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.
84 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
87 JH/02 The smtp transport option "multi_domain" is now expanded.
89 JH/03 The smtp transport now requests PRDR by default, if the server offers
92 JH/04 Certificate name checking on server certificates, when exim is a client,
93 is now done by default. The transport option tls_verify_cert_hostnames
94 can be used to disable this per-host. The build option
95 EXPERIMENTAL_CERTNAMES is withdrawn.
97 JH/05 The value of the tls_verify_certificates smtp transport and main options
98 default to the word "system" to access the system default CA bundle.
99 For GnuTLS, only version 3.0.20 or later.
101 JH/06 Verification of the server certificate for a TLS connection is now tried
102 (but not required) by default. The verification status is now logged by
103 default, for both outbound TLS and client-certificate supplying inbound
106 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
109 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
110 Status Notification (bounce) messages are now MIME format per RFC 3464.
111 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
112 under the control of the dsn_advertise_hosts option, and routers may
113 have a dsn_lasthop option.
115 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
116 default, modifiable by a malware= option. The list separator for
117 the options can now be changed in the usual way. Bug 68.
119 JH/10 The smtp_receive_timeout main option is now expanded before use.
121 JH/11 The incoming_interface log option now also enables logging of the
122 local interface on delivery outgoing connections.
124 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
125 if the interface and destination host and port all match.
127 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
130 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
131 Patch from Andrew Lewis.
133 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
134 now supports optional time-restrictions, weighting, and priority
135 modifiers per server. Patch originally by <rommer@active.by>.
137 JH/16 The spamd_address main option now supports a mixed list of local
138 and remote servers. Remote servers can be IPv6 addresses, and
139 specify a port-range.
141 JH/17 Bug 68: The spamd_address main option now supports an optional
142 timeout value per server.
144 JH/18 Bug 1581: Router and transport options headers_add/remove can
145 now have the list separator specified.
147 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
150 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
153 JH/21 Support for the A6 type of dns record is withdrawn.
155 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
156 rather than the verbs used.
158 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
159 from 255 to 1024 chars.
161 JH/24 Verification callouts now attempt to use TLS by default.
163 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
164 are generic router options now. The defaults didn't change.
166 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
167 Original patch from Alexander Shikoff, worked over by JH.
169 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
170 files if ZCAT_COMMAND is not executable.
172 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
174 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
176 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
177 Normally benign, it bites when the pair was led to by a CNAME;
178 modern usage is to not canoicalize the domain to a CNAME target
179 (and we were inconsistent anyway for A-only vs AAAA+A).
181 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
183 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
184 when evaluating $sender_host_dnssec.
186 JH/31 Check the HELO verification lookup for DNSSEC, adding new
187 $sender_helo_dnssec variable.
189 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
191 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
193 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
195 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
196 documented as working, but never had. Support all but $spam_report.
198 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
199 added for tls authenticator.
204 TL/01 When running the test suite, the README says that variables such as
205 no_msglog_check are global and can be placed anywhere in a specific
206 test's script, however it was observed that placement needed to be near
207 the beginning for it to behave that way. Changed the runtest perl
208 script to read through the entire script once to detect and set these
209 variables, reset to the beginning of the script, and then run through
210 the script parsing/test process like normal.
212 TL/02 The BSD's have an arc4random API. One of the functions to induce
213 adding randomness was arc4random_stir(), but it has been removed in
214 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
215 function when detected.
217 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
218 cause callback expansion.
220 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
221 syntax errors in an expansion can be treated as a string instead of
222 logging or causing an error, due to the internal use of bool_lax
223 instead of bool when processing it.
225 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
226 server certificates when making smtp deliveries.
228 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
230 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
232 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
234 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
235 Merged patch from Sebastian Wiedenroth.
237 JH/05 Fix results-pipe from transport process. Several recipients, combined
238 with certificate use, exposed issues where response data items split
239 over buffer boundaries were not parsed properly. This eventually
240 resulted in duplicates being sent. This issue only became common enough
241 to notice due to the introduction of conection certificate information,
242 the item size being so much larger. Found and fixed by Wolfgang Breyha.
244 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
245 size buffer was used, resulting in syntax errors when an expansion
248 JH/07 Add support for directories of certificates when compiled with a GnuTLS
249 version 3.3.6 or later.
251 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
252 is EXPERIMENTAL_EVENT, the main-configuration and transport options
253 both become "event_action", the variables become $event_name, $event_data
254 and $event_defer_errno. There is a new variable $verify_mode, usable in
255 routers, transports and related events. The tls:cert event is now also
256 raised for inbound connections, if the main configuration event_action
259 TL/06 In test suite, disable OCSP for old versions of openssl which contained
260 early OCSP support, but no stapling (appears to be less than 1.0.0).
262 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
263 server certificate names available under the smtp transport option
264 "tls_verify_cert_hostname" now do not permit multi-component wildcard
267 JH/10 Time-related extraction expansions from certificates now use the main
268 option "timezone" setting for output formatting, and are consistent
269 between OpenSSL and GnuTLS compilations. Bug 1541.
271 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
272 encoded parameter in the incoming message. Bug 1558.
274 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
275 include certificate info, eximon was claiming there were spoolfile
278 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
280 JH/14 Log delivery-related information more consistently, using the sequence
281 "H=<name> [<ip>]" wherever possible.
283 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
284 are problematic for Debian distribution, omit them from the release
287 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
289 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
292 JH/17 Fix a null-indirection in certextract expansions when a nondefault
293 output list separator was used.
298 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
299 checkers that were complaining about end of non-void function with no
302 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
303 This was a regression intruduced in 4.83 by another bugfix.
305 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
307 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
308 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
314 TF/01 Correctly close the server side of TLS when forking for delivery.
316 When a message was received over SMTP with TLS, Exim failed to clear up
317 the incoming connection properly after forking off the child process to
318 deliver the message. In some situations the subsequent outgoing
319 delivery connection happened to have the same fd number as the incoming
320 connection previously had. Exim would try to use TLS and fail, logging
321 a "Bad file descriptor" error.
323 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
324 utilities have not been installed.
326 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
327 temporary space as the ACL may create new global variables.
329 TL/01 LDAP support uses per connection or global context settings, depending
330 upon the detected version of the libraries at build time.
332 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
333 to extract and use the src ip:port in logging and expansions as if it
334 were a direct connection from the outside internet. PPv2 support was
335 updated based on HAProxy spec change in May 2014.
337 JH/02 Add ${listextract {number}{list}{success}{fail}}.
339 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
340 Properly escape header and check for NULL return.
342 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
345 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
347 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
348 characters in header names, implemented as a verify condition.
349 Contributed by Michael Fischer v. Mollard.
351 TL/05 Rename SPF condition results err_perm and err_temp to standardized
352 results permerror and temperror. Previous values are deprecated but
353 still accepted. In a future release, err_perm and err_temp will be
354 completely removed, which will be a backward incompatibility if the
355 ACL tests for either of these two old results. Patch contributed by
356 user bes-internal on the mailing list.
358 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
360 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
361 selectors, in both main and reject logs.
363 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
366 JH/07 Add malware type "sock" for talking to simple daemon.
368 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
370 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
371 routers/transports under cutthrough routing.
373 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
374 numbers. Touch up "bool" conditional to keep the same definition.
376 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
378 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
380 JH/12 Expand items in router/transport headers_add or headers_remove lists
381 individually rather than the list as a whole. Bug 1452.
383 Required for reasonable handling of multiple headers_ options when
384 they may be empty; requires that headers_remove items with embedded
385 colons must have them doubled (or the list-separator changed).
387 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
388 view the policy declared in the DMARC record. Currently, $dmarc_status
389 is a combined value of both the record presence and the result of the
392 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
394 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
395 dnslookup router and the smtp transport (applying to the forward
398 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
399 of ldap servers used for a specific lookup. Patch provided by Heiko
402 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
403 New variable $lookup_dnssec_authenticated for observability.
405 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
406 Patch submitted by Lars Timman.
408 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
410 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
411 Requires trusted mode and valid format message id, aborts otherwise.
412 Patch contributed by Heiko Schlichting.
414 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
415 certextract with support for various fields. Bug 1358.
417 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
418 is requested by default, modifiable by smtp transport option
421 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
422 operate on certificate variables to give certificate fingerprints
423 Also new ${sha256:cert_variable}.
425 JH/23 The PRDR feature is moved from being Experimental into the mainline.
427 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
428 Christian Aistleitner.
430 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
432 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
433 file. Patch from Wolfgang Breyha.
435 JH/25 Expand the coverage of the delivery $host and $host_address to
436 client authenticators run in verify callout. Bug 1476.
438 JH/26 Port service names are now accepted for tls_on_connect_ports, to
439 align with daemon_smtp_ports. Bug 72.
441 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
442 support and error reporting did not work properly.
444 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
445 and is readable. Patch from Andrew Colin Kissa.
447 TL/14 Enhance documentation of ${run expansion and how it parses the
448 commandline after expansion, particularly in the case when an
449 unquoted variable expansion results in an empty value.
451 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
453 PP/02 Fix internal collision of T_APL on systems which support RFC3123
454 by renaming away from it. Addresses GH issue 15, reported by
457 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
459 TL/15 SECURITY: prevent double expansion in math comparison functions
460 (can expand unsanitized data). Not remotely exploitable.
467 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
469 PP/02 Make -n do something, by making it not do something.
470 When combined with -bP, the name of an option is not output.
472 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
475 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
476 $sender_host_name and config options to manage this, and basic check
479 PP/05 DSCP support for outbound connections and control modifier for inbound.
481 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
482 (Only plugin which currently uses this is kerberos4, which nobody should
483 be using, but we should make it available and other future plugins might
484 conceivably use it, even though it would break NAT; stuff *should* be
485 using channel bindings instead).
487 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
488 name; added for Sendmail compatibility; requires admin caller.
489 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
490 just ignore it); requires trusted caller.
491 Also parse but ignore: -Ac -Am -X<logfile>
494 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
496 TL/02 Add +smtp_confirmation as a default logging option.
498 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
499 Patch by Magnus Holmgren from 2007-02-20.
501 TL/04 Bugzilla 1281 - Spec typo.
502 Bugzilla 1283 - Spec typo.
503 Bugzilla 1290 - Spec grammar fixes.
505 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
507 TL/06 Add Experimental DMARC support using libopendmarc libraries.
509 TL/07 Fix an out of order global option causing a segfault. Reported to dev
510 mailing list by by Dmitry Isaikin.
512 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
514 JH/02 Support "G" suffix to numbers in ${if comparisons.
516 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
518 NM/01 Bugzilla 1197 - Spec typo
519 Bugzilla 1196 - Spec examples corrections
521 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
523 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
524 gnutls_enable_pkcs11, but renamed to more accurately indicate its
527 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
528 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
530 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
531 "acl {{name}{arg}...}", and optional args on acl condition
534 JH/05 Permit multiple router/transport headers_add/remove lines.
536 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
538 JH/07 Avoid using a waiting database for a single-message-only transport.
539 Performance patch from Paul Fisher. Bugzilla 1262.
541 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
544 JH/09 Add $headers_added variable, with content from use of ACL modifier
545 add_header (but not yet added to the message). Bugzilla 199.
547 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
548 Pulled from Bugzilla 817 by Wolfgang Breyha.
550 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
552 (nb: this is the same fix as in Exim 4.80.1)
554 JH/11 Add A= logging on delivery lines, and a client_set_id option on
557 JH/12 Add optional authenticated_sender logging to A= and a log_selector
560 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
562 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
563 advertise SMTP AUTH mechanism to us, instead of a generic
564 protocol violation error. Also, make Exim more robust to bad
565 data from the Dovecot auth socket.
567 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
569 When a queue runner is handling a message, Exim first routes the
570 recipient addresses, during which it prunes them based on the retry
571 hints database. After that it attempts to deliver the message to
572 any remaining recipients. It then updates the hints database using
575 So if a recipient address works intermittently, it can get repeatedly
576 deferred at routing time. The retry hints record remains fresh so the
577 address never reaches the final cutoff time.
579 This is a fairly common occurrence when a user is bumping up against
580 their storage quota. Exim had some logic in its local delivery code
581 to deal with this. However it did not apply to per-recipient defers
582 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
584 This change adds a proper retry rule check during routing so that the
585 final cutoff time is checked against the message's age. We only do
586 this check if there is an address retry record and there is not a
587 domain retry record; this implies that previous attempts to handle
588 the address had the retry_use_local_parts option turned on. We use
589 this as an approximation for the destination being like a local
590 delivery, as in LMTP.
592 I suspect this new check makes the old local delivery cutoff check
593 redundant, but I have not verified this so I left the code in place.
595 TF/02 Correct gecos expansion when From: is a prefix of the username.
597 Test 0254 submits a message to Exim with the header
601 When I ran the test suite under the user fanf2, Exim expanded
602 the header to contain my full name, whereas it should have added
603 a Resent-Sender: header. It erroneously treats any prefix of the
604 username as equal to the username.
606 This change corrects that bug.
608 GF/01 DCC debug and logging tidyup
609 Error conditions log to paniclog rather than rejectlog.
610 Debug lines prefixed by "DCC: " to remove any ambiguity.
612 TF/03 Avoid unnecessary rebuilds of lookup-related code.
614 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
615 Bug spotted by Jeremy Harris; was flawed since initial commit.
616 Would have resulted in OCSP responses post-SNI triggering an Exim
617 NULL dereference and crash.
619 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
621 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
622 Bug detection, analysis and fix by Samuel Thibault.
623 Bugzilla 1331, Debian bug #698092.
625 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
627 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
628 Server implementation by Todd Lyons, client by JH.
629 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
630 config variable "prdr_enable" controls whether the server
631 advertises the facility. If the client requests PRDR a new
632 acl_data_smtp_prdr ACL is called once for each recipient, after
633 the body content is received and before the acl_smtp_data ACL.
634 The client is controlled by bolth of: a hosts_try_prdr option
635 on the smtp transport, and the server advertisement.
636 Default client logging of deliveries and rejections involving
637 PRDR are flagged with the string "PRDR".
639 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
640 fclose(). Diagnosis by Todd Lyons.
642 PP/17 Update configure.default to handle IPv6 localhost better.
643 Patch by Alain Williams (plus minor tweaks).
646 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
647 This is now consistent with GnuTLS, and is now documented: the
648 previous undocumented portable approach to treating the option as
649 unset was to force an expansion failure. That still works, and
650 an empty string is now equivalent.
652 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
653 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
654 not performing validation itself.
656 PP/20 Added force_command boolean option to pipe transport.
657 Patch from Nick Koston, of cPanel Inc.
659 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
662 TF/04 Added udpsend ACL modifer and hexquote expansion operator
664 PP/21 Fix eximon continuous updating with timestamped log-files.
665 Broken in a format-string cleanup in 4.80, missed when I repaired the
666 other false fix of the same issue.
667 Report and fix from Heiko Schlichting.
670 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
671 Report from Prashanth Katuri.
673 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
674 It's SecureTransport, so affects any MacOS clients which use the
675 system-integrated TLS libraries, including email clients.
677 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
678 using a MIME ACL for non-SMTP local injection.
679 Report and assistance in diagnosis by Warren Baker.
681 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
683 JH/16 Fix comparisons for 64b. Bugzilla 1385.
685 TL/09 Add expansion variable $authenticated_fail_id to keep track of
686 last id that failed so it may be referenced in subsequent ACL's.
688 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
691 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
692 ldap library initialization, allowing self-signed CA's to be
693 used. Also properly sets require_cert option later in code by
694 using NULL (global ldap config) instead of ldap handle (per
695 session). Bug diagnosis and testing by alxgomz.
697 TL/12 Enhanced documentation in the ratelimit.pl script provided in
698 the src/util/ subdirectory.
700 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
701 renamed to Transport Post Delivery Action by Jeremy Harris, as
704 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
705 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
706 redis_servers = needs to be configured which will be used by the redis
707 lookup. Patch from Warren Baker, of The Packet Hub.
709 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
711 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
712 hostname or reverse DNS when processing a host list. Used suggestions
713 from multiple comments on this bug.
715 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
717 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
718 Missed a few lines, added it to make the runtest require no keyboard
721 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
722 contains upper case chars. Make router use caseful_local_part.
724 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
725 support when GnuTLS has been built with p11-kit.
731 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
733 This, or similar/improved, will also be change PP/11 of 4.82.
739 PP/01 Handle short writes when writing local log-files.
740 In practice, only affects FreeBSD (8 onwards).
741 Bugzilla 1053, with thanks to Dmitry Isaikin.
743 NM/01 Bugzilla 949 - Documentation tweak
745 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
748 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
750 PP/02 Implemented gsasl authenticator.
752 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
754 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
755 `pkg-config foo` for cflags/libs.
757 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
758 with rest of GSASL and with heimdal_gssapi.
760 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
761 `pkg-config foo` for cflags/libs for the TLS implementation.
763 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
764 properties get this fed in as external SSF. A number of robustness
765 and debugging improvements to the cyrus_sasl authenticator.
767 PP/08 cyrus_sasl server now expands the server_realm option.
769 PP/09 Bugzilla 1214 - Log authentication information in reject log.
770 Patch by Jeremy Harris.
772 PP/10 Added dbmjz lookup type.
774 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
776 PP/12 MAIL args handles TAB as well as SP, for better interop with
777 non-compliant senders.
778 Analysis and variant patch by Todd Lyons.
780 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
781 Bug report from Lars Müller <lars@samba.org> (via SUSE),
782 Patch from Dirk Mueller <dmueller@suse.com>
784 PP/13 tls_peerdn now print-escaped for spool files.
785 Observed some $tls_peerdn in wild which contained \n, which resulted
786 in spool file corruption.
788 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
789 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
790 or write after TLS renegotiation, which otherwise led to messages
793 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
794 as a tracking header (ie: a signed header comes before the signature).
795 Patch from Wolfgang Breyha.
797 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
798 comma-sep list; embedded commas doubled.
800 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
802 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
804 Report and patch from Dmitry Banschikov.
806 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
807 Removed SSL_clear() after SSL_new() which led to protocol negotiation
808 failures. We appear to now support TLS1.1+ with Exim.
810 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
811 lets Exim select keys and certificates based upon TLS SNI from client.
812 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
813 before an outbound SMTP session. New log_selector, +tls_sni.
815 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
816 NULL dereference. Report and patch from Alun Jones.
818 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
819 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
820 Not seeing resolver debug output on NetBSD, but suspect this is a
821 resolver implementation change.
823 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
824 Left warnings. Added "eximon gdb" invocation mode.
826 PP/21 Defaulting "accept_8bitmime" to true, not false.
828 PP/22 Added -bw for inetd wait mode support.
830 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
831 locate the relevant includes and libraries. Made this the default.
833 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
834 Bugzilla 1246, report and most of solution from Tomasz Kusy.
836 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
837 This may cause build issues on older platforms.
839 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
840 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
841 gnutls_require_mac & gnutls_require_protocols (no longer supported).
842 Added SNI support via GnuTLS too.
843 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
845 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
847 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
848 Applied second patch from Janne, implementing suggestion to default
849 multiple-strings-in-record handling to match SPF spec.
851 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
853 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
854 read-only, out of scope).
855 Patch from Wolfgang Breyha, report from Stuart Northfield.
857 PP/29 Fix three issues highlighted by clang analyser static analysis.
858 Only crash-plausible issue would require the Cambridge-specific
859 iplookup router and a misconfiguration.
860 Report from Marcin Mirosław.
862 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
864 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
865 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
866 As part of this, removing so much warning spew let me fix some minor
867 real issues in debug logging.
869 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
870 assignment on my part. Fixed.
872 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
873 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
874 Janne Snabb (who went above and beyond: thank you).
876 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
877 string otherwise requires a connection and a bunch more work and it's
878 relatively easy to get wrong. Should also expose TLS library linkage
881 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
882 64-bit ${eval} (JH/03).
884 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
885 GNU libc to support some of the 64-bit stuff, should not lead to
886 conflicts. Defined before os.h is pulled in, so if a given platform
887 needs to override this, it can.
889 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
890 protection layer was required, which is not implemented.
891 Bugzilla 1254, patch from Wolfgang Breyha.
893 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
894 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
895 tls_dhparam take prime identifiers. Also unbreak combination of
896 OpenSSL+DH_params+TLSSNI.
898 PP/39 Disable SSLv2 by default in OpenSSL support.
904 PP/01 Solaris build fix for Oracle's LDAP libraries.
905 Bugzilla 1109, patch from Stephen Usher.
907 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
909 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
912 TF/02 Fix a couple more cases where we did not log the error message
913 when unlink() failed. See also change 4.74-TF/03.
915 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
916 lock up or crash if it happened to be inside a call to libc when it
917 got a SIGUSR1 from exiwhat.
919 The SIGUSR1 handler appends the current process status to the process
920 log which is later printed by exiwhat. It used to use the general
921 purpose logging code to do this, but several functions it calls are
922 not safe for signals.
924 The new output code in the SIGUSR1 handler is specific to the process
925 log, and simple enough that it's easy to inspect for signal safety.
926 Removing some special cases also simplifies the general logging code.
927 Removing the spurious timestamps from the process log simplifies
930 TF/04 Improved ratelimit ACL condition.
932 The /noupdate option has been deprecated in favour of /readonly which
933 has clearer semantics. The /leaky, /strict, and /readonly update modes
934 are mutually exclusive. The update mode is no longer included in the
935 database key; it just determines when the database is updated. (This
936 means that when you upgrde Exim will forget old rate measurements.)
938 Exim now checks that the per_* options are used with an update mode that
939 makes sense for the current ACL. For example, when Exim is processing a
940 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
941 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
942 must specify per_mail/readonly. If you omit the update mode it defaults to
943 /leaky where that makes sense (as before) or /readonly where required.
945 The /noupdate option is now undocumented but still supported for
946 backwards compatibility. It is equivalent to /readonly except that in
947 ACLs where /readonly is required you may specify /leaky/noupdate or
948 /strict/noupdate which are treated the same as /readonly.
950 A useful new feature is the /count= option. This is a generalization
951 of the per_byte option, so that you can measure the throughput of other
952 aggregate values. For example, the per_byte option is now equivalent
953 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
955 The per_rcpt option has been generalized using the /count= mechanism
956 (though it's more complicated than the per_byte equivalence). When it is
957 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
958 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
959 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
960 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
961 Note that using per_rcpt with a non-readonly update mode in more than
962 one ACL will cause the recipients to be double-counted. (The per_mail
963 and per_byte options don't have this problem.)
965 The handling of very low rates has changed slightly. If the computed rate
966 is less than the event's count (usually one) then this event is the first
967 after a long gap. In this case the rate is set to the same as this event's
968 count, so that the first message of a spam run is counted properly.
970 The major new feature is a mechanism for counting the rate of unique
971 events. The new per_addr option counts the number of different
972 recipients that someone has sent messages to in the last time period. It
973 behaves like per_rcpt if all the recipient addresses are different, but
974 duplicate recipient addresses do not increase the measured rate. Like
975 the /count= option this is a general mechanism, so the per_addr option
976 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
977 example, measure the rate that a client uses different sender addresses
978 with the options per_mail/unique=$sender_address. There are further
979 details in the main documentation.
981 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
983 TF/06 Removed a few PCRE remnants.
985 TF/07 Automatically extract Exim's version number from tags in the git
986 repository when doing development or release builds.
988 PP/02 Raise smtp_cmd_buffer_size to 16kB.
989 Bugzilla 879. Patch from Paul Fisher.
991 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
992 Heavily based on revision 40f9a89a from Simon Arlott's tree.
995 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
997 PP/05 Variable $av_failed, true if the AV scanner deferred.
998 Bugzilla 1078. Patch from John Horne.
1000 PP/06 Stop make process more reliably on build failure.
1001 Bugzilla 1087. Patch from Heiko Schlittermann.
1003 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1004 Bugzilla 1089. Patch from Heiko Schlittermann.
1006 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1007 Bugzilla 1131. Patch from Holger Weiß.
1009 PP/09 Handle IPv6 addresses with SPF.
1010 Bugzilla 860. Patch from Wolfgang Breyha.
1012 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1014 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1017 PP/11 match_* no longer expand right-hand-side by default.
1018 New compile-time build option, EXPAND_LISTMATCH_RHS.
1019 New expansion conditions, "inlist", "inlisti".
1021 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1023 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1025 PP/14 fix log_write() format string regression from TF/03.
1026 Bugzilla 1152. Patch from Dmitry Isaikin.
1032 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1034 PP/02 Harmonised TLS library version reporting; only show if debugging.
1035 Layout now matches that introduced for other libraries in 4.74 PP/03.
1037 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1039 PP/04 New "dns_use_edns0" global option.
1041 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1044 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1045 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1047 TK/01 Updated PolarSSL code to 0.14.2.
1048 Bugzilla 1097. Patch from Andreas Metzler.
1050 PP/07 Catch divide-by-zero in ${eval:...}.
1051 Fixes bugzilla 1102.
1053 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1056 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1057 format-string attack -- SECURITY: remote arbitrary code execution.
1059 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1060 time unintentionally subject to list matching rules, letting the header
1061 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1062 arbitrary string expansion). This allowed for information disclosure.
1064 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1065 INT_MIN/-1 -- value coerced to INT_MAX.
1071 NM/01 Workround for PCRE version dependency in version reporting
1074 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1075 This fixes portability to compilers other than gcc, notably
1076 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1078 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1079 makefiles for portability to HP-UX and POSIX correctness.
1081 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1082 Also via indented variable definition in the Makefile.
1083 (Debugging by Oliver Heesakkers).
1085 PP/02 Restore caching of spamd results with expanded spamd_address.
1086 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1088 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1089 Improves build reliability. Fix from: Frank Elsner
1091 NM/02 Fix wide character breakage in the rfc2047 coding
1092 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1094 NM/03 Allow underscore in dnslist lookups
1095 Fixes bug 1026. Patch from Graeme Fowler
1097 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1098 Code patches from Adam Ciarcinski of NetBSD.
1100 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1103 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1104 is logged, to avoid truncation. Patch from John Horne.
1106 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1107 Patch from Jakob Hirsch.
1109 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1110 SQL string expansion failure details.
1111 Patch from Andrey Oktyabrski.
1113 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1114 Patch from Simon Arlott.
1116 PP/09 New lookups functionality failed to compile on old gcc which rejects
1117 extern declarations in function scope.
1118 Patch from Oliver Fleischmann
1120 PP/10 Use sig_atomic_t for flags set from signal handlers.
1121 Check getgroups() return and improve debugging.
1122 Fixed developed for diagnosis in bug 927 (which turned out to be
1125 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1126 Patch from Mark Zealey.
1128 PP/12 Bugzilla 1056: Improved spamd server selection.
1129 Patch from Mark Zealey.
1131 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1132 Based on patch from Heiko Schlittermann.
1134 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1135 Patch from Uwe Doering, sign-off by Michael Haardt.
1137 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1138 variable declaration deep within a block. Bug and patch from
1141 PP/15 lookups-Makefile IRIX compatibilty coercion.
1143 PP/16 Make DISABLE_DKIM build knob functional.
1145 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1146 Patch by Simon Arlott
1148 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1149 variable argument macros. Our copy now differs from upstream.
1155 TF/01 Failure to get a lock on a hints database can have serious
1156 consequences so log it to the panic log.
1158 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1159 controlled using the smtp_confirmation log selector.
1161 TF/03 Include the error message when we fail to unlink a spool file.
1163 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1164 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1165 for maintaining out-of-tree patches for some time.
1167 PP/01 Bugzilla 139: Documentation and portability issues.
1168 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1169 Handle per-OS dynamic-module compilation flags.
1171 PP/02 Let /dev/null have normal permissions.
1172 The 4.73 fixes were a little too stringent and complained about the
1173 permissions on /dev/null. Exempt it from some checks.
1174 Reported by Andreas M. Kirchwitz.
1176 PP/03 Report version information for many libraries, including
1177 Exim version information for dynamically loaded libraries. Created
1178 version.h, now support a version extension string for distributors
1179 who patch heavily. Dynamic module ABI change.
1181 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1182 privilege escalation vulnerability whereby the Exim run-time user
1183 can cause root to append content of the attacker's choosing to
1186 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1189 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1190 If dropping privileges for untrusted macros, we disabled normal logging
1191 on the basis that it would fail; for the Exim run-time user, this is not
1192 the case, and it resulted in successful deliveries going unlogged.
1193 Fixed. Reported by Andreas Metzler.
1199 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1200 only prepend for the Resent-* case. Fixes regression introduced in
1201 Exim 4.70 by NM/22 for Bugzilla 607.
1203 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1204 increasing numbers of administrators be bitten by this.
1206 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1208 PP/03 Bugzilla 994: added openssl_options main configuration option.
1210 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1212 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1214 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1216 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1217 without a peer certificate, leading to a segfault because of an
1218 assumption that peers always have certificates. Be a little more
1219 paranoid. Problem reported by Martin Tscholak.
1221 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1222 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1223 NB: ClamAV planning to remove STREAM in "middle of 2010".
1224 CL also introduces -bmalware, various -d+acl logging additions and
1225 more caution in buffer sizes.
1227 PP/09 Implemented reverse_ip expansion operator.
1229 PP/10 Bugzilla 937: provide a "debug" ACL control.
1231 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1233 PP/12 Bugzilla 973: Implement --version.
1235 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1237 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1239 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1241 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1242 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1243 ignore trailing whitespace.
1245 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1247 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1250 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1251 Notification from Dr Andrew Aitchison.
1253 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1254 ExtendedDetectionInfo response format.
1255 Notification from John Horne.
1257 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1260 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1261 XSL and documented dependency on system catalogs, with examples of how
1264 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1267 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1268 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1269 configuration file which is writeable by the Exim user or group.
1271 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1272 of configuration files to cover files specified with the -C option if
1273 they are going to be used with root privileges, not just the default
1276 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1277 option (effectively making it always true).
1279 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1280 files to be used while preserving root privileges.
1282 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1283 that rogue child processes cannot use them.
1285 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1286 run-time user, instead of root.
1288 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1289 Exim run-time user without dropping privileges.
1291 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1292 result string, instead of calling string_vformat() twice with the same
1295 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1296 for other users. Others should always drop root privileges if they use
1297 -C on the command line, even for a whitelisted configure file.
1299 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1301 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1307 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1308 $data_path, and $header_path variables; fixed documentation bugs and
1311 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1312 exipick to access non-standard spools, including the "frozen" queue
1315 NM/01 Bugzilla 965: Support mysql stored procedures.
1316 Patch from Alain Williams
1318 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1320 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1321 Patch from Andreas Metzler
1323 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1324 Patch from Kirill Miazine
1326 NM/05 Bugzilla 671: Added umask to procmail example.
1328 JJ/03 installed exipick 20100323.0, fixing doc bug
1330 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1331 directory. Notification and patch from Dan Rosenberg.
1333 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1335 TK/02 Improve log output when DKIM signing operation fails.
1337 MH/01 Treat the transport option dkim_domain as a colon separated
1338 list, not as a single string, and sign the message with each element,
1339 omitting multiple occurences of the same signer.
1341 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1342 Bugzilla 985, 986. Patch by Simon Arlott
1344 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1345 Patch by Simon Arlott
1347 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1348 MBX locking. Notification from Dan Rosenberg.
1354 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1356 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1358 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1360 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1362 NM/04 Bugzilla 915: Flag broken perl installation during build.
1368 TK/01 Added patch by Johannes Berg that expands the main option
1369 "spamd_address" if it starts with a dollar sign.
1371 TK/02 Write list of recipients to X-Envelope-Sender header when building
1372 the mbox-format spool file for content scanning (suggested by Jakob
1375 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1376 (http://www.dcc-servers.net/) support via dccifd. Activated by
1377 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1379 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1380 by Mark Daniel Reidel <mr@df.eu>.
1382 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1383 When building exim an external PCRE library is now needed -
1384 PCRE is a system library on the majority of modern systems.
1385 See entry on PCRE_LIBS in EDITME file.
1387 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1388 conversation. Added nologin parameter to request.
1389 Patch contributed by Kirill Miazine.
1391 TF/01 Do not log submission mode rewrites if they do not change the address.
1393 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1395 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1396 log files in place. Contributed by Roberto Lima.
1398 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1400 TF/03 Bugzilla 615: When checking the local_parts router precondition
1401 after a local_part_suffix or local_part_prefix option, Exim now
1402 does not use the address's named list lookup cache, since this
1403 contains cached lookups for the whole local part.
1405 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1406 Robert Millan. Documentation is in experimental-spec.txt.
1408 TF/04 Bugzilla 668: Fix parallel build (make -j).
1410 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1412 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1413 Patch provided by Jan Srzednicki.
1415 TF/05 Leading white space used to be stripped from $spam_report which
1416 wrecked the formatting. Now it is preserved.
1418 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1419 that they are available at delivery time.
1421 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1423 TF/08 TLS error reporting now respects the incoming_interface and
1424 incoming_port log selectors.
1426 TF/09 Produce a more useful error message if an SMTP transport's hosts
1427 setting expands to an empty string.
1429 NM/06 Bugzilla 744: EXPN did not work under TLS.
1430 Patch provided by Phil Pennock.
1432 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1433 Patch provided by Richard Godbee.
1435 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1436 acl_smtp_notquit, added index entry.
1438 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1439 Patch provided by Eugene Bujak.
1441 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1442 accept(). Patch provided by Maxim Dounin.
1444 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1445 Patch provided by Phil Pennock.
1447 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1449 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1450 Patch provided by Brad "anomie" Jorsch.
1452 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1453 Patch provided by Dean Brooks.
1455 TK/05 Add native DKIM support (does not depend on external libraries).
1457 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1458 Patch provided by Graeme Fowler.
1460 NM/16 Bugzilla 851: Documentation example syntax fix.
1462 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1464 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1467 NM/19 Bugzilla 745: TLS version reporting.
1468 Patch provided by Phil Pennock.
1470 NM/20 Bugzilla 167: bool: condition support.
1471 Patch provided by Phil Pennock.
1473 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1474 clients. Patch provided by Phil Pennock.
1476 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1477 Patch provided by Brad "anomie" Jorsch.
1479 NM/23 Bugzilla 687: Fix misparses in eximstats.
1480 Patch provided by Heiko Schlittermann.
1482 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1483 Patch provided by Heiko Schlittermann.
1485 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1486 plus update to original patch.
1488 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1490 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1491 Patch provided by David Brownlee.
1493 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1495 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1497 NM/30 Bugzilla 888: TLS documentation bugfixes.
1499 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1501 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1502 Unlike the original bugzilla I have changed all shell scripts in src tree.
1504 NM/33 Bugzilla 898: Transport filter timeout fix.
1505 Patch by Todd Rinaldo.
1507 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1508 Patch by Serge Demonchaux.
1510 NM/35 Bugzilla 39: Base64 decode bug fixes.
1511 Patch by Jakob Hirsch.
1513 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1515 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1517 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1519 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1525 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1526 ALT-N's libdkim that I have put here:
1527 http://duncanthrax.net/exim-experimental/
1529 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1530 were called 'true' and it seems that C99 defines that as a reserved
1531 keyword to be used with 'bool' variable types. That means you could
1532 not include C99-style headers which use bools without triggering
1533 build errors in sieve.c.
1535 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1536 as mailq or other aliases. Changed the --help handling significantly
1537 to do whats expected. exim_usage() emits usage/help information.
1539 SC/01 Added the -bylocaldomain option to eximstats.
1541 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1543 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1545 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1551 PH/01 Another patch from the Sieve maintainer.
1553 PH/02 When an IPv6 address is converted to a string for single-key lookup
1554 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1555 dots are used instead of colons so that keys in lsearch files need not
1556 contain colons. This was done some time before quoting was made available
1557 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1558 (notated using the quote facility) so as to distinguish them from IPv4
1559 keys. This meant that lookups for IP addresses in host lists did not work
1560 for iplsearch lookups.
1562 This has been fixed by arranging for IPv6 addresses to be expressed with
1563 colons if the lookup type is iplsearch. This is not incompatible, because
1564 previously such lookups could never work.
1566 The situation is now rather anomolous, since one *can* have colons in
1567 ordinary lsearch keys. However, making the change in all cases is
1568 incompatible and would probably break a number of configurations.
1570 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1573 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1574 conversion specification without a maximum field width, thereby enabling
1575 a rogue spamd server to cause a buffer overflow. While nobody in their
1576 right mind would setup Exim to query an untrusted spamd server, an
1577 attacker that gains access to a server running spamd could potentially
1578 exploit this vulnerability to run arbitrary code as the Exim user.
1580 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1581 $primary_hostname instead of what libspf2 thinks the hosts name is.
1583 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1584 a directory entry by the name of the lookup key. Previously, if a
1585 symlink pointed to a non-existing file or a file in a directory that
1586 Exim lacked permissions to read, a lookup for a key matching that
1587 symlink would fail. Now it is enough that a matching directory entry
1588 exists, symlink or not. (Bugzilla 503.)
1590 PH/03 The body_linecount and body_zerocount variables are now exported in the
1593 PH/04 Added the $dnslist_matched variable.
1595 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1596 This means they are set thereafter only if the connection becomes
1599 PH/06 Added the client_condition to authenticators so that some can be skipped
1600 by clients under certain conditions.
1602 PH/07 The error message for a badly-placed control=no_multiline_responses left
1603 "_responses" off the end of the name.
1605 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1607 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1608 (without spaces) instead of just copying the configuration text.
1610 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1612 PH/11 Added $max_received_linelength.
1614 PH/12 Added +ignore_defer and +include_defer to host lists.
1616 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1617 way in which PCRE > 7.0 is built.
1619 PH/14 Implemented queue_only_load_latch.
1621 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1622 MAIL command. The effect was to mangle the value on 64-bit systems.
1624 PH/16 Another patch from the Sieve maintainer.
1626 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1628 PH/18 If a system quota error occurred while trying to create the file for
1629 a maildir delivery, the message "Mailbox is full" was not appended to the
1630 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1631 only to a quota excession during the actual writing of the file.
1633 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1634 characters?) which causes problems in log lines. The DN values are now
1635 passed through string_printing() before being added to log lines.
1637 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1638 and InterBase are left for another time.)
1640 PH/21 Added message_body_newlines option.
1642 PH/22 Guard against possible overflow in moan_check_errorcopy().
1644 PH/23 POSIX allows open() to be a macro; guard against that.
1646 PH/24 If the recipient of an error message contained an @ in the local part
1647 (suitably quoted, of course), incorrect values were put in $domain and
1648 $local_part during the evaluation of errors_copy.
1654 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1655 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1658 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1659 issue a MAIL command.
1661 PH/02 In an ACL statement such as
1663 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1665 if a client was not listed at all, or was listed with a value other than
1666 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1667 the condition was not true (as it should be), so access was not denied.
1668 The bug was that the ! inversion was incorrectly passed on to the second
1669 item. This has been fixed.
1671 PH/03 Added additional dnslists conditions == and =& which are different from
1672 = and & when the dns lookup returns more than one IP address.
1674 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1675 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1677 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1678 FSYNC, which compiles an option called disable_fsync that allows for
1679 bypassing fsync(). The documentation is heavily laced with warnings.
1681 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1683 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1684 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1685 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1686 including adding "make clean"; (3) Added -fPIC when compiling the test
1687 dynamically loaded module, to get rid of a warning.
1689 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1690 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1691 = 0s. The bug is otherwise harmless.
1693 PH/07 There was a bug in the dovecot authenticator such that the value of
1694 $auth1 could be overwritten, and so not correctly preserved, after a
1695 successful authentication. This usually meant that the value preserved by
1696 the server_setid option was incorrect.
1698 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1700 PH/09 Installed PCRE release 7.0.
1702 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1703 run for batched SMTP input. It is now run at the start of every message
1704 in the batch. While fixing this I discovered that the process information
1705 (output by running exiwhat) was not always getting set for -bs and -bS
1706 input. This is fixed, and it now also says "batched" for BSMTP.
1708 PH/11 Added control=no_pipelining.
1710 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1711 patch, slightly modified), and move the expansion of helo_data till after
1712 the connection is made in the smtp transport (so it can use these
1715 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1717 PH/14 Added log_selector = +pid.
1719 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1721 PH/16 Add ${if forany and ${if forall.
1723 PH/17 Added dsn_from option to vary the From: line in DSNs.
1725 PH/18 Flush SMTP output before performing a callout, unless control =
1726 no_callout_flush is set.
1728 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1729 was true (the default) a successful delivery failed to delete the retry
1730 item, thus causing premature timeout of the address. The bug is now
1733 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1735 PH/21 Long custom messages for fakedefer and fakereject are now split up
1736 into multiline reponses in the same way that messages for "deny" and
1737 other ACL rejections are.
1739 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1740 with slight modification.
1742 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1743 draft, changing the syntax and factoring some duplicate code".
1745 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1746 for deliveries of the second and subsequent messages over the same SMTP
1749 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1750 ${reduce, with only minor "tidies".
1752 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1754 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1755 expansion side effects.
1757 PH/27 When a message times out after an over-quota error from an Exim-imposed
1758 quota, the bounce message says "mailbox is full". This message was not
1759 being given when it was a system quota that was exceeded. It now should
1762 MH/03 Made $recipients available in local_scan(). local_scan() already has
1763 better access to the recipient list through recipients_list[], but
1764 $recipients can be useful in postmaster-provided expansion strings.
1766 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1767 in the case of a MAIL command with additional options following the
1768 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1769 were accidentally chopped off.
1771 PH/29 SMTP synchronization checks are implemented when a command is read -
1772 there is a check that no more input is waiting when there shouldn't be
1773 any. However, for some commands, a delay in an ACL can mean that it is
1774 some time before the response is written. In this time, more input might
1775 arrive, invalidly. So now there are extra checks after an ACL has run for
1776 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1777 pipelining has not been advertised.
1779 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1781 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1782 returned in the SMTP response when smtp_return_error_details was false.
1783 This has been fixed.
1785 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1786 instead of the C I/O that was originally supplied, because problems were
1787 reported on Solaris.
1789 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1790 Exim which did not show up earlier: it was assuming that a call to
1791 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1792 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1793 was a macro that became an assignment, so it seemed to work. This has
1794 changed to a proper function call with a void return, hence the compile
1795 error. Exim's code has been fixed.
1797 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1800 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1802 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1804 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1805 because it turns out to be too big or there is a timeout) there is no
1806 "Completed" line in the log. When some messages of this type were
1807 selected by exigrep, they were listed as "not completed". Others were
1808 picked up by some special patterns. I have improved the selection
1809 criteria to be more general.
1811 PH/38 The host_find_failed option in the manualroute router can now be set
1812 to "ignore", to completely ignore a host whose IP address cannot be
1813 found. If all hosts are ignored, the behaviour is controlled by the new
1814 host_all_ignored option.
1816 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1817 homing or because of multiple MX records with /mx) generated more than
1818 one IP address, and the following item turned out to be the local host,
1819 all the secondary addresses of the first item were incorrectly removed
1820 from the list, along with the local host and any following hosts (which
1821 is what is supposed to happen).
1823 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1824 whoever called Exim into the -H file. In the case of the daemon it was
1825 behaving confusingly. When first started, it used values for whoever
1826 started the daemon, but after a SIGHUP it used the Exim user (because it
1827 calls itself on a restart). I have changed the code so that it now always
1830 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1831 message are rejected with the same error (e.g. no authentication or bad
1832 sender address), and a DATA command is nevertheless sent (as can happen
1833 with PIPELINING or a stupid MUA), the error message that was given to the
1834 RCPT commands is included in the rejection of the DATA command. This is
1835 intended to be helpful for MUAs that show only the final error to their
1838 PH/42 Another patch from the Sieve maintainer.
1840 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1841 Eximstats - Fixed some broken HTML links and added missing column headers
1843 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1844 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1846 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1848 PH/43 Yet another patch from the Sieve maintainer.
1850 PH/44 I found a way to check for a TCP/IP connection going away before sending
1851 the response to the final '.' that terminates a message, but only in the
1852 case where the client has not sent further data following the '.'
1853 (unfortunately, this is allowed). However, in many cases there won't be
1854 any further data because there won't be any more messages to send. A call
1855 to select() can be used: if it shows that the input is "ready", there is
1856 either input waiting, or the socket has been closed. An attempt to read
1857 the next input character can distinguish the two cases. Previously, Exim
1858 would have sent an OK response which the client would never have see.
1859 This could lead to message repetition. This fix should cure that, at
1860 least in a lot of common cases.
1862 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1863 advertised in response to EHLO.
1869 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1870 fixed by 4.65/MH/01 (is this a record?) are fixed:
1872 (i) An empty string was always treated as zero by the numeric comparison
1873 operators. This behaviour has been restored.
1875 (ii) It is documented that the numeric comparison operators always treat
1876 their arguments as decimal numbers. This was broken in that numbers
1877 starting with 0 were being interpreted as octal.
1879 While fixing these problems I realized that there was another issue that
1880 hadn't been noticed. Values of message_size_limit (both the global option
1881 and the transport option) were treated as octal if they started with 0.
1882 The documentation was vague. These values are now always treated as
1883 decimal, and I will make that clear in the documentation.
1889 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1890 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1893 MH/01 Don't check that the operands of numeric comparison operators are
1894 integers when their expansion is in "skipping" mode (fixes bug
1895 introduced by 4.64-PH/07).
1897 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1898 child addresses, Exim now panics and dies. Previously, because the count
1899 is held in a short int, deliveries were likely to be lost. As such a
1900 large number of recipients for a single message is ridiculous
1901 (performance will be very, very poor), I have chosen to impose a limit
1902 rather than extend the field.
1908 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1909 leftover -K file (the existence of which was triggered by #402).
1910 While we were at it, introduced process PID as part of the -K
1911 filename. This should rule out race conditions when creating
1914 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1915 processing considerably. Previous code took too long for large mails,
1916 triggering a timeout which in turn triggers #401.
1918 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1919 in the DK code in transports.c. sendfile() is not really portable,
1920 hence the _LINUX specificness.
1922 TF/01 In the add_headers option to the mail command in an Exim filter,
1923 there was a bug that Exim would claim a syntax error in any
1924 header after the first one which had an odd number of characters
1927 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1928 callout verification, Exim cached a "reject" for the entire domain. This
1929 is correct for most verifications, but it is not correct for a recipient
1930 verification with use_sender or use_postmaster set, because in that case
1931 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1932 case of MAIL FROM:<> rejection from other early rejections (e.g.
1933 rejection of HELO). When verifying a recipient using a non-null MAIL
1934 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1935 Whatever the result of the callout, the value of the domain cache is
1936 left unchanged (for any other kind of callout, getting as far as trying
1937 RCPT means that the domain itself is ok).
1939 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1942 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1943 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1944 session, the daemon ignores ECONNECT errors and logs others; it now
1945 ignores EPIPE as well.
1947 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1948 (quoted-printable decoding).
1950 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1951 later the small subsequent patch to fix an introduced bug.
1953 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1955 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1957 PH/08 An error is now given if message_size_limit is specified negative.
1959 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1960 to be given (somewhat) arbitrary names.
1962 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1965 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1966 miscellaneous code fixes
1968 PH/10 Added the log_reject_target ACL modifier to specify where to log
1971 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1972 hostname. This is wrong, because it relates to the incoming message (and
1973 probably the interface on which it is arriving) and not to the outgoing
1974 callout (which could be using a different interface). This has been
1975 changed to use the value of the helo_data option from the smtp transport
1976 instead - this is what is used when a message is actually being sent. If
1977 there is no remote transport (possible with a router that sets up host
1978 addresses), $smtp_active_hostname is used.
1980 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1981 tweaks were necessary in order to get it to work (see also 21 below):
1982 (a) The code assumed that strncpy() returns a negative number on buffer
1983 overflow, which isn't the case. Replaced with Exim's string_format()
1985 (b) There were several signed/unsigned issues. I just did the minimum
1986 hacking in of casts. There is scope for a larger refactoring.
1987 (c) The code used strcasecmp() which is not a standard C function.
1988 Replaced with Exim's strcmpic() function.
1989 (d) The code set only $1; it now sets $auth1 as well.
1990 (e) A simple test gave the error "authentication client didn't specify
1991 service in request". It would seem that Dovecot has changed its
1992 interface. Fortunately there's a specification; I followed it and
1993 changed what the client sends and it appears to be working now.
1995 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1998 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1999 address A is aliased to B and C, where B exists and C does not. Without
2000 -v the output is "A verified" because verification stops after a
2001 successful redirection if more than one address is generated. However,
2002 with -v the child addresses are also verified. Exim was outputting "A
2003 failed to verify" and then showing the successful verification for C,
2004 with its parentage. It now outputs "B failed to verify", showing B's
2005 parentage before showing the successful verification of C.
2007 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2008 look up a TXT record in a specific list after matching in a combined
2011 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2012 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2013 they consult the DNS. I had assumed they would set it the way they
2014 wanted; and indeed my experiments on Linux seem to show that in some
2015 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2016 To be on the safe side, however, I have now made the interface to
2017 host_find_byname() similar to host_find_bydns(), with an argument
2018 containing the DNS resolver options. The host_find_byname() function now
2019 sets these options at its start, just as host_find_bydns() does. The smtp
2020 transport options dns_qualify_single and dns_search_parents are passed to
2021 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2022 of host_find_byname() use the default settings of RES_DEFNAMES
2023 (qualify_single) but not RES_DNSRCH (search_parents).
2025 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2026 spool_read_header() do less string testing, by means of a preliminary
2027 switch on the second character of optional "-foo" lines. (This is
2028 overdue, caused by the large number of possibilities that now exist.
2029 Originally there were few.) While I was there, I also converted the
2030 str(n)cmp tests so they don't re-test the leading "-" and the first
2031 character, in the hope this might squeeze out yet more improvement.
2033 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2034 flag allowing group syntax was set by the header_syntax check but not
2035 turned off, possible causing trouble later; (2) The flag was not being
2036 set at all for the header_verify test, causing "group"-style headers to
2037 be rejected. I have now set it in this case, and also caused header_
2038 verify to ignore an empty address taken from a group. While doing this, I
2039 came across some other cases where the code for allowing group syntax
2040 while scanning a header line wasn't quite right (mostly, not resetting
2041 the flag correctly in the right place). These bugs could have caused
2042 trouble for malformed header lines. I hope it is now all correct.
2044 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2045 with the "reply" argument non-NULL. The code, however (which originally
2046 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2047 but it didn't always do it. This confused somebody who was copying the
2048 code for some other use. I have removed all the tests.
2050 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2051 feature that was used to support insecure browsers during the U.S. crypto
2052 embargo. It requires special client support, and Exim is probably the
2053 only MTA that supported it -- and would never use it because real RSA is
2054 always available. This code has been removed, because it had the bad
2055 effect of slowing Exim down by computing (never used) parameters for the
2056 RSA_EXPORT functionality.
2058 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2059 authenticator to fail if there's a tab character in the incoming data
2060 (there should never be unless someone is messing about, as it's supposed
2061 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2062 if the connection is using TLS or if the remote IP is the same as the
2063 local IP, and the "valid-client-cert option" if a client certificate has
2066 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2067 authenticators. This can be used for authorization after authentication
2068 succeeds. (In the case of plaintext, it servers for both authentication
2071 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2072 if any retry times were supplied.
2074 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2075 connection, where there is no client IP address to check. In this
2076 situation, the verify now always succeeds.
2078 PH/25 Applied John Jetmore's -Mset patch.
2080 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2082 PH/27 In a string expansion for a processed (not raw) header when multiple
2083 headers of the same name were present, leading whitespace was being
2084 removed from all of them, but trailing whitespace was being removed only
2085 from the last one. Now trailing whitespace is removed from each header
2086 before concatenation. Completely empty headers in a concatenation (as
2087 before) are ignored.
2089 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2090 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2092 PH/29 [Removed. This was a change that I later backed out, and forgot to
2093 correct the ChangeLog entry (that I had efficiently created) before
2094 committing the later change.]
2096 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2097 address errors (4xx response to RCPT) over the same connection as other
2098 messages routed to the same hosts. Such deliveries are always "forced",
2099 so retry times are not inspected. This resulted in far too many retries
2100 for the affected addresses. The effect occurred only when there were more
2101 hosts than the hosts_max_try setting in the smtp transport when it had
2102 the 4xx errors. Those hosts that it had tried were not added to the list
2103 of hosts for which the message was waiting, so if all were tried, there
2104 was no problem. Two fixes have been applied:
2106 (i) If there are any address or message errors in an SMTP delivery, none
2107 of the hosts (tried or untried) are now added to the list of hosts
2108 for which the message is waiting, so the message should not be a
2109 candidate for sending over the same connection that was used for a
2110 successful delivery of some other message. This seems entirely
2111 reasonable: after all the message is NOT "waiting for some host".
2112 This is so "obvious" that I'm not sure why it wasn't done
2113 previously. Hope I haven't missed anything, but it can't do any
2114 harm, as the worst effect is to miss an optimization.
2116 (ii) If, despite (i), such a delivery is accidentally attempted, the
2117 routing retry time is respected, so at least it doesn't keep
2118 hammering the server.
2120 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2121 in ${readsocket because some servers need this prod.
2123 PH/32 Added some extra debug output when updating a wait-xxx database.
2125 PH/33 The hint "could be header name not terminated by colon", which has been
2126 given for certain expansion errors for a long time, was not being given
2127 for the ${if def:h_colon_omitted{... case.
2129 PH/34 The spec says: "With one important exception, whenever a domain list is
2130 being scanned, $domain contains the subject domain." There was at least
2131 one case where this was not true.
2133 PH/35 The error "getsockname() failed: connection reset by peer" was being
2134 written to the panic log as well as the main log, but it isn't really
2135 panic-worthy as it just means the connection died rather early on. I have
2136 removed the panic log writing for the ECONNRESET error when getsockname()
2139 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2140 runs only) independently of the message's sender address. This meant
2141 that, if the 4xx error was in fact related to the sender, a different
2142 message to the same recipient with a different sender could confuse
2143 things. In particualar, this can happen when sending to a greylisting
2144 server, but other circumstances could also provoke similar problems.
2145 I have changed the default so that the retry time for these errors is now
2146 based a combination of the sender and recipient addresses. This change
2147 can be overridden by setting address_retry_include_sender=false in the
2150 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2151 remote server are returned as part of bounce messages. This was not
2152 happening for LMTP over a pipe (the lmtp transport), but now it is the
2153 same for both kinds of LMTP.
2155 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2156 in header lines that were in fact CNAMEs. This is no longer the case.
2158 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2159 and queue runs started by the daemon processed all messages. This has
2160 been fixed so that -R and -S can now usefully be given with -q<time>.
2162 PH/40 Import PCRE release 6.7 (fixes some bugs).
2164 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2166 PH/42 Give an error if -q is specified more than once.
2168 PH/43 Renamed the variables $interface_address and $interface_port as
2169 $received_ip_address and $received_port, to make it clear that these
2170 values apply to message reception, and not to the outgoing interface when
2171 a message is delivered. (The old names remain recognized, of course.)
2173 PH/44 There was no timeout on the connect() call when using a Unix domain
2174 socket in the ${readsocket expansion. There now is.
2176 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2177 be meaningful with "accept".
2179 SC/01 Eximstats V1.43
2180 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2182 SC/02 Eximstats V1.44
2183 Use a glob alias rather than an array ref in the generated
2184 parser. This improves both readability and performance.
2186 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2187 Collect SpamAssassin and rejection statistics.
2188 Don't display local sender or destination tables unless
2189 there is data to show.
2190 Added average volumes into the top table text output.
2192 SC/04 Eximstats V1.46
2193 Collect data on the number of addresses (recipients)
2194 as well as the number of messages.
2196 SC/05 Eximstats V1.47
2197 Added 'Message too big' to the list of mail rejection
2198 reasons (thanks to Marco Gaiarin).
2200 SC/06 Eximstats V1.48
2201 Mainlog lines which have GMT offsets and are too short to
2202 have a flag are now skipped.
2204 SC/07 Eximstats V1.49 (Alain Williams)
2205 Added the -emptyok flag.
2207 SC/08 Eximstats V1.50
2208 Fixes for obtaining the IP address from reject messages.
2210 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2211 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2212 whitesspace changes from 4.64-PH/27
2214 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2217 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2218 are found, allow negative numbers in numeric criteria)
2220 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2222 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2223 to match changes made in 4.64-PH/43
2225 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2227 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2228 "domain" is actually an IP address, and force a failure. This locks out
2229 those revolvers/nameservers that support "A-for-A" lookups, in
2230 contravention of the specifications.
2232 PH/48 When a host name was looked up from an IP address, and the subsequent
2233 forward lookup of the name timed out, the host name was left in
2234 $sender_host_name, contrary to the specification.
2236 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2237 restricted to single-key lookups, Exim was not diagnosing an error if
2238 * or *@ was used with a query-style lookup.
2240 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2242 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2243 long ago, but noone interested enough thought of it. Let's just say that
2244 the "1.1" means that there are some new functions that weren't there at
2245 some point in the past.
2247 PH/51 Error processing for expansion failure of helo_data from an smtp
2248 transport during callout processing was broken.
2250 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2251 tested/used via the -bh/-bhc/-bs options.
2253 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2254 bug, fixed in subsequent PCRE releases).
2256 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2257 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2259 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2265 SC/01 Use a glob alias rather than an array ref in eximstats generated
2266 parser. This improves both readability and performance.
2268 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2269 Don't display local sender or destination tables in eximstats unless
2270 there is data to show.
2271 Added average volumes into the eximstats top table text output.
2273 SC/03 Collect data on the number of addresses (recipients) as well
2274 as the number of messages in eximstats.
2276 TF/01 Correct an error in the documentation for the redirect router. Exim
2277 does (usually) call initgroups() when daemonizing.
2279 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2280 with consistent privilege compared to when running as a daemon.
2282 TF/03 Note in the spec that $authenticated_id is not set for local
2283 submissions from trusted users.
2285 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2286 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2288 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2289 by adding some example configuration directives to the default
2290 configuration file. A little bit of work is required to uncomment the
2291 directives and define how usernames and passwords are checked, but
2292 there is now a framework to start from.
2294 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2295 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2296 without this. I don't know how relevant this is to other LDAP libraries.
2298 PH/02 Add the verb name to the "unknown ACL verb" error.
2300 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2302 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2304 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2305 directory not expanded when it should be if an expanded home directory
2306 was set for the address (which is overridden by the transport).
2308 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2311 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2312 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2313 because it is too late at that time, and has no effect.
2315 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2316 security issue with \' (bugzilla #107). I could not use the
2317 PQescapeStringConn() function, because it needs a PGconn value as one of
2320 PH/08 When testing addresses using -bt, indicate those final addresses that
2321 are duplicates that would not cause an additional delivery. At least one
2322 person was confused, thinking that -bt output corresponded to deliveries.
2323 (Suppressing duplicates isn't a good idea as you lose the information
2324 about possibly different redirections that led to the duplicates.)
2326 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2327 systems where poll() doesn't work, in particular OS X.
2329 PH/10 Added more information to debugging output for retry time not reached.
2331 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2332 operations in malware.c.
2334 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2337 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2338 syslog_duplication set false, log lines that would normally be written
2339 both the the main log and to the reject log were not written to syslog at
2342 PH/14 In the default configuration, change the use of "message" in ACL warn
2343 statements to "add_header".
2345 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2346 not followed by a command (e.g. "seen endif").
2348 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2349 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2352 PH/17 Added extra conditions to the default value of delay_warning_condition
2356 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2357 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2358 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2361 The Auto-Submitted: and various List- headers are standardised, whereas I
2362 don't think Precedence: ever was.
2364 PH/18 Refactored debugging code in route_finduser() to show more information,
2365 in particular, the error code if getpwnam() issues one.
2367 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2368 This is apparently needed in addition to the PH/07 change above to avoid
2369 any possible encoding problems.
2371 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2372 but not after initializing Perl.
2374 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2375 output them only if debugging. By default they are written stderr,
2376 apparently, which is not desirable.
2378 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2381 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2384 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2386 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2387 authenticated or an ident call has been made. Suppress the default
2388 values for $authenticated_id and $authenticated_sender (but permit -oMai
2389 and -oMas) when testing with -bh.
2391 PH/24 Re-jigged the order of the tests in the default configuration so that the
2392 tests for valid domains and recipients precede the DNS black list and CSA
2393 tests, on the grounds that those ones are more expensive.
2395 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2396 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2397 This bug exists in every version of Exim that I still have, right back to
2400 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2401 However, an attempt to turn on case-sensitivity in a regex key by
2402 including (?-i) didn't work because the subject string was already
2403 lowercased, and the effects were non-intuitive. It turns out that a
2404 one-line patch can be used to allow (?-i) to work as expected.
2410 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2411 other effects) broke the use of negated acl sub-conditions.
2413 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2416 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2417 "Deny" causes Exim to reject the incoming connection with a 554 error.
2418 Unfortunately, if there is a major crisis, such as a disk failure,
2419 tcp-wrappers gives "deny", whereas what one would like would be some
2420 kind of temporary error. A kludge has been added to help with this.
2421 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2422 554 error is used if errno is still zero or contains ENOENT (which occurs
2423 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2426 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2428 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2429 errors. Otherwise a message that provokes a temporary error (when other
2430 messages do not) can cause a whole host to time out.
2432 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2433 addresses were routed directly to files or pipes from a redirect router.
2434 File deliveries just didn't batch; pipe deliveries might have suffered
2437 PH/06 A failure to get a lock for a hints database would erroneously always say
2438 "Failed to get write lock", even when it was really a read lock.
2440 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2441 of 0600. This has been changed to use the value of the lockfile_mode
2442 option (which defaults to 0600).
2444 PH/08 Applied small patch from the Sieve maintainer.
2446 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2447 folder from quota calculations, a direct delivery into this folder messed
2448 up the contents of the maildirsize file. This was because the regex was
2449 used only to exclude .Trash (or whatever) when the size of the mailbox
2450 was calculated. There was no check that a delivery was happening into an
2451 excluded directory. This bug has been fixed by ignoring all quota
2452 processing for deliveries into excluded directories.
2454 PH/10 Added the maildirfolder_create_regex option to appendfile.
2460 PH/01 The code for finding all the local interface addresses on a FreeBSD
2461 system running IPv6 was broken. This may well have applied to all BSD
2462 systems, as well as to others that have similar system calls. The broken
2463 code found IPv4 interfaces correctly, but gave incorrect values for the
2464 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2465 that it would not match correctly against @[] and not recognize the IPv6
2468 PH/02 The ipliteral router was not recognizing addresses of the form user@
2469 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2471 PH/03 Added disable_ipv6.
2473 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2474 decoded form, because it is most often used to construct To: headers
2475 lines in autoreplies, and the decoded form may well be syntactically
2476 invalid. However, $reply_address has leading white space removed, and all
2477 newlines turned into spaces so that the autoreply transport does not
2480 PH/05 If group was specified without a user on a router, and no group or user
2481 was specified on a transport, the group from the router was ignored.
2483 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2484 for visible compile-time settings that can be used to change these
2485 numbers, for those that want even more. Backwards compatibility with old
2486 spool files has been maintained. However, going back to a previous Exim
2487 release will lost any variables that are in spool files.
2489 PH/07 Two small changes when running in the test harness: increase delay when
2490 passing a TCP/IP connection to a new process, in case the original
2491 process has to generate a bounce, and remove special handling of
2492 127.0.0.2 (sic), which is no longer necessary.
2494 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2495 be the same on different OS.
2497 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2500 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2501 whether --show-vars was specified or not
2503 JJ/02 exipick: Added support for new ACL variable spool format introduced
2506 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2507 syntactically invalid From: or Reply-to: line, and a filter used this to
2508 generate an autoreply, and therefore failed to obtain an address for the
2509 autoreply, Exim could try to deliver to a non-existent relative file
2510 name, causing unrelated and misleading errors. What now happens is that
2511 it logs this as a hard delivery error, but does not attempt to create a
2514 PH/11 The exinext utility has a -C option for testing purposes, but although
2515 the given file was scanned by exinext itself; it wasn't being passed on
2516 when Exim was called.
2518 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2519 an end-of-file indication when reading a command response.
2521 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2522 compiled. In many other places in Exim, IPv6 addresses are always
2523 recognized, so I have changed this. It also means that IPv4 domain
2524 literals of the form [IPV4:n.n.n.n] are now always recognized.
2526 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2527 used if the router is not running as root, for example, when verifying at
2528 ACL time, or when using -bh. The debugging output from this situation was
2529 non-existent - all you got was a failure to exec. I have made two
2532 (a) Failures to set uid/gid, the current directory, or a process leader
2533 in a subprocess such as that created by queryprogram now generate
2534 suitable debugging ouput when -d is set.
2536 (b) The queryprogram router detects when it is not running as root,
2537 outputs suitable debugging information if -d is set, and then runs
2538 the subprocess without attempting to change uid/gid.
2540 PH/15 Minor change to Makefile for building test_host (undocumented testing
2543 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2544 additional section of a DNS packet that returns MX or SRV records.
2545 Instead, it always explicitly searches for A/AAAA records. This avoids
2546 major problems that occur when a DNS server includes only records of one
2547 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2548 fixed another bug: if SRV records were looked up and the corresponding
2549 address records were *not* found in the additional section, the port
2550 values from the SRV records were lost.
2552 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2553 using the correct key (the original address) when searching the retry
2554 rules in order to find which one to use for generating the retry hint.
2556 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2557 adding the default one. Similarly, if it contains a Reply-To: header, the
2558 errors_reply_to option, if set, is not used.
2560 PH/19 When calculating a retry time, Exim used to measure the "time since
2561 failure" by looking at the "first failed" field in the retry record. Now
2562 it does not use this if it is later than than the arrival time of the
2563 message. Instead it uses the arrival time. This makes for better
2564 behaviour in cases where some deliveries succeed, thus re-setting the
2565 "first failed" field. An example is a quota failure for a huge message
2566 when small messages continue to be delivered. Without this change, the
2567 "time since failure" will always be short, possible causing more frequent
2568 delivery attempts for the huge message than are intended.
2569 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2571 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2572 $1, $2, $3) because the numerical variables can be reset during some
2573 expansion items (e.g. "match"), thereby losing the authentication data.
2575 PH/21 Make -bV show the size of off_t variables so that the test suite can
2576 decide whether to run tests for quotas > 2G.
2578 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2579 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2580 filecount value is greater than 2G or if a quota value is greater than 2G
2581 on a system where the size of off_t is not greater than 4, a panic error
2584 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2585 never match. The debug and -bh output now contains an explicit error
2586 message indicating a malformed IPv4 address or mask.
2588 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2589 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2590 PH/23 above applies.
2592 PH/25 Do not write to syslog when running in the test harness. The only
2593 occasion when this arises is a failure to open the main or panic logs
2594 (for which there is an explicit test).
2596 PH/26 Added the /no_tell option to "control=freeze".
2598 PH/27 If a host name lookup failed very early in a connection, for example, if
2599 the IP address matched host_lookup and the reverse lookup yielded a name
2600 that did not have a forward lookup, an error message of the form "no IP
2601 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2602 could be logged. Now it outputs the IP address instead of "NULL".
2604 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2605 allows the sender and the authenticated sender to be set when
2606 submitting a message from within Exim. Since child_open_exim() is
2607 documented for local_scan(), the new function should be too.
2609 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2610 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2611 results in an empty string is now treated as unset.
2613 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2615 PH/31 Added sender_verify_fail logging option.
2617 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2618 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2619 tidied the source and removed it altogether.
2621 PH/33 When a queue run was abandoned because the load average was too high, a
2622 log line was always written; now it is written only if the queue_run log
2623 selector is set. In addition, the log line for abandonment now contains
2624 information about the queue run such as the pid. This is always present
2625 in "start" and "stop" lines but was omitted from the "abandon" line.
2627 PH/34 Omit spaces between a header name and the colon in the error message that
2628 is given when verify = headers_syntax fails (if there are lots of them,
2629 the message gets confusing).
2631 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2632 names, as there are now some PTR records that contain slashes. This check
2633 is only to protect against broken name servers that fall over on strange
2634 characters, so the fact that it applies to all lookups doesn't matter.
2636 PH/36 Now that the new test suite is complete, we can remove some of the
2637 special code in Exim that was needed for the old test suite. For example,
2638 sorting DNS records because real resolvers return them in an arbitrary
2639 order. The new test suite's fake resolver always returns records in the
2642 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2643 bounces) except when queue_only is set, to avoid logging races between
2644 the different processes.
2646 PH/38 Panic-die if .include specifies a non-absolute path.
2648 PH/39 A tweak to the "H" retry rule from its user.
2650 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2651 a label. They prevented compilation on older perls.
2653 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2654 a warning to be raised on newish perls.
2656 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2657 on queue. Changes to match documented behaviour of showing count of
2658 messages matching specified criteria.
2660 PH/40 Changed the default ident timeout from 30s to 5s.
2662 PH/41 Added support for the use of login_cap features, on those BSD systems
2663 that have them, for controlling the resources used by pipe deliveries.
2665 PH/42 The content-scanning code uses fopen() to create files in which to put
2666 message data. Previously it was not paying any attention to the mode of
2667 the files. Exim runs with umask(0) because the rest of the code creates
2668 files with open(), and sets the required mode explicitly. Thus, these
2669 files were ending up world-writeable. This was not a big issue, because,
2670 being within the spool directory, they were not world-accessible. I have
2671 created a function called modefopen, which takes an additional mode
2672 argument. It sets umask(777), creates the file, chmods it to the required
2673 mode, then resets the umask. All the relevant calls to fopen() in the
2674 content scanning code have been changed to use this function.
2676 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2677 to 24 hours. This avoids potential overflow problems when processing G
2678 and H retry rules. I suspect nobody ever tinkers with this value.
2680 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2682 PH/45 When the plaintext authenticator is running as a client, the server's
2683 challenges are checked to ensure they are valid base64 strings. By
2684 default, the authentication attempt is cancelled if an invalid string is
2685 received. Setting client_ignore_invalid_base64 true ignores these errors.
2686 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2687 they are received. Thus, the responses can be made to depend on the
2688 challenges. If an invalid string is ignored, an empty string is placed in
2691 PH/46 Messages that are created by the autoreply transport now contains a
2692 References: header, in accordance with RFCs 2822 and 3834.
2694 PH/47 Added authenticated_sender_force to the smtp transport.
2696 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2698 PH/49 Installed latest patch from the Sieve maintainer.
2700 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2701 was also set, the appendfile transport was unnecessarily scanning a
2702 directory of message files (e.g. for maildir delivery) to find the count
2703 of files (along with the size), even though it did not need this
2704 information. It now does the scan only if it needs to find either the
2705 size of the count of files.
2707 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2709 PH/52 Two bugs concerned with error handling when the smtp transport is
2712 (i) Exim was not creating retry information for temporary errors given
2713 for individual recipients after the DATA command when the smtp transport
2714 was used in LMTP mode. This meant that they could be retried too
2715 frequently, and not timed out correctly.
2717 (ii) Exim was setting the flag that allows error details to be returned
2718 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2719 recipients that were returned after the DATA command.
2721 PH/53 This is related to PH/52, but is more general: for any failing address,
2722 when detailed error information was permitted to be returned to the
2723 sender, but the error was temporary, then after the final timeout, only
2724 "retry timeout exceeded" was returned. Now it returns the full error as
2725 well as "retry timeout exceeded".
2727 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2728 do this, and (what is worse) MTAs that accept it.
2730 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2731 will now be deprecated.
2733 PH/56 New os.c-cygwin from the Cygwin maintainer.
2735 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2736 formats (previously only available in exim formats via -bpr, -bpru,
2737 and -bpra. Now also available in native and exiqgrep formats)
2739 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2740 with very large, slow to parse queues
2742 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2744 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2746 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2747 responses to authentication challenges, though it was showing the
2748 challenges; (ii) I've removed the CR characters from the debug output for
2751 PH/58 Allow for the insertion of a newline as well as a space when a string
2752 is turned into more than one encoded-word during RFC 2047 encoding. The
2753 Sieve code now uses this.
2755 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2756 data_4xx, lost_connection, tls_required.
2758 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2759 message was being sent as an SMTP response.
2761 PH/61 Add -l and -k options to exicyclog.
2763 PH/62 When verifying, if an address was redirected to one new address, so that
2764 verification continued, and the new address failed or deferred after
2765 having set something in $address_data, the value of $address_data was not
2766 passed back to the ACL. This was different to the case when no
2767 redirection occurred. The value is now passed back in both cases.
2769 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2770 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2771 use login_cap.h, so on its own it isn't the distinguishing feature. The
2772 new name refers directly to the setclassresources() function.
2774 PH/65 Added configuration files for NetBSD3.
2776 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2778 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2779 is preferred over IPv4.
2781 PH/68 The bounce_return_message and bounce_return_body options were not being
2782 honoured for bounces generated during the reception of non-SMTP messages.
2783 In particular, this applied to messages rejected by the ACL. This bug has
2784 been fixed. However, if bounce_return_message is true and bounce_return_
2785 body is false, the headers that are returned for a non-SMTP message
2786 include only those that have been read before the error was detected.
2787 (In the case of an ACL rejection, they have all been read.)
2789 PH/69 The HTML version of the specification is now built in a directory called
2790 spec_html instead of spec.html, because the latter looks like a path with
2791 a MIME-type, and this confuses some software.
2793 PH/70 Catch two compiler warnings in sieve.c.
2795 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2796 function verify_get_ident() calls ip_connect() to connect a socket, but
2797 if the "connect()" function timed out, ip_connect() used to close the
2798 socket. However, verify_get_ident() also closes the socket later, and in
2799 between Exim writes to the log, which may get opened at this point. When
2800 the socket was closed in ip_connect(), the log could get the same file
2801 descriptor number as the socket. This naturally causes chaos. The fix is
2802 not to close the socket in ip_connect(); the socket should be closed by
2803 the function that creates it. There was only one place in the code where
2804 this was missing, in the iplookup router, which I don't think anybody now
2805 uses, but I've fixed it anyway.
2807 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2808 well as to direct DNS lookups. Otherwise the handling of names in host
2809 lists is inconsistent and therefore confusing.
2815 PH/01 Two changes to the default runtime configuration:
2817 (1) Move the checks for relay_from_hosts and authenticated clients from
2818 after to before the (commented out) DNS black list checks.
2820 (2) Add control=submission to the relay_from_hosts and authenticated
2821 clients checks, on the grounds that messages accepted by these
2822 statements are most likely to be submissions.
2824 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2826 (1) Generate an error if the third argument for the ${prvs expansion is
2829 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2832 (3) Reset the variables that are obtained from the first argument of
2833 ${prvscheck and used in the second argument before leaving the code,
2834 because their memory is reclaimed, so using them afterwards may do
2837 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2838 one (it's much easier than Tom thought :-).
2840 (5) Because of (4), we can now allow for the use of $prvscheck_result
2841 inside the third argument.
2843 PH/03 For some reason, the default setting of PATH when running a command from
2844 a pipe transport was just "/usr/bin". I have changed it to
2847 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2848 anything to be listed in the output from -bV.
2850 PH/05 When a filter generated an autoreply, the entire To: header line was
2851 quoted in the delivery log line, like this:
2853 => >A.N.Other <ano@some.domain> <original@ddress> ...
2855 This has been changed so that it extracts the operative address. There
2856 may be more than one such address. If so, they are comma-separated, like
2859 => >ano@some.domain,ona@other.domain <original@ddress> ...
2861 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2862 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2863 not being looked up in the rDNS to get a host name, Exim was showing the
2864 IP address twice in Received: lines, even though the IP addresses were
2865 identical. For example:
2867 Received: from [1.2.3.4] (helo=[1.2.3.4])
2869 However, if the real host name was known, it was omitting the HELO data
2870 if it matched the actual IP address. This has been tidied up so that it
2871 doesn't show the same IP address twice.
2873 PH/07 When both +timestamp and +memory debugging was on, the value given by
2874 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2875 called by the debug printing, thereby overwriting the timestamp buffer.
2876 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2878 PH/08 When the original message was included in an autoreply transport, it
2879 always said "this is a copy of the message, including all the headers",
2880 even if body_only or headers_only was set. It now gives an appropriate
2883 PH/09 Applied a patch from the Sieve maintainer which:
2885 o fixes some comments
2886 o adds the (disabled) notify extension core
2887 o adds some debug output for the result of if/elsif tests
2888 o points to the current vacation draft in the documentation
2889 and documents the missing references header update
2893 o fixes a bug in processing the envelope test (when testing
2894 multiple envelope elements, the last element determinted the
2897 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2898 Electronic Mail") by including:
2900 Auto-submitted: auto-generated
2902 in the messages that it generates (bounce messages and others, such as
2903 warnings). In the case of bounce messages for non-SMTP mesages, there was
2904 also a typo: it was using "Auto_submitted" (underscore instead of
2905 hyphen). Since every message generated by Exim is necessarily in response
2906 to another message, thes have all been changed to:
2908 Auto-Submitted: auto-replied
2910 in accordance with these statements in the RFC:
2912 The auto-replied keyword:
2914 - SHOULD be used on messages sent in direct response to another
2915 message by an automatic process,
2917 - MUST NOT be used on manually-generated messages,
2919 - MAY be used on Delivery Status Notifications (DSNs) and Message
2920 Disposition Notifications (MDNs),
2922 - MUST NOT be used on messages generated by automatic or periodic
2923 processes, except for messages which are automatic responses to
2926 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2927 to the default Received: header definition.
2929 PH/12 Added log selector acl_warn_skipped (default on).
2931 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2932 variables because (a) they are in the wrong storage pool and (b) even if
2933 they were copied, it wouldn't work properly because of the caching.
2935 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2936 checking when decoding. Apparently there are clients that generate
2937 overlong encoded strings. Why am I not surprised?
2939 PH/15 If the first argument of "${if match_address" was not empty, but did not
2940 contain an "@" character, Exim crashed. Now it writes a panic log message
2941 and treats the condition as false.
2943 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2945 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2946 extension "envelope-auth". The code is finished and in agreement with
2947 other implementations, but there is no documentation so far and in fact,
2948 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2949 not changing the active code.
2951 Print executed "if" and "elsif" statements when debugging is used. This
2952 helps a great deal to understand what a filter does.
2954 Document more things not specified clearly in RFC3028. I had all this
2955 sorted out, when out of a sudden new issues came to my mind. Oops."
2957 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2960 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2961 canonical form (as documented). However, after a host name lookup from
2962 the IP address, check_host() was doing a simple string comparison with
2963 addresses acquired from the DNS when checking that the found name did
2964 have the original IP as one of its addresses. Since any found IPv6
2965 addresses are likely to be in abbreviated form, the comparison could
2966 fail. Luckily, there already exists a function for doing the comparison
2967 by converting both addresses to binary, so now that is used instead of
2968 the text comparison.
2970 PH/20 There was another similar case to PH/19, when a complete host name was
2971 given in a host list; looking up its IP address could give an abbreviated
2972 form, whereas the current host's name might or might not be abbreviated.
2973 The same fix has been applied.
2979 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2980 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2983 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2984 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2986 PH/03 Typo: missing ".o" in src/pcre/Makefile.
2988 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2989 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2990 Also, for "Auto-Submitted", treat anything other than "no" as
2991 non-personal, in accordance with RFC 3834. (Previously it treated
2992 anything starting "auto-" as non-personal.)
2994 TF/01 The control=submission/name=... option had a problem with syntax
2995 errors if the name included a slash character. The /name= option
2996 now slurps the rest of the string, so it can include any characters
2997 but it must come last in the list of options (after /sender_retain
3000 PH/05 Some modifications to the interface to the fake nameserver for the new
3008 TK/01 Added the "success_on_redirect" address verification option. See
3009 NewStuff for rationale and an example.
3011 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3013 PH/02 Patch to exigrep to allow it to work on syslog lines.
3015 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3016 fread() to skip over the body file's header line, because in Cygwin the
3017 header line is locked and is inaccessible.
3019 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3020 co-exist for some time) to make it clear that it is the Exim ID that is
3021 referenced, not the Message-ID: header line.
3023 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3024 string_format() function, because snprintf() does not exist on all
3027 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3028 ${stat: expansion item.
3030 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3031 protocol synchronization error", to keep the pedants happy.
3033 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3034 well as for IRIX systems, when gcc is being used. See the host.c source
3037 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3039 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3042 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3043 transport and to the smtp transport in LMTP mode.
3045 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3047 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3048 run a filter in a subprocess. This could lead to confusion in subsequent
3049 lookups in the parent process. There should also be a search_tidyup() at
3050 the end of the subprocess.
3052 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3053 only if the host matched helo_try_verify_hosts, which caused the
3054 verification to occur when the EHLO/HELO command was issued. The ACL just
3055 tested the remembered result. Now, if a previous verification attempt has
3056 not happened, "verify = helo" does it there and then.
3058 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3060 TK/03 Fix log output including CR from clamd.
3062 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3063 exist provoked a memory error which could cause a segfault.
3065 PH/15 Installed PCRE 6.2
3067 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3069 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3070 of the problem. Specifically, suggested +O2 rather than +O1 for the
3073 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3075 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3076 record, the port was not correctly logged when the outgoing_port log
3077 selector was set (it logged the transort's default port).
3079 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3080 fallback_hosts, and "hosts" in the smtp transport.
3082 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3083 host errors such as "Connection refused".
3085 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3086 authentication with radiusclient 0.4.9:
3088 - Error returned from rc_read_config was caught wrongly
3089 - Username/password not passed on to radius server due to wrong length.
3091 The presumption is that some radiusclient API changes for 4.51/PH/17
3092 were not taken care of correctly. The code is still untested by me (my
3093 Linux distribution still has 0.3.2 of radiusclient), but it was
3094 contributed by a Radius user.
3096 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3097 expanding the "port" option of the smtp transport.
3099 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3100 while reading a MIME header. Thanks to Tom Hughes for a patch.
3102 PH/24 Include config.h inside local_scan.h so that configuration settings are
3105 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3106 that in an ACL for RCPT (for example), you can examine exactly what was
3109 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3110 commands, but it was not correctly comparing the address with the actual
3111 client host address. Thus, it would show the EHLO address in Received:
3112 header lines when this was not necessary.
3114 PH/27 Added the % operator to ${eval:}.
3116 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3117 it should be ignoring failures (because with -C, for example, it has lost
3118 privilege). It wasn't ignoring creation failures other than "already
3121 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3124 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3125 because an input line was too long, either on its own, or by virtue of
3126 too many continuations, the temporary file was not being removed, and the
3127 return code was incorrect.
3129 PH/31 Missing "BOOL" in function definition in filtertest.c.
3131 PH/32 Applied Sieve patches from the maintainer.
3133 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3135 PH/33 Added "verify = not_blind".
3137 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3138 Local/Makefile (with some defaults set). These are used in built scripts
3139 such as exicyclog, but they have never been used in the exim_install
3140 script (though there are many overriding facilities there). I have
3141 arranged that the exim_install script now takes note of these two
3144 PH/35 Installed configuration files for Dragonfly.
3146 PH/36 When a locally submitted message by a trusted user did not contain a
3147 From: header, and the sender address was obtained from -f or from an SMTP
3148 MAIL command, and the trusted user did not use -F to supply a sender
3149 name, $originator_name was incorrectly used when constructing a From:
3150 header. Furthermore, $originator_name was used for submission mode
3151 messages from external hosts without From: headers in a similar way,
3152 which is clearly wrong.
3154 PH/37 Added control=suppress_local_fixups.
3156 PH/38 When log_selector = +received_sender was set, and the addition of the
3157 sender made the log line's construction buffer exactly full, or one byte
3158 less than full, an overflow happened when the terminating "\n" was
3161 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3162 when the result of a list match is failure because a DNS lookup failed.
3164 PH/40 RM_COMMAND is now used in the building process.
3166 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3167 the "build-* directories that it finds.
3169 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3170 address in a domain literal was a prefix of an interface address.
3172 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3173 when verifying a sender address, unless rewrite_headers is false.
3175 PH/44 Wrote a long comment about why errors_to addresses are verified as
3176 recipients, not senders.
3178 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3179 the ratelimit ACL was added.
3181 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3183 PH/46 Added extra information about PostgreSQL errors to the error string.
3185 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3186 avoiding the need to install special zones in a real server. This is
3187 backwards compatible; if it can't find the fake resolver, it drops back.
3188 Thus, both old and new test suites can be run.
3190 TF/02 Added util/ratelimit.pl
3192 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3193 clock is set back in time.
3195 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3196 Candler <B.Candler@pobox.com>.
3198 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3199 OK for addresses that are the result of redirections.
3201 PH/48 A number of further additions for the benefit of the new test suite,
3202 including a fake gethostbyname() that interfaces to the fake DNS resolver
3205 TF/06 The fix for widen_domains has also been applied to qualify_single and
3206 search_parents which are the other dnslookup options that can cause
3209 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3212 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3214 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3215 service" if it encountered an MX record with an empty target hostname.
3216 The message is now "an MX or SRV record indicated no SMTP service".
3218 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3219 if the DNS of the sending site is misconfigured. This is quite a
3220 common situation. This change restores the behaviour of treating a
3221 helo verification defer as a failure.
3223 PH/51 If self=fail was set on a router, the bounce message did not include the
3224 actual error message.
3230 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3232 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3233 command itself ended in error, the underlying message about the transport
3234 filter timeout was being overwritten with the pipe command error. Now the
3235 underlying error message should be appended to the second error message.
3237 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3239 PH/02 Reduce the amount of output that "make" produces by default. Full output
3240 can still be requested.
3242 PH/03 The warning log line about a condition test deferring for a "warn" verb
3243 was being output only once per connection, rather than after each
3244 occurrence (because it was using the same function as for successful
3245 "warn" verbs). This seems wrong, so I have changed it.
3247 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3248 it should not have, which might have caused a crash in the right
3249 circumstances, but probably never did.
3251 PH/04 Installed a modified version of Tony Finch's patch to make submission
3252 mode fix the return path as well as the Sender: header line, and to
3253 add a /name= option so that you can make the user's friendly name appear
3256 TF/03 Added the control = fakedefer ACL modifier.
3258 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3259 Mark Lowes for thorough testing.
3261 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3263 TK/03 Merged latest SRS patch from Miles Wilton.
3265 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3266 with the definition in sysexits.h (which is #included earlier).
3267 Fortunately, Exim does not actually use EX_OK. The code used to try to
3268 preserve the sysexits.h value, by assumimg that macro definitions were
3269 scanned for macro replacements. I have been disabused of this notion,
3270 so now the code just undefines EX_OK before #including unistd.h.
3272 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3273 in the smtp transport. When a block could not be written in a single
3274 write() function, the timeout was being re-applied to each part-write.
3275 This seems wrong - if the receiver was accepting one byte at a time it
3276 would take for ever. The timeout is now adjusted when this happens. It
3277 doesn't have to be particularly precise.
3279 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3280 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3282 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3283 without a domain if the check to <postmaster@domain> fails.
3285 SC/01 Eximstats: added -xls and the ability to specify output files
3286 (patch written by Frank Heydlauf).
3288 SC/02 Eximstats: use FileHandles for outputing results.
3290 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3292 SC/04 Eximstats: fixed display of large numbers with -nvr option
3294 SC/05 Eximstats: fixed merging of reports with empty tables.
3296 SC/06 Eximstats: added the -include_original_destination flag
3298 SC/07 Eximstats: removed tabs and trailing whitespace.
3300 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3302 TK/06 MBOX spool code: Add real "From " MBOX separator line
3303 so the .eml file is really in mbox format (even though
3304 most programs do not really care). Patch from Alex Miller.
3306 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3307 The latter is generated from $received_to and is only set if the
3308 message has one envelope recipient. SA can use these headers,
3309 obviously out-of-the-box. Patch from Alex Miller.
3311 PH/08 The ${def test on a variable was returning false if the variable's
3312 value was "0", contrary to what the specification has always said!
3313 The result should be true unless the variable is empty.
3315 PH/09 The syntax error of a character other than { following "${if
3316 def:variable_name" (after optional whitespace) was not being diagnosed.
3317 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3318 accidental colon was present, for example, could give incorrect results.
3320 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3321 result is used (not including appendfile, where other changes are about
3324 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3325 This involved changing a lot of size variables from int to off_t. It
3326 should work with maildirs and everything.
3328 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3329 spamd dying while we are connected to it.
3331 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3334 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3335 function for BDB 4.3.
3337 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3339 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3340 the sources for PCRE have been re-arranged and more files are now
3343 PH/15 The code I had for printing potentially long long variables in PH/11
3344 above was not the best (it lost precision). The length of off_t variables
3345 is now inspected at build time, and an appropriate printing format (%ld
3346 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3347 to be "long long int" or "long int". This is needed for the internal
3348 formatting function string_vformat().
3350 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3351 the configuration file to be ":syslog", then the script "guesses" where
3352 the logs files are, rather than using the compiled in default. In our
3353 case the guess is not the same as the compiled default, so the script
3354 suddenly stopped working when I started to use syslog. The patch checks
3355 to see if log_file_path is "". If so, it attempts to read it from exim
3356 with no configuration file to get the compiled in version, before it
3357 falls back to the previous guessing code."
3359 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3360 implementing BATV in an Exim configuration. See NewStuff for the gory
3363 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3364 Makefile that are specific to HP-UX.
3366 PH/18 If the "use_postmaster" option was set for a recipient callout together
3367 with the "random" option, the postmaster address was used as the MAIL
3368 FROM address for the random test, but not for the subsequent recipient
3369 test. It is now used for both.
3371 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3372 patch removes a few documentation additions to RFC 3028, because the
3373 latest draft now contains them. It adds the new en;ascii-case comparator
3374 and a new error check for 8bit text in MIME parts. Comparator and
3375 require names are now matched exactly. I enabled the subaddress
3376 extension, but it is not well tested yet (read: it works for me)."
3378 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3379 rework some of the code of TK/09 above to avoid the hardwired use of
3380 "%lld" and "long long". Replaced the call to snprintf() with a call to
3383 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3384 records point to non-existent hosts", "retry timeout exceeded", and
3385 "retry time not reached for any host after a long failure period".
3387 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3388 experimental DomainKeys support:
3390 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3391 (2) On an error such as an illegally used "control", the wrong name for
3392 the control was given.
3394 These problems did NOT occur unless DomainKeys support was compiled.
3396 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3398 PH/24 Added ${if match_ip condition.
3400 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3401 databases so that it will be absolutely obvious if a crash occurs in the
3402 DB library. This is a regular occurrence (often caused by mis-matched
3405 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3406 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3407 picked up on a user's system that detects such things. There doesn't seem
3408 to be a gcc warning option for this - only an attribute that has to be
3409 put on the function's prototype. It seems that in Fedora Core 4 they have
3410 set this on a number of new functions. No doubt there will be more in due
3413 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3414 specify a transport. However, if an address that was verified by such a
3415 router was the subject of a callout, Exim crashed because it tried to
3416 read the rcpt_include_affixes from the non-existent transport. Now it
3417 just assumes that the setting of that option is false. This bug was
3418 introduced by 4.51/PH/31.
3420 PH/28 Changed -d+all to exclude +memory, because that information is very
3421 rarely of interest, but it makes the output a lot bigger. People tend to
3422 do -d+all out of habit.
3424 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3425 code in os-type was giving problems when libc.so lives in lib64, like on
3428 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3429 aren't the modern standard, and it seems that some systems' include files
3430 don't always have them. Exim was already checking for some of the newer
3431 ones like T_AAAA, and defining it itself. I've added checks for all the
3432 record types that Exim uses.
3434 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3435 not automatically generating a new one, as it is supposed to. This
3436 prevented TLS from working. If the file did exist, but contained invalid
3437 data, a new version was generated, as expected. It was only the case of a
3438 non-existent file that was broken.
3440 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3441 with a change in libdomainkeys > 0.64.
3443 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3444 from DNS. If the selector record carries the flag, it now has
3445 precedence over the domain-wide flag.
3447 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3449 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3450 the use of a port name that isn't defined in /etc/services) occurred, the
3451 message was deferred as in a normal delivery, and thus remained on the
3452 spool, instead of being failed because of the mua_wrapper setting. This
3453 is now fixed, and I tidied up some of the mua_wrapper messages at the
3456 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3457 the messages in a hash of arrays rather than using individual hashes.
3458 This is a bit cleaner and results in dramatic memory savings, albeit
3459 at a slight CPU cost.
3461 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3462 as requested by Marc Sherman.
3464 SC/10 Eximstats: added histograms for user specified patterns as requested
3467 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3469 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3470 fopen() in the content-scanning modules that did not already have it.
3476 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3477 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3479 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3481 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3483 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3484 to test Sieve filters that use "vacation".
3486 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3487 that changes the way the GnuTLS parameters are stored in the cache file.
3488 The new format can be generated externally. For backward compatibility,
3489 if the data in the cache doesn't make sense, Exim assumes it has read an
3490 old-format file, and it generates new data and writes a new file. This
3491 means that you can't go back to an older release without removing the
3494 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3495 work if there are any delivery delays because "one_time" forces the
3496 parent to be marked "delivered", so its unseen clone is never tried
3497 again. For this reason, Exim now forbids the simultaneous setting of
3500 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3501 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3502 entry if you want to know the details. The fix, however, neglected to
3503 consider the case when local delivery batching is involved. The test for
3504 "previously delivered" was not happening when checking to see if an
3505 address could be batched with a previous (undelivered) one; under
3506 certain circumstances this could lead to multiple deliveries to the same
3509 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3510 in its include files, and this causes problems building Exim.
3512 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3513 header_syntax) but Exim was just ignoring anything given after a slash.
3514 In particular, this caused confusion with an attempt to use "verify =
3515 reverse_host_lookup/defer_ok". An error is now given when options are
3516 supplied for verify items that do not have them. (Maybe reverse_host_
3517 lookup should have a defer_ok option, but that's a different point.)
3519 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3520 defined by RFC 821) to 2048, because there were problems with some AUTH
3521 commands, and RFC 1869 says the size should be increased for extended
3522 SMTP commands that take arguments.
3524 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3527 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3528 "unknown" error; now it says that the functionality isn't in the binary.
3530 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3531 an address' error message when a string expansion fails (syntax or
3532 whatever). Otherwise the password may appear in the log. Following change
3533 PH/42 below, there is no longer a chance of it appearing in a bounce
3536 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3538 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3539 of its addresses was ever tried. (Bugzilla bug #2.)
3541 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3542 the result incorrectly in the debug output. (It correctly added a newline
3543 to what was transported.)
3545 TF/01 Added $received_time.
3547 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3548 commented out examples of how to interface to a virus scanner and to
3549 SpamAssassin. Also added commented examples of av_scanner and
3550 spamd_address settings.
3552 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3553 and controls are allowed in which ACLs. There were a couple of minor
3554 errors. Some of the entries in the conditions table (which is a table of
3555 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3556 negation of where the condition IS allowed.
3558 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3560 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3561 header file does not have a version number, so I've had to invent a new
3562 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3563 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3564 radiusclient), but it was contributed by a Radius user.
3566 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3567 files or directories, for OpenSSL.
3569 PH/19 When an Exim process that is running as root has to create an Exim log
3570 file, it does so in a subprocess that runs as exim:exim so as to get the
3571 ownership right at creation (otherwise, other Exim processes might see
3572 the file with the wrong ownership). There was no test for failure of this
3573 fork() call, which would lead to the process getting stuck as it waited
3574 for a non-existent subprocess. Forks do occasionally fail when resources
3575 run out. I reviewed all the other calls to fork(); they all seem to check
3578 PH/20 When checking for unexpected SMTP input at connect time (before writing
3579 the banner), Exim was not dealing correctly with a non-positive return
3580 from the read() function. If the client had disconnected by this time,
3581 the result was a log entry for a synchronization error with an empty
3582 string after "input=" when read() returned zero. If read() returned -1
3583 (an event I could not check), uninitialized data bytes were printed.
3584 There were reports of junk text (parts of files, etc) appearing after
3587 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3589 PH/22 Added support for macro redefinition, and (re)definition in between
3590 driver and ACL definitions.
3592 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3593 forgetting to use the resulting value; it was using the unexpanded value.
3595 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3596 hadn't been configured. The fix is from Juergen Kreileder, who
3597 understands it better than I do:
3599 "Here's what I see happening with three configured cyrus_sasl
3600 authenticators configured (plain, login, cram-md5):
3602 On startup auth_cyrus_sasl_init() gets called for each of these.
3603 This means three calls to sasl_listmech() without a specified mech_list.
3604 => SASL tests which mechs of all available mechs actually work
3605 => three warnings about OTP not working
3606 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3608 With the patch, sasl_listmech() also gets called three times. But now
3609 SASL's mech_list option is set to the server_mech specified in the the
3610 authenticator. Or in other words, the answer from sasl_listmech()
3611 gets limited to just the mech you're testing for (which is different
3613 => the return list contains just 'plain' or 'login', 'cram-md5' or
3614 nothing depending on the value of ob->server_mech.
3616 I've just tested the patch: Authentication still works fine,
3617 unavailable mechs specified in the exim configuration are still
3618 caught, and the auth.log warnings about OTP are gone."
3620 PH/25 When debugging is enabled, the contents of the command line are added
3621 to the debugging output, even when log_selector=+arguments is not
3624 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3625 answer is "GNU", and only if the return is "GNU/something" is the answer
3628 PH/27 $acl_verify_message is now set immediately after the failure of a
3629 verification in an ACL, and so is available in subsequent modifiers. In
3630 particular, the message can be preserved by coding like this:
3632 warn !verify = sender
3633 set acl_m0 = $acl_verify_message
3635 Previously, $acl_verify_message was set only while expanding "message"
3636 and "log_message" when a very denied access.
3638 PH/28 Modified OS/os.c-Linux with
3640 -#ifndef OS_LOAD_AVERAGE
3641 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3643 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3644 nomenclature these days.)
3646 PH/29 Installed patch from the Sieve maintainer that adds the options
3647 sieve_useraddress and sieve_subaddress to the redirect router.
3649 PH/30 In these circumstances:
3650 . Two addresses routed to the same list of hosts;
3651 . First host does not offer TLS;
3652 . First host accepts first address;
3653 . First host gives temporary error to second address;
3654 . Second host offers TLS and a TLS session is established;
3655 . Second host accepts second address.
3656 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3657 and peerdn, if requested) that were in fact used only for the second
3660 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3661 attention to any local part prefix or suffix that was matched by the
3662 router that accepted the address. It now behaves in the same way as it
3663 does for delivery: the affixes are removed from the local part unless
3664 rcpt_include_affixes is set on the transport.
3666 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3667 timeout during the DATA phase of an incoming message.
3669 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3670 applied a patch sanctioned by the Sieve maintainer.
3672 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3673 the uid or gid is negative. A case of a negative gid caused this to be
3674 noticed. The fix allows for either to be negative.
3676 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3677 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3680 PH/36 The change PH/12 above was broken. Fixed it.
3682 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3683 the grounds that routing the same address twice would always produce the
3684 same answer. This might have been true once, but it is certainly no
3685 longer true now. Routing a child address may depend on the previous
3686 routing that produced that child. Some complicated redirection strategies
3687 went wrong when messages had multiple recipients, and made Exim's
3688 behaviour dependent on the order in which the addresses were given.
3690 I have moved the duplicate checking until after the routing is complete.
3691 Exim scans the addresses that are assigned to local and remote
3692 transports, and removes any duplicates. This means that more work will be
3693 done, as duplicates will always all be routed, but duplicates are
3694 presumably rare, so I don't expect this is of any significance.
3696 For deliveries to pipes, files, and autoreplies, the duplicate checking
3697 still happens during the routing process, since they are not going to be
3700 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3701 It corrects a timeout issue with spamd. This is Ian's comment: "The
3702 background is that sometimes spamd either never reads data from a
3703 connection it has accepted, or it never writes response data. The exiscan
3704 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3705 blindly assumes that writes won't block so it may never time out."
3707 PH/39 Allow G after quota size as well as K and M.
3709 PH/40 The value set for $authenticated_id in an authenticator may not contain
3710 binary zeroes or newlines because the value is written to log lines and
3711 to spool files. There was no check on this. Now the value is run through
3712 the string_printing() function so that such characters are converted to
3713 printable escape sequences.
3715 PH/41 $message_linecount is a new variable that contains the total number of
3716 lines in the message. Compare $body_linecount, which is the count for the
3719 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3720 bounce and delay warning messages, except in certain special cases, which
3723 (a) An SMTP error message from a remote host;
3724 (b) A message specified in a :fail: redirection;
3725 (c) A message specified in a "fail" command in a system filter;
3726 (d) A message specified in a FAIL return from the queryprogram router;
3727 (e) A message specified by the cannot_route_message router option.
3729 In these cases only, Exim does include the error details in bounce and
3730 warning messages. There are also a few cases where bland messages such
3731 as "unrouteable address" or "local delivery error" are given.
3733 PH/43 $value is now also set for the "else" part of a ${run expansion.
3735 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3736 being worked on, but at least Exim now implements the latest version to
3739 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3740 process to complete was treated as a delivery failure, a timeout while
3741 writing the message to the pipe was logged, but erroneously treated as a
3742 successful delivery. Such timeouts include transport filter timeouts. For
3743 consistency with the overall process timeout, these timeouts are now
3744 treated as errors, giving rise to delivery failures by default. However,
3745 there is now a new Boolean option for the pipe transport called
3746 timeout_defer, which, if set TRUE, converts the failures into defers for
3747 both kinds of timeout. A transport filter timeout is now identified in
3750 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3751 systems where "make" and "gmake" are different, calling "gmake" at top
3752 level broke things. I've arranged for the value of $(MAKE) to be passed
3753 from the Makefile to this script so that it can call the same version of
3757 A note about Exim versions 4.44 and 4.50
3758 ----------------------------------------
3760 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3761 changes of various kinds. As a consequence, a big documentation update was
3762 needed. This delayed the release for rather longer than seemed good, especially
3763 in the light of a couple of (minor) security issues. Therefore, the changes
3764 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3765 release. So 4.44 and 4.50 are in effect two different branches that both start
3768 I have left the 4.50 change log unchanged; it contains all the changes since
3769 4.43. The change log for 4.44 is below; many of its items are identical to
3770 those for 4.50. This seems to be the most sensible way to preserve the
3771 historical information.
3777 1. Minor wording change to the doc/README.SIEVE file.
3779 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3780 computation of the current number of files was incorrect.
3782 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3783 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3786 4. Give more explanation in the error message when the command for a transport
3787 filter fails to execute.
3789 5. There are several places where Exim runs a non-Exim command in a
3790 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3791 was being done only for the command run by the queryprogram router. It is
3792 now done for all such subprocesses. The other cases are: ${run, transport
3793 filters, and the commands run by the lmtp and pipe transports.
3795 6. Added CONFIGURE_GROUP build-time option.
3797 7. Some older OS have a limit of 256 on the maximum number of file
3798 descriptors. Exim was using setrlimit() to set 1000 as a large value
3799 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3800 systems. I've change it so that if it can't get 1000, it tries for 256.
3802 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3803 was an oversight, and furthermore, ever since the addition of extra
3804 controls (e.g. 4.43/32), the checks on when to allow different forms of
3805 "control" were broken. There should now be diagnostics for all cases when a
3806 control that does not make sense is encountered.
3808 9. Added the /retain_sender option to "control=submission".
3810 10. $recipients is now available in the predata ACL (oversight).
3812 11. Tidy the search cache before the fork to do a delivery from a message
3813 received from the command line. Otherwise the child will trigger a lookup
3814 failure and thereby defer the delivery if it tries to use (for example) a
3815 cached ldap connection that the parent has called unbind on.
3817 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3818 of $address_data from the recipient verification was clobbered by the
3819 sender verification.
3821 13. The value of address_data from a sender verification is now available in
3822 $sender_address_data in subsequent conditions in the ACL statement.
3824 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3826 15. Added a new option "connect=<time>" to callout options, to set a different
3829 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3830 was its contents. (It was OK if the option was not defined at all.)
3832 17. A "Completed" log line is now written for messages that are removed from
3833 the spool by the -Mrm option.
3835 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3836 information about exactly what failed.
3838 19. Added -dd to debug only the daemon process.
3840 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3841 handles timeouts, both on the server side and network timeouts. Renamed the
3842 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3844 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3845 It is now set to "smtps".
3847 22. $host_address is now set to the target address during the checking of
3848 ignore_target_hosts.
3850 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3851 being passed; this would have caused $sender_host_name to have been used if
3852 matching the list had actually called for a host name (not very likely,
3853 since this list is usually IP addresses). A host name is now passed as
3856 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3857 code that specifies a non-restarting handler (typically sigaction() in
3858 modern systems) in an attempt to fix a rare and obscure crash bug.
3860 25. Narrowed the window for a race in the daemon that could cause it to ignore
3861 SIGCHLD signals. This is not a major problem, because they are used only to
3862 wake it up if nothing else does.
3864 26. A malformed maildirsize file could cause Exim to calculate negative values
3865 for the mailbox size or file count. Odd effects could occur as a result.
3866 The maildirsize information is now recalculated if the size or filecount
3869 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3870 support for a long time. Removed HAVE_SYS_VFS_H.
3872 28. Installed the latest version of exipick from John Jetmore.
3874 29. In an address list, if the pattern was not a regular expression, an empty
3875 subject address (from a bounce message) matched only if the pattern was an
3876 empty string. Non-empty patterns were not even tested. This was the wrong
3877 because it is perfectly reasonable to use an empty address as part of a
3878 database query. An empty address is now tested by patterns that are
3879 lookups. However, all the other forms of pattern expect the subject to
3880 contain a local part and a domain, and therefore, for them, an empty
3881 address still always fails if the pattern is not itself empty.
3883 30. Exim went into a mad DNS loop when attempting to do a callout where the
3884 host was specified on an smtp transport, and looking it up yielded more
3885 than one IP address.
3887 31. Re-factored the code for checking spool and log partition space into a
3888 function that finds that data and another that does the check. The former
3889 is then used to implement four new variables: $spool_space, $log_space,
3890 $spool_inodes, and $log_inodes.
3892 32. The RFC2047 encoding function was originally intended for short strings
3893 such as real names; it was not keeping to the 75-character limit for
3894 encoded words that the RFC imposes. It now respects the limit, and
3895 generates multiple encoded words if necessary. To be on the safe side, I
3896 have increased the buffer size for the ${rfc2047: expansion operator from
3899 33. It is now permitted to omit both strings after an "if" condition; if the
3900 condition is true, the result is "true". As before, when the second string
3901 is omitted, a false condition yields an empty string. This makes it less
3902 cumbersome to write custom ACL and router conditions.
3904 34. Failure to deliver a bounce message always caused it to be frozen, even if
3905 there was an errors_to setting on the router. The errors_to setting is now
3908 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3909 canonical form (fully expanded) before being placed in
3910 $sender_host_address.
3912 36. The table in the code that translates DNS record types into text (T_A to
3913 "A" for instance) was missing entries for NS and CNAME. It is just possible
3914 that this could have caused confusion if both these types were looked up
3915 for the same domain, because the text type is used as part of Exim's
3916 per-process caching. But the chance of anyone hitting this buglet seems
3919 37. The dnsdb lookup has been extended in a number of ways.
3921 (1) There is a new type, "zns", which walks up the domain tree until it
3922 finds some nameserver records. It should be used with care.
3924 (2) There is a new type, "mxh", which is like "mx" except that it returns
3925 just the host names, not the priorities.
3927 (3) It is now possible to give a list of domains (or IP addresses) to be
3928 looked up. The behaviour when one of the lookups defers can be
3929 controlled by a keyword.
3931 (4) It is now possible to specify the separator character for use when
3932 multiple records are returned.
3934 38. The dnslists ACL condition has been extended: it is now possible to supply
3935 a list of IP addresses and/or domains to be looked up in a particular DNS
3938 39. Added log_selector=+queue_time_overall.
3940 40. When running the queue in the test harness, wait just a tad after forking a
3941 delivery process, to get repeatability of debugging output.
3943 41. Include certificate and key file names in error message when GnuTLS fails
3944 to set them up, because the GnuTLS error message doesn't include the name
3945 of the failing file when there is a problem reading it.
3947 42. Allow both -bf and -bF in the same test run.
3949 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3951 44. The "Exiscan patch" is now merged into the mainline Exim source.
3953 45. Sometimes the final signoff response after QUIT could fail to get
3954 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3955 < 0 before doing a fflush(). This bug looks as though it goes back to the
3956 introduction of TLS in release 3.20, but "sometimes" must have been rare
3957 because the tests only now provoked it.
3959 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3960 (this can affect the format of dates).
3962 47. exim_tidydb, when checking for the continued existence of a message for
3963 which it has found a message-specific retry record, was not finding
3964 messages that were in split spool directories. Consequently, it was
3965 deleting retry records that should have stayed in existence.
3967 48. Steve fixed some bugs in eximstats.
3969 49. The SPA authentication driver was not abandoning authentication and moving
3970 on to the next authenticator when an expansion was forced to fail,
3971 contradicting the general specification for all authenticators. Instead it
3972 was generating a temporary error. It now behaves as specified.
3974 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3975 (the order specifies the preference for clients). The order is now AES256,
3976 AES128, 3DES, ARCFOUR128.
3978 51. Small patch to Sieve code - explicitly set From: when generating an
3981 52. Exim crashed if a remote delivery caused a very long error message to be
3982 recorded - for instance if somebody sent an entire SpamAssassin report back
3983 as a large number of 550 error lines. This bug was coincidentally fixed by
3984 increasing the size of one of Exim's internal buffers (big_buffer) that
3985 happened as part of the Exiscan merge. However, to be on the safe side, I
3986 have made the code more robust (and fixed the comments that describe what
3989 53. Now that there can be additional text after "Completed" in log lines (if
3990 the queue_time_overall log selector is set), a one-byte patch to exigrep
3991 was needed to allow it to recognize "Completed" as not the last thing in
3994 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3995 patch that reportedly fixes this has been added. I am not expert enough to
3996 create a test for it. This is what the patch creator wrote:
3998 "I found a little strange behaviour of ldap code when working with
3999 Windows 2003 AD Domain, where users was placed in more than one
4000 Organization Units. When I tried to give exim partial DN, the exit code
4001 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4002 But simultaneously result of request was absolutely normal ldap result,
4003 so I produce this patch..."
4005 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4006 REFERENCE, so I have modified the code to exclude the patch when that macro
4009 55. Some experimental protocols are using DNS PTR records for new purposes. The
4010 keys for these records are domain names, not reversed IP addresses. The
4011 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4012 leaves it alone. Component reversal etc. now happens only for IP addresses.
4015 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4017 57. Double the size of the debug message buffer (to 2048) so that more of very
4018 long debug lines gets shown.
4020 58. The exicyclog utility now does better if the number of log files to keep
4021 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4023 59. Two changes related to the smtp_active_hostname option:
4025 (1) $smtp_active_hostname is now available as a variable.
4026 (2) The default for smtp_banner uses $smtp_active_hostname instead
4027 of $primary_hostname.
4029 60. The host_aton() function is supposed to be passed a string that is known
4030 to be a valid IP address. However, in the case of IPv6 addresses, it was
4031 not checking this. This is a hostage to fortune. Exim now panics and dies
4032 if the condition is not met. A case was found where this could be provoked
4033 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4034 components; fortuitously, this particular loophole had already been fixed
4035 by change 4.50/55 above.
4037 If there are any other similar loopholes, the new check in host_aton()
4038 itself should stop them being exploited. The report I received stated that
4039 data on the command line could provoke the exploit when Exim was running as
4040 exim, but did not say which command line option was involved. All I could
4041 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4042 running as the user.
4045 61. There was a buffer overflow vulnerability in the SPA authentication code
4046 (which came originally from the Samba project). I have added a test to the
4047 spa_base64_to_bits() function which I hope fixes it.
4050 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4051 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4053 63. The daemon start-up calls getloadavg() while still root for those OS that
4054 need the first call to be done as root, but it missed one case: when
4055 deliver_queue_load_max is set with deliver_drop_privilege. This is
4056 necessary for the benefit of the queue runner, because there is no re-exec
4057 when deliver_drop_privilege is set.
4059 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4060 This has been fixed.
4062 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4063 was in use, was not putting the data itself into the right store pool;
4064 consequently, it could be overwritten for a subsequent message in the same
4065 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4068 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4070 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4071 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4072 the function were treating the return as a boolean value, which happened to
4073 work because 0=false and not-0=true, but is not correct code.
4075 68. The host_aton() function was not handling scoped IPv6 addresses (those
4076 with, for example, "%eth0" on the end) correctly.
4078 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4079 negated items (that is, ~something) in unsigned ints. Some compilers
4080 apparently mutter when there is no cast.
4082 70. If an address verification called from an ACL failed, and did not produce a
4083 user-specific message (i.e. there was only a "system" message), nothing was
4084 put in $acl_verify_message. In this situation, it now puts the system
4087 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4088 session; change 4.31/43 added the unwanted input to the log line - except
4089 that it did not do this in the start of session case. It now does.
4091 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4092 This is wrong and can cause the other end to generate a synchronization
4093 error if it is another Exim or anything else that does the synchronization
4094 check. A QUIT command is no longer sent after a timeout.
4096 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4097 during host lookups.
4099 74. The defer_ok option of callout verification was not working if it was used
4100 when verifying addresses in header lines, that is, for this case:
4102 verify = header_sender/callout=defer_ok
4104 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4105 those file descriptors could be used for SMTP connections. If anything
4106 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4107 could be sent to the SMTP client, causing chaos. The daemon now opens
4108 stdin, stdout, and stderr to /dev/null when it puts itself into the
4111 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4112 log by default. The user can override this with suitable Perl magic.
4114 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4115 the log message when discard triggers, was not working for the DATA ACL or
4116 for the non-SMTP ACL.
4118 78. Error message wording change in sieve.c.
4120 79. If smtp_accept_max_per_host was set, the number of connections could be
4121 restricted to fewer than expected, because the daemon was trying to set up
4122 a new connection before checking whether the processes handling previous
4123 connections had finished. The check for completed processes is now done
4124 earlier. On busy systems, this bug wouldn't be noticed because something
4125 else would have woken the daemon, and it would have reaped the completed
4128 80. If a message was submitted locally by a user whose login name contained one
4129 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4130 It caused a spool format error. I have fixed the spool reading code. A
4131 related problem was that the "from" clause in the Received: line became
4132 illegal because of the space(s). It is now covered by ${quote_local_part.
4134 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4137 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4139 83. Patch from Sieve maintainer to fix unterminated string problem in
4140 "vacation" handling.
4142 84. Some minor changes to the Linux configuration files to help with other
4143 OS variants using glibc.
4145 85. One more patch for Sieve to update vacation handling to latest spec.
4148 ----------------------------------------------------
4149 See the note above about the 4.44 and 4.50 releases.
4150 ----------------------------------------------------
4156 1. Change 4.43/35 introduced a bug that caused file counts to be
4157 incorrectly computed when quota_filecount was set in an appendfile
4160 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4161 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4164 3. Give more explanation in the error message when the command for a transport
4165 filter fails to execute.
4167 4. There are several places where Exim runs a non-Exim command in a
4168 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4169 was being done only for the command run by the queryprogram router. It is
4170 now done for all such subprocesses. The other cases are: ${run, transport
4171 filters, and the commands run by the lmtp and pipe transports.
4173 5. Some older OS have a limit of 256 on the maximum number of file
4174 descriptors. Exim was using setrlimit() to set 1000 as a large value
4175 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4176 systems. I've change it so that if it can't get 1000, it tries for 256.
4178 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4179 was an oversight, and furthermore, ever since the addition of extra
4180 controls (e.g. 4.43/32), the checks on when to allow different forms of
4181 "control" were broken. There should now be diagnostics for all cases when a
4182 control that does not make sense is encountered.
4184 7. $recipients is now available in the predata ACL (oversight).
4186 8. Tidy the search cache before the fork to do a delivery from a message
4187 received from the command line. Otherwise the child will trigger a lookup
4188 failure and thereby defer the delivery if it tries to use (for example) a
4189 cached ldap connection that the parent has called unbind on.
4191 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4192 of $address_data from the recipient verification was clobbered by the
4193 sender verification.
4195 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4196 was its contents. (It was OK if the option was not defined at all.)
4198 11. A "Completed" log line is now written for messages that are removed from
4199 the spool by the -Mrm option.
4201 12. $host_address is now set to the target address during the checking of
4202 ignore_target_hosts.
4204 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4205 being passed; this would have caused $sender_host_name to have been used if
4206 matching the list had actually called for a host name (not very likely,
4207 since this list is usually IP addresses). A host name is now passed as
4210 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4211 code that specifies a non-restarting handler (typically sigaction() in
4212 modern systems) in an attempt to fix a rare and obscure crash bug.
4214 15. Narrowed the window for a race in the daemon that could cause it to ignore
4215 SIGCHLD signals. This is not a major problem, because they are used only to
4216 wake it up if nothing else does.
4218 16. A malformed maildirsize file could cause Exim to calculate negative values
4219 for the mailbox size or file count. Odd effects could occur as a result.
4220 The maildirsize information is now recalculated if the size or filecount
4223 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4224 support for a long time. Removed HAVE_SYS_VFS_H.
4226 18. Updated exipick to current release from John Jetmore.
4228 19. Allow an empty sender to be matched against a lookup in an address list.
4229 Previously the only cases considered were a regular expression, or an
4232 20. Exim went into a mad DNS lookup loop when doing a callout where the
4233 host was specified on the transport, if the DNS lookup yielded more than
4236 21. The RFC2047 encoding function was originally intended for short strings
4237 such as real names; it was not keeping to the 75-character limit for
4238 encoded words that the RFC imposes. It now respects the limit, and
4239 generates multiple encoded words if necessary. To be on the safe side, I
4240 have increased the buffer size for the ${rfc2047: expansion operator from
4243 22. Failure to deliver a bounce message always caused it to be frozen, even if
4244 there was an errors_to setting on the router. The errors_to setting is now
4247 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4248 canonical form (fully expanded) before being placed in
4249 $sender_host_address.
4251 24. Updated eximstats to version 1.33
4253 25. Include certificate and key file names in error message when GnuTLS fails
4254 to set them up, because the GnuTLS error message doesn't include the name
4255 of the failing file when there is a problem reading it.
4257 26. Expand error message when OpenSSL has problems setting up cert/key files.
4260 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4261 (this can affect the format of dates).
4263 28. exim_tidydb, when checking for the continued existence of a message for
4264 which it has found a message-specific retry record, was not finding
4265 messages that were in split spool directories. Consequently, it was
4266 deleting retry records that should have stayed in existence.
4268 29. eximstats updated to version 1.35
4269 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4270 1.35 - bugfix such that pie charts by volume are generated correctly
4272 30. The SPA authentication driver was not abandoning authentication and moving
4273 on to the next authenticator when an expansion was forced to fail,
4274 contradicting the general specification for all authenticators. Instead it
4275 was generating a temporary error. It now behaves as specified.
4277 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4278 (the order specifies the preference for clients). The order is now AES256,
4279 AES128, 3DES, ARCFOUR128.
4281 31. Small patch to Sieve code - explicitly set From: when generating an
4284 32. Exim crashed if a remote delivery caused a very long error message to be
4285 recorded - for instance if somebody sent an entire SpamAssassin report back
4286 as a large number of 550 error lines. This bug was coincidentally fixed by
4287 increasing the size of one of Exim's internal buffers (big_buffer) that
4288 happened as part of the Exiscan merge. However, to be on the safe side, I
4289 have made the code more robust (and fixed the comments that describe what
4292 33. Some experimental protocols are using DNS PTR records for new purposes. The
4293 keys for these records are domain names, not reversed IP addresses. The
4294 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4295 leaves it alone. Component reversal etc. now happens only for IP addresses.
4298 34. The host_aton() function is supposed to be passed a string that is known
4299 to be a valid IP address. However, in the case of IPv6 addresses, it was
4300 not checking this. This is a hostage to fortune. Exim now panics and dies
4301 if the condition is not met. A case was found where this could be provoked
4302 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4303 components; fortuitously, this particular loophole had already been fixed
4304 by change 4.50/55 or 4.44/33 above.
4306 If there are any other similar loopholes, the new check in host_aton()
4307 itself should stop them being exploited. The report I received stated that
4308 data on the command line could provoke the exploit when Exim was running as
4309 exim, but did not say which command line option was involved. All I could
4310 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4311 running as the user.
4314 35. There was a buffer overflow vulnerability in the SPA authentication code
4315 (which came originally from the Samba project). I have added a test to the
4316 spa_base64_to_bits() function which I hope fixes it.
4319 36. The daemon start-up calls getloadavg() while still root for those OS that
4320 need the first call to be done as root, but it missed one case: when
4321 deliver_queue_load_max is set with deliver_drop_privilege. This is
4322 necessary for the benefit of the queue runner, because there is no re-exec
4323 when deliver_drop_privilege is set.
4325 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4326 was in use, was not putting the data itself into the right store pool;
4327 consequently, it could be overwritten for a subsequent message in the same
4328 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4331 38. Sometimes the final signoff response after QUIT could fail to get
4332 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4333 < 0 before doing a fflush(). This bug looks as though it goes back to the
4334 introduction of TLS in release 3.20, but "sometimes" must have been rare
4335 because the tests only now provoked it.
4341 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4342 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4343 Now it returns an integer. A number of calls were still expecting a T/F
4344 return. Fortuitously, in all cases, the tests worked in OK situations,
4345 which is the norm. However, things would have gone wrong on any write
4346 failures on the smtp file descriptor. This function is used when sending
4347 messages over SMTP and also when doing verify callouts.
4349 2. When Exim is called to do synchronous delivery of a locally submitted
4350 message (the -odf or -odi options), it no longer closes stderr before doing
4353 3. Implemented the mua_wrapper option.
4355 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4357 5. Implemented the functions header_remove(), header_testname(),
4358 header_add_at_position(), and receive_remove_recipient(), and exported them
4361 6. If an ACL "warn" statement specified the addition of headers, Exim already
4362 inserted X-ACL-Warn: at the start if there was no header name. However, it
4363 was not making this test for the second and subsequent header lines if
4364 there were newlines in the string. This meant that an invalid header could
4365 be inserted if Exim was badly configured.
4367 7. Allow an ACL "warn" statement to add header lines at the start or after all
4368 the Received: headers, as well as at the end.
4370 8. Added the rcpt_4xx retry error code.
4372 9. Added postmaster_mailfrom=xxx to callout verification option.
4374 10. Added mailfrom=xxxx to the callout verification option, for verify=
4377 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4378 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4380 12. Inserted some casts to stop certain compilers warning when using pointer
4381 differences as field lengths or precisions in printf-type calls (mostly
4382 affecting debugging statements).
4384 13. Added optional readline() support for -be (dynamically loaded).
4386 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4387 same clock tick as a message's arrival, so that its received time was the
4388 same as the "first fail" time on the retry record, and that message
4389 remained on the queue past the ultimate address timeout, every queue runner
4390 would try a delivery (because it was past the ultimate address timeout) but
4391 after another failure, the ultimate address timeout, which should have then
4392 bounced the address, did not kick in. This was a "< instead of <=" error;
4393 in most cases the first failure would have been in the next clock tick
4394 after the received time, and all would be well.
4396 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4397 being recognized when the domain list was tested by the match_domain
4398 condition in an expansion string.
4400 16. Added the ${str2b64: operator.
4402 17. Exim was always calling setrlimit() to set a large limit for the number of
4403 processes, without checking whether the existing limit was already
4404 adequate. (It did check for the limit on file descriptors.) Furthermore,
4405 errors from getrlimit() and setrlimit() were being ignored. Now they are
4406 logged to the main and panic logs, but Exim does carry on, to try to do its
4407 job under whatever limits there are.
4409 18. Imported PCRE 5.0.
4411 19. Trivial typo in log message " temporarily refused connection" (the leading
4414 20. If the log selector return_path_on_delivery was set and an address was
4415 redirected to /dev/null, the delivery process crashed because it assumed
4416 that a return path would always be set for a "successful" delivery. In this
4417 case, the whole delivery is bypassed as an optimization, and therefore no
4420 21. Internal re-arrangement: the function for sending a challenge and reading
4421 a response while authentication was assuming a zero-terminated challenge
4422 string. It's now changed to take a pointer and a length, to allow for
4423 binary data in such strings.
4425 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4427 23. Exim was not respecting finduser_retries when seeking the login of the
4428 uid under which it was called; it was always trying 10 times. (The default
4429 setting of finduser_retries is zero.) Also, it was sleeping after the final
4430 failure, which is pointless.
4432 24. Implemented tls_on_connect_ports.
4434 25. Implemented acl_smtp_predata.
4436 26. If the domain in control=submission is set empty, Exim assumes that the
4437 authenticated id is a complete email address when it generates From: or
4438 Sender: header lines.
4440 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4441 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4442 chown and chgrp in /bin and hostname in /usr/bin.
4444 28. Exim was keeping the "process log" file open after each use, just as it
4445 does for the main log. This opens the possibility of it remaining open for
4446 long periods when the USR1 signal hits a daemon. Occasional processlog
4447 errors were reported, that could have been caused by this. Anyway, it seems
4448 much more sensible not to leave this file open at all, so that is what now
4451 29. The long-running daemon process does not normally write to the log once it
4452 has entered its main loop, and it closes the log before doing so. This is
4453 so that log files can straightforwardly be renamed and moved. However,
4454 there are a couple of unusual error situations where the daemon does write
4455 log entries, and I had neglected to close the log afterwards.
4457 30. The text of an SMTP error response that was received during a remote
4458 delivery was being truncated at 512 bytes. This is too short for some of
4459 the long messages that one sometimes sees. I've increased the limit to
4462 31. It is now possible to make retry rules that apply only when a message has a
4463 specific sender, in particular, an empty sender.
4465 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4466 it possible to be selective about when SMTP synchronization is enforced.
4468 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4470 32. Implemented hosts_connection_nolog.
4472 33. Added an ACL for QUIT.
4474 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4477 35. Added mailbox_size and mailbox_filecount to appendfile.
4479 36. Added control = no_multiline_responses to ACLs.
4481 37. There was a bug in the logic of the code that waits for the clock to tick
4482 in the case where the clock went backwards by a substantial amount such
4483 that the microsecond fraction of "now" was more than the microsecond
4484 fraction of "then" (but the whole seconds number was less).
4486 38. Added support for the libradius Radius client library this is found on
4487 FreeBSD (previously only the radiusclient library was supported).
4493 1. When certain lookups returned multiple values in the form name=value, the
4494 quoting of the values was not always being done properly. Specifically:
4495 (a) If the value started with a double quote, but contained no whitespace,
4497 (b) If the value contained whitespace other than a space character (i.e.
4498 tabs or newlines or carriage returns) it was not quoted.
4499 This fix has been applied to the mysql and pgsql lookups by writing a
4500 separate quoting function and calling it from the lookup code. The fix
4501 should probably also be applied to nisplus, ibase and oracle lookups, but
4502 since I cannot test any of those, I have not disturbed their existing code.
4504 2. A hit in the callout cache for a specific address caused a log line with no
4505 reason for rejecting RCPT. Now it says "Previous (cached) callout
4506 verification failure".
4508 3. There was an off-by-one bug in the queryprogram router. An over-long
4509 return line was truncated at 256 instead of 255 characters, thereby
4510 overflowing its buffer with the terminating zero. As well as fixing this, I
4511 have increased the buffer size to 1024 (and made a note to document this).
4513 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4514 when Exim is waiting for an SMTP response from a remote server, Exim
4515 restarts its select() call on the socket, thereby resetting its timeout.
4516 This is not a problem when such interrupts are rare. Somebody set up a cron
4517 job to run exiwhat every 2 minutes, which is less than the normal select()
4518 timeout (5 or 10 minutes). This meant that the select() timeout never
4519 kicked in because it was always reset. I have fixed this by comparing the
4520 time when an interrupt arrives with the time at the start of the first call
4521 to select(). If more time than the timeout has elapsed, the interrupt is
4522 treated as a timeout.
4524 5. Some internal re-factoring in preparation for the addition of Sieve
4525 extensions (by MH). In particular, the "personal" test is moved to a
4526 separate function, and given an option for scanning Cc: and Bcc: (which is
4527 not set for Exim filters).
4529 6. When Exim created an email address using the login of the caller as the
4530 local part (e.g. when creating a From: or Sender: header line), it was not
4531 quoting the local part when it contained special characters such as @.
4533 7. Installed new OpenBSD configuration files.
4535 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4536 try to make them clearer.
4538 9. Callout options, other than the timeout value, were being ignored when
4539 verifying sender addresses in header lines. For example, when using
4541 verify = header_sender/callout=no_cache
4543 the cache was (incorrectly) being used.
4545 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4546 only the Cygwin environment.
4548 11. When return_path_on_delivery was set as a log selector, if different remote
4549 addresses in the same message used different return paths and parallel
4550 remote delivery occurred, the wrong values would sometimes be logged.
4551 (Whenever a remote delivery process finished, the return path value from
4552 the most recently started remote delivery process was logged.)
4554 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4555 lines when AUTH and/or TLS are in use. This is the "received protocol"
4556 field. Exim used to use "asmtp" for authenticated SMTP, without any
4557 indication (in the protocol name) for TLS use. Now it follows the RFC and
4558 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4559 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4560 names appear in log lines as well as in Received: header lines.
4562 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4563 extensions, and comparison tests, and to fix some bugs.
4565 14. Changes to the "personal" filter test:
4567 (1) The test was buggy in that it was just doing the equivalent of
4568 "contains" tests on header lines. For example, if a user's address was
4569 anne@some.where, the "personal" test would incorrectly be true for
4571 To: susanne@some.where
4573 This test is now done by extracting each address from the header in turn,
4574 and checking the entire address. Other tests that are part of "personal"
4575 are now done using regular expressions (for example, to check local parts
4576 of addresses in From: header lines).
4578 (2) The list of non-personal local parts in From: addresses has been
4579 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4580 taken from the Sieve specification recommendations.
4582 (3) If the message contains any header line starting with "List-" it is
4583 treated as non-personal.
4585 (4) The test for "circular" in the Subject: header line has been removed
4586 because it now seems ill-conceived.
4588 15. Minor typos in src/EDITME comments corrected.
4590 16. Installed latest exipick from John Jetmore.
4592 17. If headers_add on a router specified a text string that was too long for
4593 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4594 of string_sprintf() is now avoided.
4596 18. $message_body_size was not set (it was always zero) when running the DATA
4597 ACL and the local_scan() function.
4599 19. For the "mail" command in an Exim filter, no default was being set for
4600 the once_repeat time, causing a random time value to be used if "once" was
4601 specified. (If the value happened to be <= 0, no repeat happened.) The
4602 default is now 0s, meaning "never repeat". The "vacation" command was OK
4603 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4604 (I found it when inspecting the code).
4606 20. There is now an overall timeout for performing a callout verification. It
4607 defaults to 4 times the callout timeout, which applies to individual SMTP
4608 commands during the callout. The overall timeout applies when there is more
4609 than one host that can be tried. The timeout is checked before trying the
4610 next host. This prevents very long delays if there are a large number of
4611 hosts and all are timing out (e.g. when the network connections are timing
4612 out). The value of the overall timeout can be changed by specifying an
4613 additional sub-option for "callout", called "maxwait". For example:
4615 verify = sender/callout=5s,maxwait=20s
4617 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4618 to the end before writing, but this should make it even safer).
4620 22. Exim was forgetting that it had advertised PIPELINING for the second and
4621 subsequent messages on an SMTP connection. It was also not resetting its
4622 memory on STARTTLS and an internal HELO.
4624 23. When Exim logs an SMTP synchronization error within a session, it now
4625 records whether PIPELINING has been advertised or not.
4627 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4628 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4629 rather than long int.
4631 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4633 26. Added the never_mail option to autoreply.
4639 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4640 crash if the getsockname() call failed; this can happen if a connection is
4641 closed very soon after it is established. The problem was simply in the
4642 order in which certain operations were done, causing Exim to try to write
4643 to the SMTP stream before it had set up the file descriptor. The bug has
4644 been fixed by making things happen in the correct order.
4650 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4651 before the connection was closed, thus losing the rejection response.
4653 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4654 some early Solaris releases, but causes trouble in current releases where
4655 socklen_t is defined.
4657 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4660 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4663 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4664 that releasing the top of it at the end releases what was used for sub-
4665 expansions (unless the block got too big). However, discard this block if
4666 the first thing is a variable or header, so that we can use its block when
4667 it is dynamic (useful for very large $message_headers, for example).
4669 6. Lookups now cache *every* query, not just the most recent. A new, separate
4670 store pool is used for this. It can be recovered when all lookup caches are
4671 flushed. Lookups now release memory at the end of their result strings.
4672 This has involved some general refactoring of the lookup sources.
4674 7. Some code has been added to the store_xxx() functions to reduce the amount
4675 of flapping under certain conditions.
4677 8. log_incoming_interface used to affect only the <= reception log lines. Now
4678 it causes the local interface and port to be added to several more SMTP log
4679 lines, for example "SMTP connection from", and rejection lines.
4681 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4683 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4685 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4686 could be overwritten at the end of the current message (or the start of a
4687 new message if it was set in a HELO ACL). The value is now preserved for
4688 the duration of the SMTP connection.
4690 12. If a transport had a headers_rewrite setting, and a matching header line
4691 contained an unqualified address, that address was qualified, even if it
4692 did not match any rewriting rules. The underlying bug was that the values
4693 of the flags that permit the existence of unqualified sender and recipient
4694 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4695 non-local messages, and by -bnq for local messages) were not being
4696 preserved with the message after it was received.
4698 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4699 "next input=" as part of the text comprising the host identity instead of
4700 the correct text. The code was using the same buffer for two different
4701 strings. However, depending on which order the printing function evaluated
4702 its arguments, the bug did not always show up. Under Linux, for example, my
4703 test suite worked just fine.
4705 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4706 doesn't work with some older versions of Perl. It has been changed to "my",
4707 which in any case is probably the better facility to use.
4709 15. A really picky compiler found some instances of statements for creating
4710 error messages that either had too many or two few arguments for the format
4713 16. The size of the buffer for calls to the DNS resolver has been increased
4714 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4715 for addresses that have a lot of PTR records. This alleviates a problem; it
4716 does not fully solve it.
4718 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4719 buffer now truncates the list and logs the incident, which is the same
4720 action as happens when Exim is looking up a host name and its aliases.
4721 Previously in this situation something unpredictable would happen;
4722 sometimes it was "internal error: store_reset failed".
4724 18. If a server dropped the connection unexpectedly when an Exim client was
4725 using GnuTLS and trying to read a response, the client delivery process
4726 crashed while trying to generate an error log message.
4728 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4729 string, for example:
4731 warn message = H1: something\nH2: something
4733 the text was added as a single header line from Exim's point of view
4734 though it ended up OK in the delivered message. However, searching for the
4735 second and subsequent header lines using $h_h2: did not work. This has been
4736 fixed. Similarly, if a system filter added multiple headers in this way,
4737 the routers could not see them.
4739 20. Expanded the error message when iplsearch is called with an invalid key to
4740 suggest using net-iplsearch in a host list.
4742 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4743 ACLs are no longer actually imposed (and a message to that effect is
4746 22. If a "gecos" field in a passwd entry contained escaped characters, in
4747 particular, if it contained a \" sequence, Exim got it wrong when building
4748 a From: or a Sender: header from that name. A second bug also caused
4749 incorrect handling when an unquoted " was present following a character
4750 that needed quoting.
4752 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4753 was not being matched caselessly.
4755 24. Arranged for all hyphens in the exim.8 source to be escaped with
4758 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4759 properly. Recipient callout cache records were still being keyed to include
4760 the sender, even when use_sender was set false. This led to far more
4761 callouts that were necessary. The sender is no longer included in the key
4762 when use_sender is false.
4764 26. Added "control = submission" modifier to ACLs.
4766 27. Added the ${base62d: operator to decode base 62 numbers.
4768 28. dnsdb lookups can now access SRV records.
4770 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4771 the configuration file.
4773 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4774 (-v) mode. This makes the output for a verbose queue run more intelligible.
4776 31. Added a use_postmaster feature to recipient callouts.
4778 32. Added the $body_zerocount variable, containing the number of binary zero
4779 bytes in the message body.
4781 33. The time of last modification of the "new" subdirectory is now used as the
4782 "mailbox time last read" when there is a quota error for a maildir
4785 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4787 35. Added +ignore_unknown as a special item in host lists.
4789 36. Code for decoding IPv6 addresses in host lists is now included, even if
4790 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4791 address was recognized as an IP address, but was then not correctly decoded
4792 into binary, causing unexpected and incorrect effects when compared with
4799 1. Very minor rewording of debugging text in manualroute to say "list of
4800 hosts" instead of "hostlist".
4802 2. If verify=header_syntax was set, and a header line with an unqualified
4803 address (no domain) and a large number of spaces between the end of the
4804 name and the colon was received, the reception process suffered a buffer
4805 overflow, and (when I tested it) crashed. This was caused by some obsolete
4806 code that should have been removed. The fix is to remove it!
4808 3. When running in the test harness, delay a bit after writing a bounce
4809 message to get a bit more predictability in the log output.
4811 4. Added a call to search_tidyup() just before forking a reception process. In
4812 theory, someone could use a lookup in the expansion of smtp_accept_max_
4813 per_host which, without the tidyup, could leave open a database connection.
4815 5. Added the variables $recipient_data and $sender_data which get set from a
4816 lookup success in an ACL "recipients" or "senders" condition, or a router
4817 "senders" option, similar to $domain_data and $local_part_data.
4819 6. Moved the writing of debug_print from before to after the "senders" test
4822 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4823 problems for message scanning, either using a data ACL, or using
4824 local_scan() because the Received: header was not generated till after they
4825 were called (in order to set the time as the time of reception completion).
4826 I have revised the way this works. The header is now generated after the
4827 body is received, but before the ACL or local_scan() are called. After they
4828 are run, the timestamp in the header is updated.
4834 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4835 before starting a queue runner without re-exec. This happened only when
4836 deliver_drop_privilege was set or when the Exim user was set to root. The
4837 effect of the bug was that timeouts during subsequent deliveries caused
4838 crashes instead of being properly handled. The handler is now left at its
4839 default (and expected) setting.
4841 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4842 message, again when deliver_drop_privilege is set or Exim is run as root.
4843 The bug described in (1) was not present in this case, but the tidying up
4844 of the other signals was missing. I have made the two cases consistent.
4846 3. The ignore_target_hosts setting on a manualroute router was being ignored
4847 for hosts that were looked up using the /MX notation.
4849 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4852 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4853 operated on the sender address. After changing the $sender_address to <>
4854 for the sender address verify, Exim was re-instated it as the original
4855 (before rewriting) address, but remembering that it had rewritten it, so it
4856 wasn't rewriting it again. This bug also had the effect of breaking the
4857 sender address verification caching when the sender address was rewritten.
4859 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4860 This has been changed so that if the ip literal address matches
4861 ignore_target_hosts, the router declines.
4863 7. Added expansion conditions match_domain, match_address, and match_local_
4864 part (NOT match_host).
4866 8. The placeholder for the Received: header didn't have a length field set.
4868 9. Added code to Exim itself and to exim_lock to test for a specific race
4869 condition that could lead to file corruption when using MBX delivery. The
4870 issue is with the lockfile that is created in /tmp. If this file is removed
4871 after a process has opened it but before that process has acquired a lock,
4872 there is the potential for a second process to recreate the file and also
4873 acquire a lock. This could lead to two Exim processes writing to the file
4874 at the same time. The added code performs the same test as UW imapd; it
4875 checks after acquiring the lock that its file descriptor still refers to
4876 the same named file.
4878 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4879 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4882 11. Added the smtp_active_hostname option. If used, this will typically be made
4883 to depend on the incoming interface address. Because $interface_address is
4884 not set up until the daemon has forked a reception process, error responses
4885 that can happen earlier (such as "too many connections") no longer contain
4888 12. If an expansion in a condition on a "warn" statement fails because a lookup
4889 defers, the "warn" statement is abandoned, and the next ACL statement is
4890 processed. Previously this caused the whole ACL to be aborted.
4892 13. Added the iplsearch lookup type.
4894 14. Added ident_timeout as a log selector.
4896 15. Added tls_certificate_verified as a log selector.
4898 16. Added a global option tls_require_ciphers (compare the smtp transport
4899 option of the same name). This controls incoming TLS connections.
4901 17. I finally figured out how to make tls_require_ciphers do a similar thing
4902 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4903 before starting the TLS session.
4905 18. Tabs are now shown as \t in -bP output.
4907 19. If the log selector return_path_on_delivery was set, Exim crashed when
4908 bouncing a message because it had too many Received: header lines.
4910 20. If two routers both had headers_remove settings, and the first one included
4911 a superfluous trailing colon, the final name in the first list and the
4912 first name in the second list were incorrectly joined into one item (with a
4913 colon in the middle).
4919 1. Added -C and -D options to the exinext utility, mainly to make it easier
4920 to include in the automated testing, but these could be helpful when
4921 multiple configurations are in use.
4923 2. The exinext utility was not formatting the output nicely when there was
4924 an alternate port involved in the retry record key, nor when there was a
4925 message id as well (for retries that were specific to a specific message
4926 and a specific host). It was also confused by IPv6 addresses, because of
4927 the additional colons they contain. I have fixed the IPv4 problem, and
4928 patched it up to do a reasonable job for IPv6.
4930 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4931 delivery, the log line now contains "pipelined" if PIPELINING was used.
4933 4. An SMTP transport process used to panic and die if the bind() call to set
4934 an explicit outgoing interface failed. This has been changed; it is now
4935 treated in the same way as a connect() failure.
4937 5. A reference to $sender_host_name in the part of a conditional expansion
4938 that was being skipped was still causing a DNS lookup. This no longer
4941 6. The def: expansion condition was not recognizing references to header lines
4942 that used bh_ and bheader_.
4944 7. Added the _cache feature to named lists.
4946 8. The code for checking quota_filecount in the appendfile transport was
4947 allowing one more file than it should have been.
4949 9. For compatibility with Sendmail, the command line option
4957 and sets the incoming protocol and host name (for trusted callers). The
4958 host name and its colon can be omitted when only the protocol is to be set.
4959 Note the Exim already has two private options, -pd and -ps, that refer to
4960 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4961 "s", but I don't think that's a major issue.
4963 10. A number of refactoring changes to the code, none of which should affect
4966 (a) The number of logging options was getting close to filling up the
4967 32-bit word that was used as a bit map. I have split them into two classes:
4968 those that are passed in the argument to log_write(), and those that are
4969 only ever tested independently outside of that function. These are now in
4970 separate 32-bit words, so there is plenty of room for expansion again.
4971 There is no change in the user interface or the logging behaviour.
4973 (b) When building, for example, log lines, the code previously used a
4974 macro that called string_cat() twice, in order to add two strings. This is
4975 not really sufficiently general. Furthermore, there was one instance where
4976 it was actually wrong because one of the argument was used twice, and in
4977 one call a function was used. (As it happened, calling the function twice
4978 did not affect the overall behaviour.) The macro has been replaced by a
4979 function that can join an arbitrary number of extra strings onto a growing
4982 (c) The code for expansion conditions now uses a table and a binary chop
4983 instead of a serial search (which was left over from when there were very
4984 few conditions). Also, it now recognizes conditions like "pam" even when
4985 the relevant support is not compiled in: a suitably worded error message is
4986 given if an attempt is made to use such a condition.
4988 11. Added ${time_interval:xxxxx}.
4990 12. A bug was causing one of the ddress fields not to be passed back correctly
4991 from remote delivery subprocesses. The field in question was not being
4992 subsequently used, so this caused to problems in practice.
4994 13. Added new log selectors queue_time and deliver_time.
4996 14. Might have fixed a bug in maildirsizefile handling that threw up
4997 "unexpected character" debug warnings, and recalculated the data
4998 unnecessarily. In any case, I expanded the warning message to give more
5001 15. Added the message "Restricted characters in address" to the statements in
5002 the default ACL that block characters like @ and % in local parts.
5004 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5005 Three changes have been made:
5007 (a) There was a serious bug; a negative response to MAIL caused the whole
5008 recipient domain to be cached as invalid, thereby blocking all messages
5009 to all local parts at the same domain, from all senders. This bug has
5010 been fixed. The domain is no longer cached after a negative response to
5011 MAIL if the sender used is not empty.
5013 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5016 (c) A new callout option, "use_sender" has been added for people who want
5017 the modified behaviour.
5023 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5026 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5027 indeed breaks things for older releases.
5029 3. Added additional logging to the case where there is a problem reading data
5030 from a filter that is running in a subprocess using a pipe, in order to
5031 try to track down a specific problem.
5033 4. Testing facility fudge: when running in the test harness and attempting
5034 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5035 getting "No route to host". Convert this to a timeout.
5037 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5040 6. Some OS don't have socklen_t but use size_t instead. This affects the
5041 fifth argument of getsockopt() amongst other things. This is now
5042 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5043 can be set for individual OS. I have set it for SunOS5, OSF1, and
5044 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5045 some earlier ones do not.
5047 7. Change 4.30/15 was not doing the test caselessly.
5049 8. The standard form for an IPv6 address literal was being rejected by address
5050 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5051 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5052 this, as well as the form without the "IPv6" on the front (but only when
5053 address literals are enabled, of course).
5055 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5057 10. Exim crashed if a message with an empty sender address specified by -f
5058 encountered a router with an errors_to setting. This could be provoked only
5059 by a command such as
5063 where an empty string was supplied; "<>" did not hit this bug.
5065 11. Installed PCRE release 4.5.
5067 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5068 remained set. It is now erased.
5070 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5071 times from message ids (which are base 36 rather than the normal 62).
5073 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5074 were being counted as actual protocol errors, and logged if the log
5075 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5076 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5077 and DATA following a set of rejected RCPTs do not count as protocol errors.
5078 In other words, Exim assumes they were pipelined, though this may not
5079 actually be the case. Of course, in all cases the client gets an
5080 appropriate error code.
5082 15. If a lookup fails in an ACL condition, a message about the failure may
5083 be available; it is used if testing the ACL cannot continue, because most
5084 such messages specify what the cause of the deferral is. However, some
5085 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5086 that caused an old message to be retained and used if a later statement
5087 caused a defer, replacing the real cause of the deferral.
5089 16. If an IP address had so many PTR records that the DNS lookup buffer
5090 was not large enough to hold them, Exim could crash while trying to process
5091 the truncated data. It now detects and logs this case.
5093 17. Further to 4.21/58, another change has been made: if (and only if) the
5094 first line of a message (the first header line) ends with CRLF, a bare LF
5095 in a subsequent header line has a space inserted after it, so as not to
5096 terminate the header.
5098 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5099 unnecessarily, used atoi() instead of strtol(), and didn't check the
5100 termination when getting file sizes from file names by regex.
5102 19. Completely re-implemented the support for maildirsize files, in the light
5103 of a number of problems with the previous contributed implementation
5104 (4.30/29). In particular:
5106 . If the quota is zero, the maildirsize file is maintained, but no quota is
5109 . If the maildir directory does not exist, it is created before any attempt
5110 to write a maildirsize file.
5112 . The quota value in the file is just a cache; if the quota is changed in
5113 the transport, the new value overrides.
5115 . A regular expression is available for excluding directories from the
5118 20. The autoreply transport checks the characters in options that define the
5119 message's headers; it allows continued headers, but it was checking with
5120 isspace() after an embedded newline instead of explicitly looking for a
5123 21. If all the "regular" hosts to which an address was routed had passed their
5124 expiry times, and had not reached their retry times, the address was
5125 bounced, even if fallback hosts were defined. Now Exim should go on to try
5128 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5129 equivalent code in the SMTP transport. Some hosts send humungous responses
5130 to HELO/EHLO, more than 1024 it seems.
5132 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5133 gives compiler warnings in some environments. I've now done it "properly",
5136 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5137 (because of problems with the built-in one) was declared to return uschar *
5138 instead of char *, causing compiler failure.
5140 25. Fixed a file descriptor leak when processing alias/forward files.
5142 26. Fixed a minor format string issue in dbfn.c.
5144 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5146 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5147 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5150 29. When a sender address is verified, it is cached, to save repeating the test
5151 when there is more than one recipient in a message. However, when the
5152 verification involves a callout, it is possible for different callout
5153 options to be set for different recipients. It is too complicated to keep
5154 track of this in the cache, so now Exim always runs a verification when a
5155 callout is required, relying on the callout cache for the optimization.
5156 The overhead is duplication of the address routing, but this should not be
5159 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5160 to be verified with callout=postmaster, and the main callout worked but the
5161 postmaster check failed, the verification correctly failed. However, if a
5162 subsequent RCPT command asked for sender verification *without* the
5163 postmaster check, incorrect caching caused this verification also to fail,
5166 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5167 it was not caching the DNS options (qualify_single, search_parents) that
5168 were used when the lookup failed. A subsequent lookup with different
5169 options therefore always gave the same answer, though there were cases
5170 where it should not have. (Example: a "domains = !$mx_any" option on a
5171 dnslookup router: the "domains" option is always processed without any
5172 widening, but the router might have qualify_single set.) Now Exim uses the
5173 cached value only when the same options are set.
5175 32. Added John Jetmore's "exipick" utility to the distribution.
5177 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5178 than a timeout (e.g. a certificate is required, and is not provided), an
5179 Exim server now closes the connection immediately. Previously it waited for
5180 the client to close - but if the client is SSL, it seems that they each
5181 wait for each other, leading to a delay before one of them times out.
5183 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5184 maintained 0.8.x compatibility because I don't think many are using it, and
5185 it is clearly obsolete.
5187 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5190 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5191 client certificate was expired. A simple patch fixes this, though I don't
5192 understand the full logic of why the verify callback is called multiple
5195 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5196 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5197 which causes problems with some clients (such as the Certicom SSL Plus
5198 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5199 disables the coutermeasure allowing Eudora to connect."
5201 38. Exim was not checking that a write() to a log file succeeded. This could
5202 lead to Bad Things if a log got too big, in particular if it hit a file
5203 size limit. Exim now panics and dies if it cannot write to a log file, just
5204 as it does if it cannot open a log file.
5206 39. Modified OS/Makefile-Linux so that it now contains
5208 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5210 The two -D definitions ensure that Exim is compiled with large file
5211 support, which makes it possible to handle log files that are bigger than
5214 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5215 instance) a domain was checked against a named list that involved a lookup,
5216 causing $domain_data to be set, then another domain was checked against the
5217 same list, then the first domain was re-checked, the value of $domain_data
5218 after the final check could be wrong. In particular, if the second check
5219 failed, it could be set empty. This bug probably also applied to
5222 41. The strip_trailing_dot option was not being applied to the address given
5223 with the -f command-line option.
5225 42. The code for reading a message's header from the spool was incrementing
5226 $received_count, but never initializing it. This meant that the value was
5227 incorrect (doubled) while delivering a message in the same process in which
5228 it was received. In the most common configuration of Exim, this never
5229 happens - a fresh exec is done - but it can happen when
5230 deliver_drop_privilege is set.
5232 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5233 it now includes up to 150 characters of the unexpected data in the log
5236 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5237 and building data strings. The size of both of these buffers was 10 000
5238 bytes - far larger than anybody would *ever* want, thought I. Needless to
5239 say, somebody hit the limit. I have increased the maximum line length to
5240 20 000 and the maximum data length of concatenated lines to 100 000. I have
5241 also fixed two bugs, because there was no checking on these buffers. Tsk,
5242 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5243 buffer is too small.
5245 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5246 lsearch lookups. Now it does.
5248 46. When parsing a route_list item in a manualroute router, a fixed-length
5249 buffer was used for the list of hosts. I made this 1024 bytes long,
5250 thinking that nobody would ever have a list of hosts that long. Wrong.
5251 Somebody had a whole pile of complicated expansion conditions, and the
5252 string was silently truncated, leading to an expansion error. It turns out
5253 that it is easier to change to an unlimited length (owing to other changes
5254 that have happened since this code was originally written) than to build
5255 structure for giving a limitation error. The length of the item that
5256 expands into the list of hosts is now unlimited.
5258 47. The lsearch lookup could not handle data where the length of text line was
5259 more than 4095 characters. Such lines were truncated, leading to shortened
5260 data being returned. It should now handle lines of any length.
5262 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5263 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5266 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5267 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5268 no longer happens in comment lines. A list of replacements is now placed
5269 at the head of all of the source files, except those whose only change is
5270 to replace PERL_COMMAND in the very first #! line.
5272 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5273 messages on the queue, with a bottom-up merge sort, using code contributed
5274 by Michael Haardt. This should make operations like -bp somewhat faster on
5275 large queues. It won't affect queue runners, except when queue_run_in_order
5278 51. Installed eximstats 1.31 in the distribution.
5280 52. Added support for SRV lookups to the dnslookup router.
5282 53. If an ACL referred to $message_body or $message_body_end, the value was not
5283 reset for any messages that followed in the same SMTP session.
5285 54. The store-handling optimization for building very long strings was not
5286 differentiating between the different store pools. I don't think this
5287 actually made any difference in practice, but I've tidied it.
5289 55. While running the routers to verify a sender address, $sender_address
5290 was still set to the sender address. This is wrong, because when routing to
5291 send a bounce to the sender, it would be empty. Therefore, I have changed
5292 it so that, while verifying a sender address, $sender_address is set to <>.
5293 (There is no change to what happens when verifying a recipient address.)
5295 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5296 target A or AAAA records (if not already returned) without resetting the
5297 qualify_single or search_parents options of the DNS resolver. These are
5298 inappropriate in this case because the targets of MX and SRV records must
5299 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5300 target that, when qualified, matched something in the local domain. These
5301 two options are now turned off when doing these lookups.
5303 57. It seems that at least some releases of Reiserfs (which does not have the
5304 concept of a fixed number of inodes) returns zero and not -1 for the
5305 number of available inodes. This interacted badly with check_spool_inodes,
5306 which assumed that -1 was the "no such thing" setting. What I have done is
5307 to check that the total number of inodes is greater than zero before doing
5308 the test of how many are available.
5310 58. When a "warn" ACL statement has a log_message modifier, the message is
5311 remembered, and not repeated. This is to avoid a lot of repetition when a
5312 message has many recipients that cause the same warning to be written.
5313 Howewer, Exim was preserving the list of already written lines for an
5314 entire SMTP session, which doesn't seem right. The memory is now reset if a
5315 new message is started.
5317 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5318 debugging output unless log_rewrite was also set.
5320 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5321 of (int)(handle) when we know that handle contains (void *)(-1).
5323 61. The Exim daemon panic-logs an error return when it closes the incoming
5324 connection. However "connection reset by peer" seems to be common, and
5325 isn't really an error worthy of noting specially, so that particular error
5328 62. When Exim is trying to find all the local interfaces, it used to panic and
5329 die if the ioctl to get the interface flags failed. However, it seems that
5330 on at least one OS (Solaris 9) it is possible to have an interface that is
5331 included in the list of interfaces, but for which you get a failure error
5332 for this call. This happens when the interface is not "plumbed" into a
5333 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5334 failure of the "get flags" call assumes that the interface is down.
5336 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5337 makes life easier for people who are doing arithmetic on fields extracted
5338 from dates, where you often get leading zeros that should not be
5339 interpreted as octal.
5341 64. Added qualify_domain to the redirect router, to override the global
5344 65. If a pathologically long header line contained very many addresses (the
5345 report of this problem mentioned 10 000) and each of them was rewritten,
5346 Exim could use up a very large amount of memory. (It kept on making new
5347 copies of the header line as it rewrote, and never released the old ones.)
5348 At the expense of a bit more processing, the header rewriting function has
5349 been changed so that it no longer eats memory in this way.
5351 66. The generation of the Received: header has been moved from the time that a
5352 message starts to be received, to the time that it finishes. The timestamp
5353 in the Received: header should now be very close to that of the <= log
5354 line. There are two side-effects of this change:
5356 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5357 logged header lines no longer include the local Received: line, because
5358 it has not yet been created. The same applies to a copy of the message
5359 that is returned to a non-SMTP sender when a message is rejected.
5361 (b) When a filter file is tested using -bf, no additional Received: header
5362 is added to the test message. After some thought, I decided that this
5365 This change does not affect the value of $received_for. It is still set
5366 after address rewriting, but before local_scan() is called.
5368 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5370 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5371 gave an unhelpful panic error message, and a defer error. I have managed to
5372 change this behaviour so that it now rejects any supplied certificate,
5373 which seems right, as the list of acceptable certificates is empty.
5375 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5376 gave an unhelpful defer error. I have not managed to make this reject any
5377 supplied certificates, but the error message it gives is "no certificate
5378 supplied", which is not helpful.
5380 70. exigrep's output now also includes lines that are not associated with any
5381 message, but which match the given pattern. Implemented by a patch from
5382 Martin Sluka, which also tidied up the Perl a bit.
5384 71. Recipient callout verification, like sender verification, was using <> in
5385 the MAIL FROM command. This isn't really the right thing, since the actual
5386 sender may affect whether the remote host accepts the recipient or not. I
5387 have changed it to use the actual sender in the callout; this means that
5388 the cache record is now keyed on a recipient/sender pair, not just the
5389 recipient address. There doesn't seem to be a real danger of callout loops,
5390 since a callout by the remote host to check the sender would use <>.
5391 [SEE ABOVE: changed after hitting problems.]
5393 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5394 temporary errors. However, in the case of such a code being given after
5395 the end of a data transmission (i.e. after ".") Exim was failing to write
5396 a retry record for the message. (Yes, there was some broken host that was
5397 actually sending 8xx at this point.)
5399 73. An unknown lookup type in a host list could cause Exim to panic-die when
5400 the list was checked. (An example that provoked this was putting <; in the
5401 middle of a list instead of at the start.) If this happened during a DATA
5402 ACL check, a -D file could be left lying around. This kind of configuration
5403 error no longer causes Exim to die; instead it causes a defer errror. The
5404 incident is still logged to the main and panic logs.
5406 74. Buglet left over from Exim 3 conversion. The message "too many messages
5407 in one connection" was written to the rejectlog but not the mainlog, except
5408 when address rewriting (yes!) was being logged.
5410 75. Added write_rejectlog option.
5412 76. When a system filter was run not as root (that is, when system_filter_user
5413 was set), the values of the $n variables were not being returned to the
5414 main process; thus, they were not subsequently available in the $sn
5417 77. Added +return_path_on_delivery log selector.
5419 78. A connection timeout was being treated differently from recipients deferred
5420 when testing hosts_max_try with a message that was older than the host's
5421 retry timeout. (The host should not be counted, thus allowing all hosts to
5422 be tried at least once before bouncing.) This may have been the cause of an
5423 occasionally reported bug whereby a message would remain on the queue
5424 longer than the retry timeout, but would be bounced if a delivery was
5425 forced. I say "may" because I never totally pinned down the problem;
5426 setting up timeout/retry tests is difficult. See also the next item.
5428 79. The ultimate address timeout was not being applied to errors that involved
5429 a combination of host plus message (for example, a timeout on a MAIL
5430 command). When an address resolved to a number of possible hosts, and they
5431 were not all tried for each delivery (e.g. because of hosts_max_try), a
5432 message could remain on the queue longer than the retry timeout.
5434 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5437 81. Fixed an obscure SMTP outgoing bug which required at least the following
5438 conditions: (a) there was another message waiting for the same server;
5439 (b) the server returned 5xx to all RCPT commands in the first message so
5440 that the message was not completed; (c) the server dropped the connection
5441 or gave a negative response to the RSET that Exim sends to abort the
5442 transaction. The observed case was a dropped connection after DATA that had
5443 been sent in pipelining mode. That is, the server had advertised PIPELINING
5444 but was not implementing it correctly. The effect of the bug was incorrect
5445 behaviour, such as trying another host, and this could lead to a crash.
5451 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5452 and daemon.c were passed as pointers to ints; they should have been
5453 pointers to socklen_t variables (which are typically unsigned ints).
5455 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5458 3. Fixed a really odd bug that affected only the testing scheme; patching a
5459 certain fixed string in the binary changed the value of another string that
5460 happened to be identical to the end of the original first string.
5462 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5463 name", it returns that address as the IP address. On some operating
5464 systems (e.g. Solaris), it also passes back the IP address string as the
5465 "host name". However, on others (e.g. Linux), it passes back an empty
5466 string. Exim wasn't checking for this, and was changing the host name to an
5467 empty string, assuming it had been canonicized.
5469 5. Although rare, it is permitted to have more than one PTR record for a given
5470 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5471 all the names associated with an address, because they do in Solaris.
5472 However, it seems that they do not in Linux for data that comes from the
5473 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5474 I found this out when I moved to a new Linux workstation and tried to run
5475 the Exim test suite.
5477 To get round this problem I have changed the code so that it now does its
5478 own call to the DNS to look up PTR records when searching for a host name.
5479 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5480 addresses that are only in /etc/hosts are still found.
5482 This behaviour is, however, controlled by an option called host_lookup_
5483 order, which defaults to "bydns:byaddr". If people want to use the other
5484 order, or indeed, just use one or the other means of lookup, they can
5485 specify it in this variable.
5487 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5488 some operating systems, this comes back from gethostbyaddr() as an empty
5489 string, and this is what Exim used to test for. However, it seems that in
5490 other systems, "." is yielded. Exim now tests for this case too.
5492 7. The values of check_spool_space and check_log_space are now held internally
5493 as a number of kilobytes instead of an absolute number of bytes. If a
5494 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5495 kilobyte. This means that much larger values can be stored.
5497 8. Exim monitor: an attempt to get the action menu when not actually pointing
5498 at a message produces an empty menu entitled "No message selected". This
5499 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5500 no entries in it ("Shell widget menu has zero width and/or height"). So I
5501 have added a single, blank menu entry in this case.
5503 9. Added ${quote_local_part.
5505 10. MIME decoding is now applied to the contents of Subject: header lines when
5508 11. Now that a reference to $sender_host_address automatically causes a reverse
5509 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5510 host lookup before query-style lookups in lists that might use this
5511 variable. This has therefore been abolished, and the "net-" prefix is no
5512 longer necessary for query-style lookups.
5514 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5515 have been a typo for LFLAGS, so it has been changed.
5517 13. The install script calls Exim with "-C /dev/null" in order to find the
5518 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5519 to be output. Howeve, since Exim outputs its version number before the
5520 error, it didn't break the script. It just looked ugly. I fixed this by
5521 always allowing "-C /dev/null" if the caller is root.
5523 14. Ignore overlarge ACL variable number when reading spool file - insurance
5524 against a later release with more variables having written the file.
5526 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5527 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5528 this, as well as the form without the "IPv6" on the front.
5530 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5531 OS/Makefile-Darwin file.
5533 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5534 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5536 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5537 variable. However, some OS use ints for inodes. I've added cast to long int
5538 to get rid of the compiler warning.
5540 19. I had forgotten to lock out "/../" in configuration file names when
5541 ALT_CONFIG_PREFIX was set.
5543 20. Routers used for verification do not need to specify transports. However,
5544 if such a router generated a host list, and callout was configured, Exim
5545 crashed, because it could not find a port number from the (non-existent)
5546 transport. It now assumes port 25 in this circumstance.
5548 21. Added the -t option to exigrep.
5550 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5551 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5552 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5553 them is set, LOOKUP_LSEARCH is forced.
5555 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5557 24. Added sender and host information to the "rejected by local_scan()" log
5558 line; previously there was no indication of these.
5560 25. Added .include_if_exists.
5562 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5563 Linux. It turns out that not all file systems support this. Apparently some
5564 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5565 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5566 not supported on the file descriptor, is now ignored when Exim is trying to
5567 sync a directory. This applies only to Linux.
5569 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5571 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5572 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5575 29. Added support for maildirsize files from supplied patch (modified a bit).
5577 30. The use of :fail: followed by an empty string could lead Exim to respond to
5578 sender verification failures with (e.g.):
5580 550 Verification failed for <xxx>
5581 550 Sender verify failed
5583 where the first response line was missing the '-' that indicates it is not
5584 the final line of the response.
5586 31. The loop for finding the name of the user that called Exim had a hardwired
5587 limit of 10; it now uses the value of finduser_retries, which is used for
5588 all other user lookups.
5590 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5593 33. Exim was neglecting to zero errno before one call of strtol() when
5594 expanding a string and expecting an integer value. On some systems this
5595 resulted in spurious "integer overflow" errors. Also, it was casting the
5596 result into an int without checking.
5598 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5599 did not work. The code looks as if it has *never* worked, though it appears
5600 to have been documented since at least releast 1.62. I have made it work.
5602 35. The "timeout_DNS" error in retry rules, also documented since at least
5603 1.62, also never worked. As it isn't clear exactly what this means, and
5604 clearly it isn't a major issue, I have abolished the feature by treating it
5605 as "timeout", and writing a warning to the main and panic logs.
5607 36. The display of retry rules for -brt wasn't always showing the error code
5610 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5611 timeout_connect_A, timeout_connect_MX.
5613 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5614 to the empty sender.
5616 39. The daemon was not analysing the content of -oX till after it had closed
5617 stderr and disconnected from the controlling terminal. This meant that any
5618 syntax errors were only noted on the panic log, and the return code from
5619 the command was 0. By re-arranging the code a little, I've made the
5620 decoding happen first, so such errors now appear on stderr, and the return
5621 code is 1. However, the actual setting up of the sockets still happens in
5622 the disconnected process, so errors there are still only recorded on the
5625 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5626 connections (as happens on some IP stacks) was logged at start up time as
5627 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5628 it from "IPv6 and IPv4", which means that two separate sockets are being
5631 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5632 says whether AF_INET or AF_INET6 was passed as an argument.
5634 42. Exiwhat output was messed up when time zones were included in log
5637 43. Exiwhat now gives more information about the daemon's listening ports,
5638 and whether -tls-on-connect was used.
5640 44. The "port" option of the smtp transport is now expanded.
5642 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5643 silently ignored. Now an error message is written to the main and panic
5646 46. There's a new ACL modifier called "logwrite" which writes to a log file
5647 as soon as it is encountered.
5649 47. Added $local_user_uid and $local_user_gid at routing time.
5651 48. Exim crashed when trying to verify a sender address that was being
5654 49. Exim was recognizing only a space character after ".include". It now also
5655 recognizes a tab character.
5657 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5658 extracting the relevant information from the specification. The man page no
5659 longer contains scrambled data for the -d option, and I've added a section
5660 at the front about calling Exim under different names.
5662 51. Added "extra_headers" argument to the "mail" command in filter files.
5664 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5667 53. Installed eximstats 1.29.
5669 54. Added transport_filter_timeout as a generic transport option.
5671 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5672 Cc: header lines. This was required by RFC 822, but it not required by RFC
5675 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5676 incoming messages that did not have them. Now it does so only if the
5677 message originates locally, that is, if there is no associated remote host
5678 address. When Resent- header lines are present, this applies to the Resent-
5679 lines rather than the non-Resent- lines.
5681 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5682 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5684 58. Messages for configuration errors now include the name of the main
5685 configuration file - useful now that there may be more than one file in a
5686 list (.included file names were always shown).
5688 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5689 for those rare installations that do not start the daemon as root or run it
5690 setuid root. I've cut out the call to initgroups() if the daemon is not
5693 60. The Exim user and group can now be bound into the binary as text strings
5694 that are looked up at the start of Exim's processing.
5696 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5698 62. Added $mailstore_basename variable.
5700 63. Installed patch to sieve.c from Michael Haardt.
5702 64. When Exim failed to open the panic log after failing to open the main log,
5703 the original message it was trying to log was written to stderr and debug
5704 output, but if they were not available (the usual case in production), it
5705 was lost. Now it is written to syslog before the two lines that record the
5706 failures to open the logs.
5708 65. Users' Exim filters run in subprocesses under the user's uid. It is
5709 possible for a "deliver" command or an alias in a "personal" command to
5710 provoke an address rewrite. If logging of address rewriting is configured,
5711 this fails because the process is not running as root or exim. There may be
5712 a better way of dealing with this, but for the moment (because 4.30 needs
5713 to be released), I have disabled address rewrite logging when running a
5714 filter in a non-root, non-exim process.
5720 1. The buildconfig auxiliary program wasn't quoting the value set for
5721 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5722 not defined. This bug was masked in 4.22 by the effect that was fixed in
5725 2. Some messages that were rejected after a message id was allocated were
5726 shown as "incomplete" by exigrep. It no longer does this for messages that
5727 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5729 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5730 have allow_domain_literals set, the ID did not get logged in the <= line.
5731 Domain literals are now always recognized in Message-ID: header lines.
5733 4. The first argument for a ${extract expansion item is the key name or field
5734 number. Leading and trailing spaces in this item were not being ignored,
5735 causing some misleading effects.
5737 5. When deliver_drop_privilege was set, single queue runner processes started
5738 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5739 same command in the process it spins off) were not dropping privilege.
5741 6. When the daemon running as "exim" started a queue runner, it always
5742 re-executed Exim in the spun-off process. This is a waste of effort when
5743 deliver_drop_privilege is set. The new process now just calls the
5744 queue-runner function directly.
5750 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5753 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5754 ignored. Though the use of -oP was forcing the writing of a pid file, it
5755 was always written to the default place.
5757 3. If the message "no IP address found for host xxxx" is generated during
5758 incoming verification, it is now followed by identification of the incoming
5759 connection (so you can more easily find what provoked it).
5761 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5763 5. Added some features to "harden" Exim a bit more against certain attacks:
5765 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5766 be put in Local/Makefile. This is like the never_users runtime option,
5767 but it cannot be overridden. The default setting is "root".
5769 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5770 prefix string with which any file named in a -C command line option
5773 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5774 is retained for -C and -D only if the caller of Exim is root. Without
5775 it, the exim user may also use -C and -D and retain privilege.
5777 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5778 command line option is disabled.
5780 6. Macro names set by the -D option must start with an upper case letter, just
5781 like macro names defined in the configuration file.
5783 7. Added "dereference=" facility to LDAP.
5785 8. Two instances of the typo "uknown" in the source files are fixed.
5787 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5788 the Configure-Makefile script screwed up while processing it.
5790 10. Incorporated PCRE 4.4.
5792 11. The SMTP synchronization check was not operating right at the start of an
5793 SMTP session. For example, it could not catch a HELO sent before the client
5794 waited for the greeting. There is now a check for outstanding input at the
5795 point when the greeting is written. Because of the duplex, asynchronous
5796 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5797 way, but not yet received, when the check is performed.
5799 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5800 on TCP/IP sockets, because this apparently causes some broken clients to
5803 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5804 unchanged) from the Cygwin maintainer.
5806 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5807 was supported instead of testing for mbx. Effectively a typo.
5809 15. The spa authenticator server code was not checking that the input it
5810 received was valid base64.
5812 16. The debug output line for the "set" modifier in ACLs was not showing the
5813 name of the variable that was being set.
5815 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5817 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5818 lookup on its own. Something else was needed to trigger the lookup. For
5819 example, a match in host_lookup or the need for a host name in a host list.
5820 Now, if $sender_host_name is referenced and the host name has not yet been
5821 looked up, a lookup is performed. If the lookup fails, the variable remains
5822 empty, and $host_lookup_failed is set to "1".
5824 19. Added "eqi" as a case-independent comparison operator.
5826 20. The saslauthd authentication condition could segfault if neither service
5827 nor realm was specified.
5829 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5830 error message that was logged was misleading, and incoming SMTP
5831 connections were dropped. The message is now more accurate, and temporary
5832 errors are given to SMTP connections.
5834 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5835 (except RSET) with a 421 error, until QUIT is received. However, it was
5836 failing to send a response to QUIT.
5838 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5839 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5840 ACL is now run after the helo_try_verify_hosts code.
5842 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5843 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5844 case-independent, so other case variants are also recognized). Apparently
5845 some systems use these upper case variants.
5847 25. If more than two messages were waiting for the same host, and a transport
5848 filter was specified for the transport, Exim sent two messages over the
5849 same TCP/IP connection, and then failed with "socket operation on non-
5850 socket" when it tried to send the third.
5852 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5854 27. The extern definition of crypt16() in expand.c was not being excluded when
5855 the OS had its own crypt16() function.
5857 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5858 as a preferred synonym for return_size_limit, both as an option and as an
5861 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5863 30. Changed the default configuration ACL to relax the local part checking rule
5864 for addresses that are not in any local domains. For these addresses,
5865 slashes and pipe symbols are allowed within local parts, but the sequence
5866 /../ is explicitly forbidden.
5868 31. SPA server authentication was not clearing the challenge buffer before
5871 32. log_message in a "warn" ACL statement was writing to the reject log as
5872 well as to the main log, which contradicts the documentation and doesn't
5873 seem right (because no rejection is happening). So I have stopped it.
5875 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5876 However, I am unable to do any testing of this.
5878 34. Fixed an infelicity in the appendfile transport. When checking directories
5879 for a mailbox, to see if any needed to be created, it was accidentally
5880 using path names with one or more superfluous leading slashes; tracing
5881 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5883 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5884 contents are added to the log line that is written for every discarded
5885 recipient. (Previously a log_message setting was ignored.)
5887 36. The ${quote: operator now quotes the string if it is empty.
5889 37. The install script runs exim in order to find its version number. If for
5890 some reason other than non-existence or emptiness, which it checks, it
5891 could not run './exim', it was installing it with an empty version number,
5892 i.e. as "exim-". This error state is now caught, and the installation is
5895 38. An argument was missing from the function that creates an error message
5896 when Exim fails to connect to the socket for saslauthd authentication.
5897 This could cause Exim to crash, or give a corrupted message.
5899 39. Added isip, isip4, and isip6 to ${if conditions.
5901 40. The ACL variables $acl_xx are now saved with the message, and can be
5902 accessed later in routers, transports, and filters.
5904 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5905 strings in the file are not string-expanded.
5907 42. If a MAIL command specified a SIZE value that was too large to fit into an
5908 int variable, the check against message_size_limit failed. Such values are
5909 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5910 day this will have to be increased, but I don't think I want to be around
5911 when emails are that large.
5918 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5919 iconv() is not standard in FreeBSD.
5921 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5922 IPv6 enabled. The observed symptom was a segmentation fault on return from
5923 the function os_common_find_running_interfaces() in src/os.c.
5925 3. In the check_special_case() function in daemon.c I had used "errno" as an
5926 argument name, which causes warnings on some systems. This was basically a
5927 typo, since it was named "eno" in the comments!
5929 4. The code that waits for the clock to tick (at a resolution of some fraction
5930 of a second) so as to ensure message-id uniqueness was always waiting for
5931 at least one whole tick, when it could have waited for less. [This is
5932 almost certainly not relevant at current processor speeds, where it is
5933 unlikely to ever wait at all. But we try to future-proof.]
5935 5. The function that sleeps for a time interval that includes fractions of a
5936 second contained a race. It did not block SIGALRM between setting the
5937 timer, and suspending (a couple of lines later). If the interval was short
5938 and the sigsuspend() was delayed until after it had expired, the suspension
5939 never ended. On busy systems this could lead to processes getting stuck for
5942 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5943 process, before it forks any delivery processes. The open lookup caching
5944 mechanism meant that the open file or database connection was passed into
5945 the delivery process. The problem was that delivery processes always tidy
5946 up cached lookup data. This could cause a problem for the next delivery
5947 process started by the queue runner, because the external queue runner
5948 process does not know about the closure. So the next delivery process
5949 still has data in the lookup cache. In the case of a file lookup, there was
5950 no problem because closing a file descriptor in a subprocess doesn't affect
5951 the parent. However, if the lookup was caching a connection to a database,
5952 the connection was closed, and the second delivery process was likely to
5953 see errors such as "PGSQL: query failed: server closed the connection
5954 unexpectedly". The problem has been fixed by closing all cached lookups
5955 in a queue runner before running a delivery process.
5957 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5958 seem to have the "const" qualifier which it has on other OS. I've
5961 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5962 *of the same type* (client or server) with the same public name that an
5963 error should be diagnosed.
5965 9. When Exim looked up a host name for an IP address, but failed to find the
5966 original IP address when looking up the host name (a safety check), it
5967 output the message "<ip address> does not match any IP for NULL", which was
5968 confusing, to say the least. The bug was that the host name should have
5969 appeared instead of "NULL".
5971 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5972 user that is built into the binary, and the -C or -D options is used, root
5973 privilege is dropped before the configuration file is read. In addition,
5974 logging is switched to stderr instead of the normal log files. If the
5975 configuration then re-defines the Exim user, the unprivileged environment
5976 is probably not what is expected, so Exim logs a panic warning message (but
5979 However, if deliver_drop_privilege is set, the unprivileged state may well
5980 be exactly what is intended, so the warning has been cut out in that case,
5981 and Exim is allowed to try to write to its normal log files.
5987 1. smtp_return_error_details was not giving details for temporary sender
5988 or receiver verification errors.
5990 2. Diagnose a configuration error if two authenticators have the same public
5993 3. Exim used not to create the message log file for a message until the first
5994 delivery attempt. This could be confusing when incoming messages were held
5995 for policy or load reasons. The message log file is now created at the time
5996 the message is received, and an initial "Received" line is written to it.
5998 4. The automatically generated man page for command line options had a minor
5999 bug that caused no ill effects; however, a more serious problem was that
6000 the procedure for building the man page automatically didn't always
6001 operate. Consequently, release 4.20 contains an out-of-date version. This
6002 shouldn't happen again.
6004 5. When building Exim with embedded Perl support, the script that builds the
6005 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6006 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6008 6. The freeze_tell option was not being used for messages that were frozen on
6009 arrival, either by an ACL or by local_scan().
6011 7. Added the smtp_incomplete_transaction log selector.
6013 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6014 was accepting AUTH without a new EHLO.
6016 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6017 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6020 10. Logging of TCP/IP connections (when configured) now happens in the main
6021 daemon process instead of the child process, so that the TCP/IP connection
6022 count is more accurate (but it can never be perfect).
6024 11. The use of "drop" in a nested ACL was not being handled correctly in the
6025 outer ACL. Now, if condition failure induced by the nested "drop" causes
6026 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6027 or "require"), the connection is dropped.
6029 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6030 that yield "discard" can now be used with an "accept" or a "discard" verb,
6031 but an error is generated for any others (because I can't see a useful way
6032 to define what should happen).
6034 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6035 are now processed in the same way as lines in the Exim configuration file.
6036 In particular, continuation lines are supported.
6038 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6040 15. Added -ti meaning -t -i.
6042 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6043 domains, and warn by logging if others are found.
6045 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6046 returned from GIFCONF when Exim is trying to find the list of interfaces on
6047 a host. The code in os.c has been modified to copy each ifreq to an aligned
6048 structure in all cases.
6050 Also, in some cases, the returned ifreq's were being copied to a 'struct
6051 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6052 means the last couple of bytes of an IPv6 address could be chopped if the
6053 ifreq contained only a normal sockaddr (14 bytes storage).
6055 18. Named domain lists were not supported in the hosts_treat_as_local option.
6056 An entry such as +xxxx was not recognized, and was treated as a literal
6059 19. Ensure that header lines added by a DATA ACL are included in the reject log
6060 if the ACL subsequently rejects the message.
6062 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6063 MD5 (which is deprecated).
6065 21. When testing a filter file using -bf, Exim was writing a message when it
6066 took the sender from a "From " line in the message, but it was not doing so
6067 when it took $return_path from a Return-Path: header line. It now does.
6069 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6070 with a valid header line field name (a series of printing characters
6071 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6073 23. Changed "disc" in the source to "disk" to conform to the documentation and
6074 the book and for uniformity.
6076 24. Ignore Sendmail's -Ooption=value command line item.
6078 25. When execve() failed while trying to run a command in a pipe transport,
6079 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6080 could be confused with a return value of 69 from the command itself. This
6081 has been changed to 127, the value the shell returns if it is asked to run
6082 a non-existent command. The wording for the related log line suggests a
6083 non-existent command as the problem.
6085 26. If received_header_text expands to an empty string, do not add a Received:
6086 header line to the message. (Well, it adds a token one on the spool, but
6087 marks it "old" so that it doesn't get used or transmitted.)
6089 27. Installed eximstats 1.28 (addition of -nt option).
6091 28. There was no check for failure on the call to getsockname() in the daemon
6092 code. This can fail if there is a shortage of resources on the system, with
6093 ENOMEM, for example. A temporary error is now given on failure.
6095 29. Contrary to the C standard, it seems that in some environments, the
6096 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6097 program. Exim now does this explicitly; it affects the formatting of
6098 timestamps using strftime().
6100 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6101 complaints. I've now initialized all the variables, to avoid this.
6103 32. Header lines added by a system filter were not being "seen" during
6104 transport-time rewrites.
6106 33. The info_callback() function passed to OpenSSL is set up with type void
6107 (*)(SSL *, int, int), as described somewhere. However, when calling the
6108 function (actually a macro) that sets it up, the type void(*)() is
6109 expected. I've put in a cast to prevent warnings from picky compilers.
6111 34. If a DNS black list lookup found a CNAME record, but there were no A
6112 records associated with the domain it pointed at, Exim crashed.
6114 35. If a DNS black list lookup returned more than one A record, Exim ignored
6115 all but the first. It now scans all returned addresses if a particular IP
6116 value is being sought. In this situation, the contents of the
6117 $dnslist_value variable are a list of all the addresses, separated by a
6120 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6121 to accept a host name and all its aliases if the forward lookup for any of
6122 them yielded the IP address of the incoming connection. Now it accepts only
6123 those names whose forward lookup yields the correct IP address. Any other
6124 names are discarded. This closes a loophole whereby a rogue DNS
6125 administrator could create reverse DNS records to break through a
6126 wildcarded host restriction in an ACL.
6128 37. If a user filter or a system filter that ran in a subprocess used any of
6129 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6130 the wrong values were passed to the pipe command ($thisaddress had the
6131 value of $0, $0 had the value of $1, etc). This bug was introduced by
6132 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6134 38. Improved the line breaking for long SMTP error messages from ACLs.
6135 Previously, if there was no break point between 40 and 75 characters, Exim
6136 left the rest of the message alone. Two changes have been made: (a) I've
6137 reduced the minimum length to 35 characters; (b) if it can't find a break
6138 point between 35 and 75 characters, it looks ahead and uses the first one
6139 that it finds. This may give the occasional overlong line, but at least the
6140 remaining text gets split now.
6142 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6143 file descriptors might be low, and that setting 1000 would always raise it.
6144 It turns out that in some environments, the limit is already over 1000 and
6145 that lowering it causes trouble. So now Exim takes care not to decrease it.
6147 40. When delivering a message, the value of $return_path is set to $sender_
6148 address at the start of routing (routers may change the value). By an
6149 oversight, this default was not being set up when an address was tested by
6150 -bt or -bv, which affected the outcome if any router or filter referred to
6153 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6154 log only when "message" or "log_message" are set. However, if one of the
6155 conditions was an address verification, or a call to a nested ACL, the
6156 messages generated by the underlying test were being passed through. This
6157 no longer happens. The underlying message is available in $acl_verify_
6158 message for both "message" and "log_message" expansions, so it can be
6159 passed through if needed.
6161 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6162 new expansion $bh_ to give the encoded byte string without charset
6163 translation. Translation happens only if iconv() is available; HAVE_ICONV
6164 indicates this at build time. HEADERS_CHARSET gives the charset to
6165 translate to; headers_charset can change it in the configuration, and
6166 "headers charset" can change it in an individual filter file.
6168 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6169 that creates RFC 2047 encoded "words" labels them as that charset instead
6170 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6171 expansion operator; (ii) when Exim creates a From: line for a local
6172 message; (iii) when a header line is rewritten to include a "phrase" part.
6174 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6175 buggy, causing it to skip the first lines of messages whose message ID
6176 ended in 'D'. This would not have bitten before Exim release 4.14, because
6177 message IDs were unlikely to end in 'D' before then. The effect was to have
6178 incorrect size information for certain domains.
6180 45. #include "config.h" was missing at the start of the crypt16.c module. This
6181 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6184 46. If there was a timeout during a "random" callout check, Exim treated it as
6185 a failure of the random address, and carried on sending RSET and the real
6186 address. If the delay was just some slowness somewhere, the response to the
6187 original RCPT would be taken as a response to RSET and so on, causing
6188 mayhem of various kinds.
6190 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6191 when I implemented it. It didn't allow for the fact that some option values
6192 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6193 the right test for positive values.
6195 48. Domain names in DNS records are case-independent. Exim always looks them up
6196 in lower case. Some resolvers return domain names in exactly the case they
6197 appear in the zone file, that is, they may contain uppercase letters. Not
6198 all resolvers do this - some return always lower case. Exim was treating a
6199 change of case by a resolver as a change of domain, similar to a widening
6200 of a domain abbreviation. This triggered its re-routing code and so it was
6201 trying to route what was effectively the same domain again. This normally
6202 caused routing to fail (because the router wouldn't handle the domain
6203 twice). Now Exim checks for this case specially, and just changes the
6204 casing of the domain that it ultimately uses when it transmits the message
6207 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6210 50. If a filter generated a file delivery with a non-absolute name (possible if
6211 no home directory exists for the router), the forbid_file option was not
6214 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6215 the existing equality matching.
6217 52. Exim was using ints instead of ino_t variables in some places where it was
6218 dealing with inode numbers.
6220 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6221 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6222 called TMPDIR, and if it finds it is different, it changes its value.
6224 54. The smtp_printf() function is now made available to local_scan() so
6225 additional output lines can be written before returning. There is also an
6226 smtp_fflush() function to enable the detection of a dropped connection.
6227 The variables smtp_input and smtp_batched_input are exported to
6230 55. Changed the default runtime configuration: the message "Unknown user"
6231 has been removed from the ACL, and instead placed on the localuser router,
6232 using the cannot_route_message feature. This means that any verification
6233 failures that generate their own messages won't get overridden. Similarly,
6234 the "Unrouteable address" message that was in the ACL for unverifiable
6235 relay addresses has also been removed.
6237 56. Added hosts_avoid_esmtp to the smtp transport.
6239 57. The exicyclog script was not checking for the esoteric option
6240 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6241 will work only if exicyclog is run under the appropriate euid.
6243 58. Following a discussion on the list, the rules by which Exim recognises line
6244 endings on incoming messages have been changed. The -dropcr and drop_cr
6245 options are now no-ops, retained only for backwards compatibility. The
6246 following line terminators are recognized: LF CRLF CR. However, special
6247 processing applies to CR:
6249 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6250 nor a local message in the state where . is a terminator.
6252 (ii) If a bare CR is encountered in a header line, an extra space is added
6253 after the line terminator so as not to end the header. The reasoning
6254 behind this is that bare CRs in header lines are most likely either
6255 to be mistakes, or people trying to play silly games.
6257 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6258 was being incorrectly given as 18 bytes larger than it should have been.
6259 This is a VOB (very old bug).
6261 60. This may never have affected anything current, but just in case it has:
6262 When the local host is found other than at the start of a list of hosts,
6263 the local host, those with the same MX, and any that follow, are discarded.
6264 When the list in question was part of a longer list of hosts, the following
6265 hosts (not currently being processed) were also being discarded. This no
6266 longer happens. I'm not sure if this situation could ever has previously
6269 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6272 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6273 header. This is something that is recommended in a new Internet Draft, and
6274 is something that is documented as being done by Sendmail. There are two
6275 possible values. For messages generated by the autoreply transport, Exim
6278 Auto-Submitted: auto-replied
6280 whereas for all other generated messages (e.g. bounces) it adds
6282 Auto-Submitted: auto-generated
6284 63. The "personal" condition in filters now includes a test for the
6285 Auto-Submitted: header. If it contains the string "auto-" the message it
6286 not considered personal.
6288 64. Added rcpt_include_affixes as a generic transport option.
6290 65. Added queue_only_override (default true).
6292 66. Added the syslog_duplication option.
6294 67. If what should have been the first header line of a message consisted of
6295 a space followed by a colon, Exim was mis-interpreting it as a header line.
6296 It isn't of course - it is syntactically invalid and should therefore be
6297 treated as the start of the message body. The misbehaviour could have
6298 caused a number of strange effects, including loss of data in subsequent
6299 header lines, and spool format errors.
6301 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6302 client host had authenticated. This control can now be exercised by an ACL
6303 for more flexibility.
6305 69. By default, callouts do not happen when testing with -bh. There is now a
6306 variant, -bhc, which does actually run the callout code, including
6307 consulting and updating the callout cache.
6309 70. Added support for saslauthd authentication, courtesy of Alexander
6312 71. If statvfs() failed on the spool or log directories while checking their
6313 size for availability, Exim confusingly gave the error "space shortage".
6314 Furthermore, in debugging mode it crashed with a floating point exception.
6315 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6316 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6317 serious problem, Exim now writes to the main and panic logs when this
6318 happens, with details of the failure. It then refuses to accept the
6319 incoming message, giving the message "spool directory problem" or "log
6320 directory problem" with a 421 code for SMTP messages.
6322 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6323 0, 1, and 2 exist, because some OS complain for execs without them (see
6324 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6325 descriptors. However, the code omitted to check that the open succeeded,
6326 causing mysterious errors if for some reason the permissions on /dev/null
6327 got screwed. Now Exim writes a message to the main and panic logs, and
6328 bombs out if it can't open /dev/null.
6330 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6331 interact so that it is all more flexible. It is supposed to remain
6332 backwards compatible. Also added extra_local_interfaces.
6334 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6335 to bomb out with an assertion failure - to the client this appears as a
6336 connection drop. This problem occurs in the part of the code that was taken
6337 from the Samba project. Fortunately, the assertion is in a very simple
6338 function, so I have fixed this by reproducing the function inline in the
6339 one place where it is called, and arranging for authentication to fail
6340 instead of killing the process with assert().
6342 75. The SPA client code was not working when the server requested OEM rather
6343 than Unicode encoding.
6345 76. Added code to make require_files with a specific uid setting more usable in
6346 the case where statting the file as root fails - usually a non-root-mounted
6347 NFS file system. When this happens and the failure is EACCES, Exim now
6348 forks a subprocess and does the per-uid checking as the relevant uid.
6350 77. Added process_log_path.
6352 78. If log_file_path was not explicitly set, a setting of check_log_space or
6353 check_log_inodes was ignored.
6355 79. If a space check for the spool or log partitions fails, the incident is now
6356 logged. Of course, in the latter case the data may get lost...
6358 80. Added the %p formatting code to string_format() so that it can be used to
6359 print addresses in debug_print(). Adjusted all the address printing in the
6360 debugging in store.c to use %p rather than %d.
6362 81. There was a concern that a line of code in smtp_in.c could overflow a
6363 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6364 initially expressed, the concern was not well-founded, because trailing
6365 spaces are removed early. However, if the trailing spaces were followed by
6366 a NULL, they did not get removed, so the overflow was possible. Two fixes
6369 (a) I re-wrote the offending code in a cleaner fashion.
6370 (b) If an incoming SMTP command contains a NULL character, it is rejected
6373 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6374 runs initgroups(), so that if the Exim user is in any additional groups,
6375 they will be used during message reception.
6381 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / jgh / exim.git / blob