog when running in the test harness. The only
2611 occasion when this arises is a failure to open the main or panic logs
2612 (for which there is an explicit test).
2614 PH/26 Added the /no_tell option to "control=freeze".
2616 PH/27 If a host name lookup failed very early in a connection, for example, if
2617 the IP address matched host_lookup and the reverse lookup yielded a name
2618 that did not have a forward lookup, an error message of the form "no IP
2619 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2620 could be logged. Now it outputs the IP address instead of "NULL".
2622 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2623 allows the sender and the authenticated sender to be set when
2624 submitting a message from within Exim. Since child_open_exim() is
2625 documented for local_scan(), the new function should be too.
2627 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2628 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2629 results in an empty string is now treated as unset.
2631 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2633 PH/31 Added sender_verify_fail logging option.
2635 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2636 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2637 tidied the source and removed it altogether.
2639 PH/33 When a queue run was abandoned because the load average was too high, a
2640 log line was always written; now it is written only if the queue_run log
2641 selector is set. In addition, the log line for abandonment now contains
2642 information about the queue run such as the pid. This is always present
2643 in "start" and "stop" lines but was omitted from the "abandon" line.
2645 PH/34 Omit spaces between a header name and the colon in the error message that
2646 is given when verify = headers_syntax fails (if there are lots of them,
2647 the message gets confusing).
2649 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2650 names, as there are now some PTR records that contain slashes. This check
2651 is only to protect against broken name servers that fall over on strange
2652 characters, so the fact that it applies to all lookups doesn't matter.
2654 PH/36 Now that the new test suite is complete, we can remove some of the
2655 special code in Exim that was needed for the old test suite. For example,
2656 sorting DNS records because real resolvers return them in an arbitrary
2657 order. The new test suite's fake resolver always returns records in the
2660 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2661 bounces) except when queue_only is set, to avoid logging races between
2662 the different processes.
2664 PH/38 Panic-die if .include specifies a non-absolute path.
2666 PH/39 A tweak to the "H" retry rule from its user.
2668 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2669 a label. They prevented compilation on older perls.
2671 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2672 a warning to be raised on newish perls.
2674 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2675 on queue. Changes to match documented behaviour of showing count of
2676 messages matching specified criteria.
2678 PH/40 Changed the default ident timeout from 30s to 5s.
2680 PH/41 Added support for the use of login_cap features, on those BSD systems
2681 that have them, for controlling the resources used by pipe deliveries.
2683 PH/42 The content-scanning code uses fopen() to create files in which to put
2684 message data. Previously it was not paying any attention to the mode of
2685 the files. Exim runs with umask(0) because the rest of the code creates
2686 files with open(), and sets the required mode explicitly. Thus, these
2687 files were ending up world-writeable. This was not a big issue, because,
2688 being within the spool directory, they were not world-accessible. I have
2689 created a function called modefopen, which takes an additional mode
2690 argument. It sets umask(777), creates the file, chmods it to the required
2691 mode, then resets the umask. All the relevant calls to fopen() in the
2692 content scanning code have been changed to use this function.
2694 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2695 to 24 hours. This avoids potential overflow problems when processing G
2696 and H retry rules. I suspect nobody ever tinkers with this value.
2698 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2700 PH/45 When the plaintext authenticator is running as a client, the server's
2701 challenges are checked to ensure they are valid base64 strings. By
2702 default, the authentication attempt is cancelled if an invalid string is
2703 received. Setting client_ignore_invalid_base64 true ignores these errors.
2704 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2705 they are received. Thus, the responses can be made to depend on the
2706 challenges. If an invalid string is ignored, an empty string is placed in
2709 PH/46 Messages that are created by the autoreply transport now contains a
2710 References: header, in accordance with RFCs 2822 and 3834.
2712 PH/47 Added authenticated_sender_force to the smtp transport.
2714 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2716 PH/49 Installed latest patch from the Sieve maintainer.
2718 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2719 was also set, the appendfile transport was unnecessarily scanning a
2720 directory of message files (e.g. for maildir delivery) to find the count
2721 of files (along with the size), even though it did not need this
2722 information. It now does the scan only if it needs to find either the
2723 size of the count of files.
2725 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2727 PH/52 Two bugs concerned with error handling when the smtp transport is
2730 (i) Exim was not creating retry information for temporary errors given
2731 for individual recipients after the DATA command when the smtp transport
2732 was used in LMTP mode. This meant that they could be retried too
2733 frequently, and not timed out correctly.
2735 (ii) Exim was setting the flag that allows error details to be returned
2736 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2737 recipients that were returned after the DATA command.
2739 PH/53 This is related to PH/52, but is more general: for any failing address,
2740 when detailed error information was permitted to be returned to the
2741 sender, but the error was temporary, then after the final timeout, only
2742 "retry timeout exceeded" was returned. Now it returns the full error as
2743 well as "retry timeout exceeded".
2745 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2746 do this, and (what is worse) MTAs that accept it.
2748 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2749 will now be deprecated.
2751 PH/56 New os.c-cygwin from the Cygwin maintainer.
2753 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2754 formats (previously only available in exim formats via -bpr, -bpru,
2755 and -bpra. Now also available in native and exiqgrep formats)
2757 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2758 with very large, slow to parse queues
2760 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2762 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2764 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2765 responses to authentication challenges, though it was showing the
2766 challenges; (ii) I've removed the CR characters from the debug output for
2769 PH/58 Allow for the insertion of a newline as well as a space when a string
2770 is turned into more than one encoded-word during RFC 2047 encoding. The
2771 Sieve code now uses this.
2773 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2774 data_4xx, lost_connection, tls_required.
2776 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2777 message was being sent as an SMTP response.
2779 PH/61 Add -l and -k options to exicyclog.
2781 PH/62 When verifying, if an address was redirected to one new address, so that
2782 verification continued, and the new address failed or deferred after
2783 having set something in $address_data, the value of $address_data was not
2784 passed back to the ACL. This was different to the case when no
2785 redirection occurred. The value is now passed back in both cases.
2787 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2788 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2789 use login_cap.h, so on its own it isn't the distinguishing feature. The
2790 new name refers directly to the setclassresources() function.
2792 PH/65 Added configuration files for NetBSD3.
2794 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2796 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2797 is preferred over IPv4.
2799 PH/68 The bounce_return_message and bounce_return_body options were not being
2800 honoured for bounces generated during the reception of non-SMTP messages.
2801 In particular, this applied to messages rejected by the ACL. This bug has
2802 been fixed. However, if bounce_return_message is true and bounce_return_
2803 body is false, the headers that are returned for a non-SMTP message
2804 include only those that have been read before the error was detected.
2805 (In the case of an ACL rejection, they have all been read.)
2807 PH/69 The HTML version of the specification is now built in a directory called
2808 spec_html instead of spec.html, because the latter looks like a path with
2809 a MIME-type, and this confuses some software.
2811 PH/70 Catch two compiler warnings in sieve.c.
2813 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2814 function verify_get_ident() calls ip_connect() to connect a socket, but
2815 if the "connect()" function timed out, ip_connect() used to close the
2816 socket. However, verify_get_ident() also closes the socket later, and in
2817 between Exim writes to the log, which may get opened at this point. When
2818 the socket was closed in ip_connect(), the log could get the same file
2819 descriptor number as the socket. This naturally causes chaos. The fix is
2820 not to close the socket in ip_connect(); the socket should be closed by
2821 the function that creates it. There was only one place in the code where
2822 this was missing, in the iplookup router, which I don't think anybody now
2823 uses, but I've fixed it anyway.
2825 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2826 well as to direct DNS lookups. Otherwise the handling of names in host
2827 lists is inconsistent and therefore confusing.
2833 PH/01 Two changes to the default runtime configuration:
2835 (1) Move the checks for relay_from_hosts and authenticated clients from
2836 after to before the (commented out) DNS black list checks.
2838 (2) Add control=submission to the relay_from_hosts and authenticated
2839 clients checks, on the grounds that messages accepted by these
2840 statements are most likely to be submissions.
2842 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2844 (1) Generate an error if the third argument for the ${prvs expansion is
2847 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2850 (3) Reset the variables that are obtained from the first argument of
2851 ${prvscheck and used in the second argument before leaving the code,
2852 because their memory is reclaimed, so using them afterwards may do
2855 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2856 one (it's much easier than Tom thought :-).
2858 (5) Because of (4), we can now allow for the use of $prvscheck_result
2859 inside the third argument.
2861 PH/03 For some reason, the default setting of PATH when running a command from
2862 a pipe transport was just "/usr/bin". I have changed it to
2865 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2866 anything to be listed in the output from -bV.
2868 PH/05 When a filter generated an autoreply, the entire To: header line was
2869 quoted in the delivery log line, like this:
2871 => >A.N.Other <ano@some.domain> <original@ddress> ...
2873 This has been changed so that it extracts the operative address. There
2874 may be more than one such address. If so, they are comma-separated, like
2877 => >ano@some.domain,ona@other.domain <original@ddress> ...
2879 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2880 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2881 not being looked up in the rDNS to get a host name, Exim was showing the
2882 IP address twice in Received: lines, even though the IP addresses were
2883 identical. For example:
2885 Received: from [1.2.3.4] (helo=[1.2.3.4])
2887 However, if the real host name was known, it was omitting the HELO data
2888 if it matched the actual IP address. This has been tidied up so that it
2889 doesn't show the same IP address twice.
2891 PH/07 When both +timestamp and +memory debugging was on, the value given by
2892 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2893 called by the debug printing, thereby overwriting the timestamp buffer.
2894 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2896 PH/08 When the original message was included in an autoreply transport, it
2897 always said "this is a copy of the message, including all the headers",
2898 even if body_only or headers_only was set. It now gives an appropriate
2901 PH/09 Applied a patch from the Sieve maintainer which:
2903 o fixes some comments
2904 o adds the (disabled) notify extension core
2905 o adds some debug output for the result of if/elsif tests
2906 o points to the current vacation draft in the documentation
2907 and documents the missing references header update
2911 o fixes a bug in processing the envelope test (when testing
2912 multiple envelope elements, the last element determinted the
2915 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2916 Electronic Mail") by including:
2918 Auto-submitted: auto-generated
2920 in the messages that it generates (bounce messages and others, such as
2921 warnings). In the case of bounce messages for non-SMTP mesages, there was
2922 also a typo: it was using "Auto_submitted" (underscore instead of
2923 hyphen). Since every message generated by Exim is necessarily in response
2924 to another message, thes have all been changed to:
2926 Auto-Submitted: auto-replied
2928 in accordance with these statements in the RFC:
2930 The auto-replied keyword:
2932 - SHOULD be used on messages sent in direct response to another
2933 message by an automatic process,
2935 - MUST NOT be used on manually-generated messages,
2937 - MAY be used on Delivery Status Notifications (DSNs) and Message
2938 Disposition Notifications (MDNs),
2940 - MUST NOT be used on messages generated by automatic or periodic
2941 processes, except for messages which are automatic responses to
2944 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2945 to the default Received: header definition.
2947 PH/12 Added log selector acl_warn_skipped (default on).
2949 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2950 variables because (a) they are in the wrong storage pool and (b) even if
2951 they were copied, it wouldn't work properly because of the caching.
2953 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2954 checking when decoding. Apparently there are clients that generate
2955 overlong encoded strings. Why am I not surprised?
2957 PH/15 If the first argument of "${if match_address" was not empty, but did not
2958 contain an "@" character, Exim crashed. Now it writes a panic log message
2959 and treats the condition as false.
2961 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2963 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2964 extension "envelope-auth". The code is finished and in agreement with
2965 other implementations, but there is no documentation so far and in fact,
2966 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2967 not changing the active code.
2969 Print executed "if" and "elsif" statements when debugging is used. This
2970 helps a great deal to understand what a filter does.
2972 Document more things not specified clearly in RFC3028. I had all this
2973 sorted out, when out of a sudden new issues came to my mind. Oops."
2975 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2978 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2979 canonical form (as documented). However, after a host name lookup from
2980 the IP address, check_host() was doing a simple string comparison with
2981 addresses acquired from the DNS when checking that the found name did
2982 have the original IP as one of its addresses. Since any found IPv6
2983 addresses are likely to be in abbreviated form, the comparison could
2984 fail. Luckily, there already exists a function for doing the comparison
2985 by converting both addresses to binary, so now that is used instead of
2986 the text comparison.
2988 PH/20 There was another similar case to PH/19, when a complete host name was
2989 given in a host list; looking up its IP address could give an abbreviated
2990 form, whereas the current host's name might or might not be abbreviated.
2991 The same fix has been applied.
2997 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2998 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3001 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3002 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3004 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3006 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3007 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3008 Also, for "Auto-Submitted", treat anything other than "no" as
3009 non-personal, in accordance with RFC 3834. (Previously it treated
3010 anything starting "auto-" as non-personal.)
3012 TF/01 The control=submission/name=... option had a problem with syntax
3013 errors if the name included a slash character. The /name= option
3014 now slurps the rest of the string, so it can include any characters
3015 but it must come last in the list of options (after /sender_retain
3018 PH/05 Some modifications to the interface to the fake nameserver for the new
3026 TK/01 Added the "success_on_redirect" address verification option. See
3027 NewStuff for rationale and an example.
3029 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3031 PH/02 Patch to exigrep to allow it to work on syslog lines.
3033 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3034 fread() to skip over the body file's header line, because in Cygwin the
3035 header line is locked and is inaccessible.
3037 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3038 co-exist for some time) to make it clear that it is the Exim ID that is
3039 referenced, not the Message-ID: header line.
3041 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3042 string_format() function, because snprintf() does not exist on all
3045 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3046 ${stat: expansion item.
3048 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3049 protocol synchronization error", to keep the pedants happy.
3051 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3052 well as for IRIX systems, when gcc is being used. See the host.c source
3055 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3057 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3060 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3061 transport and to the smtp transport in LMTP mode.
3063 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3065 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3066 run a filter in a subprocess. This could lead to confusion in subsequent
3067 lookups in the parent process. There should also be a search_tidyup() at
3068 the end of the subprocess.
3070 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3071 only if the host matched helo_try_verify_hosts, which caused the
3072 verification to occur when the EHLO/HELO command was issued. The ACL just
3073 tested the remembered result. Now, if a previous verification attempt has
3074 not happened, "verify = helo" does it there and then.
3076 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3078 TK/03 Fix log output including CR from clamd.
3080 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3081 exist provoked a memory error which could cause a segfault.
3083 PH/15 Installed PCRE 6.2
3085 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3087 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3088 of the problem. Specifically, suggested +O2 rather than +O1 for the
3091 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3093 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3094 record, the port was not correctly logged when the outgoing_port log
3095 selector was set (it logged the transort's default port).
3097 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3098 fallback_hosts, and "hosts" in the smtp transport.
3100 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3101 host errors such as "Connection refused".
3103 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3104 authentication with radiusclient 0.4.9:
3106 - Error returned from rc_read_config was caught wrongly
3107 - Username/password not passed on to radius server due to wrong length.
3109 The presumption is that some radiusclient API changes for 4.51/PH/17
3110 were not taken care of correctly. The code is still untested by me (my
3111 Linux distribution still has 0.3.2 of radiusclient), but it was
3112 contributed by a Radius user.
3114 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3115 expanding the "port" option of the smtp transport.
3117 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3118 while reading a MIME header. Thanks to Tom Hughes for a patch.
3120 PH/24 Include config.h inside local_scan.h so that configuration settings are
3123 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3124 that in an ACL for RCPT (for example), you can examine exactly what was
3127 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3128 commands, but it was not correctly comparing the address with the actual
3129 client host address. Thus, it would show the EHLO address in Received:
3130 header lines when this was not necessary.
3132 PH/27 Added the % operator to ${eval:}.
3134 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3135 it should be ignoring failures (because with -C, for example, it has lost
3136 privilege). It wasn't ignoring creation failures other than "already
3139 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3142 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3143 because an input line was too long, either on its own, or by virtue of
3144 too many continuations, the temporary file was not being removed, and the
3145 return code was incorrect.
3147 PH/31 Missing "BOOL" in function definition in filtertest.c.
3149 PH/32 Applied Sieve patches from the maintainer.
3151 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3153 PH/33 Added "verify = not_blind".
3155 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3156 Local/Makefile (with some defaults set). These are used in built scripts
3157 such as exicyclog, but they have never been used in the exim_install
3158 script (though there are many overriding facilities there). I have
3159 arranged that the exim_install script now takes note of these two
3162 PH/35 Installed configuration files for Dragonfly.
3164 PH/36 When a locally submitted message by a trusted user did not contain a
3165 From: header, and the sender address was obtained from -f or from an SMTP
3166 MAIL command, and the trusted user did not use -F to supply a sender
3167 name, $originator_name was incorrectly used when constructing a From:
3168 header. Furthermore, $originator_name was used for submission mode
3169 messages from external hosts without From: headers in a similar way,
3170 which is clearly wrong.
3172 PH/37 Added control=suppress_local_fixups.
3174 PH/38 When log_selector = +received_sender was set, and the addition of the
3175 sender made the log line's construction buffer exactly full, or one byte
3176 less than full, an overflow happened when the terminating "\n" was
3179 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3180 when the result of a list match is failure because a DNS lookup failed.
3182 PH/40 RM_COMMAND is now used in the building process.
3184 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3185 the "build-* directories that it finds.
3187 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3188 address in a domain literal was a prefix of an interface address.
3190 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3191 when verifying a sender address, unless rewrite_headers is false.
3193 PH/44 Wrote a long comment about why errors_to addresses are verified as
3194 recipients, not senders.
3196 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3197 the ratelimit ACL was added.
3199 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3201 PH/46 Added extra information about PostgreSQL errors to the error string.
3203 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3204 avoiding the need to install special zones in a real server. This is
3205 backwards compatible; if it can't find the fake resolver, it drops back.
3206 Thus, both old and new test suites can be run.
3208 TF/02 Added util/ratelimit.pl
3210 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3211 clock is set back in time.
3213 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3214 Candler <B.Candler@pobox.com>.
3216 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3217 OK for addresses that are the result of redirections.
3219 PH/48 A number of further additions for the benefit of the new test suite,
3220 including a fake gethostbyname() that interfaces to the fake DNS resolver
3223 TF/06 The fix for widen_domains has also been applied to qualify_single and
3224 search_parents which are the other dnslookup options that can cause
3227 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3230 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3232 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3233 service" if it encountered an MX record with an empty target hostname.
3234 The message is now "an MX or SRV record indicated no SMTP service".
3236 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3237 if the DNS of the sending site is misconfigured. This is quite a
3238 common situation. This change restores the behaviour of treating a
3239 helo verification defer as a failure.
3241 PH/51 If self=fail was set on a router, the bounce message did not include the
3242 actual error message.
3248 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3250 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3251 command itself ended in error, the underlying message about the transport
3252 filter timeout was being overwritten with the pipe command error. Now the
3253 underlying error message should be appended to the second error message.
3255 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3257 PH/02 Reduce the amount of output that "make" produces by default. Full output
3258 can still be requested.
3260 PH/03 The warning log line about a condition test deferring for a "warn" verb
3261 was being output only once per connection, rather than after each
3262 occurrence (because it was using the same function as for successful
3263 "warn" verbs). This seems wrong, so I have changed it.
3265 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3266 it should not have, which might have caused a crash in the right
3267 circumstances, but probably never did.
3269 PH/04 Installed a modified version of Tony Finch's patch to make submission
3270 mode fix the return path as well as the Sender: header line, and to
3271 add a /name= option so that you can make the user's friendly name appear
3274 TF/03 Added the control = fakedefer ACL modifier.
3276 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3277 Mark Lowes for thorough testing.
3279 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3281 TK/03 Merged latest SRS patch from Miles Wilton.
3283 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3284 with the definition in sysexits.h (which is #included earlier).
3285 Fortunately, Exim does not actually use EX_OK. The code used to try to
3286 preserve the sysexits.h value, by assumimg that macro definitions were
3287 scanned for macro replacements. I have been disabused of this notion,
3288 so now the code just undefines EX_OK before #including unistd.h.
3290 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3291 in the smtp transport. When a block could not be written in a single
3292 write() function, the timeout was being re-applied to each part-write.
3293 This seems wrong - if the receiver was accepting one byte at a time it
3294 would take for ever. The timeout is now adjusted when this happens. It
3295 doesn't have to be particularly precise.
3297 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3298 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3300 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3301 without a domain if the check to <postmaster@domain> fails.
3303 SC/01 Eximstats: added -xls and the ability to specify output files
3304 (patch written by Frank Heydlauf).
3306 SC/02 Eximstats: use FileHandles for outputing results.
3308 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3310 SC/04 Eximstats: fixed display of large numbers with -nvr option
3312 SC/05 Eximstats: fixed merging of reports with empty tables.
3314 SC/06 Eximstats: added the -include_original_destination flag
3316 SC/07 Eximstats: removed tabs and trailing whitespace.
3318 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3320 TK/06 MBOX spool code: Add real "From " MBOX separator line
3321 so the .eml file is really in mbox format (even though
3322 most programs do not really care). Patch from Alex Miller.
3324 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3325 The latter is generated from $received_to and is only set if the
3326 message has one envelope recipient. SA can use these headers,
3327 obviously out-of-the-box. Patch from Alex Miller.
3329 PH/08 The ${def test on a variable was returning false if the variable's
3330 value was "0", contrary to what the specification has always said!
3331 The result should be true unless the variable is empty.
3333 PH/09 The syntax error of a character other than { following "${if
3334 def:variable_name" (after optional whitespace) was not being diagnosed.
3335 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3336 accidental colon was present, for example, could give incorrect results.
3338 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3339 result is used (not including appendfile, where other changes are about
3342 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3343 This involved changing a lot of size variables from int to off_t. It
3344 should work with maildirs and everything.
3346 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3347 spamd dying while we are connected to it.
3349 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3352 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3353 function for BDB 4.3.
3355 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3357 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3358 the sources for PCRE have been re-arranged and more files are now
3361 PH/15 The code I had for printing potentially long long variables in PH/11
3362 above was not the best (it lost precision). The length of off_t variables
3363 is now inspected at build time, and an appropriate printing format (%ld
3364 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3365 to be "long long int" or "long int". This is needed for the internal
3366 formatting function string_vformat().
3368 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3369 the configuration file to be ":syslog", then the script "guesses" where
3370 the logs files are, rather than using the compiled in default. In our
3371 case the guess is not the same as the compiled default, so the script
3372 suddenly stopped working when I started to use syslog. The patch checks
3373 to see if log_file_path is "". If so, it attempts to read it from exim
3374 with no configuration file to get the compiled in version, before it
3375 falls back to the previous guessing code."
3377 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3378 implementing BATV in an Exim configuration. See NewStuff for the gory
3381 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3382 Makefile that are specific to HP-UX.
3384 PH/18 If the "use_postmaster" option was set for a recipient callout together
3385 with the "random" option, the postmaster address was used as the MAIL
3386 FROM address for the random test, but not for the subsequent recipient
3387 test. It is now used for both.
3389 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3390 patch removes a few documentation additions to RFC 3028, because the
3391 latest draft now contains them. It adds the new en;ascii-case comparator
3392 and a new error check for 8bit text in MIME parts. Comparator and
3393 require names are now matched exactly. I enabled the subaddress
3394 extension, but it is not well tested yet (read: it works for me)."
3396 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3397 rework some of the code of TK/09 above to avoid the hardwired use of
3398 "%lld" and "long long". Replaced the call to snprintf() with a call to
3401 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3402 records point to non-existent hosts", "retry timeout exceeded", and
3403 "retry time not reached for any host after a long failure period".
3405 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3406 experimental DomainKeys support:
3408 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3409 (2) On an error such as an illegally used "control", the wrong name for
3410 the control was given.
3412 These problems did NOT occur unless DomainKeys support was compiled.
3414 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3416 PH/24 Added ${if match_ip condition.
3418 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3419 databases so that it will be absolutely obvious if a crash occurs in the
3420 DB library. This is a regular occurrence (often caused by mis-matched
3423 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3424 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3425 picked up on a user's system that detects such things. There doesn't seem
3426 to be a gcc warning option for this - only an attribute that has to be
3427 put on the function's prototype. It seems that in Fedora Core 4 they have
3428 set this on a number of new functions. No doubt there will be more in due
3431 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3432 specify a transport. However, if an address that was verified by such a
3433 router was the subject of a callout, Exim crashed because it tried to
3434 read the rcpt_include_affixes from the non-existent transport. Now it
3435 just assumes that the setting of that option is false. This bug was
3436 introduced by 4.51/PH/31.
3438 PH/28 Changed -d+all to exclude +memory, because that information is very
3439 rarely of interest, but it makes the output a lot bigger. People tend to
3440 do -d+all out of habit.
3442 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3443 code in os-type was giving problems when libc.so lives in lib64, like on
3446 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3447 aren't the modern standard, and it seems that some systems' include files
3448 don't always have them. Exim was already checking for some of the newer
3449 ones like T_AAAA, and defining it itself. I've added checks for all the
3450 record types that Exim uses.
3452 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3453 not automatically generating a new one, as it is supposed to. This
3454 prevented TLS from working. If the file did exist, but contained invalid
3455 data, a new version was generated, as expected. It was only the case of a
3456 non-existent file that was broken.
3458 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3459 with a change in libdomainkeys > 0.64.
3461 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3462 from DNS. If the selector record carries the flag, it now has
3463 precedence over the domain-wide flag.
3465 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3467 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3468 the use of a port name that isn't defined in /etc/services) occurred, the
3469 message was deferred as in a normal delivery, and thus remained on the
3470 spool, instead of being failed because of the mua_wrapper setting. This
3471 is now fixed, and I tidied up some of the mua_wrapper messages at the
3474 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3475 the messages in a hash of arrays rather than using individual hashes.
3476 This is a bit cleaner and results in dramatic memory savings, albeit
3477 at a slight CPU cost.
3479 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3480 as requested by Marc Sherman.
3482 SC/10 Eximstats: added histograms for user specified patterns as requested
3485 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3487 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3488 fopen() in the content-scanning modules that did not already have it.
3494 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3495 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3497 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3499 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3501 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3502 to test Sieve filters that use "vacation".
3504 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3505 that changes the way the GnuTLS parameters are stored in the cache file.
3506 The new format can be generated externally. For backward compatibility,
3507 if the data in the cache doesn't make sense, Exim assumes it has read an
3508 old-format file, and it generates new data and writes a new file. This
3509 means that you can't go back to an older release without removing the
3512 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3513 work if there are any delivery delays because "one_time" forces the
3514 parent to be marked "delivered", so its unseen clone is never tried
3515 again. For this reason, Exim now forbids the simultaneous setting of
3518 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3519 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3520 entry if you want to know the details. The fix, however, neglected to
3521 consider the case when local delivery batching is involved. The test for
3522 "previously delivered" was not happening when checking to see if an
3523 address could be batched with a previous (undelivered) one; under
3524 certain circumstances this could lead to multiple deliveries to the same
3527 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3528 in its include files, and this causes problems building Exim.
3530 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3531 header_syntax) but Exim was just ignoring anything given after a slash.
3532 In particular, this caused confusion with an attempt to use "verify =
3533 reverse_host_lookup/defer_ok". An error is now given when options are
3534 supplied for verify items that do not have them. (Maybe reverse_host_
3535 lookup should have a defer_ok option, but that's a different point.)
3537 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3538 defined by RFC 821) to 2048, because there were problems with some AUTH
3539 commands, and RFC 1869 says the size should be increased for extended
3540 SMTP commands that take arguments.
3542 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3545 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3546 "unknown" error; now it says that the functionality isn't in the binary.
3548 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3549 an address' error message when a string expansion fails (syntax or
3550 whatever). Otherwise the password may appear in the log. Following change
3551 PH/42 below, there is no longer a chance of it appearing in a bounce
3554 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3556 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3557 of its addresses was ever tried. (Bugzilla bug #2.)
3559 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3560 the result incorrectly in the debug output. (It correctly added a newline
3561 to what was transported.)
3563 TF/01 Added $received_time.
3565 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3566 commented out examples of how to interface to a virus scanner and to
3567 SpamAssassin. Also added commented examples of av_scanner and
3568 spamd_address settings.
3570 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3571 and controls are allowed in which ACLs. There were a couple of minor
3572 errors. Some of the entries in the conditions table (which is a table of
3573 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3574 negation of where the condition IS allowed.
3576 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3578 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3579 header file does not have a version number, so I've had to invent a new
3580 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3581 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3582 radiusclient), but it was contributed by a Radius user.
3584 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3585 files or directories, for OpenSSL.
3587 PH/19 When an Exim process that is running as root has to create an Exim log
3588 file, it does so in a subprocess that runs as exim:exim so as to get the
3589 ownership right at creation (otherwise, other Exim processes might see
3590 the file with the wrong ownership). There was no test for failure of this
3591 fork() call, which would lead to the process getting stuck as it waited
3592 for a non-existent subprocess. Forks do occasionally fail when resources
3593 run out. I reviewed all the other calls to fork(); they all seem to check
3596 PH/20 When checking for unexpected SMTP input at connect time (before writing
3597 the banner), Exim was not dealing correctly with a non-positive return
3598 from the read() function. If the client had disconnected by this time,
3599 the result was a log entry for a synchronization error with an empty
3600 string after "input=" when read() returned zero. If read() returned -1
3601 (an event I could not check), uninitialized data bytes were printed.
3602 There were reports of junk text (parts of files, etc) appearing after
3605 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3607 PH/22 Added support for macro redefinition, and (re)definition in between
3608 driver and ACL definitions.
3610 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3611 forgetting to use the resulting value; it was using the unexpanded value.
3613 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3614 hadn't been configured. The fix is from Juergen Kreileder, who
3615 understands it better than I do:
3617 "Here's what I see happening with three configured cyrus_sasl
3618 authenticators configured (plain, login, cram-md5):
3620 On startup auth_cyrus_sasl_init() gets called for each of these.
3621 This means three calls to sasl_listmech() without a specified mech_list.
3622 => SASL tests which mechs of all available mechs actually work
3623 => three warnings about OTP not working
3624 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3626 With the patch, sasl_listmech() also gets called three times. But now
3627 SASL's mech_list option is set to the server_mech specified in the the
3628 authenticator. Or in other words, the answer from sasl_listmech()
3629 gets limited to just the mech you're testing for (which is different
3631 => the return list contains just 'plain' or 'login', 'cram-md5' or
3632 nothing depending on the value of ob->server_mech.
3634 I've just tested the patch: Authentication still works fine,
3635 unavailable mechs specified in the exim configuration are still
3636 caught, and the auth.log warnings about OTP are gone."
3638 PH/25 When debugging is enabled, the contents of the command line are added
3639 to the debugging output, even when log_selector=+arguments is not
3642 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3643 answer is "GNU", and only if the return is "GNU/something" is the answer
3646 PH/27 $acl_verify_message is now set immediately after the failure of a
3647 verification in an ACL, and so is available in subsequent modifiers. In
3648 particular, the message can be preserved by coding like this:
3650 warn !verify = sender
3651 set acl_m0 = $acl_verify_message
3653 Previously, $acl_verify_message was set only while expanding "message"
3654 and "log_message" when a very denied access.
3656 PH/28 Modified OS/os.c-Linux with
3658 -#ifndef OS_LOAD_AVERAGE
3659 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3661 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3662 nomenclature these days.)
3664 PH/29 Installed patch from the Sieve maintainer that adds the options
3665 sieve_useraddress and sieve_subaddress to the redirect router.
3667 PH/30 In these circumstances:
3668 . Two addresses routed to the same list of hosts;
3669 . First host does not offer TLS;
3670 . First host accepts first address;
3671 . First host gives temporary error to second address;
3672 . Second host offers TLS and a TLS session is established;
3673 . Second host accepts second address.
3674 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3675 and peerdn, if requested) that were in fact used only for the second
3678 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3679 attention to any local part prefix or suffix that was matched by the
3680 router that accepted the address. It now behaves in the same way as it
3681 does for delivery: the affixes are removed from the local part unless
3682 rcpt_include_affixes is set on the transport.
3684 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3685 timeout during the DATA phase of an incoming message.
3687 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3688 applied a patch sanctioned by the Sieve maintainer.
3690 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3691 the uid or gid is negative. A case of a negative gid caused this to be
3692 noticed. The fix allows for either to be negative.
3694 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3695 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3698 PH/36 The change PH/12 above was broken. Fixed it.
3700 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3701 the grounds that routing the same address twice would always produce the
3702 same answer. This might have been true once, but it is certainly no
3703 longer true now. Routing a child address may depend on the previous
3704 routing that produced that child. Some complicated redirection strategies
3705 went wrong when messages had multiple recipients, and made Exim's
3706 behaviour dependent on the order in which the addresses were given.
3708 I have moved the duplicate checking until after the routing is complete.
3709 Exim scans the addresses that are assigned to local and remote
3710 transports, and removes any duplicates. This means that more work will be
3711 done, as duplicates will always all be routed, but duplicates are
3712 presumably rare, so I don't expect this is of any significance.
3714 For deliveries to pipes, files, and autoreplies, the duplicate checking
3715 still happens during the routing process, since they are not going to be
3718 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3719 It corrects a timeout issue with spamd. This is Ian's comment: "The
3720 background is that sometimes spamd either never reads data from a
3721 connection it has accepted, or it never writes response data. The exiscan
3722 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3723 blindly assumes that writes won't block so it may never time out."
3725 PH/39 Allow G after quota size as well as K and M.
3727 PH/40 The value set for $authenticated_id in an authenticator may not contain
3728 binary zeroes or newlines because the value is written to log lines and
3729 to spool files. There was no check on this. Now the value is run through
3730 the string_printing() function so that such characters are converted to
3731 printable escape sequences.
3733 PH/41 $message_linecount is a new variable that contains the total number of
3734 lines in the message. Compare $body_linecount, which is the count for the
3737 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3738 bounce and delay warning messages, except in certain special cases, which
3741 (a) An SMTP error message from a remote host;
3742 (b) A message specified in a :fail: redirection;
3743 (c) A message specified in a "fail" command in a system filter;
3744 (d) A message specified in a FAIL return from the queryprogram router;
3745 (e) A message specified by the cannot_route_message router option.
3747 In these cases only, Exim does include the error details in bounce and
3748 warning messages. There are also a few cases where bland messages such
3749 as "unrouteable address" or "local delivery error" are given.
3751 PH/43 $value is now also set for the "else" part of a ${run expansion.
3753 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3754 being worked on, but at least Exim now implements the latest version to
3757 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3758 process to complete was treated as a delivery failure, a timeout while
3759 writing the message to the pipe was logged, but erroneously treated as a
3760 successful delivery. Such timeouts include transport filter timeouts. For
3761 consistency with the overall process timeout, these timeouts are now
3762 treated as errors, giving rise to delivery failures by default. However,
3763 there is now a new Boolean option for the pipe transport called
3764 timeout_defer, which, if set TRUE, converts the failures into defers for
3765 both kinds of timeout. A transport filter timeout is now identified in
3768 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3769 systems where "make" and "gmake" are different, calling "gmake" at top
3770 level broke things. I've arranged for the value of $(MAKE) to be passed
3771 from the Makefile to this script so that it can call the same version of
3775 A note about Exim versions 4.44 and 4.50
3776 ----------------------------------------
3778 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3779 changes of various kinds. As a consequence, a big documentation update was
3780 needed. This delayed the release for rather longer than seemed good, especially
3781 in the light of a couple of (minor) security issues. Therefore, the changes
3782 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3783 release. So 4.44 and 4.50 are in effect two different branches that both start
3786 I have left the 4.50 change log unchanged; it contains all the changes since
3787 4.43. The change log for 4.44 is below; many of its items are identical to
3788 those for 4.50. This seems to be the most sensible way to preserve the
3789 historical information.
3795 1. Minor wording change to the doc/README.SIEVE file.
3797 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3798 computation of the current number of files was incorrect.
3800 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3801 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3804 4. Give more explanation in the error message when the command for a transport
3805 filter fails to execute.
3807 5. There are several places where Exim runs a non-Exim command in a
3808 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3809 was being done only for the command run by the queryprogram router. It is
3810 now done for all such subprocesses. The other cases are: ${run, transport
3811 filters, and the commands run by the lmtp and pipe transports.
3813 6. Added CONFIGURE_GROUP build-time option.
3815 7. Some older OS have a limit of 256 on the maximum number of file
3816 descriptors. Exim was using setrlimit() to set 1000 as a large value
3817 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3818 systems. I've change it so that if it can't get 1000, it tries for 256.
3820 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3821 was an oversight, and furthermore, ever since the addition of extra
3822 controls (e.g. 4.43/32), the checks on when to allow different forms of
3823 "control" were broken. There should now be diagnostics for all cases when a
3824 control that does not make sense is encountered.
3826 9. Added the /retain_sender option to "control=submission".
3828 10. $recipients is now available in the predata ACL (oversight).
3830 11. Tidy the search cache before the fork to do a delivery from a message
3831 received from the command line. Otherwise the child will trigger a lookup
3832 failure and thereby defer the delivery if it tries to use (for example) a
3833 cached ldap connection that the parent has called unbind on.
3835 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3836 of $address_data from the recipient verification was clobbered by the
3837 sender verification.
3839 13. The value of address_data from a sender verification is now available in
3840 $sender_address_data in subsequent conditions in the ACL statement.
3842 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3844 15. Added a new option "connect=<time>" to callout options, to set a different
3847 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3848 was its contents. (It was OK if the option was not defined at all.)
3850 17. A "Completed" log line is now written for messages that are removed from
3851 the spool by the -Mrm option.
3853 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3854 information about exactly what failed.
3856 19. Added -dd to debug only the daemon process.
3858 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3859 handles timeouts, both on the server side and network timeouts. Renamed the
3860 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3862 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3863 It is now set to "smtps".
3865 22. $host_address is now set to the target address during the checking of
3866 ignore_target_hosts.
3868 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3869 being passed; this would have caused $sender_host_name to have been used if
3870 matching the list had actually called for a host name (not very likely,
3871 since this list is usually IP addresses). A host name is now passed as
3874 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3875 code that specifies a non-restarting handler (typically sigaction() in
3876 modern systems) in an attempt to fix a rare and obscure crash bug.
3878 25. Narrowed the window for a race in the daemon that could cause it to ignore
3879 SIGCHLD signals. This is not a major problem, because they are used only to
3880 wake it up if nothing else does.
3882 26. A malformed maildirsize file could cause Exim to calculate negative values
3883 for the mailbox size or file count. Odd effects could occur as a result.
3884 The maildirsize information is now recalculated if the size or filecount
3887 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3888 support for a long time. Removed HAVE_SYS_VFS_H.
3890 28. Installed the latest version of exipick from John Jetmore.
3892 29. In an address list, if the pattern was not a regular expression, an empty
3893 subject address (from a bounce message) matched only if the pattern was an
3894 empty string. Non-empty patterns were not even tested. This was the wrong
3895 because it is perfectly reasonable to use an empty address as part of a
3896 database query. An empty address is now tested by patterns that are
3897 lookups. However, all the other forms of pattern expect the subject to
3898 contain a local part and a domain, and therefore, for them, an empty
3899 address still always fails if the pattern is not itself empty.
3901 30. Exim went into a mad DNS loop when attempting to do a callout where the
3902 host was specified on an smtp transport, and looking it up yielded more
3903 than one IP address.
3905 31. Re-factored the code for checking spool and log partition space into a
3906 function that finds that data and another that does the check. The former
3907 is then used to implement four new variables: $spool_space, $log_space,
3908 $spool_inodes, and $log_inodes.
3910 32. The RFC2047 encoding function was originally intended for short strings
3911 such as real names; it was not keeping to the 75-character limit for
3912 encoded words that the RFC imposes. It now respects the limit, and
3913 generates multiple encoded words if necessary. To be on the safe side, I
3914 have increased the buffer size for the ${rfc2047: expansion operator from
3917 33. It is now permitted to omit both strings after an "if" condition; if the
3918 condition is true, the result is "true". As before, when the second string
3919 is omitted, a false condition yields an empty string. This makes it less
3920 cumbersome to write custom ACL and router conditions.
3922 34. Failure to deliver a bounce message always caused it to be frozen, even if
3923 there was an errors_to setting on the router. The errors_to setting is now
3926 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3927 canonical form (fully expanded) before being placed in
3928 $sender_host_address.
3930 36. The table in the code that translates DNS record types into text (T_A to
3931 "A" for instance) was missing entries for NS and CNAME. It is just possible
3932 that this could have caused confusion if both these types were looked up
3933 for the same domain, because the text type is used as part of Exim's
3934 per-process caching. But the chance of anyone hitting this buglet seems
3937 37. The dnsdb lookup has been extended in a number of ways.
3939 (1) There is a new type, "zns", which walks up the domain tree until it
3940 finds some nameserver records. It should be used with care.
3942 (2) There is a new type, "mxh", which is like "mx" except that it returns
3943 just the host names, not the priorities.
3945 (3) It is now possible to give a list of domains (or IP addresses) to be
3946 looked up. The behaviour when one of the lookups defers can be
3947 controlled by a keyword.
3949 (4) It is now possible to specify the separator character for use when
3950 multiple records are returned.
3952 38. The dnslists ACL condition has been extended: it is now possible to supply
3953 a list of IP addresses and/or domains to be looked up in a particular DNS
3956 39. Added log_selector=+queue_time_overall.
3958 40. When running the queue in the test harness, wait just a tad after forking a
3959 delivery process, to get repeatability of debugging output.
3961 41. Include certificate and key file names in error message when GnuTLS fails
3962 to set them up, because the GnuTLS error message doesn't include the name
3963 of the failing file when there is a problem reading it.
3965 42. Allow both -bf and -bF in the same test run.
3967 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3969 44. The "Exiscan patch" is now merged into the mainline Exim source.
3971 45. Sometimes the final signoff response after QUIT could fail to get
3972 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3973 < 0 before doing a fflush(). This bug looks as though it goes back to the
3974 introduction of TLS in release 3.20, but "sometimes" must have been rare
3975 because the tests only now provoked it.
3977 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3978 (this can affect the format of dates).
3980 47. exim_tidydb, when checking for the continued existence of a message for
3981 which it has found a message-specific retry record, was not finding
3982 messages that were in split spool directories. Consequently, it was
3983 deleting retry records that should have stayed in existence.
3985 48. Steve fixed some bugs in eximstats.
3987 49. The SPA authentication driver was not abandoning authentication and moving
3988 on to the next authenticator when an expansion was forced to fail,
3989 contradicting the general specification for all authenticators. Instead it
3990 was generating a temporary error. It now behaves as specified.
3992 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3993 (the order specifies the preference for clients). The order is now AES256,
3994 AES128, 3DES, ARCFOUR128.
3996 51. Small patch to Sieve code - explicitly set From: when generating an
3999 52. Exim crashed if a remote delivery caused a very long error message to be
4000 recorded - for instance if somebody sent an entire SpamAssassin report back
4001 as a large number of 550 error lines. This bug was coincidentally fixed by
4002 increasing the size of one of Exim's internal buffers (big_buffer) that
4003 happened as part of the Exiscan merge. However, to be on the safe side, I
4004 have made the code more robust (and fixed the comments that describe what
4007 53. Now that there can be additional text after "Completed" in log lines (if
4008 the queue_time_overall log selector is set), a one-byte patch to exigrep
4009 was needed to allow it to recognize "Completed" as not the last thing in
4012 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4013 patch that reportedly fixes this has been added. I am not expert enough to
4014 create a test for it. This is what the patch creator wrote:
4016 "I found a little strange behaviour of ldap code when working with
4017 Windows 2003 AD Domain, where users was placed in more than one
4018 Organization Units. When I tried to give exim partial DN, the exit code
4019 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4020 But simultaneously result of request was absolutely normal ldap result,
4021 so I produce this patch..."
4023 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4024 REFERENCE, so I have modified the code to exclude the patch when that macro
4027 55. Some experimental protocols are using DNS PTR records for new purposes. The
4028 keys for these records are domain names, not reversed IP addresses. The
4029 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4030 leaves it alone. Component reversal etc. now happens only for IP addresses.
4033 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4035 57. Double the size of the debug message buffer (to 2048) so that more of very
4036 long debug lines gets shown.
4038 58. The exicyclog utility now does better if the number of log files to keep
4039 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4041 59. Two changes related to the smtp_active_hostname option:
4043 (1) $smtp_active_hostname is now available as a variable.
4044 (2) The default for smtp_banner uses $smtp_active_hostname instead
4045 of $primary_hostname.
4047 60. The host_aton() function is supposed to be passed a string that is known
4048 to be a valid IP address. However, in the case of IPv6 addresses, it was
4049 not checking this. This is a hostage to fortune. Exim now panics and dies
4050 if the condition is not met. A case was found where this could be provoked
4051 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4052 components; fortuitously, this particular loophole had already been fixed
4053 by change 4.50/55 above.
4055 If there are any other similar loopholes, the new check in host_aton()
4056 itself should stop them being exploited. The report I received stated that
4057 data on the command line could provoke the exploit when Exim was running as
4058 exim, but did not say which command line option was involved. All I could
4059 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4060 running as the user.
4063 61. There was a buffer overflow vulnerability in the SPA authentication code
4064 (which came originally from the Samba project). I have added a test to the
4065 spa_base64_to_bits() function which I hope fixes it.
4068 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4069 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4071 63. The daemon start-up calls getloadavg() while still root for those OS that
4072 need the first call to be done as root, but it missed one case: when
4073 deliver_queue_load_max is set with deliver_drop_privilege. This is
4074 necessary for the benefit of the queue runner, because there is no re-exec
4075 when deliver_drop_privilege is set.
4077 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4078 This has been fixed.
4080 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4081 was in use, was not putting the data itself into the right store pool;
4082 consequently, it could be overwritten for a subsequent message in the same
4083 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4086 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4088 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4089 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4090 the function were treating the return as a boolean value, which happened to
4091 work because 0=false and not-0=true, but is not correct code.
4093 68. The host_aton() function was not handling scoped IPv6 addresses (those
4094 with, for example, "%eth0" on the end) correctly.
4096 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4097 negated items (that is, ~something) in unsigned ints. Some compilers
4098 apparently mutter when there is no cast.
4100 70. If an address verification called from an ACL failed, and did not produce a
4101 user-specific message (i.e. there was only a "system" message), nothing was
4102 put in $acl_verify_message. In this situation, it now puts the system
4105 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4106 session; change 4.31/43 added the unwanted input to the log line - except
4107 that it did not do this in the start of session case. It now does.
4109 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4110 This is wrong and can cause the other end to generate a synchronization
4111 error if it is another Exim or anything else that does the synchronization
4112 check. A QUIT command is no longer sent after a timeout.
4114 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4115 during host lookups.
4117 74. The defer_ok option of callout verification was not working if it was used
4118 when verifying addresses in header lines, that is, for this case:
4120 verify = header_sender/callout=defer_ok
4122 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4123 those file descriptors could be used for SMTP connections. If anything
4124 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4125 could be sent to the SMTP client, causing chaos. The daemon now opens
4126 stdin, stdout, and stderr to /dev/null when it puts itself into the
4129 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4130 log by default. The user can override this with suitable Perl magic.
4132 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4133 the log message when discard triggers, was not working for the DATA ACL or
4134 for the non-SMTP ACL.
4136 78. Error message wording change in sieve.c.
4138 79. If smtp_accept_max_per_host was set, the number of connections could be
4139 restricted to fewer than expected, because the daemon was trying to set up
4140 a new connection before checking whether the processes handling previous
4141 connections had finished. The check for completed processes is now done
4142 earlier. On busy systems, this bug wouldn't be noticed because something
4143 else would have woken the daemon, and it would have reaped the completed
4146 80. If a message was submitted locally by a user whose login name contained one
4147 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4148 It caused a spool format error. I have fixed the spool reading code. A
4149 related problem was that the "from" clause in the Received: line became
4150 illegal because of the space(s). It is now covered by ${quote_local_part.
4152 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4155 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4157 83. Patch from Sieve maintainer to fix unterminated string problem in
4158 "vacation" handling.
4160 84. Some minor changes to the Linux configuration files to help with other
4161 OS variants using glibc.
4163 85. One more patch for Sieve to update vacation handling to latest spec.
4166 ----------------------------------------------------
4167 See the note above about the 4.44 and 4.50 releases.
4168 ----------------------------------------------------
4174 1. Change 4.43/35 introduced a bug that caused file counts to be
4175 incorrectly computed when quota_filecount was set in an appendfile
4178 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4179 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4182 3. Give more explanation in the error message when the command for a transport
4183 filter fails to execute.
4185 4. There are several places where Exim runs a non-Exim command in a
4186 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4187 was being done only for the command run by the queryprogram router. It is
4188 now done for all such subprocesses. The other cases are: ${run, transport
4189 filters, and the commands run by the lmtp and pipe transports.
4191 5. Some older OS have a limit of 256 on the maximum number of file
4192 descriptors. Exim was using setrlimit() to set 1000 as a large value
4193 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4194 systems. I've change it so that if it can't get 1000, it tries for 256.
4196 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4197 was an oversight, and furthermore, ever since the addition of extra
4198 controls (e.g. 4.43/32), the checks on when to allow different forms of
4199 "control" were broken. There should now be diagnostics for all cases when a
4200 control that does not make sense is encountered.
4202 7. $recipients is now available in the predata ACL (oversight).
4204 8. Tidy the search cache before the fork to do a delivery from a message
4205 received from the command line. Otherwise the child will trigger a lookup
4206 failure and thereby defer the delivery if it tries to use (for example) a
4207 cached ldap connection that the parent has called unbind on.
4209 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4210 of $address_data from the recipient verification was clobbered by the
4211 sender verification.
4213 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4214 was its contents. (It was OK if the option was not defined at all.)
4216 11. A "Completed" log line is now written for messages that are removed from
4217 the spool by the -Mrm option.
4219 12. $host_address is now set to the target address during the checking of
4220 ignore_target_hosts.
4222 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4223 being passed; this would have caused $sender_host_name to have been used if
4224 matching the list had actually called for a host name (not very likely,
4225 since this list is usually IP addresses). A host name is now passed as
4228 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4229 code that specifies a non-restarting handler (typically sigaction() in
4230 modern systems) in an attempt to fix a rare and obscure crash bug.
4232 15. Narrowed the window for a race in the daemon that could cause it to ignore
4233 SIGCHLD signals. This is not a major problem, because they are used only to
4234 wake it up if nothing else does.
4236 16. A malformed maildirsize file could cause Exim to calculate negative values
4237 for the mailbox size or file count. Odd effects could occur as a result.
4238 The maildirsize information is now recalculated if the size or filecount
4241 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4242 support for a long time. Removed HAVE_SYS_VFS_H.
4244 18. Updated exipick to current release from John Jetmore.
4246 19. Allow an empty sender to be matched against a lookup in an address list.
4247 Previously the only cases considered were a regular expression, or an
4250 20. Exim went into a mad DNS lookup loop when doing a callout where the
4251 host was specified on the transport, if the DNS lookup yielded more than
4254 21. The RFC2047 encoding function was originally intended for short strings
4255 such as real names; it was not keeping to the 75-character limit for
4256 encoded words that the RFC imposes. It now respects the limit, and
4257 generates multiple encoded words if necessary. To be on the safe side, I
4258 have increased the buffer size for the ${rfc2047: expansion operator from
4261 22. Failure to deliver a bounce message always caused it to be frozen, even if
4262 there was an errors_to setting on the router. The errors_to setting is now
4265 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4266 canonical form (fully expanded) before being placed in
4267 $sender_host_address.
4269 24. Updated eximstats to version 1.33
4271 25. Include certificate and key file names in error message when GnuTLS fails
4272 to set them up, because the GnuTLS error message doesn't include the name
4273 of the failing file when there is a problem reading it.
4275 26. Expand error message when OpenSSL has problems setting up cert/key files.
4278 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4279 (this can affect the format of dates).
4281 28. exim_tidydb, when checking for the continued existence of a message for
4282 which it has found a message-specific retry record, was not finding
4283 messages that were in split spool directories. Consequently, it was
4284 deleting retry records that should have stayed in existence.
4286 29. eximstats updated to version 1.35
4287 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4288 1.35 - bugfix such that pie charts by volume are generated correctly
4290 30. The SPA authentication driver was not abandoning authentication and moving
4291 on to the next authenticator when an expansion was forced to fail,
4292 contradicting the general specification for all authenticators. Instead it
4293 was generating a temporary error. It now behaves as specified.
4295 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4296 (the order specifies the preference for clients). The order is now AES256,
4297 AES128, 3DES, ARCFOUR128.
4299 31. Small patch to Sieve code - explicitly set From: when generating an
4302 32. Exim crashed if a remote delivery caused a very long error message to be
4303 recorded - for instance if somebody sent an entire SpamAssassin report back
4304 as a large number of 550 error lines. This bug was coincidentally fixed by
4305 increasing the size of one of Exim's internal buffers (big_buffer) that
4306 happened as part of the Exiscan merge. However, to be on the safe side, I
4307 have made the code more robust (and fixed the comments that describe what
4310 33. Some experimental protocols are using DNS PTR records for new purposes. The
4311 keys for these records are domain names, not reversed IP addresses. The
4312 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4313 leaves it alone. Component reversal etc. now happens only for IP addresses.
4316 34. The host_aton() function is supposed to be passed a string that is known
4317 to be a valid IP address. However, in the case of IPv6 addresses, it was
4318 not checking this. This is a hostage to fortune. Exim now panics and dies
4319 if the condition is not met. A case was found where this could be provoked
4320 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4321 components; fortuitously, this particular loophole had already been fixed
4322 by change 4.50/55 or 4.44/33 above.
4324 If there are any other similar loopholes, the new check in host_aton()
4325 itself should stop them being exploited. The report I received stated that
4326 data on the command line could provoke the exploit when Exim was running as
4327 exim, but did not say which command line option was involved. All I could
4328 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4329 running as the user.
4332 35. There was a buffer overflow vulnerability in the SPA authentication code
4333 (which came originally from the Samba project). I have added a test to the
4334 spa_base64_to_bits() function which I hope fixes it.
4337 36. The daemon start-up calls getloadavg() while still root for those OS that
4338 need the first call to be done as root, but it missed one case: when
4339 deliver_queue_load_max is set with deliver_drop_privilege. This is
4340 necessary for the benefit of the queue runner, because there is no re-exec
4341 when deliver_drop_privilege is set.
4343 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4344 was in use, was not putting the data itself into the right store pool;
4345 consequently, it could be overwritten for a subsequent message in the same
4346 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4349 38. Sometimes the final signoff response after QUIT could fail to get
4350 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4351 < 0 before doing a fflush(). This bug looks as though it goes back to the
4352 introduction of TLS in release 3.20, but "sometimes" must have been rare
4353 because the tests only now provoked it.
4359 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4360 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4361 Now it returns an integer. A number of calls were still expecting a T/F
4362 return. Fortuitously, in all cases, the tests worked in OK situations,
4363 which is the norm. However, things would have gone wrong on any write
4364 failures on the smtp file descriptor. This function is used when sending
4365 messages over SMTP and also when doing verify callouts.
4367 2. When Exim is called to do synchronous delivery of a locally submitted
4368 message (the -odf or -odi options), it no longer closes stderr before doing
4371 3. Implemented the mua_wrapper option.
4373 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4375 5. Implemented the functions header_remove(), header_testname(),
4376 header_add_at_position(), and receive_remove_recipient(), and exported them
4379 6. If an ACL "warn" statement specified the addition of headers, Exim already
4380 inserted X-ACL-Warn: at the start if there was no header name. However, it
4381 was not making this test for the second and subsequent header lines if
4382 there were newlines in the string. This meant that an invalid header could
4383 be inserted if Exim was badly configured.
4385 7. Allow an ACL "warn" statement to add header lines at the start or after all
4386 the Received: headers, as well as at the end.
4388 8. Added the rcpt_4xx retry error code.
4390 9. Added postmaster_mailfrom=xxx to callout verification option.
4392 10. Added mailfrom=xxxx to the callout verification option, for verify=
4395 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4396 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4398 12. Inserted some casts to stop certain compilers warning when using pointer
4399 differences as field lengths or precisions in printf-type calls (mostly
4400 affecting debugging statements).
4402 13. Added optional readline() support for -be (dynamically loaded).
4404 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4405 same clock tick as a message's arrival, so that its received time was the
4406 same as the "first fail" time on the retry record, and that message
4407 remained on the queue past the ultimate address timeout, every queue runner
4408 would try a delivery (because it was past the ultimate address timeout) but
4409 after another failure, the ultimate address timeout, which should have then
4410 bounced the address, did not kick in. This was a "< instead of <=" error;
4411 in most cases the first failure would have been in the next clock tick
4412 after the received time, and all would be well.
4414 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4415 being recognized when the domain list was tested by the match_domain
4416 condition in an expansion string.
4418 16. Added the ${str2b64: operator.
4420 17. Exim was always calling setrlimit() to set a large limit for the number of
4421 processes, without checking whether the existing limit was already
4422 adequate. (It did check for the limit on file descriptors.) Furthermore,
4423 errors from getrlimit() and setrlimit() were being ignored. Now they are
4424 logged to the main and panic logs, but Exim does carry on, to try to do its
4425 job under whatever limits there are.
4427 18. Imported PCRE 5.0.
4429 19. Trivial typo in log message " temporarily refused connection" (the leading
4432 20. If the log selector return_path_on_delivery was set and an address was
4433 redirected to /dev/null, the delivery process crashed because it assumed
4434 that a return path would always be set for a "successful" delivery. In this
4435 case, the whole delivery is bypassed as an optimization, and therefore no
4438 21. Internal re-arrangement: the function for sending a challenge and reading
4439 a response while authentication was assuming a zero-terminated challenge
4440 string. It's now changed to take a pointer and a length, to allow for
4441 binary data in such strings.
4443 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4445 23. Exim was not respecting finduser_retries when seeking the login of the
4446 uid under which it was called; it was always trying 10 times. (The default
4447 setting of finduser_retries is zero.) Also, it was sleeping after the final
4448 failure, which is pointless.
4450 24. Implemented tls_on_connect_ports.
4452 25. Implemented acl_smtp_predata.
4454 26. If the domain in control=submission is set empty, Exim assumes that the
4455 authenticated id is a complete email address when it generates From: or
4456 Sender: header lines.
4458 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4459 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4460 chown and chgrp in /bin and hostname in /usr/bin.
4462 28. Exim was keeping the "process log" file open after each use, just as it
4463 does for the main log. This opens the possibility of it remaining open for
4464 long periods when the USR1 signal hits a daemon. Occasional processlog
4465 errors were reported, that could have been caused by this. Anyway, it seems
4466 much more sensible not to leave this file open at all, so that is what now
4469 29. The long-running daemon process does not normally write to the log once it
4470 has entered its main loop, and it closes the log before doing so. This is
4471 so that log files can straightforwardly be renamed and moved. However,
4472 there are a couple of unusual error situations where the daemon does write
4473 log entries, and I had neglected to close the log afterwards.
4475 30. The text of an SMTP error response that was received during a remote
4476 delivery was being truncated at 512 bytes. This is too short for some of
4477 the long messages that one sometimes sees. I've increased the limit to
4480 31. It is now possible to make retry rules that apply only when a message has a
4481 specific sender, in particular, an empty sender.
4483 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4484 it possible to be selective about when SMTP synchronization is enforced.
4486 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4488 32. Implemented hosts_connection_nolog.
4490 33. Added an ACL for QUIT.
4492 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4495 35. Added mailbox_size and mailbox_filecount to appendfile.
4497 36. Added control = no_multiline_responses to ACLs.
4499 37. There was a bug in the logic of the code that waits for the clock to tick
4500 in the case where the clock went backwards by a substantial amount such
4501 that the microsecond fraction of "now" was more than the microsecond
4502 fraction of "then" (but the whole seconds number was less).
4504 38. Added support for the libradius Radius client library this is found on
4505 FreeBSD (previously only the radiusclient library was supported).
4511 1. When certain lookups returned multiple values in the form name=value, the
4512 quoting of the values was not always being done properly. Specifically:
4513 (a) If the value started with a double quote, but contained no whitespace,
4515 (b) If the value contained whitespace other than a space character (i.e.
4516 tabs or newlines or carriage returns) it was not quoted.
4517 This fix has been applied to the mysql and pgsql lookups by writing a
4518 separate quoting function and calling it from the lookup code. The fix
4519 should probably also be applied to nisplus, ibase and oracle lookups, but
4520 since I cannot test any of those, I have not disturbed their existing code.
4522 2. A hit in the callout cache for a specific address caused a log line with no
4523 reason for rejecting RCPT. Now it says "Previous (cached) callout
4524 verification failure".
4526 3. There was an off-by-one bug in the queryprogram router. An over-long
4527 return line was truncated at 256 instead of 255 characters, thereby
4528 overflowing its buffer with the terminating zero. As well as fixing this, I
4529 have increased the buffer size to 1024 (and made a note to document this).
4531 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4532 when Exim is waiting for an SMTP response from a remote server, Exim
4533 restarts its select() call on the socket, thereby resetting its timeout.
4534 This is not a problem when such interrupts are rare. Somebody set up a cron
4535 job to run exiwhat every 2 minutes, which is less than the normal select()
4536 timeout (5 or 10 minutes). This meant that the select() timeout never
4537 kicked in because it was always reset. I have fixed this by comparing the
4538 time when an interrupt arrives with the time at the start of the first call
4539 to select(). If more time than the timeout has elapsed, the interrupt is
4540 treated as a timeout.
4542 5. Some internal re-factoring in preparation for the addition of Sieve
4543 extensions (by MH). In particular, the "personal" test is moved to a
4544 separate function, and given an option for scanning Cc: and Bcc: (which is
4545 not set for Exim filters).
4547 6. When Exim created an email address using the login of the caller as the
4548 local part (e.g. when creating a From: or Sender: header line), it was not
4549 quoting the local part when it contained special characters such as @.
4551 7. Installed new OpenBSD configuration files.
4553 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4554 try to make them clearer.
4556 9. Callout options, other than the timeout value, were being ignored when
4557 verifying sender addresses in header lines. For example, when using
4559 verify = header_sender/callout=no_cache
4561 the cache was (incorrectly) being used.
4563 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4564 only the Cygwin environment.
4566 11. When return_path_on_delivery was set as a log selector, if different remote
4567 addresses in the same message used different return paths and parallel
4568 remote delivery occurred, the wrong values would sometimes be logged.
4569 (Whenever a remote delivery process finished, the return path value from
4570 the most recently started remote delivery process was logged.)
4572 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4573 lines when AUTH and/or TLS are in use. This is the "received protocol"
4574 field. Exim used to use "asmtp" for authenticated SMTP, without any
4575 indication (in the protocol name) for TLS use. Now it follows the RFC and
4576 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4577 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4578 names appear in log lines as well as in Received: header lines.
4580 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4581 extensions, and comparison tests, and to fix some bugs.
4583 14. Changes to the "personal" filter test:
4585 (1) The test was buggy in that it was just doing the equivalent of
4586 "contains" tests on header lines. For example, if a user's address was
4587 anne@some.where, the "personal" test would incorrectly be true for
4589 To: susanne@some.where
4591 This test is now done by extracting each address from the header in turn,
4592 and checking the entire address. Other tests that are part of "personal"
4593 are now done using regular expressions (for example, to check local parts
4594 of addresses in From: header lines).
4596 (2) The list of non-personal local parts in From: addresses has been
4597 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4598 taken from the Sieve specification recommendations.
4600 (3) If the message contains any header line starting with "List-" it is
4601 treated as non-personal.
4603 (4) The test for "circular" in the Subject: header line has been removed
4604 because it now seems ill-conceived.
4606 15. Minor typos in src/EDITME comments corrected.
4608 16. Installed latest exipick from John Jetmore.
4610 17. If headers_add on a router specified a text string that was too long for
4611 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4612 of string_sprintf() is now avoided.
4614 18. $message_body_size was not set (it was always zero) when running the DATA
4615 ACL and the local_scan() function.
4617 19. For the "mail" command in an Exim filter, no default was being set for
4618 the once_repeat time, causing a random time value to be used if "once" was
4619 specified. (If the value happened to be <= 0, no repeat happened.) The
4620 default is now 0s, meaning "never repeat". The "vacation" command was OK
4621 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4622 (I found it when inspecting the code).
4624 20. There is now an overall timeout for performing a callout verification. It
4625 defaults to 4 times the callout timeout, which applies to individual SMTP
4626 commands during the callout. The overall timeout applies when there is more
4627 than one host that can be tried. The timeout is checked before trying the
4628 next host. This prevents very long delays if there are a large number of
4629 hosts and all are timing out (e.g. when the network connections are timing
4630 out). The value of the overall timeout can be changed by specifying an
4631 additional sub-option for "callout", called "maxwait". For example:
4633 verify = sender/callout=5s,maxwait=20s
4635 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4636 to the end before writing, but this should make it even safer).
4638 22. Exim was forgetting that it had advertised PIPELINING for the second and
4639 subsequent messages on an SMTP connection. It was also not resetting its
4640 memory on STARTTLS and an internal HELO.
4642 23. When Exim logs an SMTP synchronization error within a session, it now
4643 records whether PIPELINING has been advertised or not.
4645 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4646 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4647 rather than long int.
4649 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4651 26. Added the never_mail option to autoreply.
4657 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4658 crash if the getsockname() call failed; this can happen if a connection is
4659 closed very soon after it is established. The problem was simply in the
4660 order in which certain operations were done, causing Exim to try to write
4661 to the SMTP stream before it had set up the file descriptor. The bug has
4662 been fixed by making things happen in the correct order.
4668 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4669 before the connection was closed, thus losing the rejection response.
4671 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4672 some early Solaris releases, but causes trouble in current releases where
4673 socklen_t is defined.
4675 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4678 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4681 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4682 that releasing the top of it at the end releases what was used for sub-
4683 expansions (unless the block got too big). However, discard this block if
4684 the first thing is a variable or header, so that we can use its block when
4685 it is dynamic (useful for very large $message_headers, for example).
4687 6. Lookups now cache *every* query, not just the most recent. A new, separate
4688 store pool is used for this. It can be recovered when all lookup caches are
4689 flushed. Lookups now release memory at the end of their result strings.
4690 This has involved some general refactoring of the lookup sources.
4692 7. Some code has been added to the store_xxx() functions to reduce the amount
4693 of flapping under certain conditions.
4695 8. log_incoming_interface used to affect only the <= reception log lines. Now
4696 it causes the local interface and port to be added to several more SMTP log
4697 lines, for example "SMTP connection from", and rejection lines.
4699 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4701 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4703 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4704 could be overwritten at the end of the current message (or the start of a
4705 new message if it was set in a HELO ACL). The value is now preserved for
4706 the duration of the SMTP connection.
4708 12. If a transport had a headers_rewrite setting, and a matching header line
4709 contained an unqualified address, that address was qualified, even if it
4710 did not match any rewriting rules. The underlying bug was that the values
4711 of the flags that permit the existence of unqualified sender and recipient
4712 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4713 non-local messages, and by -bnq for local messages) were not being
4714 preserved with the message after it was received.
4716 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4717 "next input=" as part of the text comprising the host identity instead of
4718 the correct text. The code was using the same buffer for two different
4719 strings. However, depending on which order the printing function evaluated
4720 its arguments, the bug did not always show up. Under Linux, for example, my
4721 test suite worked just fine.
4723 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4724 doesn't work with some older versions of Perl. It has been changed to "my",
4725 which in any case is probably the better facility to use.
4727 15. A really picky compiler found some instances of statements for creating
4728 error messages that either had too many or two few arguments for the format
4731 16. The size of the buffer for calls to the DNS resolver has been increased
4732 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4733 for addresses that have a lot of PTR records. This alleviates a problem; it
4734 does not fully solve it.
4736 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4737 buffer now truncates the list and logs the incident, which is the same
4738 action as happens when Exim is looking up a host name and its aliases.
4739 Previously in this situation something unpredictable would happen;
4740 sometimes it was "internal error: store_reset failed".
4742 18. If a server dropped the connection unexpectedly when an Exim client was
4743 using GnuTLS and trying to read a response, the client delivery process
4744 crashed while trying to generate an error log message.
4746 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4747 string, for example:
4749 warn message = H1: something\nH2: something
4751 the text was added as a single header line from Exim's point of view
4752 though it ended up OK in the delivered message. However, searching for the
4753 second and subsequent header lines using $h_h2: did not work. This has been
4754 fixed. Similarly, if a system filter added multiple headers in this way,
4755 the routers could not see them.
4757 20. Expanded the error message when iplsearch is called with an invalid key to
4758 suggest using net-iplsearch in a host list.
4760 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4761 ACLs are no longer actually imposed (and a message to that effect is
4764 22. If a "gecos" field in a passwd entry contained escaped characters, in
4765 particular, if it contained a \" sequence, Exim got it wrong when building
4766 a From: or a Sender: header from that name. A second bug also caused
4767 incorrect handling when an unquoted " was present following a character
4768 that needed quoting.
4770 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4771 was not being matched caselessly.
4773 24. Arranged for all hyphens in the exim.8 source to be escaped with
4776 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4777 properly. Recipient callout cache records were still being keyed to include
4778 the sender, even when use_sender was set false. This led to far more
4779 callouts that were necessary. The sender is no longer included in the key
4780 when use_sender is false.
4782 26. Added "control = submission" modifier to ACLs.
4784 27. Added the ${base62d: operator to decode base 62 numbers.
4786 28. dnsdb lookups can now access SRV records.
4788 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4789 the configuration file.
4791 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4792 (-v) mode. This makes the output for a verbose queue run more intelligible.
4794 31. Added a use_postmaster feature to recipient callouts.
4796 32. Added the $body_zerocount variable, containing the number of binary zero
4797 bytes in the message body.
4799 33. The time of last modification of the "new" subdirectory is now used as the
4800 "mailbox time last read" when there is a quota error for a maildir
4803 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4805 35. Added +ignore_unknown as a special item in host lists.
4807 36. Code for decoding IPv6 addresses in host lists is now included, even if
4808 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4809 address was recognized as an IP address, but was then not correctly decoded
4810 into binary, causing unexpected and incorrect effects when compared with
4817 1. Very minor rewording of debugging text in manualroute to say "list of
4818 hosts" instead of "hostlist".
4820 2. If verify=header_syntax was set, and a header line with an unqualified
4821 address (no domain) and a large number of spaces between the end of the
4822 name and the colon was received, the reception process suffered a buffer
4823 overflow, and (when I tested it) crashed. This was caused by some obsolete
4824 code that should have been removed. The fix is to remove it!
4826 3. When running in the test harness, delay a bit after writing a bounce
4827 message to get a bit more predictability in the log output.
4829 4. Added a call to search_tidyup() just before forking a reception process. In
4830 theory, someone could use a lookup in the expansion of smtp_accept_max_
4831 per_host which, without the tidyup, could leave open a database connection.
4833 5. Added the variables $recipient_data and $sender_data which get set from a
4834 lookup success in an ACL "recipients" or "senders" condition, or a router
4835 "senders" option, similar to $domain_data and $local_part_data.
4837 6. Moved the writing of debug_print from before to after the "senders" test
4840 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4841 problems for message scanning, either using a data ACL, or using
4842 local_scan() because the Received: header was not generated till after they
4843 were called (in order to set the time as the time of reception completion).
4844 I have revised the way this works. The header is now generated after the
4845 body is received, but before the ACL or local_scan() are called. After they
4846 are run, the timestamp in the header is updated.
4852 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4853 before starting a queue runner without re-exec. This happened only when
4854 deliver_drop_privilege was set or when the Exim user was set to root. The
4855 effect of the bug was that timeouts during subsequent deliveries caused
4856 crashes instead of being properly handled. The handler is now left at its
4857 default (and expected) setting.
4859 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4860 message, again when deliver_drop_privilege is set or Exim is run as root.
4861 The bug described in (1) was not present in this case, but the tidying up
4862 of the other signals was missing. I have made the two cases consistent.
4864 3. The ignore_target_hosts setting on a manualroute router was being ignored
4865 for hosts that were looked up using the /MX notation.
4867 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4870 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4871 operated on the sender address. After changing the $sender_address to <>
4872 for the sender address verify, Exim was re-instated it as the original
4873 (before rewriting) address, but remembering that it had rewritten it, so it
4874 wasn't rewriting it again. This bug also had the effect of breaking the
4875 sender address verification caching when the sender address was rewritten.
4877 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4878 This has been changed so that if the ip literal address matches
4879 ignore_target_hosts, the router declines.
4881 7. Added expansion conditions match_domain, match_address, and match_local_
4882 part (NOT match_host).
4884 8. The placeholder for the Received: header didn't have a length field set.
4886 9. Added code to Exim itself and to exim_lock to test for a specific race
4887 condition that could lead to file corruption when using MBX delivery. The
4888 issue is with the lockfile that is created in /tmp. If this file is removed
4889 after a process has opened it but before that process has acquired a lock,
4890 there is the potential for a second process to recreate the file and also
4891 acquire a lock. This could lead to two Exim processes writing to the file
4892 at the same time. The added code performs the same test as UW imapd; it
4893 checks after acquiring the lock that its file descriptor still refers to
4894 the same named file.
4896 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4897 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4900 11. Added the smtp_active_hostname option. If used, this will typically be made
4901 to depend on the incoming interface address. Because $interface_address is
4902 not set up until the daemon has forked a reception process, error responses
4903 that can happen earlier (such as "too many connections") no longer contain
4906 12. If an expansion in a condition on a "warn" statement fails because a lookup
4907 defers, the "warn" statement is abandoned, and the next ACL statement is
4908 processed. Previously this caused the whole ACL to be aborted.
4910 13. Added the iplsearch lookup type.
4912 14. Added ident_timeout as a log selector.
4914 15. Added tls_certificate_verified as a log selector.
4916 16. Added a global option tls_require_ciphers (compare the smtp transport
4917 option of the same name). This controls incoming TLS connections.
4919 17. I finally figured out how to make tls_require_ciphers do a similar thing
4920 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4921 before starting the TLS session.
4923 18. Tabs are now shown as \t in -bP output.
4925 19. If the log selector return_path_on_delivery was set, Exim crashed when
4926 bouncing a message because it had too many Received: header lines.
4928 20. If two routers both had headers_remove settings, and the first one included
4929 a superfluous trailing colon, the final name in the first list and the
4930 first name in the second list were incorrectly joined into one item (with a
4931 colon in the middle).
4937 1. Added -C and -D options to the exinext utility, mainly to make it easier
4938 to include in the automated testing, but these could be helpful when
4939 multiple configurations are in use.
4941 2. The exinext utility was not formatting the output nicely when there was
4942 an alternate port involved in the retry record key, nor when there was a
4943 message id as well (for retries that were specific to a specific message
4944 and a specific host). It was also confused by IPv6 addresses, because of
4945 the additional colons they contain. I have fixed the IPv4 problem, and
4946 patched it up to do a reasonable job for IPv6.
4948 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4949 delivery, the log line now contains "pipelined" if PIPELINING was used.
4951 4. An SMTP transport process used to panic and die if the bind() call to set
4952 an explicit outgoing interface failed. This has been changed; it is now
4953 treated in the same way as a connect() failure.
4955 5. A reference to $sender_host_name in the part of a conditional expansion
4956 that was being skipped was still causing a DNS lookup. This no longer
4959 6. The def: expansion condition was not recognizing references to header lines
4960 that used bh_ and bheader_.
4962 7. Added the _cache feature to named lists.
4964 8. The code for checking quota_filecount in the appendfile transport was
4965 allowing one more file than it should have been.
4967 9. For compatibility with Sendmail, the command line option
4975 and sets the incoming protocol and host name (for trusted callers). The
4976 host name and its colon can be omitted when only the protocol is to be set.
4977 Note the Exim already has two private options, -pd and -ps, that refer to
4978 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4979 "s", but I don't think that's a major issue.
4981 10. A number of refactoring changes to the code, none of which should affect
4984 (a) The number of logging options was getting close to filling up the
4985 32-bit word that was used as a bit map. I have split them into two classes:
4986 those that are passed in the argument to log_write(), and those that are
4987 only ever tested independently outside of that function. These are now in
4988 separate 32-bit words, so there is plenty of room for expansion again.
4989 There is no change in the user interface or the logging behaviour.
4991 (b) When building, for example, log lines, the code previously used a
4992 macro that called string_cat() twice, in order to add two strings. This is
4993 not really sufficiently general. Furthermore, there was one instance where
4994 it was actually wrong because one of the argument was used twice, and in
4995 one call a function was used. (As it happened, calling the function twice
4996 did not affect the overall behaviour.) The macro has been replaced by a
4997 function that can join an arbitrary number of extra strings onto a growing
5000 (c) The code for expansion conditions now uses a table and a binary chop
5001 instead of a serial search (which was left over from when there were very
5002 few conditions). Also, it now recognizes conditions like "pam" even when
5003 the relevant support is not compiled in: a suitably worded error message is
5004 given if an attempt is made to use such a condition.
5006 11. Added ${time_interval:xxxxx}.
5008 12. A bug was causing one of the ddress fields not to be passed back correctly
5009 from remote delivery subprocesses. The field in question was not being
5010 subsequently used, so this caused to problems in practice.
5012 13. Added new log selectors queue_time and deliver_time.
5014 14. Might have fixed a bug in maildirsizefile handling that threw up
5015 "unexpected character" debug warnings, and recalculated the data
5016 unnecessarily. In any case, I expanded the warning message to give more
5019 15. Added the message "Restricted characters in address" to the statements in
5020 the default ACL that block characters like @ and % in local parts.
5022 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5023 Three changes have been made:
5025 (a) There was a serious bug; a negative response to MAIL caused the whole
5026 recipient domain to be cached as invalid, thereby blocking all messages
5027 to all local parts at the same domain, from all senders. This bug has
5028 been fixed. The domain is no longer cached after a negative response to
5029 MAIL if the sender used is not empty.
5031 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5034 (c) A new callout option, "use_sender" has been added for people who want
5035 the modified behaviour.
5041 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5044 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5045 indeed breaks things for older releases.
5047 3. Added additional logging to the case where there is a problem reading data
5048 from a filter that is running in a subprocess using a pipe, in order to
5049 try to track down a specific problem.
5051 4. Testing facility fudge: when running in the test harness and attempting
5052 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5053 getting "No route to host". Convert this to a timeout.
5055 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5058 6. Some OS don't have socklen_t but use size_t instead. This affects the
5059 fifth argument of getsockopt() amongst other things. This is now
5060 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5061 can be set for individual OS. I have set it for SunOS5, OSF1, and
5062 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5063 some earlier ones do not.
5065 7. Change 4.30/15 was not doing the test caselessly.
5067 8. The standard form for an IPv6 address literal was being rejected by address
5068 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5069 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5070 this, as well as the form without the "IPv6" on the front (but only when
5071 address literals are enabled, of course).
5073 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5075 10. Exim crashed if a message with an empty sender address specified by -f
5076 encountered a router with an errors_to setting. This could be provoked only
5077 by a command such as
5081 where an empty string was supplied; "<>" did not hit this bug.
5083 11. Installed PCRE release 4.5.
5085 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5086 remained set. It is now erased.
5088 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5089 times from message ids (which are base 36 rather than the normal 62).
5091 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5092 were being counted as actual protocol errors, and logged if the log
5093 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5094 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5095 and DATA following a set of rejected RCPTs do not count as protocol errors.
5096 In other words, Exim assumes they were pipelined, though this may not
5097 actually be the case. Of course, in all cases the client gets an
5098 appropriate error code.
5100 15. If a lookup fails in an ACL condition, a message about the failure may
5101 be available; it is used if testing the ACL cannot continue, because most
5102 such messages specify what the cause of the deferral is. However, some
5103 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5104 that caused an old message to be retained and used if a later statement
5105 caused a defer, replacing the real cause of the deferral.
5107 16. If an IP address had so many PTR records that the DNS lookup buffer
5108 was not large enough to hold them, Exim could crash while trying to process
5109 the truncated data. It now detects and logs this case.
5111 17. Further to 4.21/58, another change has been made: if (and only if) the
5112 first line of a message (the first header line) ends with CRLF, a bare LF
5113 in a subsequent header line has a space inserted after it, so as not to
5114 terminate the header.
5116 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5117 unnecessarily, used atoi() instead of strtol(), and didn't check the
5118 termination when getting file sizes from file names by regex.
5120 19. Completely re-implemented the support for maildirsize files, in the light
5121 of a number of problems with the previous contributed implementation
5122 (4.30/29). In particular:
5124 . If the quota is zero, the maildirsize file is maintained, but no quota is
5127 . If the maildir directory does not exist, it is created before any attempt
5128 to write a maildirsize file.
5130 . The quota value in the file is just a cache; if the quota is changed in
5131 the transport, the new value overrides.
5133 . A regular expression is available for excluding directories from the
5136 20. The autoreply transport checks the characters in options that define the
5137 message's headers; it allows continued headers, but it was checking with
5138 isspace() after an embedded newline instead of explicitly looking for a
5141 21. If all the "regular" hosts to which an address was routed had passed their
5142 expiry times, and had not reached their retry times, the address was
5143 bounced, even if fallback hosts were defined. Now Exim should go on to try
5146 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5147 equivalent code in the SMTP transport. Some hosts send humungous responses
5148 to HELO/EHLO, more than 1024 it seems.
5150 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5151 gives compiler warnings in some environments. I've now done it "properly",
5154 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5155 (because of problems with the built-in one) was declared to return uschar *
5156 instead of char *, causing compiler failure.
5158 25. Fixed a file descriptor leak when processing alias/forward files.
5160 26. Fixed a minor format string issue in dbfn.c.
5162 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5164 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5165 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5168 29. When a sender address is verified, it is cached, to save repeating the test
5169 when there is more than one recipient in a message. However, when the
5170 verification involves a callout, it is possible for different callout
5171 options to be set for different recipients. It is too complicated to keep
5172 track of this in the cache, so now Exim always runs a verification when a
5173 callout is required, relying on the callout cache for the optimization.
5174 The overhead is duplication of the address routing, but this should not be
5177 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5178 to be verified with callout=postmaster, and the main callout worked but the
5179 postmaster check failed, the verification correctly failed. However, if a
5180 subsequent RCPT command asked for sender verification *without* the
5181 postmaster check, incorrect caching caused this verification also to fail,
5184 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5185 it was not caching the DNS options (qualify_single, search_parents) that
5186 were used when the lookup failed. A subsequent lookup with different
5187 options therefore always gave the same answer, though there were cases
5188 where it should not have. (Example: a "domains = !$mx_any" option on a
5189 dnslookup router: the "domains" option is always processed without any
5190 widening, but the router might have qualify_single set.) Now Exim uses the
5191 cached value only when the same options are set.
5193 32. Added John Jetmore's "exipick" utility to the distribution.
5195 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5196 than a timeout (e.g. a certificate is required, and is not provided), an
5197 Exim server now closes the connection immediately. Previously it waited for
5198 the client to close - but if the client is SSL, it seems that they each
5199 wait for each other, leading to a delay before one of them times out.
5201 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5202 maintained 0.8.x compatibility because I don't think many are using it, and
5203 it is clearly obsolete.
5205 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5208 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5209 client certificate was expired. A simple patch fixes this, though I don't
5210 understand the full logic of why the verify callback is called multiple
5213 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5214 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5215 which causes problems with some clients (such as the Certicom SSL Plus
5216 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5217 disables the coutermeasure allowing Eudora to connect."
5219 38. Exim was not checking that a write() to a log file succeeded. This could
5220 lead to Bad Things if a log got too big, in particular if it hit a file
5221 size limit. Exim now panics and dies if it cannot write to a log file, just
5222 as it does if it cannot open a log file.
5224 39. Modified OS/Makefile-Linux so that it now contains
5226 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5228 The two -D definitions ensure that Exim is compiled with large file
5229 support, which makes it possible to handle log files that are bigger than
5232 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5233 instance) a domain was checked against a named list that involved a lookup,
5234 causing $domain_data to be set, then another domain was checked against the
5235 same list, then the first domain was re-checked, the value of $domain_data
5236 after the final check could be wrong. In particular, if the second check
5237 failed, it could be set empty. This bug probably also applied to
5240 41. The strip_trailing_dot option was not being applied to the address given
5241 with the -f command-line option.
5243 42. The code for reading a message's header from the spool was incrementing
5244 $received_count, but never initializing it. This meant that the value was
5245 incorrect (doubled) while delivering a message in the same process in which
5246 it was received. In the most common configuration of Exim, this never
5247 happens - a fresh exec is done - but it can happen when
5248 deliver_drop_privilege is set.
5250 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5251 it now includes up to 150 characters of the unexpected data in the log
5254 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5255 and building data strings. The size of both of these buffers was 10 000
5256 bytes - far larger than anybody would *ever* want, thought I. Needless to
5257 say, somebody hit the limit. I have increased the maximum line length to
5258 20 000 and the maximum data length of concatenated lines to 100 000. I have
5259 also fixed two bugs, because there was no checking on these buffers. Tsk,
5260 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5261 buffer is too small.
5263 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5264 lsearch lookups. Now it does.
5266 46. When parsing a route_list item in a manualroute router, a fixed-length
5267 buffer was used for the list of hosts. I made this 1024 bytes long,
5268 thinking that nobody would ever have a list of hosts that long. Wrong.
5269 Somebody had a whole pile of complicated expansion conditions, and the
5270 string was silently truncated, leading to an expansion error. It turns out
5271 that it is easier to change to an unlimited length (owing to other changes
5272 that have happened since this code was originally written) than to build
5273 structure for giving a limitation error. The length of the item that
5274 expands into the list of hosts is now unlimited.
5276 47. The lsearch lookup could not handle data where the length of text line was
5277 more than 4095 characters. Such lines were truncated, leading to shortened
5278 data being returned. It should now handle lines of any length.
5280 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5281 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5284 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5285 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5286 no longer happens in comment lines. A list of replacements is now placed
5287 at the head of all of the source files, except those whose only change is
5288 to replace PERL_COMMAND in the very first #! line.
5290 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5291 messages on the queue, with a bottom-up merge sort, using code contributed
5292 by Michael Haardt. This should make operations like -bp somewhat faster on
5293 large queues. It won't affect queue runners, except when queue_run_in_order
5296 51. Installed eximstats 1.31 in the distribution.
5298 52. Added support for SRV lookups to the dnslookup router.
5300 53. If an ACL referred to $message_body or $message_body_end, the value was not
5301 reset for any messages that followed in the same SMTP session.
5303 54. The store-handling optimization for building very long strings was not
5304 differentiating between the different store pools. I don't think this
5305 actually made any difference in practice, but I've tidied it.
5307 55. While running the routers to verify a sender address, $sender_address
5308 was still set to the sender address. This is wrong, because when routing to
5309 send a bounce to the sender, it would be empty. Therefore, I have changed
5310 it so that, while verifying a sender address, $sender_address is set to <>.
5311 (There is no change to what happens when verifying a recipient address.)
5313 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5314 target A or AAAA records (if not already returned) without resetting the
5315 qualify_single or search_parents options of the DNS resolver. These are
5316 inappropriate in this case because the targets of MX and SRV records must
5317 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5318 target that, when qualified, matched something in the local domain. These
5319 two options are now turned off when doing these lookups.
5321 57. It seems that at least some releases of Reiserfs (which does not have the
5322 concept of a fixed number of inodes) returns zero and not -1 for the
5323 number of available inodes. This interacted badly with check_spool_inodes,
5324 which assumed that -1 was the "no such thing" setting. What I have done is
5325 to check that the total number of inodes is greater than zero before doing
5326 the test of how many are available.
5328 58. When a "warn" ACL statement has a log_message modifier, the message is
5329 remembered, and not repeated. This is to avoid a lot of repetition when a
5330 message has many recipients that cause the same warning to be written.
5331 Howewer, Exim was preserving the list of already written lines for an
5332 entire SMTP session, which doesn't seem right. The memory is now reset if a
5333 new message is started.
5335 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5336 debugging output unless log_rewrite was also set.
5338 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5339 of (int)(handle) when we know that handle contains (void *)(-1).
5341 61. The Exim daemon panic-logs an error return when it closes the incoming
5342 connection. However "connection reset by peer" seems to be common, and
5343 isn't really an error worthy of noting specially, so that particular error
5346 62. When Exim is trying to find all the local interfaces, it used to panic and
5347 die if the ioctl to get the interface flags failed. However, it seems that
5348 on at least one OS (Solaris 9) it is possible to have an interface that is
5349 included in the list of interfaces, but for which you get a failure error
5350 for this call. This happens when the interface is not "plumbed" into a
5351 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5352 failure of the "get flags" call assumes that the interface is down.
5354 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5355 makes life easier for people who are doing arithmetic on fields extracted
5356 from dates, where you often get leading zeros that should not be
5357 interpreted as octal.
5359 64. Added qualify_domain to the redirect router, to override the global
5362 65. If a pathologically long header line contained very many addresses (the
5363 report of this problem mentioned 10 000) and each of them was rewritten,
5364 Exim could use up a very large amount of memory. (It kept on making new
5365 copies of the header line as it rewrote, and never released the old ones.)
5366 At the expense of a bit more processing, the header rewriting function has
5367 been changed so that it no longer eats memory in this way.
5369 66. The generation of the Received: header has been moved from the time that a
5370 message starts to be received, to the time that it finishes. The timestamp
5371 in the Received: header should now be very close to that of the <= log
5372 line. There are two side-effects of this change:
5374 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5375 logged header lines no longer include the local Received: line, because
5376 it has not yet been created. The same applies to a copy of the message
5377 that is returned to a non-SMTP sender when a message is rejected.
5379 (b) When a filter file is tested using -bf, no additional Received: header
5380 is added to the test message. After some thought, I decided that this
5383 This change does not affect the value of $received_for. It is still set
5384 after address rewriting, but before local_scan() is called.
5386 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5388 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5389 gave an unhelpful panic error message, and a defer error. I have managed to
5390 change this behaviour so that it now rejects any supplied certificate,
5391 which seems right, as the list of acceptable certificates is empty.
5393 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5394 gave an unhelpful defer error. I have not managed to make this reject any
5395 supplied certificates, but the error message it gives is "no certificate
5396 supplied", which is not helpful.
5398 70. exigrep's output now also includes lines that are not associated with any
5399 message, but which match the given pattern. Implemented by a patch from
5400 Martin Sluka, which also tidied up the Perl a bit.
5402 71. Recipient callout verification, like sender verification, was using <> in
5403 the MAIL FROM command. This isn't really the right thing, since the actual
5404 sender may affect whether the remote host accepts the recipient or not. I
5405 have changed it to use the actual sender in the callout; this means that
5406 the cache record is now keyed on a recipient/sender pair, not just the
5407 recipient address. There doesn't seem to be a real danger of callout loops,
5408 since a callout by the remote host to check the sender would use <>.
5409 [SEE ABOVE: changed after hitting problems.]
5411 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5412 temporary errors. However, in the case of such a code being given after
5413 the end of a data transmission (i.e. after ".") Exim was failing to write
5414 a retry record for the message. (Yes, there was some broken host that was
5415 actually sending 8xx at this point.)
5417 73. An unknown lookup type in a host list could cause Exim to panic-die when
5418 the list was checked. (An example that provoked this was putting <; in the
5419 middle of a list instead of at the start.) If this happened during a DATA
5420 ACL check, a -D file could be left lying around. This kind of configuration
5421 error no longer causes Exim to die; instead it causes a defer errror. The
5422 incident is still logged to the main and panic logs.
5424 74. Buglet left over from Exim 3 conversion. The message "too many messages
5425 in one connection" was written to the rejectlog but not the mainlog, except
5426 when address rewriting (yes!) was being logged.
5428 75. Added write_rejectlog option.
5430 76. When a system filter was run not as root (that is, when system_filter_user
5431 was set), the values of the $n variables were not being returned to the
5432 main process; thus, they were not subsequently available in the $sn
5435 77. Added +return_path_on_delivery log selector.
5437 78. A connection timeout was being treated differently from recipients deferred
5438 when testing hosts_max_try with a message that was older than the host's
5439 retry timeout. (The host should not be counted, thus allowing all hosts to
5440 be tried at least once before bouncing.) This may have been the cause of an
5441 occasionally reported bug whereby a message would remain on the queue
5442 longer than the retry timeout, but would be bounced if a delivery was
5443 forced. I say "may" because I never totally pinned down the problem;
5444 setting up timeout/retry tests is difficult. See also the next item.
5446 79. The ultimate address timeout was not being applied to errors that involved
5447 a combination of host plus message (for example, a timeout on a MAIL
5448 command). When an address resolved to a number of possible hosts, and they
5449 were not all tried for each delivery (e.g. because of hosts_max_try), a
5450 message could remain on the queue longer than the retry timeout.
5452 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5455 81. Fixed an obscure SMTP outgoing bug which required at least the following
5456 conditions: (a) there was another message waiting for the same server;
5457 (b) the server returned 5xx to all RCPT commands in the first message so
5458 that the message was not completed; (c) the server dropped the connection
5459 or gave a negative response to the RSET that Exim sends to abort the
5460 transaction. The observed case was a dropped connection after DATA that had
5461 been sent in pipelining mode. That is, the server had advertised PIPELINING
5462 but was not implementing it correctly. The effect of the bug was incorrect
5463 behaviour, such as trying another host, and this could lead to a crash.
5469 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5470 and daemon.c were passed as pointers to ints; they should have been
5471 pointers to socklen_t variables (which are typically unsigned ints).
5473 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5476 3. Fixed a really odd bug that affected only the testing scheme; patching a
5477 certain fixed string in the binary changed the value of another string that
5478 happened to be identical to the end of the original first string.
5480 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5481 name", it returns that address as the IP address. On some operating
5482 systems (e.g. Solaris), it also passes back the IP address string as the
5483 "host name". However, on others (e.g. Linux), it passes back an empty
5484 string. Exim wasn't checking for this, and was changing the host name to an
5485 empty string, assuming it had been canonicized.
5487 5. Although rare, it is permitted to have more than one PTR record for a given
5488 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5489 all the names associated with an address, because they do in Solaris.
5490 However, it seems that they do not in Linux for data that comes from the
5491 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5492 I found this out when I moved to a new Linux workstation and tried to run
5493 the Exim test suite.
5495 To get round this problem I have changed the code so that it now does its
5496 own call to the DNS to look up PTR records when searching for a host name.
5497 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5498 addresses that are only in /etc/hosts are still found.
5500 This behaviour is, however, controlled by an option called host_lookup_
5501 order, which defaults to "bydns:byaddr". If people want to use the other
5502 order, or indeed, just use one or the other means of lookup, they can
5503 specify it in this variable.
5505 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5506 some operating systems, this comes back from gethostbyaddr() as an empty
5507 string, and this is what Exim used to test for. However, it seems that in
5508 other systems, "." is yielded. Exim now tests for this case too.
5510 7. The values of check_spool_space and check_log_space are now held internally
5511 as a number of kilobytes instead of an absolute number of bytes. If a
5512 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5513 kilobyte. This means that much larger values can be stored.
5515 8. Exim monitor: an attempt to get the action menu when not actually pointing
5516 at a message produces an empty menu entitled "No message selected". This
5517 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5518 no entries in it ("Shell widget menu has zero width and/or height"). So I
5519 have added a single, blank menu entry in this case.
5521 9. Added ${quote_local_part.
5523 10. MIME decoding is now applied to the contents of Subject: header lines when
5526 11. Now that a reference to $sender_host_address automatically causes a reverse
5527 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5528 host lookup before query-style lookups in lists that might use this
5529 variable. This has therefore been abolished, and the "net-" prefix is no
5530 longer necessary for query-style lookups.
5532 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5533 have been a typo for LFLAGS, so it has been changed.
5535 13. The install script calls Exim with "-C /dev/null" in order to find the
5536 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5537 to be output. Howeve, since Exim outputs its version number before the
5538 error, it didn't break the script. It just looked ugly. I fixed this by
5539 always allowing "-C /dev/null" if the caller is root.
5541 14. Ignore overlarge ACL variable number when reading spool file - insurance
5542 against a later release with more variables having written the file.
5544 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5545 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5546 this, as well as the form without the "IPv6" on the front.
5548 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5549 OS/Makefile-Darwin file.
5551 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5552 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5554 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5555 variable. However, some OS use ints for inodes. I've added cast to long int
5556 to get rid of the compiler warning.
5558 19. I had forgotten to lock out "/../" in configuration file names when
5559 ALT_CONFIG_PREFIX was set.
5561 20. Routers used for verification do not need to specify transports. However,
5562 if such a router generated a host list, and callout was configured, Exim
5563 crashed, because it could not find a port number from the (non-existent)
5564 transport. It now assumes port 25 in this circumstance.
5566 21. Added the -t option to exigrep.
5568 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5569 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5570 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5571 them is set, LOOKUP_LSEARCH is forced.
5573 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5575 24. Added sender and host information to the "rejected by local_scan()" log
5576 line; previously there was no indication of these.
5578 25. Added .include_if_exists.
5580 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5581 Linux. It turns out that not all file systems support this. Apparently some
5582 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5583 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5584 not supported on the file descriptor, is now ignored when Exim is trying to
5585 sync a directory. This applies only to Linux.
5587 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5589 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5590 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5593 29. Added support for maildirsize files from supplied patch (modified a bit).
5595 30. The use of :fail: followed by an empty string could lead Exim to respond to
5596 sender verification failures with (e.g.):
5598 550 Verification failed for <xxx>
5599 550 Sender verify failed
5601 where the first response line was missing the '-' that indicates it is not
5602 the final line of the response.
5604 31. The loop for finding the name of the user that called Exim had a hardwired
5605 limit of 10; it now uses the value of finduser_retries, which is used for
5606 all other user lookups.
5608 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5611 33. Exim was neglecting to zero errno before one call of strtol() when
5612 expanding a string and expecting an integer value. On some systems this
5613 resulted in spurious "integer overflow" errors. Also, it was casting the
5614 result into an int without checking.
5616 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5617 did not work. The code looks as if it has *never* worked, though it appears
5618 to have been documented since at least releast 1.62. I have made it work.
5620 35. The "timeout_DNS" error in retry rules, also documented since at least
5621 1.62, also never worked. As it isn't clear exactly what this means, and
5622 clearly it isn't a major issue, I have abolished the feature by treating it
5623 as "timeout", and writing a warning to the main and panic logs.
5625 36. The display of retry rules for -brt wasn't always showing the error code
5628 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5629 timeout_connect_A, timeout_connect_MX.
5631 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5632 to the empty sender.
5634 39. The daemon was not analysing the content of -oX till after it had closed
5635 stderr and disconnected from the controlling terminal. This meant that any
5636 syntax errors were only noted on the panic log, and the return code from
5637 the command was 0. By re-arranging the code a little, I've made the
5638 decoding happen first, so such errors now appear on stderr, and the return
5639 code is 1. However, the actual setting up of the sockets still happens in
5640 the disconnected process, so errors there are still only recorded on the
5643 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5644 connections (as happens on some IP stacks) was logged at start up time as
5645 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5646 it from "IPv6 and IPv4", which means that two separate sockets are being
5649 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5650 says whether AF_INET or AF_INET6 was passed as an argument.
5652 42. Exiwhat output was messed up when time zones were included in log
5655 43. Exiwhat now gives more information about the daemon's listening ports,
5656 and whether -tls-on-connect was used.
5658 44. The "port" option of the smtp transport is now expanded.
5660 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5661 silently ignored. Now an error message is written to the main and panic
5664 46. There's a new ACL modifier called "logwrite" which writes to a log file
5665 as soon as it is encountered.
5667 47. Added $local_user_uid and $local_user_gid at routing time.
5669 48. Exim crashed when trying to verify a sender address that was being
5672 49. Exim was recognizing only a space character after ".include". It now also
5673 recognizes a tab character.
5675 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5676 extracting the relevant information from the specification. The man page no
5677 longer contains scrambled data for the -d option, and I've added a section
5678 at the front about calling Exim under different names.
5680 51. Added "extra_headers" argument to the "mail" command in filter files.
5682 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5685 53. Installed eximstats 1.29.
5687 54. Added transport_filter_timeout as a generic transport option.
5689 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5690 Cc: header lines. This was required by RFC 822, but it not required by RFC
5693 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5694 incoming messages that did not have them. Now it does so only if the
5695 message originates locally, that is, if there is no associated remote host
5696 address. When Resent- header lines are present, this applies to the Resent-
5697 lines rather than the non-Resent- lines.
5699 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5700 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5702 58. Messages for configuration errors now include the name of the main
5703 configuration file - useful now that there may be more than one file in a
5704 list (.included file names were always shown).
5706 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5707 for those rare installations that do not start the daemon as root or run it
5708 setuid root. I've cut out the call to initgroups() if the daemon is not
5711 60. The Exim user and group can now be bound into the binary as text strings
5712 that are looked up at the start of Exim's processing.
5714 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5716 62. Added $mailstore_basename variable.
5718 63. Installed patch to sieve.c from Michael Haardt.
5720 64. When Exim failed to open the panic log after failing to open the main log,
5721 the original message it was trying to log was written to stderr and debug
5722 output, but if they were not available (the usual case in production), it
5723 was lost. Now it is written to syslog before the two lines that record the
5724 failures to open the logs.
5726 65. Users' Exim filters run in subprocesses under the user's uid. It is
5727 possible for a "deliver" command or an alias in a "personal" command to
5728 provoke an address rewrite. If logging of address rewriting is configured,
5729 this fails because the process is not running as root or exim. There may be
5730 a better way of dealing with this, but for the moment (because 4.30 needs
5731 to be released), I have disabled address rewrite logging when running a
5732 filter in a non-root, non-exim process.
5738 1. The buildconfig auxiliary program wasn't quoting the value set for
5739 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5740 not defined. This bug was masked in 4.22 by the effect that was fixed in
5743 2. Some messages that were rejected after a message id was allocated were
5744 shown as "incomplete" by exigrep. It no longer does this for messages that
5745 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5747 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5748 have allow_domain_literals set, the ID did not get logged in the <= line.
5749 Domain literals are now always recognized in Message-ID: header lines.
5751 4. The first argument for a ${extract expansion item is the key name or field
5752 number. Leading and trailing spaces in this item were not being ignored,
5753 causing some misleading effects.
5755 5. When deliver_drop_privilege was set, single queue runner processes started
5756 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5757 same command in the process it spins off) were not dropping privilege.
5759 6. When the daemon running as "exim" started a queue runner, it always
5760 re-executed Exim in the spun-off process. This is a waste of effort when
5761 deliver_drop_privilege is set. The new process now just calls the
5762 queue-runner function directly.
5768 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5771 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5772 ignored. Though the use of -oP was forcing the writing of a pid file, it
5773 was always written to the default place.
5775 3. If the message "no IP address found for host xxxx" is generated during
5776 incoming verification, it is now followed by identification of the incoming
5777 connection (so you can more easily find what provoked it).
5779 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5781 5. Added some features to "harden" Exim a bit more against certain attacks:
5783 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5784 be put in Local/Makefile. This is like the never_users runtime option,
5785 but it cannot be overridden. The default setting is "root".
5787 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5788 prefix string with which any file named in a -C command line option
5791 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5792 is retained for -C and -D only if the caller of Exim is root. Without
5793 it, the exim user may also use -C and -D and retain privilege.
5795 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5796 command line option is disabled.
5798 6. Macro names set by the -D option must start with an upper case letter, just
5799 like macro names defined in the configuration file.
5801 7. Added "dereference=" facility to LDAP.
5803 8. Two instances of the typo "uknown" in the source files are fixed.
5805 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5806 the Configure-Makefile script screwed up while processing it.
5808 10. Incorporated PCRE 4.4.
5810 11. The SMTP synchronization check was not operating right at the start of an
5811 SMTP session. For example, it could not catch a HELO sent before the client
5812 waited for the greeting. There is now a check for outstanding input at the
5813 point when the greeting is written. Because of the duplex, asynchronous
5814 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5815 way, but not yet received, when the check is performed.
5817 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5818 on TCP/IP sockets, because this apparently causes some broken clients to
5821 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5822 unchanged) from the Cygwin maintainer.
5824 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5825 was supported instead of testing for mbx. Effectively a typo.
5827 15. The spa authenticator server code was not checking that the input it
5828 received was valid base64.
5830 16. The debug output line for the "set" modifier in ACLs was not showing the
5831 name of the variable that was being set.
5833 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5835 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5836 lookup on its own. Something else was needed to trigger the lookup. For
5837 example, a match in host_lookup or the need for a host name in a host list.
5838 Now, if $sender_host_name is referenced and the host name has not yet been
5839 looked up, a lookup is performed. If the lookup fails, the variable remains
5840 empty, and $host_lookup_failed is set to "1".
5842 19. Added "eqi" as a case-independent comparison operator.
5844 20. The saslauthd authentication condition could segfault if neither service
5845 nor realm was specified.
5847 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5848 error message that was logged was misleading, and incoming SMTP
5849 connections were dropped. The message is now more accurate, and temporary
5850 errors are given to SMTP connections.
5852 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5853 (except RSET) with a 421 error, until QUIT is received. However, it was
5854 failing to send a response to QUIT.
5856 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5857 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5858 ACL is now run after the helo_try_verify_hosts code.
5860 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5861 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5862 case-independent, so other case variants are also recognized). Apparently
5863 some systems use these upper case variants.
5865 25. If more than two messages were waiting for the same host, and a transport
5866 filter was specified for the transport, Exim sent two messages over the
5867 same TCP/IP connection, and then failed with "socket operation on non-
5868 socket" when it tried to send the third.
5870 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5872 27. The extern definition of crypt16() in expand.c was not being excluded when
5873 the OS had its own crypt16() function.
5875 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5876 as a preferred synonym for return_size_limit, both as an option and as an
5879 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5881 30. Changed the default configuration ACL to relax the local part checking rule
5882 for addresses that are not in any local domains. For these addresses,
5883 slashes and pipe symbols are allowed within local parts, but the sequence
5884 /../ is explicitly forbidden.
5886 31. SPA server authentication was not clearing the challenge buffer before
5889 32. log_message in a "warn" ACL statement was writing to the reject log as
5890 well as to the main log, which contradicts the documentation and doesn't
5891 seem right (because no rejection is happening). So I have stopped it.
5893 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5894 However, I am unable to do any testing of this.
5896 34. Fixed an infelicity in the appendfile transport. When checking directories
5897 for a mailbox, to see if any needed to be created, it was accidentally
5898 using path names with one or more superfluous leading slashes; tracing
5899 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5901 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5902 contents are added to the log line that is written for every discarded
5903 recipient. (Previously a log_message setting was ignored.)
5905 36. The ${quote: operator now quotes the string if it is empty.
5907 37. The install script runs exim in order to find its version number. If for
5908 some reason other than non-existence or emptiness, which it checks, it
5909 could not run './exim', it was installing it with an empty version number,
5910 i.e. as "exim-". This error state is now caught, and the installation is
5913 38. An argument was missing from the function that creates an error message
5914 when Exim fails to connect to the socket for saslauthd authentication.
5915 This could cause Exim to crash, or give a corrupted message.
5917 39. Added isip, isip4, and isip6 to ${if conditions.
5919 40. The ACL variables $acl_xx are now saved with the message, and can be
5920 accessed later in routers, transports, and filters.
5922 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5923 strings in the file are not string-expanded.
5925 42. If a MAIL command specified a SIZE value that was too large to fit into an
5926 int variable, the check against message_size_limit failed. Such values are
5927 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5928 day this will have to be increased, but I don't think I want to be around
5929 when emails are that large.
5936 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5937 iconv() is not standard in FreeBSD.
5939 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5940 IPv6 enabled. The observed symptom was a segmentation fault on return from
5941 the function os_common_find_running_interfaces() in src/os.c.
5943 3. In the check_special_case() function in daemon.c I had used "errno" as an
5944 argument name, which causes warnings on some systems. This was basically a
5945 typo, since it was named "eno" in the comments!
5947 4. The code that waits for the clock to tick (at a resolution of some fraction
5948 of a second) so as to ensure message-id uniqueness was always waiting for
5949 at least one whole tick, when it could have waited for less. [This is
5950 almost certainly not relevant at current processor speeds, where it is
5951 unlikely to ever wait at all. But we try to future-proof.]
5953 5. The function that sleeps for a time interval that includes fractions of a
5954 second contained a race. It did not block SIGALRM between setting the
5955 timer, and suspending (a couple of lines later). If the interval was short
5956 and the sigsuspend() was delayed until after it had expired, the suspension
5957 never ended. On busy systems this could lead to processes getting stuck for
5960 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5961 process, before it forks any delivery processes. The open lookup caching
5962 mechanism meant that the open file or database connection was passed into
5963 the delivery process. The problem was that delivery processes always tidy
5964 up cached lookup data. This could cause a problem for the next delivery
5965 process started by the queue runner, because the external queue runner
5966 process does not know about the closure. So the next delivery process
5967 still has data in the lookup cache. In the case of a file lookup, there was
5968 no problem because closing a file descriptor in a subprocess doesn't affect
5969 the parent. However, if the lookup was caching a connection to a database,
5970 the connection was closed, and the second delivery process was likely to
5971 see errors such as "PGSQL: query failed: server closed the connection
5972 unexpectedly". The problem has been fixed by closing all cached lookups
5973 in a queue runner before running a delivery process.
5975 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5976 seem to have the "const" qualifier which it has on other OS. I've
5979 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5980 *of the same type* (client or server) with the same public name that an
5981 error should be diagnosed.
5983 9. When Exim looked up a host name for an IP address, but failed to find the
5984 original IP address when looking up the host name (a safety check), it
5985 output the message "<ip address> does not match any IP for NULL", which was
5986 confusing, to say the least. The bug was that the host name should have
5987 appeared instead of "NULL".
5989 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5990 user that is built into the binary, and the -C or -D options is used, root
5991 privilege is dropped before the configuration file is read. In addition,
5992 logging is switched to stderr instead of the normal log files. If the
5993 configuration then re-defines the Exim user, the unprivileged environment
5994 is probably not what is expected, so Exim logs a panic warning message (but
5997 However, if deliver_drop_privilege is set, the unprivileged state may well
5998 be exactly what is intended, so the warning has been cut out in that case,
5999 and Exim is allowed to try to write to its normal log files.
6005 1. smtp_return_error_details was not giving details for temporary sender
6006 or receiver verification errors.
6008 2. Diagnose a configuration error if two authenticators have the same public
6011 3. Exim used not to create the message log file for a message until the first
6012 delivery attempt. This could be confusing when incoming messages were held
6013 for policy or load reasons. The message log file is now created at the time
6014 the message is received, and an initial "Received" line is written to it.
6016 4. The automatically generated man page for command line options had a minor
6017 bug that caused no ill effects; however, a more serious problem was that
6018 the procedure for building the man page automatically didn't always
6019 operate. Consequently, release 4.20 contains an out-of-date version. This
6020 shouldn't happen again.
6022 5. When building Exim with embedded Perl support, the script that builds the
6023 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6024 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6026 6. The freeze_tell option was not being used for messages that were frozen on
6027 arrival, either by an ACL or by local_scan().
6029 7. Added the smtp_incomplete_transaction log selector.
6031 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6032 was accepting AUTH without a new EHLO.
6034 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6035 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6038 10. Logging of TCP/IP connections (when configured) now happens in the main
6039 daemon process instead of the child process, so that the TCP/IP connection
6040 count is more accurate (but it can never be perfect).
6042 11. The use of "drop" in a nested ACL was not being handled correctly in the
6043 outer ACL. Now, if condition failure induced by the nested "drop" causes
6044 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6045 or "require"), the connection is dropped.
6047 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6048 that yield "discard" can now be used with an "accept" or a "discard" verb,
6049 but an error is generated for any others (because I can't see a useful way
6050 to define what should happen).
6052 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6053 are now processed in the same way as lines in the Exim configuration file.
6054 In particular, continuation lines are supported.
6056 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6058 15. Added -ti meaning -t -i.
6060 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6061 domains, and warn by logging if others are found.
6063 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6064 returned from GIFCONF when Exim is trying to find the list of interfaces on
6065 a host. The code in os.c has been modified to copy each ifreq to an aligned
6066 structure in all cases.
6068 Also, in some cases, the returned ifreq's were being copied to a 'struct
6069 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6070 means the last couple of bytes of an IPv6 address could be chopped if the
6071 ifreq contained only a normal sockaddr (14 bytes storage).
6073 18. Named domain lists were not supported in the hosts_treat_as_local option.
6074 An entry such as +xxxx was not recognized, and was treated as a literal
6077 19. Ensure that header lines added by a DATA ACL are included in the reject log
6078 if the ACL subsequently rejects the message.
6080 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6081 MD5 (which is deprecated).
6083 21. When testing a filter file using -bf, Exim was writing a message when it
6084 took the sender from a "From " line in the message, but it was not doing so
6085 when it took $return_path from a Return-Path: header line. It now does.
6087 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6088 with a valid header line field name (a series of printing characters
6089 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6091 23. Changed "disc" in the source to "disk" to conform to the documentation and
6092 the book and for uniformity.
6094 24. Ignore Sendmail's -Ooption=value command line item.
6096 25. When execve() failed while trying to run a command in a pipe transport,
6097 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6098 could be confused with a return value of 69 from the command itself. This
6099 has been changed to 127, the value the shell returns if it is asked to run
6100 a non-existent command. The wording for the related log line suggests a
6101 non-existent command as the problem.
6103 26. If received_header_text expands to an empty string, do not add a Received:
6104 header line to the message. (Well, it adds a token one on the spool, but
6105 marks it "old" so that it doesn't get used or transmitted.)
6107 27. Installed eximstats 1.28 (addition of -nt option).
6109 28. There was no check for failure on the call to getsockname() in the daemon
6110 code. This can fail if there is a shortage of resources on the system, with
6111 ENOMEM, for example. A temporary error is now given on failure.
6113 29. Contrary to the C standard, it seems that in some environments, the
6114 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6115 program. Exim now does this explicitly; it affects the formatting of
6116 timestamps using strftime().
6118 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6119 complaints. I've now initialized all the variables, to avoid this.
6121 32. Header lines added by a system filter were not being "seen" during
6122 transport-time rewrites.
6124 33. The info_callback() function passed to OpenSSL is set up with type void
6125 (*)(SSL *, int, int), as described somewhere. However, when calling the
6126 function (actually a macro) that sets it up, the type void(*)() is
6127 expected. I've put in a cast to prevent warnings from picky compilers.
6129 34. If a DNS black list lookup found a CNAME record, but there were no A
6130 records associated with the domain it pointed at, Exim crashed.
6132 35. If a DNS black list lookup returned more than one A record, Exim ignored
6133 all but the first. It now scans all returned addresses if a particular IP
6134 value is being sought. In this situation, the contents of the
6135 $dnslist_value variable are a list of all the addresses, separated by a
6138 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6139 to accept a host name and all its aliases if the forward lookup for any of
6140 them yielded the IP address of the incoming connection. Now it accepts only
6141 those names whose forward lookup yields the correct IP address. Any other
6142 names are discarded. This closes a loophole whereby a rogue DNS
6143 administrator could create reverse DNS records to break through a
6144 wildcarded host restriction in an ACL.
6146 37. If a user filter or a system filter that ran in a subprocess used any of
6147 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6148 the wrong values were passed to the pipe command ($thisaddress had the
6149 value of $0, $0 had the value of $1, etc). This bug was introduced by
6150 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6152 38. Improved the line breaking for long SMTP error messages from ACLs.
6153 Previously, if there was no break point between 40 and 75 characters, Exim
6154 left the rest of the message alone. Two changes have been made: (a) I've
6155 reduced the minimum length to 35 characters; (b) if it can't find a break
6156 point between 35 and 75 characters, it looks ahead and uses the first one
6157 that it finds. This may give the occasional overlong line, but at least the
6158 remaining text gets split now.
6160 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6161 file descriptors might be low, and that setting 1000 would always raise it.
6162 It turns out that in some environments, the limit is already over 1000 and
6163 that lowering it causes trouble. So now Exim takes care not to decrease it.
6165 40. When delivering a message, the value of $return_path is set to $sender_
6166 address at the start of routing (routers may change the value). By an
6167 oversight, this default was not being set up when an address was tested by
6168 -bt or -bv, which affected the outcome if any router or filter referred to
6171 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6172 log only when "message" or "log_message" are set. However, if one of the
6173 conditions was an address verification, or a call to a nested ACL, the
6174 messages generated by the underlying test were being passed through. This
6175 no longer happens. The underlying message is available in $acl_verify_
6176 message for both "message" and "log_message" expansions, so it can be
6177 passed through if needed.
6179 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6180 new expansion $bh_ to give the encoded byte string without charset
6181 translation. Translation happens only if iconv() is available; HAVE_ICONV
6182 indicates this at build time. HEADERS_CHARSET gives the charset to
6183 translate to; headers_charset can change it in the configuration, and
6184 "headers charset" can change it in an individual filter file.
6186 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6187 that creates RFC 2047 encoded "words" labels them as that charset instead
6188 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6189 expansion operator; (ii) when Exim creates a From: line for a local
6190 message; (iii) when a header line is rewritten to include a "phrase" part.
6192 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6193 buggy, causing it to skip the first lines of messages whose message ID
6194 ended in 'D'. This would not have bitten before Exim release 4.14, because
6195 message IDs were unlikely to end in 'D' before then. The effect was to have
6196 incorrect size information for certain domains.
6198 45. #include "config.h" was missing at the start of the crypt16.c module. This
6199 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6202 46. If there was a timeout during a "random" callout check, Exim treated it as
6203 a failure of the random address, and carried on sending RSET and the real
6204 address. If the delay was just some slowness somewhere, the response to the
6205 original RCPT would be taken as a response to RSET and so on, causing
6206 mayhem of various kinds.
6208 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6209 when I implemented it. It didn't allow for the fact that some option values
6210 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6211 the right test for positive values.
6213 48. Domain names in DNS records are case-independent. Exim always looks them up
6214 in lower case. Some resolvers return domain names in exactly the case they
6215 appear in the zone file, that is, they may contain uppercase letters. Not
6216 all resolvers do this - some return always lower case. Exim was treating a
6217 change of case by a resolver as a change of domain, similar to a widening
6218 of a domain abbreviation. This triggered its re-routing code and so it was
6219 trying to route what was effectively the same domain again. This normally
6220 caused routing to fail (because the router wouldn't handle the domain
6221 twice). Now Exim checks for this case specially, and just changes the
6222 casing of the domain that it ultimately uses when it transmits the message
6225 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6228 50. If a filter generated a file delivery with a non-absolute name (possible if
6229 no home directory exists for the router), the forbid_file option was not
6232 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6233 the existing equality matching.
6235 52. Exim was using ints instead of ino_t variables in some places where it was
6236 dealing with inode numbers.
6238 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6239 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6240 called TMPDIR, and if it finds it is different, it changes its value.
6242 54. The smtp_printf() function is now made available to local_scan() so
6243 additional output lines can be written before returning. There is also an
6244 smtp_fflush() function to enable the detection of a dropped connection.
6245 The variables smtp_input and smtp_batched_input are exported to
6248 55. Changed the default runtime configuration: the message "Unknown user"
6249 has been removed from the ACL, and instead placed on the localuser router,
6250 using the cannot_route_message feature. This means that any verification
6251 failures that generate their own messages won't get overridden. Similarly,
6252 the "Unrouteable address" message that was in the ACL for unverifiable
6253 relay addresses has also been removed.
6255 56. Added hosts_avoid_esmtp to the smtp transport.
6257 57. The exicyclog script was not checking for the esoteric option
6258 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6259 will work only if exicyclog is run under the appropriate euid.
6261 58. Following a discussion on the list, the rules by which Exim recognises line
6262 endings on incoming messages have been changed. The -dropcr and drop_cr
6263 options are now no-ops, retained only for backwards compatibility. The
6264 following line terminators are recognized: LF CRLF CR. However, special
6265 processing applies to CR:
6267 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6268 nor a local message in the state where . is a terminator.
6270 (ii) If a bare CR is encountered in a header line, an extra space is added
6271 after the line terminator so as not to end the header. The reasoning
6272 behind this is that bare CRs in header lines are most likely either
6273 to be mistakes, or people trying to play silly games.
6275 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6276 was being incorrectly given as 18 bytes larger than it should have been.
6277 This is a VOB (very old bug).
6279 60. This may never have affected anything current, but just in case it has:
6280 When the local host is found other than at the start of a list of hosts,
6281 the local host, those with the same MX, and any that follow, are discarded.
6282 When the list in question was part of a longer list of hosts, the following
6283 hosts (not currently being processed) were also being discarded. This no
6284 longer happens. I'm not sure if this situation could ever has previously
6287 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6290 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6291 header. This is something that is recommended in a new Internet Draft, and
6292 is something that is documented as being done by Sendmail. There are two
6293 possible values. For messages generated by the autoreply transport, Exim
6296 Auto-Submitted: auto-replied
6298 whereas for all other generated messages (e.g. bounces) it adds
6300 Auto-Submitted: auto-generated
6302 63. The "personal" condition in filters now includes a test for the
6303 Auto-Submitted: header. If it contains the string "auto-" the message it
6304 not considered personal.
6306 64. Added rcpt_include_affixes as a generic transport option.
6308 65. Added queue_only_override (default true).
6310 66. Added the syslog_duplication option.
6312 67. If what should have been the first header line of a message consisted of
6313 a space followed by a colon, Exim was mis-interpreting it as a header line.
6314 It isn't of course - it is syntactically invalid and should therefore be
6315 treated as the start of the message body. The misbehaviour could have
6316 caused a number of strange effects, including loss of data in subsequent
6317 header lines, and spool format errors.
6319 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6320 client host had authenticated. This control can now be exercised by an ACL
6321 for more flexibility.
6323 69. By default, callouts do not happen when testing with -bh. There is now a
6324 variant, -bhc, which does actually run the callout code, including
6325 consulting and updating the callout cache.
6327 70. Added support for saslauthd authentication, courtesy of Alexander
6330 71. If statvfs() failed on the spool or log directories while checking their
6331 size for availability, Exim confusingly gave the error "space shortage".
6332 Furthermore, in debugging mode it crashed with a floating point exception.
6333 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6334 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6335 serious problem, Exim now writes to the main and panic logs when this
6336 happens, with details of the failure. It then refuses to accept the
6337 incoming message, giving the message "spool directory problem" or "log
6338 directory problem" with a 421 code for SMTP messages.
6340 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6341 0, 1, and 2 exist, because some OS complain for execs without them (see
6342 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6343 descriptors. However, the code omitted to check that the open succeeded,
6344 causing mysterious errors if for some reason the permissions on /dev/null
6345 got screwed. Now Exim writes a message to the main and panic logs, and
6346 bombs out if it can't open /dev/null.
6348 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6349 interact so that it is all more flexible. It is supposed to remain
6350 backwards compatible. Also added extra_local_interfaces.
6352 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6353 to bomb out with an assertion failure - to the client this appears as a
6354 connection drop. This problem occurs in the part of the code that was taken
6355 from the Samba project. Fortunately, the assertion is in a very simple
6356 function, so I have fixed this by reproducing the function inline in the
6357 one place where it is called, and arranging for authentication to fail
6358 instead of killing the process with assert().
6360 75. The SPA client code was not working when the server requested OEM rather
6361 than Unicode encoding.
6363 76. Added code to make require_files with a specific uid setting more usable in
6364 the case where statting the file as root fails - usually a non-root-mounted
6365 NFS file system. When this happens and the failure is EACCES, Exim now
6366 forks a subprocess and does the per-uid checking as the relevant uid.
6368 77. Added process_log_path.
6370 78. If log_file_path was not explicitly set, a setting of check_log_space or
6371 check_log_inodes was ignored.
6373 79. If a space check for the spool or log partitions fails, the incident is now
6374 logged. Of course, in the latter case the data may get lost...
6376 80. Added the %p formatting code to string_format() so that it can be used to
6377 print addresses in debug_print(). Adjusted all the address printing in the
6378 debugging in store.c to use %p rather than %d.
6380 81. There was a concern that a line of code in smtp_in.c could overflow a
6381 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6382 initially expressed, the concern was not well-founded, because trailing
6383 spaces are removed early. However, if the trailing spaces were followed by
6384 a NULL, they did not get removed, so the overflow was possible. Two fixes
6387 (a) I re-wrote the offending code in a cleaner fashion.
6388 (b) If an incoming SMTP command contains a NULL character, it is rejected
6391 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6392 runs initgroups(), so that if the Exim user is in any additional groups,
6393 they will be used during message reception.
6399 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob